Vous êtes sur la page 1sur 481

Corrosion and Electrochemistry

of Zinc

Corrosion and Electrochemistry

of Zinc
Xiaoge Gregory Zhang
Cominco Ltd.
Product Technology Centre
Mississauga, Ontario, Canada

Springer Science+ Business Media, LLC

Library of Congress Cataloging-in-Pub1ication

Data

Zhang, Xiaoge Gregory.


C o r r o s i o n a n d e l e c t r o c h e m i s t r y of z i n c / X i a o g e G r e g o r y
p.
cm.
Includes bibliographical references a n d index.
I S B N 978-1-4757-9879-1
1. Z i n c C o r r o s i o n ,
2. E l e c t r o c h e m i s t r y .
I. T i t l e .
TA480.Z6Z45 1996
620. 1'84223dc20

ISBN 978-1-4757-9879-1
DOI 10.1007/978-1-4757-9877-7

Zhang,

96-32551
CIP

ISBN 978-1-4757-9877-7 (eBook)

1996 Springer Science+Business Media New York


Originally published by Plenum Press, New York in 1996
Softcover reprint of the hardcovrer 1st edition 1996

All rights reserved


10 9 8 7 6 5 4 3 2 1
No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written
permission from the Publisher

To my mother Youliu Zhang


and father Hongtao Zhang

Foreword
Humankind's use of zinc stretches back to antiquity, and it was a component in some of
the earliest known alloy systems. Even though metallic zinc was not "discovered" in
Europe until 1746 (by Marggral), zinc ores were used for making brass in biblical times,
and an 87% zinc alloy was found in prehistoric ruins in Transylvania. Also, zinc (the
metal) was produced in quantity in India as far back as the thirteenth century, well before
it was recognized as being a separate element.
The uses of zinc are manifold, ranging from galvanizing to die castings to electronics.
It is a preferred anode material in high-energy-density batteries (e.g., Ni/Zn, Ag/Zn,
ZnJair), so that its electrochemistry, particularly in alkaline media, has been extensively
explored. In the passive state, zinc is photoelectrochemically active, with the passive film
displaying n-type characteristics. For the same reason that zinc is considered to be an
excellent battery anode, it has found extensive use as a sacrificial anode for the protection
of ships and pipelines from corrosion. Indeed, aside from zinc's well-known attributes as
an alloying element, its widespread use is principally due to its electrochemical properties,
which include a well-placed position in the galvanic series for protecting iron and steel
in natural aqueous environments and its reversible dissolution behavior in alkaline
solutions.
Dr. Zhang has undertaken the monumental task of describing the corrosion properties
and electrochemistry of zinc in a single book. The reason why this task is "monumental"
is that the literature on this important metal is highly fragmented, no doubt reflecting
zinc's diversity of use. Furthermore, the literature stretches from the very fundamental to
the very applied, with some of the reports on the properties of the metal and its compounds
being anecdotal in nature. The task of assembling all of the relevant information into a
single monograph, in a manner that is logical and easy to read. is the task that Dr. Zhang
undertook. He has succeeded admirably, and this book will surely become an authoritative
source of information on the electrochemistry of this technologically important metal.
Digby D. Macdonald
The Pennsylvania State University
University Park, Pennsylvania

vii

Preface
Zinc is one of the most widely used metals. Its most important commercial application is
corrosion protection of steel. In the past decades, a tremendous amount of research work
has been done on the various aspects of zinc corrosion and electrochemistry. This book
provides a systematic review of the enormous volume of technical results generated from
these investigations. It is hoped that it will not only be useful to those interested in specific
information on this subject but also will stimulate those currently working in the field to
carry out further research.
This book attempts to combine fundamental information on the electrochemistry of
zinc with practical corrosion data for zinc and its alloys and to connect the academic and
industrial realms of interest, which are often detached from each other. In general, books
on corrosion written from an academic perspective usually treat individual metals or
alloys either as examples or in a rather general fashion, while those issuing from the metals
industries cover little fundamental information. However, from the viewpoint of a metals
user or researcher, it is most beneficial that all the relevant corrosion and electrochemistry
information, theoretical and practical, on one metal be systematically organized in one
single source. Additionally, only a dozen or so metals, including zinc, are used in massive
quantities in today's society, and a compilation of all the corrosion, electrochemistry, and
related information in a single book for each of them would be very useful for more
effective application of these metals in the future.
This book focuses on corrosion and does not cover other applied aspects of zinc
electrochemistry. However, as it contains a large collection of electrochemical information on zinc, it can also serve as a source of reference for electrochemical processes such
as electroplating, electrowinning, and batteries. Much of the electrochemical information
presented in this book is related to the elemental reactions such as dissolution, hydrogen
evolution, oxygen reduction, and passivation, which are also important in electrochemical
processes other than corrosion.
Two general approaches have been taken in the selection and treatment of the
information presented in this book. The first is to emphasize the properties pertaining to
zinc as a material, rather than those pertaining to specific products (such as coatings,
wires, plates, cast alloys, etc.). The behavior of the various zinc products is taken into
account in the consideration of the effect of specific physical or chemical factors, such as
alloying elements, physical dimensions, temperature, solution composition, and pH. The
ix

PREFACE

second is to emphasize the specificity of corrosion data in relation to corrosion environments. The environmental conditions pertaining to each corrosion situation are specified,
together with detailed data. Generalizations are provided when a consensus exists in the
data. Also, because the environment is as important as the material in a corrosion process,
information is provided at the beginning of each chapter to describe the corrosion
environments and to define the various factors involved.
The book consists of 15 chapters. The first chapter of the book presents the basic
physical and chemical properties of zinc. The remainder of the first half of the book is
concerned with the electrochemistry of zinc. More specifically, Chapter 2 on Electrochemical Thermodynamics and Kinetics, Chapter 3 on Passivation and Surface Film
Formation, and Chapter 4 on Electrochemistry of Zinc Oxide deal with the fundamental
electrochemistry of zinc, and Chapter 5 on Corrosion Potential and Corrosion Current,
Chapter 6 on Corrosion Products, and Chapter 7 on Corrosion Forms. These last three
chapters, in different aspects, connect the fundamental electrochemistry with the practical
corrosion behavior of zinc. The remaining chapters deal with corrosion performance in
various environments.
ACKNOWLEDGMENTS
I am profoundly grateful to the management of Cominco Ltd., particularly to Dr.
E. M. Valeriote and Mr. S. R. Wilkinson, for their support of my undertaking of such a
time-consuming task. In addition, I would like to personally thank Dr. Valeriote, who, as
the manager of the Co minco Product Technology Centre, was not only instrumental in
initiating this project but was also always keen to assist by providing advice and resources.
A very special thanks is due to Professor D. D. Macdonald of The Pennsylvania State
University, who suggested that I write this book and kindly helped at various stages during
the process. Also, I am deeply indebted to Professor M. Pourbaix of I'Universite Libre
de Bruxelles, who has had a great influence on my career and whose work and spirit have
inspired me in my writing of this book.
It would not have been possible for the book to arrive at its present form without the
constructive suggestions and criticisms of many people. I sincerely thank the following
people, who have helped in reviewing the various chapters of the manuscript:
Dr. T. D. Burleigh, University of Pittsburgh, United States
Dr. T. G. Chang, Cominco Ltd, Canada
Dr. B. R. Conard, INCO Ltd., Canada
Dr. F. E. Goodwin, International Lead and Zinc Research Organization, United States
Dr. J. A. Gonzalez, Cominco Ltd., Canada
Dr. T. E. Graedel, AT&T Bell Laboratories, United States
Professor T. M. Harris, University of Tulsa, United States
Professor M. B. Ives, McMaster University, Canada
Professor D. W. Kirk, University of Toronto, Canada
Mr. G. P. Lewis, Lewis Consulting, Canada
Professor C. Leygraf, Royal Institute of Technology, Sweden
Dr. J. H. Lindsay, General Motors Research Laboratories, United States
Professor D. D. Macdonald, The Pennsylvania State University, United States

xi

PREFACE

Dr. 1. Odnevall, Royal Institute of Technology, Sweden


Professor P. Searson, Johns Hopkins University, United States
Dr. H. E. Townsend, Bethlehem Steel Co., United States
Dr. K. Tomantschger, Cominco Ltd., Canada
Dr. E. M. Valeriote, Cominco Ltd., Canada
Professor R. Wiart, Universite Pierre et Marie Curie, France
In particular, the critical reading of the manuscript and the very helpful suggestions made
by Mr. G. P. Lewis of Lewis Consulting and Dr. H. E. Townsend of Bethlehem Steel Co.
are greatly appreciated.
Many people at the Product Technology Centre of Cominco have helped at different
stages in the development of the manuscript. To them I am most grateful. In particular, I
would like to thank Mr. J. E. Valeriote, who helped in the preparation of the figures, Mr.
J. Hwang, who provided feedback by reading the first draft of the manuscript, Mrs. V.
Rodic and Mrs. P. L. Doyle, who helped in obtaining and organizing the references, Mrs.
H. Laur for handling correspondence and mailing, and Ms. M. F. Haughton, who assisted
in preparing the tables and making manuscript corrections. Also, I would like to acknowledge The Second Cup at 292 Dundas West in Toronto, where I read the literature and
revised the manuscript during the morning hours of numerous weekends.
The work of Amelia McNamara, Arun Das, Kenneth Howell, and Jacqueline Sedman
at Plenum Press is greatly appreciated.
Finally, I wish to express my deeply felt thanks to my wife, Li, for her understanding
and moral support during the long period required for the writing of this manuscript.
Xiaoge Gregory Zhang

Mississauga, Ontario, Canada

Contents
LIST OF SYMBOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

XIX

CHAPTER 1. Properties, Products, and Processes


1.1. Introduction . . . . . . .
1.2. Basic Properties . . . . . . .
1.2.1. Physical Properties ..
1.2.2. Mechanical Properties
1.2.3. Alloying Properties ..
1.3. Main Products and Applications
1.3.1. Zinc Coatings .
1.3.2. Cast Products . . . . .
1.3.3. Rolled Zinc . . . . . .
1.3.4. Zinc Dust and Powder
1.4. Coating Processes . . . . .
1.4.1. Hot-Dip Galvanizing
1.4.2. Electroplating.
1.3. Phosphating
1.4. Chromating . . . . .

3
3
3
7
7
7
7
7
13
15
16

CHAPTER 2. Electrochemical Thermodynamics and Kinetics


2.1. Introduction . . . . . . .
2.2. Thermodynamic Stability
2.3. Ionic Properties . . . . .
2.4. Double-Layer Properties
2.5. Kinetics of Elemental Reactions
2.5.1. Dissolution . . . . .
2.5.2. Deposition . . . . .
2.5.3. Hydrogen Evolution
2.5.4. Oxygen Reduction .

19
19
25
27

29
29
36
39

48
xiii

xiv

CONTENTS

2.6. Corrosion Processes . . . . . . . . . . . .


2.6.1. General Considerations . . . . . . .
2.6.2. Impedance of Corroding Electrodes

54
54
54

CHAPTER 3. Passivation and Surface Film Formation


3.1. Introduction . . . . . . . . . .
3.2. Characteristics and Conditions.
3.3. Alkaline Solutions .. .

3.3.1. i-VCurves .. .
3.3.2. Passivation Time
3.3.3. Characteristics ..
3.3.4. Mechanisms of Formation of Passive Films
3.4. Other Solutions. . . . . . . . . . . . . . . . . .
3.4.1. Slightly Alkaline and Carbonate Solutions
3.4.2. Phosphate Solutions. . .
3.4.3. Miscellaneous Solutions
3.5. Anodization . . . . . . .
3.6. Stability of Passivation . . . .
3.6.1. Type of Passivation ..
3.6.2. Passivation Breakdown

65
65

68
68
70
73
75
77
77
80

84

85
87
87

89

CHAPTER 4. Electrochemistry of Zinc Oxide


4.1. Introduction . . . . . . .
4.2. Basic Properties . . . . . .
4.2.1. Physical Properties .
4.2.2. Electronic Properties
4.3. Semiconductor Electrochemical Behavior.
4.3.1. Basic Theories .. .
4.3.2. Flatband Potential . . . . . . .
4.3.3. Band Structure . . . . . . . .
4.3.4. Electrode Kinetics in the Dark
4.3.5. Photoelectrochemical Kinetics
4.3.6. Electroluminescence
4.4. Thin ZnO Films . . . . . . . . . . .
4.5. Stability . . . . . . . . . . . . . . .
4.5.1. Conditions of Stability and Decomposition Reactions
4.5.2. Rate of Decomposition . . . . . . . . . . . . .

93
93
93
95
97
97
100
103
105
109
114
115
119
119
121

CHAPTER 5. Corrosion Potential and Corrosion Current


5.1. Introduction . . . . . . . . . . . . . . . . . . . . . .
5.2. Relation between Corrosion Potential and Corrosion Current
5.2.1. Polarization Resistance and Corrosion Current.
5.2.2. Conversion Factors . . . . . . . . . . . . . . . . . . .

125
125
127
129

CONTENTS

5.3. Corrosion Potential and Reaction Kinetics


5.4. Ecorr and icorr under Various Conditions
5.4.l. Effect of Zinc Ions . . . . . .
5.4.2. Effect of Anions and Cations.
5.4.3. Effect of pH . . . . . . . . . .
5.4.4. Effect of Temperature . . . .
5.4.5. Effect of Aeration and Convection.
5.4.6. Effect of Surface Condition
5.5. Zinc Alloys . . . . . . . . . . . . . . . .
5.6. Effect of Time . . . . . . . . . . . . . . .
5.7. Correlation between Corrosion Current and Weight Loss Rate

xv

130
133
133
135
137
140
141
143
144
149
153

CHAPTER 6. Corrosion Products


6.l. Introduction . . . . . . . . .
6.2. In Atmospheric Environments ..
6.2.1. Composition and Structure .
6.2.2. Quantity and Morphology
6.2.3. Formation Processes
6.3. In Waters . . . . . .
6.3.l. Fresh Waters
6.3.2. Seawater ..
6.4. In Solutions . . . .
6.4.l. Effect of pH .
6.4.2. Formation Processes
6.4.3. Zinc Alloys . . . . .
6.5. In Other Environments ..
6.6. Effect of Corrosion Products on Zinc Corrosion

157
158
158
163
165
168
168
170
171
171
173
176
176
178

CHAPTER 7. Corrosion Forms


7.1. Introduction . . . .
7.2. Galvanic Corrosion
7.2.1. Introduction.
7.2.2. Theoretical Aspects .
7.2.3. Practical Factors ..
7.2.4. Polarity Reversal ..
7.2.5. Galvanic Corrosion in Natural Environments
7.2.6. Galvanic Protection of Steel by Zinc.
7.3. Pitting Corrosion . . . . . .
7.3.l. Introduction. . . . . .
7.3.2. Occurrence of Pitting.
7.3.3. Pitting Potential
7.3.4. Morphology
7.3.5. Mechanisms ..

183
183
183
185
196

.203
.208
213
.217
217
217
221
.224

.225

xvi

CONTENTS

7 A. Intergranular Corrosion
704.1. Introduction . .
704.2. Occurrence. . .
704.3. Metallurgical Effects.
70404. Effect of Environmental Factors .
704.5. Effect on Mechanical Properties.
704.6. Mechanisms . . . . . . . . . . .
7.5. Wet Storage Stain . . . . . . . . . . . .
7.6. Hydrogen Embrittlement and Corrosion Cracking

227
227
227
229
232
234
235
236
238

CHAPTER 8. Atmospheric Corrosion


8.1. Introduction . . . . . .
8.2. Atmospheric Factors . .
8.2.1. Type of Wetting
8.2.2. Air Pollutants .
8.3. Corrosion in Outdoor Environments.
8.3.1. Typical Corrosion Rates .
8.3.2. Effect of Time of Wetness . .
8.3.3. Effect of Pollutants . . . . .
8.3.4. Effect of Elevation and Distance from Seawater
8.3.5. Effect of Initial Weather Conditions.
8.3.6. Effect of Climate . . . . . . . .
8.3.7. Effect of Sample Configuration
8.3.8. Effect of Sheltering
8.3.9. Galvanized Steel . . . . . .
8.3.10. Effect of Alloying . . . . .
8.3.11. Effect of Surface Treatment
8.3.12. Effect of Corrosion Products
8.3.13. Forms of Corrosion . . . .
8.3.14. Highway Environment . . . .
804. Corrosion in Indoor Environments .
8.5. Corrosion in Simulated Environments.
8.5.1. Humidity Chamber Exposure
8.5.2. Water and Salt Spray ..
8.5.3. Cyclic Test . . . . . . .
8.504. Thin-Layer Electrolytes
8.6. Corrosion Mechanisms . . . .

241
241
241
243
245
245
248
249
252
252
254
254
255
256
258
260
261
261
262
264
266
267
270
272

274
278

CHAPTER 9. Corrosion in Waters and Aqueous Solutions


9.1. Introduction . . . . . . .
9.2. Characteristics of Waters .
9.2.1. Fresh Waters . . .
9.2.2. Seawater . . . . . .
9.3. Corrosion in Pure Water .
9.4. Corrosion in Natural Waters

283
283
283
284
286
288

CONTENTS

9.4.1. Cold Fresh Water.


9.4.2. Hot Fresh Water .
9.4.3. Seawater . . . . .
9.5. Corrosion in Aqueous Solutions
9.5.1. Effect of Dissolved Species
9.5.2. Effect of pH . . . . . . . . .
9.5.3. Effect of Immersion Conditions
9.5.4. Effect of Surface Treatments ..
9.5.5. Effect of Metallurgical Factors.

xvii

.288
.289
291
.296
.296
.298
301
.302
.302

CHAPTER 10. Corrosion in Soil


10.1. Introduction . . . . .
10.2. Characteristics of Soil
10.3. Corrosion Rates . . .
10.3.1. Effect of Soil Factors
10.3.2. Galvanic Corrosion .
10.4. Electrochemical Measurements

.305
.305
.308
.308
312
312

CHAPTER 11. Under-Paint Corrosion


11.1. Introduction . . . . . . . . .
11.2. Basic Characteristics of Paint
11.2.1. Components in Paint .
11.2.2. Barrier Properties of Paint .
11.3. Corrosion Tests. . . . . . . . . . .
11.4. Corrosion Behavior . . . . . . . .
11.4.1. Characterization of Corrosion .
11.4.2. Effect of Coating Type. .
11.4.3. Effect of Test Conditions
11.4.4. Effect of Paint System .
11.4.5. Galvanic Action
11.5. Corrosion Mechanisms . . . .

315
315
316
316
317
319
.319
.321
.325
.328
.332
333

CHAPTER 12. Zinc-Rich Coatings


12.1. Introduction . . . . . .
12.2. Coating Characteristics
12.3. Protection Mechanism.
12.4. Performance . . . . . .
12.4.1. Effect of Zinc Content .
12.4.2. Effect of Zinc Particle Size
12.4.3. Effect of Binders . . . . . .
12.4.4. Effect of Coating Thickness .
12.4.5. Effect of Additives . . . . .
12.4.6. Effect of Surface Condition
12.4.7. Other Factors . . . . . . . .

.337
.337
.339
341
341
.343
.344

.345
.345
.347
.348

xviii

CONTENTS

CHAPTER 13. Corrosion in Concrete


13.1. Introduction . . . . . . . . . .
13.2. Concrete Environment . . . .
13.2.1. Formation of Concrete
13.2.2. Characteristics . . . .
l3.3. Corrosion of Steel Reinforcement in Concrete.
13.3.1. Effect of Corrosion .
l3.3.2. Protection Methods . . . . . . . . .
13.3.3. Galvanized Coatings . . . . . . . .
13.4. Corrosion of Galvanized Steel in Concrete
13.4.1. Testing Methods . . . . . . .
13.4.2. Field Test Results . . . . . . .
l3.4.3. Results from Simulated Tests

351
352
352
353
358
358
359
359
360
360
360
365

CHAPTER 14. Corrosion in Batteries


14.1. Introduction . . . . . . .
14.2. Zinc Cells and Batteries
14.2.1. Leclanche Cell .
14.2.2. Zinc Chloride Cell
14.2.3. Zinc Alkaline Cell
14.2.4. Zinc-Air Battery .
14.2.5. Zinc-Nickel Battery
14.3. Corrosion . . . . . . . . . .
14.3.1. Effect of Testing Time
14.3.2. Effect of Electrolyte .
14.3.3. Effect of Chemical Agents
14.3.4. Zinc Electrode . . . .
14.3.5. Operating Conditions .. .

373
373
373
375
375
376
377
377
379
379
382
385
390

CHAPTER 15. Corrosion in Other Environments


15.1. Introduction . . . . .
15.2. Organic Solvents ..
15.2.1. Classification
15.2.2. Corrosion ..
15.3. Gaseous Environments
15.4. Zinc Anodes . . . . . .
15.4.1. Sacrificial Anodes
15.4.2. Anodes for Impressed Current Cathodic Protection

393
393
393
395
399
403
403
407

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 409
INDEX

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463

List of Symbols
Symbol
A
aH+' aOH-

B
b

C
Cd
C,c
CH
Co
C,
Ccrit

D
d

E
Eo
EII2
E"
E"
Ec
Ecorr
EF
ED
Etb

E~
Egc

E"
E"
E,

e-, p+
e

F
hv
fa

Definition
Surface area
Activity of hydrogen and hydroxyl ions
Stern-Geary constant
Tafel slope
Capacitance
Capacitance of double layer
Capacitance of space charge layer
Capacitance of Helmholtz double layer
Bulk concentration
Surface concentration
Critical concentration for passivation
Diffusion coefficient
Distance between anode and cathode
Electrode potential
Standard potential
Half-wave potential
Potential of anode
Breakdown potential
Lower edge of conduction band
Potential of cathode
Corrosion potential
Fermi level
Decomposition potential
Flatband potential
Width of band gap
Potential of a galvanic couple
Passivation potential
Top edge of valence band
Solution potential
Electron and hole
Electron charge
Faraday constant
Photon
Anodic current
xix

Section

5.2
2.5
5.2
2.5
2.4
2.6
4.3
4.3
3.3
3.3
3.3
3.3
7.2
2.5
2.2
2.5
7.2
3.2
4.2
7.2
5.2
4.2
4.5
4.3
4.2
7.2
3.2
4.2
4.3
4.3
4.3
2.5
4.5
7.2

xx

LIST OF SYMBOLS

Symbol

Ie
Ig

)~
i,
io

--7
I

iOa' iOe

i a , ie
icorr

ig

i gt
im

i"
iphoto

Kf
L
L,

L"

n ox ,

nred

n" Ps
PH"PO,

Q; -

t+, L
VH

Vrd

V,

VSCE
V SHE

Definition
Cathodic current
Galvanic current
Current density
Oxidation and reduction current density of a redox reaction
Exchange c'urrent density
Exchange current densities for anodic and
cathodic reactions
Anodic and cathodic current density
Corrosion current density
Galvanic current density
Gravimetric corrosion rate
Limiting current density
Passivation current
Photocurrent density
Rate constant for forward reaction
Polarization parameter
Width of space charge layer
Hole diffusion length
Molar concentration per liter
Molar mass weight
Effective density of states in conduction band
Dopant concentration
Bulk electron and hole density
Charge per atom
Number of oxidizing and reducing species
Surface electron and hole density
Partial pressure of hydrogen and oxygen
Immobile charge of space charge layer
Resistance
Gas constant
Corrosion rate
Charge-transfer resistance
Electrolyte resistance
Capacitance of corrosion product film
Metallic resistance
Polarization resistance
Capacitance of charge transfer
Potential scanning rate
Passivation time
Temperature
Thickness loss
Time
Transport number
Potential drop across Helmholtz double layer
Rest dark potential
Potential drop across space charge layer
Potential versus saturated calomel electrode
Potential versus standard hydrogen electrode

Section

7.2
7.2
2.5
5.2
2.5
7.2
3.4

5.2
7.2
5.7

2.5
3.2

4.3

2.5
7.2

4.3
4.3

2.5
5.2

4.3
4.3
4.3

5.2

4.3
4.3

2.5

4.3

7.2
2.5
8.3

2.6
2.6
2.6
7.2
5.2
2.6
3.4
3.3

2.5
5.2
3.3
2.3

4.3
4.3
4.3

xxi

LIST OF SYMBOLS
S~mbol

V CSE
AVmAVe
AVd
W

X
Z

Zw

fJ
e

eo
'I
'la' 'Ie
'II'
()

A+"L
v
p
(J

w
AC
CCT
ohp
mpy
PZC
RH
r.d.s.
rpm
SCE
SHE
SST

Definition
Potential versus copper sulfate electrode
Potential drops across anode and cathode
Potential drop between anode and cathode
Weight loss
Width of steel cathode
Galvanic protection width
Electrochemical impedance
Warburgimpedance
Charge-transfer coefficient
Photo absorption coefficient
Tafel slope
Relative permittivity
Permittivity of vacuum
Overpotential
Anodic and cathodic overpotential
Passivation overpotential
Area factor
Rate constant
Equivalent conductance
Stoichiometric coefficient
Density
Specific resistivity
Cross section
Potential difference between electrode and solution
Electrode rotation speed
Frequency of AC signal
Alternating current
Cyclic corrosion test
Outer Helmholtz double layer
Mils per year
Potential of zero charge
Relative humidity
Rate-determining step
Number of rotations per minute
Potential of saturated calomel electrode
Standard potential of hydrogen electrode
Salt spray test

Section

7.2
7.2
5.2
7.2
7.2
2.6
2.6
2.5
4.3
5.2
4.3
4.3
2.5
7.2
3.2
7.2
4.3
2.3
2.2
5.2
7.2
4.3
2.5
2.5
2.6
2.6

11.3

4.3
2.4
8.3
2.5
2.5
2.5
2.5

11.3

1
Properties, Products, and
Processes
1.1. INTRODUCTION
Zinc is 23rd among the elements in relative abundance in the earth's crust, amounting
to 0.013%, compared with aluminum's 8.13% and iron's 5.0%. However, it ranks fourth
among the metals in worldwide production and consumption, behind only iron, aluminum, and copper [527].
The uses of zinc can be divided into six major categories: (a) coatings, (b) casting
alloys, Cc) alloying element in brass and other alloys, Cd) wrought zinc alloys, (e) zinc
oxide, and (f) zinc chemicals. The use of zinc coatings for corrosion protection of steel
structures is the most important application owing to the high corrosion resistance of zinc
in atmospheric and other environments. Nearly half of the zinc produced is used for this
purpose. The position of zinc in the electromotive series of metals means that zinc coating
provides not only a barrier layer to prevent contact between the coated steel and the
environment but also a sacrificial protection if discontinuities in the coating occur.
This chapter provides background information on the physical and metallurgical
properties of zinc, its main products and applications, and the processes that are important
to its corrosion behavior. The treatment of these subjects in this chapter is rather general:
more detailed and specific information will be presented in subsequent chapters.
1.2. BASIC PROPERTIES

1.2.1. Physical Properties


Zinc is a silvery blue-gray metal with a relatively low melting point (4 19'soC) and
boiling point (907C). The physical properties of zinc are shown in Table 1.1 [218,1295].
Zinc crystals have a close-packed hexagonal structure. The lattice constants a and c
are 0.2664 and 0.4947 nm, respectively. The axial ratio cia is 1.856, which is considerably
greater than the theoretical value of 1.633 for the system. Although each zinc atom has
12 near neighbors, 6 are at a distance of 0.2664 nm and the other 6 are at 0.2907 nm.
Thus, the bonds between the atoms in the hexagonal basal layers are appreciably stronger

CHAPTER I

TABLE 1.1. Physical Properties of Zinc"


Atomic number
Atomic weight
Density
Solid,20oe
Solid,419se
Liquid, 419.5e
Velocity of sound, 20C
Melting point
Boiling point, I atm
Ionization potentials
First
Second
Third
Heat of fusion, 419.5e
Heat of vaporization, 907C
Heat capacity
Solid,25e
Liquid
Resistivity
Solid,20oe
Liquid,419.7e
Thermal conductivity
Solid, 18C
Solid,419.SOe
Liquid, 419.5e
Linear coefficient of thermal expansion
Polycrystalline
a axis
c axis
Volume coefficient of thermal expansion
Surface tension, liquid, 419.5e
Viscosity, liquid, 419Se

30
65.38
7.14 g/cm 3
6.83 g/cm 3
6.62 g/cm 3
3.67 kmls

419.5e
907C
9.39 eV
17.87 eV
40.0eV
7.28 kllmol
114.7 kllmol
25.4 llmol
31.4 llmol
5.96IlQcm
37.4/lQcm
113 W/(mK)
96 W/(m-K)
61 W/(m-K)

39.7 x 10-6 K- 1
14.3 X 10-6 K- 1
60.8 X 10-6 K- 1
0.9 X 10-6 K- 1
782 mN/m
3.85 mNlm

"Refs. 218 and 50 I.

than those between the layers. This accounts for much of the deformation behavior and
anisotropy of the zinc crystal [218].
The grain structure in a polycrystalline zinc product has preferred orientations
depending on the casting and mechanical working conditions: for cast products, the
(0001) direction is perpendicular to the axis of the cast columnar crystals; for wire, the
(0001) plane is parallel to the axis of drawn wire; and, for sheet, the (0001) plane is parallel
to the rolling plane and the (1120) direction is parallel to the rolling direction for sheet
rolled at 20 0 e [527].
1.2.2. Mechanical Properties
The strength and hardness of unalloyed zinc are greater than those of tin or lead but
appreciably less than those of aluminum or copper. The pure metal cannot be used in
stressed applications because of its low creep resistance. Except when very pure, zinc is
brittle at ordinary temperatures, but it is ductile at about lOOoe [501].
Since the distance between atoms in the basal plane is shorter than that between
atoms in adjoining layers, bonding between basal planes is relatively weak, and, under

PROPERTIES, PRODUCTS, AND PROCESSES

stress, the lattice tends to first slip along this plane. At higher temperature, slip may also
occur along the (1010) plane. Another major deformation mode of zinc crystal is twinnmg,
which tends to occur along one of the (1012) pyramidal planes [2181.
Pure zinc recrystallizes rapidly after deformation at room temperature because of the
high mobility ofthe atoms within the lattice. Thus, zinc cannot be work-hardened at room
temperature. Zinc has low resistance to creep due to grain boundary migration. The
temperature for recrystallization and the creep resistance can be increased through
alloying [218].
Superplasticity, with an extension of up to 1000%, can be obtained for Zn-AI
eutectoid alloys of very fine grain size, on the order of I ,um, at temperatures of
200-2700C [218]. The deformation under this condition appears to take place by slip of
the small grains over each other, with little distortion of the grains.

1.2.3. Alloying Properties


The binary zinc alloy systems of most interest for commercial applications are (1)
Zn-AI, which at 4% AI forms the basis of the zinc die-casting alloys, (2) Zn-Cu, which
with up to 45% zinc are brass alloys; (3) Zn-Fe, which includes the phases making up
the galvanized coatings, and (4) Zn-Pb, which plays an important role in some pyrometallurgical extraction processes [218]. Ternary and quaternary systems, involving the
above alloys with addition of such elements as Ni, Mg, Ti, and Cd, are also of commercial
importance.
Figure 1.1 is the phase diagram for the Zn-AI alloy system. While the solubility of
zinc in aluminum is reasonably high, that of aluminum in zinc is rather limited. This
system has a eutectic composition at about 5% AI and a eutectoid composition at 22%
AI. Many commercially important alloys, such as the standard die-casting alloys and the
coating alloy Galfan, have been developed at or near the eutectic composition.
The phase diagram of the Zn-Fe system is illustrated in Fig. 1.2. Iron has very little
solid solubility in zinc. When the amount of iron in zinc is above 0.001%, its presence
can already be detected micrographically by the appearance of an intermetallic phase,
possibly FeZn 7 In a typical hot-dip galvanizing process, a number ofZn-Fe intermetallic
compounds can be formed. The relative amounts and metallographic morphology of these
compounds depend on the substrate steel and the hot-dipping conditions, particularly on
the Al content in the bath. The zinc corner of the phase diagram of the Zn-Fe-AI ternary
system is important for the hot-dip galvanizing process and has been investigated in a
number of studies [1281, 1282].
1.3. MAIN PRODUCTS AND APPLICATIONS

1.3.1. Zinc Coatings


The many types of zinc and zinc alloy coatings can be classified according to the
coating composition and the production methods employed [501, 1250, 1297]. When
classified according to chemical composition, zinc-based coatings fall into several major
categories: pure zinc, zinc-iron, zinc-aluminum, zinc-nickel, and zinc composites. In
terms of methods, zinc coatings can be produced by hot-dipping, electroplating, mechanical bonding, sherardizing, and thermal spraying (metallizing). The hot-dip method can

100

200

300

400

:;00

Al

10
20

10

30

50

30
f

60

40

Weight Percent Zinc

40

20

Alomi c Per ce nl Zinc

70

.'

50

60

60

.JoT"""'"

70

90

Zn

100

'11I.Me

90 100

(Zn)

"

60

FIGURE 1.1. Zinc-aluminum equilibrium diagram. From Baker [1228]. Reprinted with permission from ASM International.

CU
b

0..

CU

...

....<II

...cu
::I

600

~2"(

700

600

tTl
;Al

-l

'"0

;l>

:c

"'"

f-

QJ

.~.~l~C

,,
,,
,

,,

-I

- " _',

,
............
4

782"C

60

-2.-'-'---_.- _._._._._'."UII:!:. ._._. _._._._.-

,,

50

70
80

,90

\-' e

We ig h l Percen l Zinc:

:l00 \ ( .. , " .. T .. .. .. , .. .. ,,. ~ , .... n-.'~~ . ,,~1'-- " 1-"~ '~ I


o
10
20
:.to
"0
:;0
60
70
60
90

400

:;00

600

700

(aFe)p

,,
,,
,,

40

100

100

Zn

FIGURE 1.2. Zinc- iron phase diagram. From Baker [1228J . Reprinted with permission from ASM International.

a.

<I.>

600

912"C
900 -

....

cO

1000

1100 ,

::I

....

<I.>

u
o

, 200

1300

,,

,,

30

;;0

til

'"
'"
tTl
'"

tTl

-0
;;0

-l
)/'
;J>

o
o

20

1400
1394"C

10

tTl

.'"-0

1638"C
1500

1600

Alo m lc P erce n l Zinc

tTl

o-0

-0
;;0

CHAPTER I

TABLE 1.2. Typical Applications of Zinc-Coated Steel Products"


Coating

Applications

Coatings by continuous process


Zn and Zn-5AI
Roofing, siding, doors, culvert. ductwork. housing. appliances. autobody
Sheet
panels and structural components
Nails. staples. guy wires. stand. tension members. rope. utility wire.
Wire
fencing
Autobody panels and structural components
Zn-Fe
Autobody panels and structural components. housings. appliances.
Zn-Ni
fasteners
Automotive small parts and fasteners
Zn-Co
Roofing, siding, ductwork, culvert. mufflers. tailpipes. heat shields. ovens.
Zn-55Al
toasters, chimneys. silo roofs
Batch gal vanized
Structural steel for power-generating plants. petrochemical facilities. heat
exchangers, cooling coils. water treatment facilities. and electrical
transmission towers and poles
Bridge structural members. culverts, corrugated steel pipe. arches
Reinforcing steel for concrete structures
Highway guard rails. lighting stands, sign structures
Marine pilings. rails
Grates. ladders. safety cages
Architectural applications of structural steel. lintels. beams. columns. and
related building materials
Painted galvanized structural steel for aesthetic. color-coded. or extendedlife applicatIOns
"Refs. 1238. 1250. and 1296.

be further divided into two processes: continuous hot -dip and batch hot -dip. In continuous
hot-dip, long strands of sheet, wire, or tubing are fed through a bath of molten zinc alloy
in a continuous process. In batch hot-dip, fabricated parts, such as fasteners, poles, or
beams, are dipped into a molten bath either individually or in discrete batches. Similarly,
zinc electroplating can be made in a continuous or a batch process.
Typical applications for zinc- and zinc-alloy-coated steel sheet products cover a wide
range in the construction, automobile, utility, and appliance industries as shown in Table
1.2. As the cost of lumber increases, additional large-scale applications of zinc-coated
steel products are also expected to develop in the residential construction markets for
roofing, siding, and framing [1250]. Among all coated steel products, continuous-hot-dip
zinc-coated steel sheet has the widest range of applications and is predominant in terms
of tonnage produced and consumed. The electroplated zinc-based coatings are applied
primarily on automotive bodies and have advantages of uniform coating thickness and
excellent surface characteristics for subsequent painting, but certain disadvantages in
terms of cost [1296].
Zinc coatings can also be produced by sherardizing and thermal spraying [527].
Sherardized coatings are produced by a cementation or diffusion process in which the steel
parts are heated with zinc dust in a slowly rotating drum. They are suitable for nuts, bolts,
hinges, nails, and similar hardware and fittings. Thermal-sprayed zinc coatings are obtained
by melting zinc powder or wire in a flame or electric arc and projecting the molten metal
by air or gas onto the surface to be coated. This process is mostly used with structural steel

PROPERTIES, PRODUCTS, AND PROCESSES

works that are too large for a galvanizing bath, for example, steel bridges. It is also used
for repairing galvanized steel surfaces on which the coating is lost due to mechanical
damage or welding. More recently. thermal-sprayed zinc coatings have been used on
concrete surfaces to serve as the anode for cathodic protection. Sprayed zinc coatings are
relatively rough and porous compared to the coatings produced by other methods.

1.3.2. Cast Products


Cast zinc products are mainly produced by the die-casting process. in which liquid
metal is forced under pressure into a cooled die and solidifies almost instantaneously to
produce a fine-grained product. Die casting is a single high-speed operation that can
produce complex but very accurate components requiring little or no tinal shaping.
Die-cast products are used for automotive parts, household appliances and fixtures, office
and computer equipment, and building hardware. A typical composition for the most
commonly used die-casting alloy (Alloy 3) is Al 4.0%, Mg 0.03%. Cu < 0.25%, Fe <
0.1 %. The aluminum content contributes to the alloy's good castability and strength [218,
501].

1.3.3. Rolled Zinc


Rolled zinc products are in the form of sheet, strip, foil, plate, rod. and wire, with a
variety of compositions. Rolled zinc sheet is widely used in building. in the form of
roofing, cladding, gutters, rainwater pipes, and flashings. The rolled zinc used for roofing
is typically a Zn-Cu-Ti alloy (0.7-0.9% Cu and 0.08-0.14% Ti), which gives a good
combination of tensile strength. creep strength, and formability. Rolled zinc foil has been
made into adhesive tape for coating the surfaces of large structures that are difficult to
galvanize. Zinc wires are primarily used for metal spraying [501].

1.3.4. Zinc Dust and Powder


Zinc dust and zinc powder are particulate materials. The word "dust" is used for fine
particles, usually 2-20 .um in diameter, and "powder" for coarser particles. A distinction
is often made whereby "zinc dust" refers to the product made by condensation of zinc
vapor, and "zinc powder" to the product of atomization of molten zinc by a jet of air or
an inert gas [SOl]. Zinc dust may also be made in the form of flakes by milling in a
nonreactive fluid such as a hydrocarbon, a process referred to as "flaking." Flake thickness
tends to be 1 .um or smaller, and the diameter-to-thickness ratio may be about 10.
Zinc dust and powder are used mainly as reagents for producing chemicals, in metal
refining processes. as a component for making zinc-rich paints. and as an active material
for zinc batteries. They are also used in smaller quantities as a material for sherardizing
and thermal spraying and as an additive in plastics [501].
1.4. COATING PROCESSES

1.4.1. Hot-Dip Galvanizing


1.4.1.1. General Considerations. Hot-dip galvanizing is a process by which an
adherent coating of zinc and zinc-iron alloys is produced on the surface of iron or steel

CHAPTER 1

products by immersing them in a bath of molten zinc. It is the oldest and the most used
process for producing zinc coatings. Hot-dip galvanizing can be further divided into two
main processes: batch galvanizing and continuous galvanizing. In general, an article to
be galvanized is cleaned, pickled, and fluxed in a batch process or heat-treated in a
reducing atmosphere to remove surface oxide in a continuous galvanizing process. It is
then immersed in a bath of molten zinc for a time sufficient for it to wet and alloy with
zinc, after which it is withdrawn and cooled. Any of these stages can be critical to coating
quality.
The coating so produced is bonded to the steel by a series ofZn-Fe alloy layers with
a layer of almost pure zinc on the surface. The engineering quality of the coating depends
on the physical and chemical nature of the Zn-Fe intermetallic layers formed. The
thickness and composition of the alloys depend on whether they are produced in a batch
or a continuous process, mainly because of the differences in the immersion time in the
molten zinc bath and the bath composition employed in the two types of processes. The
coating produced by a batch process is thicker and has clearly distinguishable alloy layers
as shown in Fig. 1.3, while that produced by a continuous process is thinner and has only
a very thin and sometimes not visible (with an optical microscope) alloy layer at the
coating/steel interface as shown in Fig. 1.4. Table 1.3 gives the characteristics of the
Zn-Fe alloys in a batch hot-dip galvanized coating.
The thickness of a given phase in a Zn coating on steel is determined by the rate of
diffusion through the phases during their growth [262]. The main diffusion process is
diffusion of Zn through the galvanized layer toward the iron interface. The diffusion of
the iron moving outward occurs at a much slower rate. During a galvanizing process, the

Fe

FIGURE 1.3. Cross section of a typical batch-galvanized coating, showing the various Zn-Fe alloy layers.

PROPERTIES, PRODUCTS, AND PROCESSES

FIGURE 1.4. Cross section of a typical coating produced in a continuous process.

( layer is fonned first, followed by the i5 layer and, finally, the rlayer. The growth rate
of these different layers in the coating is shown in Fig. 1.5 [262]. The ( layer grows rapidly
at first but then much more slowly, while the growth of the i5 layer is at first slower than
that of the ( layer but then becomes faster. The growth rate of the r layer is very slow,
and therefore this phase may not be seen under the microscope at short reaction times.
The formation of zinc-iron alloy layers greatly depends on the silicon content of the

steel, as shown in Fig. 1.6 [262]. At a nonnal galvanizing temperature, the coatings on
steels with low Si concentrations, less than 0.03%, have nonnal thicknesses (Fig. 1.3).
On the other hand, at high concentrations of Si, e.g., 3%, the reactivity of the steel is low
and thin coatings are produced. At intermediate concentrations of Si, the reactivity of the
steel is high and thick zinc-iron alloy layers are produced. The peak reactivities occur at
Si contents of 0.06-0.1 % and about 0.5% as shown in Fig. 1.6. Figure 1.7 shows the
cross-sectional structure of a typical zinc coating on an Si-containing steel. The addition

TABLE 1.3. Characteristics of Zn-Fe Intermetallic Alloys"


Phase

Formula

Fe content (wt. %)

'7
(
,)1

Zn
FeZn13
FeZnw
FeSZn21

Max. 0.003

"Refs. 262, 312, and 501.

5.7-6.3
7.0-11.5
21.0-28.0

Crystal structure
Hexagonal close-packed
Monoclinic
Hexagonal close-packed
Face-centered cubic

Densi ty (glcm 3)

7.14
7.18
7.24
7.36

10

CHAPTER 1

200 r----------------------------------------;

Time, hours

FIGURE 1.5. Rate of growth of r, "I' and (layers at 45TC. After Mackowiak and Short [262].

of small amounts of aluminum to the zinc bath may effectively inhibit the growth of the
alloy layers on silicon-containing steels.
The as-galvanized coating is typically characterized by the appearance of spangles,
which often show a strong (000 1) basal texturing [261]. Also, the surface of fresh
galvanized coating is readily oxidized in air to form a very thin oxide film. The surface
oxide film on the galvanized steel produced in an aluminum-containing bath is usually
enriched with aluminum, owing to the high affinity between aluminum and oxygen, and
has a thickness varying from 20 to 100 A depending on the content of aluminum in the
coating [253,491].
1.4.1.2. Batch Galvanizing. In batch hot-dip galvanizing, the articles to be galvanized are first degreased and then pickled to remove mill scale and rust from steel parts.
3~---r----.-----r----r----.----.

o
(),

3mln.455C
4 min. 460C
8mln. 455C
8mln. 460C

FIGURE 1.6. Comparison of the results of various studies on the effect of silicon content on
OL---"l:---...."J.;;:----:h::----,.'-:;--~--O;:;:"i6 coating weight. After Mackowiak and Short
[262].

PROPERTIES, PRODUCTS, AND PROCESSES

11

FIGURE 1.7. Cross section of a zinc-iron alloy coating structure obtained on a high-Si (0.4%) steel.

Each of the degreasing and pickling steps is followed by a water rinse. The most common
degreasing process uses heated (65-82C) alkaline solution. Aqueous solutions of 3-14
wt. % sulfuric acid or 5-15 wt. % hydrochloric acid are generally used in pickling. To
avoid overpickling, inhibitors are often used [312].
Batch galvanizing can be a wet or a dry process. Wet galvanizing involves a kettle-top
flux blanket; dry galvanizing uses a preflux but does not use a flux blanket on the kettle.
In the dry process, after the steel article is degreased and pickled, it is immersed in an
aqueous zinc ammonium chloride solution, dried, and then immersed in the molten zinc
bath. In the wet process, the article is not usually prefluxed after cleaning but is placed
directly in the molten zinc bath through the top flux blanket. Zinc ammonium chloride is
generally used as the flux blanket. The fluxing promotes the alloying process at the
steel/molten zinc interface by removal of FeO on the steel substrate and ZnO on the
surface of molten zinc through chemical reactions with ZnCI 2 and NH 4Cl.
The molten zinc bath generally operates in a temperature range of 445-454C. The
bath temperature affects the fluidity of the molten zinc, the rate of formation of oxides
on the bath surface, the rate of coating solidification, the coating thickness, and the amount
and structure of the Zn-Fe alloy layers. The immersion time is usually in the range of
3-6 min. The speed of immersion and withdrawal influences the coating uniformity,
particularly with large articles.
1.4. J.3. Continuous Galvanizing. In the continuous hot-dip coating process, coils
of steel are welded end to end and are coated at speeds of up to 200 mlmin [1254]. In
general, there are "hot" and "cold" continuous hot-dipping processes. The major differ-

12

CHAPTER I

ence between the "hot" and "cold" processes is in the preparation of the steel surface after
the first cleaning stage and before immersion in the molten zinc bath. In the "hot" process
the strip first enters an alkaline bath that removes oils, dirt, and residual iron fines from
the rolling process. This is followed by a further cleaning stage with mechanical brushing
and electrolytic alkaline cleaning. The sheet then passes into a radiant tube furnace
containing a mixture of hydrogen and nitrogen that reduces surface iron oxides. Heating
of the steel also takes place to a temperature just above that for subcritical recrystallization.
The steel is then cooled to near bath temperature before entering the zinc bath.
In the cold process, steel strip is cleaned, pickled, and fluxed in-line with no heating
beyond that required to dry an aqueous flux solution of ammonium chloride and zinc
chloride on the steel surface before entering the zinc bath.
Typically, 0.1-0.2% Al is added to the bath to prevent the formation of a thick,
continuous layer of Zn-Fe intennetallic that could lead to poor coating adhesion during
forming [1250]. As the steel strip exits the bath, a layer of molten zinc is coated on the
surface. The thickness of the layer is controlled by passing the strip between wiping dies
to remove excess metal with a stream of gas. Forced-air cooling is used to reduce the sheet
temperature, which prevents coating damage from contact with turnaround rolls. Before
the sheet is finally wound into the coil fonn, it may be subjected to one or more
post-treatments such as oiling, chromating, and phosphating.
1.4.1.4. Galvannealing. In galvannealing, the hot-dipped steel sheet is processed
further. After exiting the molten zinc bath and passing through the wiping dies, it is heated
to temperatures of 500-550C for about 10 s to generate interdiffusion of iron from the
substrate and zinc from the coating to fonn an Fe-Znalloy coating. The actual alloying
time is influenced by the coating thickness and the compositions of both the zinc bath
and the steel substrate.
Compared to the galvanized coating, the galvannealed coating is generally easier to
paint without a special pretreatment, probably because of its rougher surface. The outer
surface of a galvannealed coating is a , phase containing about 6% Fe. The intermediate
J phase contains iron in the range of 8-12%. Next to the steel substrate is a Tlayer. These
intermetallic phases differ significantly in mechanical properties and detennine the
forming properties of the coating.
1.4.1.5. Zn-AI Alloy Coatings. Zn-55% Al (Galvalume) and Zn-5% Al (Galfan)
coatings are the two major hot-dip commercial zinc-aluminum alloy coatings. Ga1valume, containing 55% AI, 1.5% Si, and 43.5% Zn, was developed by Bethlehem Steel
[353, 1250]. This alloy has properties intermediate between those of hot-dipped zinc and
aluminum coatings. Galvalume has a higher corrosion resistance but less galvanic action
than a zinc coating. The microstructure consists of an outer layer and a thin intennetallic
layer that bonds the outer layer to the steel. This thin layer consists of two intennetallic
compounds: the inner sublayer is a quaternary AI-Fe-Si-Zn compound, and the outer
sublayer is a ternary A1-Si-Fe compound. The silicon moderates the reaction during hot
dipping and serves to minimize the thickness of this intermetallic layer [238]. About 80
vol % of the outer layer is composed of cored, aluminum-rich dendrites, representing the
first solid fonned during cooling. The last liquid to freeze in the interdendritic volume
between the aluminum-rich regions is enriched in zinc.
Galfan, containing 95% Zn, 5% AI, and a small amount of misch metal, exhibits
improved fonnability and increased corrosion life. The multiphase microstructure of

PROPERTIES, PRODUCTS, AND PROCESSES

13

Galfan is characteristic of its composition, exhibiting a lamellar structure of alternating


zinc-rich and aluminum-rich phases. The fineness of the structure increases with increasing cooling rates, and the structure is completely eutectic when fast-cooled. It is also
oriented in the direction of cooling. One characteristic of the Galfan microstructure is the
virtual absence of a brittle intermetallic phase between the steel and the coating. This has
been attributed to the addition of misch metal, which allows complete wetting of the steel
surface. The lack of the alloy layer is directly responsible for the high formability of
Galfan-coated steel [353].

1.4.2. Electroplating
Electroplating is another common method for producing zinc coatings on steel
surfaces. The plating process generally comprises three stages: (1) degreasing and
cleaning, (2) electroplating, and (3) post-treatment. Various types of plating baths are used
in the plating industry, and these can be roughly classified as acid or alkaline.
Most commercial zinc plating before 1980 was done in conventional alkaline cyanide
baths. The environmental concerns related to cyanide use have led to continuing development and application of other processes. At present, acid zinc plating baths constitute
about half of all zinc baths in developed nations, and their use is rapidly increasing
throughout the world [1297]. Typical compositions for acid and alkaline plating baths are
given in Table 1.4.
1.4.2.1. Continuous Plating Process. Continuous plating is a process for plating a
metal coating onto an endless steel sheet or wire. It consists of five main sections: payoff,
pretreatment, plating, post-treatment, and delivery [1296]. In the payoff section, the coils
of cold-rolled steel are loaded onto the entry reels, with the end of a new coil welded to
the tail end of the previous coil. The strip then passes through a precleaning and rinse
station, where the bulk of the oils is removed. In the pretreatment section, the residual
oil, surface carbon, and light surface oxide are removed as the strip is passed through
one or more alkaline cleaning, brushing, pickling, and rinsing stations. The cleaned
strip then enters the plating section, which consists typically of multiple plating cells.
At the beginning of the plating process, a conditioning cell may be used to prepare
the steel surface for plating. As the strip moves through from cell to cell, the coating

TABLE lA. Typical Plating Bath Compositions and Plating


Conditionsa
Chloride bath
Composition

ZnCI 2, 15-56 gil


NH4 C\, 100-200 gil
Brighteners, 3-5%

Temperature
pH
Current density

21-27C
5.2-6.2

aRefs. 312 and 1297.

0.3-5A1dm 2

Cyanide bath
Zn(CN)z, 54-86 gil
NaCN, 30-41 gIl
NaOH, 68-105 gil
Na2C03' 15-60 gil
Sodium polysulfide, 2-3 gil
Brightener, 1-4 gil
21-40C

-13

1-5 Aldm 2

14

CHAPTER I

thicknessis gradually built up to the required value. Upon exiting the last cell, the coated
sheet is immediately rinsed and dried to prevent streaking or staining of the coated surface.
In the post-treatment section, the steel strip is treated with processes such as phosphating,
chromating, or oiling to prepare it for painting or to provide extra surface protection.
Continuous plating processes are generally classified in terms of three characteristics: anode type, electrolyte chemistry, and plating cell geometry [1158]. Either
soluble or insoluble anodes are used. Soluble anodes dissolve anodically into the electrolyte, whereas the reaction on insoluble anodes is generally the oxidation of water. Soluble
anodes are usually used in chloride baths, and insoluble anodes are usually used in sulfate
baths. There are three types of cells based on cell geometry: vertical, horizontal, and radial.
The choice of anode type, bath chemistry, and cell geometry in an electroplating line
depends on the capacity, productivity, automation. specialized components, etc.
1.4.2.2. Alloy Plating. The development of zinc alloy coatings has addressed the
need for more corrosion-resistant automotive bodies. The steel sheets used for automotive
body panels must have not only high corrosion resistance but also good paintability,
formability, and weldability. Currently, the most prevalent zinc alloy coatings are Zn-Fe
and Zn-Ni [1158]. Other alloy and composite coatings such as Zn-Co [229,246], Zn-Mn
[330, 349], Zn-Co-Mo [312], Zn-Co-Cr [425], Zn-Ni-Si0 2 [263], Zn-Ni-Ti [487],
Zn-Si0 2 [284], and Zn-Co-Cr-Alp3 [351] have also been developed.
Table 1.5 shows the typical compositions of Zn-Fe and Zn-Ni alloy plating baths.
The content of Fe and Ni in the alloy coating can be greatly varied by changing the bath
chemistry and operating conditions [227, 312]. The typical Zn-Fe coatings contain
approximately 18% Fe, and Zn-Ni coatings contain 9-13% Ni.
The deposition of Zn-Fe and Zn-Ni alloy coatings represents cases of anomalous
codeposition [37]. In an anomalous codeposition the less noble metal is predominantly
deposited. This is opposite to the normal process, where the more noble metal is
predominantly deposited. For Zn-Fe or Zn-Ni coating at low current densities, zinc and
iron or zinc and nickel codeposit by a normal process. However, at high current densities,
zinc is predominantly deposited. This arises from an increase in pH near the cathode
surface under high-current conditions. The higher pH favors the formation of Zn(OH)2
and thus reduces the deposition sites for iron or nickel.

TABLE 1.5. Typical Bath Compositions and Conditions for Plating


Zn-Fe and Zn-Ni Alloy Coatings"
Zn-Fe bath
Composition

Temperature
pH
Current density
"Ref. 312,

ZnS04,7H 20, 50 gIl


FeS04,7H 20, 250 gIl
(NH4 hS04, 120 gil
CSH S07' 0,5 gil
50C
1
30 Ndm 2

Zn-Ni bath
ZnS04,7H20 + NiS0 4,6H20, 500 gil
Na2S04, 60 gil
50C
2
10-15 Ncm 2

15

PROPERTIES, PRODUCTS, AND PROCESSES

1.3. PHOSPHATING
Phosphating is a surface treatment process in which metals such as iron, zinc, and
aluminum and their alloys are treated with a solution of phosphoric acid and other
chemicals. The reaction between the surface of the metals and the solution results in the
formation of an integral layer of insoluble crystalline phosphate. Phosphate coatings
typically range in thickness from 3 to 50 ,urn and vary in color from iridescent blue to
dark gray [163, 578].
Phosphating on zinc-coated steels is used mainly to prepare the metal surface for
painting. Phosphate coatings provide uniform surface texture and increased surface area,
and, when used as a base for paint, they promote good adhesion, increase the resistance
of the paint to humidity and water soaking, and eventually increase the corrosion
resistance of the painted system.
Phosphate coatings can be produced by spray, immersion, or a combination of the
two [578]. There are three principal types of phosphate coatings used in the industry: zinc,
iron, and manganese phosphate. In the strip galvanizing lines, zinc phosphate baths are
usually used. The zinc phosphating bath is typically operated in the pH range of 1.4-3.4
and at a temperature in the range from 30 to 100C. The process time can vary from several
seconds to several minutes. In general, the spray method produces a coating at a faster
rate than the immersion method. A phosphating bath usually contains an accelerating
agent to speed up the rate and to reduce crystal size.
The composition of phosphating solutions varies greatly and depends on the specific
application. Many phosphating solutions used in the industry are proprietary. A simple
phosphating bath can, for example, contain 6.4 g ZnO/I, 10 ml H 1POil, 4 ml HNO/I, and
I g Ni(N0 3h/1 [94]. A dark gray coating can be obtained on galvanized steel after 5 min
in this solution at about 60C. Most phosphated articles that are used as a paint base are
also given a post-treatment with a rinse of chromic acid or other solutions [5781.
Phosphate coatings, when used for corrosion protection, are usually topped with an oil
or wax to seal the pores in the coating.
Phosphate coatings are formed through a dissolution and a precipitation process. The
zinc dissolves in the phosphating solution with accompanying hydrogen evolution. As a
result of the hydrogen reduction, a thin layer of electrolyte near the surface is depleted of
hydrogen ions and becomes neutralized. Since the solubility of zinc phosphate in neutral
solution is low, zinc phosphate precipitates on the zinc surface to form the crystalline
phosphate coating. This process involves essentially three reactions [93-95, 11751:
I. Dissolution of zinc:
Zn

Zn 2+ + 2e-

2. Reduction of oxidizing agents to the reduced form:


Ox (4W, O2 , 4NO:;, etc.) + 4e- ~ Re (2H2' 20 2-, 4NO, etc.)
3. Precipitation of a zinc phosphate layer, hopeite:
3Zn2+ + 2H 2 PO:;- + 4HP ~ Zn/P0 4 )2 . 4H 2 0 (s) + 4W
and also a zinc-iron phosphate layer, phosphophyllite, in the presence of Fe 2+:

16

CHAPTER I

The kinetics of nucleation and growth is systematically discussed in the book by


Rausch [1175]. The most important reaction in controlling the thickness of the phosphate
coating is the nucleation of phosphate crystals. To obtain a large number of nuclei, and
thereby a fine-grained coating, the increase in pH at the metal surface should be as fast
as possible during the first seconds of fonnation of the phosphate coating. Dipping in a
colloidal solution of titanium phosphate is usually used as an activation process to increase
the number of nuclei. The phosphate coating process is completed when the surface of
the metal is so fully covered by the crystalline phosphate that no further significant
neutralization of the near-surface liquid layer can take place.
1.4. CHROMATING
Chromating is a process in which an aqueous solution of chromic acid, chromium
salts, and mineral acids is used to produce a thin conversion coating on a metal surface.
The chemical reactions between the metal and solution cause the dissolution of the metal
and fonnation of a protective film containing complex chromium and metal compounds.
Chromate conversion coatings can be applied to a number of metals. They are most
commonly used to protect coatings based on zinc and its alloys during storage and
transportation. They also generally enhance the corrosion resistance at zinc and zincalloy coatings on tubings, fasteners, etc.
Since its introduction in the mid-1930s [57], the chromating process has become the
most widely used process for surface post-treatment of zinc products. However, because
of environmental concerns, its applications have become increasingly limited, which has
led to new research activities in search of alternatives [1258].
Most of the chromating formulations used today are proprietary [1258]. A conventional chromating process, for example, the "Cronak process" developed by The New
Jersey Zinc Company in 1936, consists of immersion of the zinc article for 5-15 s in a
chromate solution (200 g of Na2Cr207H20 and 6-9 ml of H 2S04 in II of water at 20C),
followed by rinsing and drying [57].
The detailed fonnation mechanism for chromate coatings is not fully understood. In
general, the fonnation follows a dissolution and precipitation process similar to that in
phosphating [67]. During chromating, a set of electrochemical and chemical reactions
occur on the zinc surface in contact with the chromate solution. There are primarily three
reactions: zinc dissolution coupled with a cathodic hydrogen reduction, reduction of
hexavalent chromium ions to trivalent ions, and precipitation of trivalent chromium
hydroxide which incorporates hexavalent chromium compounds and zinc compounds.
The precipitation of the hydroxide of chromium is promoted by the rise of surface pH
through the local consumption of acid [65].
The thickness, composition, and color of chromate coatings depend mainly on
chromate concentration, pH, and dipping time [57,65]. For example, a yellow coating
typically ranges in thickness from 0.1 to 0.6 pm [57]. The freshly fonned films are gel-like
and do not reveal any crystalline components by X-ray diffraction examination [59,65].
They harden from their original soft state to a reasonable abrasion-resistant coating within
24 h in air. The hardening also results in the crystallization of the coating, which can be

17

PROPERTIES, PRODUCTS, AND PROCESSES

TABLE 1.6. Composition of the Chromate


Coating Formed by the Cronak Process"
Constituent
Chromium(VI)
Chromium(III)
Sulfate
Sodium
Zinc
Water

Relative amount (%)

7-12
25-30
2-3.5
0.2-0.5
2-2.5
15-20

"Ref. 57.

observed from X-ray diffraction patterns. Cr20 3 , ZnCr0 4 , and ZnO have been detected
in chromate coatings. Table 1.6 shows the composition of a chromate coating [57].
Chromate coatings protect the zinc metal through barrier and passivation effects. The
complex chromium oxide film serves as a barrier to the environment, while the hexavalent
chromium contained in the film serves as a passivating agent. The hexavalent chromium
leaches out when in contact with water and produces a local chromate solution that forms
a chromate film at an exposed zinc surface [65]. Through leaching, immersion of a
chromated surface in distilled water for 24 h results in a marked loss of chromate. A
similar effect is seen in outdoor exposure [71].

2
Electrochemical
Thermodynamics and Kinetics
2.1. INTRODUCTION
Electrochemical processes playa very important role in the production and application of zinc. Electrowinning in zinc refining, electroplating in the production of zinc
coatings, zinc batteries for energy storage, and zinc coatings and anodes for corrosion
protection are all essentially based on electrochemical processes.
In this chapter, the thermodynamic and kinetic properties of the zinc electrode are
reviewed. The material presented is organized according to each of the elemental reactions
that can occur on a corroding electrode: zinc dissolution, zinc deposition, hydrogen
evolution, and oxygen reduction. While zinc dissolution is discussed in detail, zinc
deposition is only treated superficially since it is not important from a corrosion perspective. The topics of oxide film formation and passivation are dealt with in a separate chapter
because of the large amount of literature on these phenomena and their particular
importance in .corrosion processes. The corrosion potential and corrosion current, which
are the two key parameters connecting the fundamental electrochemistry and practical
corrosion behaviors in various applications, are also considered in another chapter.
The information presented in this chapter is limited to aqueous solutions. Some
electrochemical information on the zinc electrode in nonaqueous electrolytes is presented
in Chapter 15. A discussion of the zinc electrode kinetics in nonaqueous electrolytes and
fused salts can be found elsewhere [532].
2.2. THERMODYNAMIC STABILITY
Zinc is divalent in all its compounds. Compounds of Zn(l) do not exist naturally
[1253J. The stability of zinc and its compounds in aqueous solutions in the absence of
complex formation is determined by the equilibrium conditions listed in Table 2.1 [1,906].
The value for the standard potential of the zinc electrode can be calculated from
thermodynamic data [I]:

Eo = LVjJ.123060n = -35,18412 x 23,060 = -0.763 V SHE (2.1)


19

20

CHAPTER 2

TABLE 2.1. Reactions of Zinc in Aqueous Solutions and Equilibrium Conditions"


Reaction

Equilibrium

Standard potential or equilibrium condition

Two dissolved substances


I
Zn 2+ + H20 =ZnOH+ + H+
2
ZnOH+ + H20 =HZnO;- + 2H+
3
Zn 2+ + 2H 20 =HZnO;- + 3H+
4
HZn02 =ZnO~- + H+
Two solid substances

ZnO + H20 =HZnO;- + H+


ZnO + H 20 =ZnO~- + 2H+
Zn Zn 2+ + 2eZn + 2H 20 =HZnO;- + 3H+ + 2eZn + 2H 20 =ZnO~- + 4H+ + 2e-

log(Zn 2+) = 10.96 - 2pH


log(HZnOz) =-16.68 + pH
log(ZnOh =-29.78 + 2pH
Eo =-0.763 + 0.0295Iog(Zn 2+)
Eo =0.054 - 0.0886pH + 0.0295 log(HZnO;)
Eo =0.441 - 0.1182 pH + 0.0295Iog(ZnO~-)

H2 =2H+ + 2e2H20 =O2 + 4H+ + 4e-

Eo =0.000 - 0.0591 pH
Eo = 1.228 - 0.0591 pH

Stability of water
(a)

(b)

Eo =-0.439 - 0.0591 pH

5
One solid and one dissolved substance
6
Zn 2+ + H20 =ZnO + 2H+

7
8
9
10
II

log (ZnOH+)/(Zn 2+) =-9.67 + pH


log (HZnO;- )/(ZnOH+) =-17.97 + 2pH
log (HZn02 )/(Zn2+) -27.63 + 3pH
log (ZnO~-)/(HZnO;- ) =-13.17 + pH

"Ref. I.

where v is the stoichiometric coefficient. n is the number of electrons involved in the


reaction, and f..l is the chemical potential of the species involved in the reaction. Temperature has slight effect on the zinc potential as shown in Fig. 2.1 [532]. At equilibrium the
potential difference between various crystal surfaces of a zinc single crystal and polycrystalline zinc is less than 10mV [532]. The potential values measured in different zinc salt
solutions generally show good agreement with the calculated values.

-0.76

+ Ref. 2
... Ref. 1

+ Ref. 7
-0.762

Ref. 6

-W

"

-0.764

-0.766 ' - - -_ _...l.-_ _ _- " -_ _ _ _' - -_ _ _. . . l . - _ - l


o
10
20
30
40
Temperature,

FIGURE 2.1. Effect of temperature on the zinc electrode potential determined experimentally. The reference
numbers in the figure are from Brodd and Leger [532]. Reprinted by courtesy of Marcel Dekker, Inc.

21

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

The equilibrium conditions listed in Table 2.1 can be represented by the Pourbaix
diagram in Fig. 2.2 [1]. The lines labeled with the letters a and b represent, respectively,
the equilibrium conditions of the reduction of water to gaseous hydrogen and of the
oxidation of water to gaseous oxygen, when the partial pressure of hydrogen or oxygen
is I atm at 25C.
According to Fig. 2.2, the stable region of zinc is below line a, and thus zinc is
thermodynamically unstable in water and aqueous solutions and tends to dissolve with
the evolution of hydrogen over the whole pH range. In solutions of pH between
approximately 8.5 and 12, zinc can be covered with a hydroxide film, which has the effect
of inhibiting zinc dissolution. The pH-potential diagram in Fig. 2.2 is valid only in the
absence of the chemical species with which zinc can form soluble complexes or insoluble
compounds.
The stability of zinc oxides and hydroxides in aqueous solutions depends on pH. As
shown in Fig. 2.3, zinc hydroxides are amphoteric. They dissolve in acid solutions to give

-2 -I

10

11

12

13

14

15

16

2,2,-,----i'---.--T-T--....;-~-T-......;---.:_;___=;...__.;~~~'---~~~~.:,'2

EM

$,

1,B

1.4

,,

1,6

1,2

1,6

1,4

1,2

I
I

-2
-6

I
I
I

I
I
I
I

-6

-....

0,8

-2

0,6

I-_
I

0,4

0,4
0,2

Z~(OH)2
I
I

-0,4

I
I
I
I

i-_
I

-~81~~9~~~~~~~~~~~=d~~i
F
0

-z

-0,2

-0,6

-4 -6

0,6

"0-_

0,8

I
I

0,2

I,B

Zn02

--- __

'

-0,2

,,

-0,4

I
I

-0,6

...,-

-0,8

I
I
I
I
I

,,

-I

-1,2

-1,2

-1,4

Zn

-1,6

-I, 8~7--;.;---:--;';---::--:---::---=--=_-::-~---:''::--"-:---:'::--:'!:--,L--,.L.-.-l-1 8
-2

-I

14

15pH16'

FIGURE 2.2. Potential-pH equilibrium diagram for the zinc-water system at 25"C [established by considering
Zn(OHh]. From Pourbaix [I]. Copyright by NACE International. All Rights Reserved by NACE; reprinted
with permission.

22

CHAPTER 2
1

12

11

10

13

-1

'/

?"
~ -2

9/

-\

,~""'/
-$> /

//,

-2

~ /

-)

-3

~ /

~/

'1,/

~ -4

-.

15 16

(-

+
..-...

14

-4

~
+ -5

L:::::J

g>

-5

-6

-6

-7

-7

-8
4

10

12

11

13

14

15

pH

-8

FIGURE 2.3. Influence of pH on the solubility of the zinc hydroxides in water at 25C. From Pourbaix [IJ.

Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.

zincic ions Zn 2+ and in alkaline solutions to give bizincate or zincate ions


HZnO;- and Znq- [1]. The solubility varies slightly with the type of hydroxides and
oxides. At room temperature, e-Zn(OHh is the most stable compound whereas amorphous
Zn(OHh is the most unstable [404]. Figure 2.4 shows the experimentally measured
solubility of ZnO as a function of KOH concentration [1131].
Zinc can form insoluble compounds with many chemical agents. The compounds
commonly found in corrosion products are zinc sulfate, chloride, and carbonate. The
stability of these compounds has been found to affect the corrosion resistance of zinc in
many environments [331]. Zinc carbonate is of particular importance because it has been

2
:E

c-

';a
~
Q)

<.l
C

.
0

'C:::
N

10

15

KOH Concentration , M

FIGURE 2.4. Solubility of ZnO in KOH at 23C. After Hampson et al. r1131].

23

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

ZnO,

- - - - -- - - - -

--;;:.
~

CUl

'

ZnCO, ,

I
I

-:- ~
I

Zn

,
I

H,CO,

-2

Zn"

- - -- - - ---

,
"t -

-: - Ln(OH),(CO,),

- - ,, - - - - - -,
ZnO,
- - ----- - L
I

IICO,-

4
pH

Z,,(OH),-

Zn(OI1).'-

I
I
I

COll-

12

FIGURE 2.5. Potential-pH diagram for the zinc-water-carbonate system at 298 K. [Zn 2-j = 10-4 molldm 3
(dissolved zinc species); [H 2CO]j + [HCO]l + [CO~-l = 10-2 molldm J (dissolved carbonate species). After
Kannangara and Conway [3[.

found to be responsible for the high corrosion resistance of zinc in atmospheric environments. Theoretically, the formation of zinc carbonate can occur in solutions containing
carbonates and bicarbonates according to the following reactions 13. 906J:
ZnO + 2H+ ~ Zn 2+ + H 2O

(2.2)

Zn 2+ + H 2CO] ~ ZnCOJ(s) + 2W

(2.3)

-1

..... -2

'"

00

.....

-3

Solution

-4

-s

10

11

pH
FIGURE 2.6. Equilibrium diagram for corrosion products of zinc in a chloride environment. After Feitknecht
[404].

CHAPTER 2

24

(2.4)

2W + 2HCO:J + HzO + 5ZnO (s)

(2.5)

Zns(OHMC03h (s)

The pH-potential diagram for the zinc-water-carbonate system is shown in Fig. 2.5. It
can be seen that the presence of carbonates and bicarbonates extends the possible
passivation region to near neutral pH values. It has been calculated that when the total
concentration of HZC0 3 + HCO:J + CO~- is larger than lO-zAM, the domain of zinc
passivation by formation of a zinc carbonate film becomes larger toward neutrality than
that of zinc passivation by formation of a zinc hydroxide film [906]. At pH values greater
than 9, zinc carbonate is less stable than zinc hydroxide.
The stability of zinc sulfate and zinc chloride are determined by the following
reactions [331]:
(y = 3 or 6) (2.6)

ZnCI

Zn (OH
)..-_=::l:=:....L_..:::::::L=-4--r-.r,,;:;:::I:::>:I...

I;

zn (OH I 2

ZnCI~

10

N'"
v'
U
r "'_Nr
o v
C

-!V
r

--

~ ~

-~
I

0
c

0
c

'"

FIGURE 2.7. Relative concentrations of basic zinc chloride species in aqueous solution as a function of pH
and Cl- concentration. After Despic [1118].

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

25

(2.7)
Figure 2.6 shows that in a chloride solution, the zinc chloride compound that can be
formed varies with the pH of the solution [404]. The relative amounts of the various
chloro-hydroxo complexes as a function of pH and concentration of Cl- are shown in Fig.
2.7 [1118]. In concentrated solutions, zinc seldom exists as a simple ion owing to the
formation of complexes.
2.3. IONIC PROPERTIES
The zinc ion has a radius of 0.74-0.83A and a hydration number of 10-12 [8, 327].
Owing to the electronic configuration of the zinc atom, zinc ions tend to form spJ -hybridized tetrahedrally coordinated complexes in solution [532]. The primary coordination
number of the zinc ions is four, although variations have been shown to exist. The
equilibrium constants for several ligands are listed in Table 2.2. Generally, complex
formation follows the pattern:
(2.8a)
(2.8b)
(2.8c)
(L8d)
The transport properties of ions in solution are characterized by the ionic diffusion
coefficient and mobility. In Table 2.3 the ionic mobilities and diffusion coefficients of

TABLE 2.2.
Complexing
agent
F
CIBr-

1-

cw
NH3
OW
SCW
Tartrate"
SO~-

Overall Formation Constants for Various Zinc Complexes"


Log formation constant"

PI
0.92
0.72
0.22
-0.47
2.32
6.31
1.57
2.30
2.08

Ih

P3

IJ4

-0.49
-0.10
-2.00
11.07
4.61
11.19
1.56
4.10

-0.19
-0.74
-0.74
26.05
6.97
14.31
1.51
5.55

-0.18
-1.00
-1.25
35.67
9.36
17.70
3.02
6.79

"Reprinted from Brodd and Leger [532], by courtesy of Marcel Dekker, Inc.

h/I,-fl. are the formation constants corresponding to the reactions given by Eqs. (2.Sa)-(2.Sd), respectively.

26

CHAPTER 2
TABLE 2.3.

Values of Equivalent Conductances (}.~) and Diffusion Coefficients (D) of


Selected Ions at Infinite Dilution at 25Ca
.0

1..;

Cation
H+
Li+
Na+
K+
NH;
Ag+
TI+
Mg2+
Ca 2+
Sr2+
Ba 2+
Cu 2+
Zn 2+
La3+

Co(NH3)~+

I"i

(cm 2/s)

(Q-I'cm2/equiv)
349.8
38.69
50.11
73.52
73.4
61.92
74.7
53.06
59.50
59.46
63.64
54
53
69.5
102.3

.0

Djx 105

9.312
1.030
1.334
1.957
1.954
1.648
1.989
0.7063
0.7920
0.7914
0.8471
0.72
0.71
0.617
0.908

Anion
OW

CI-

Br-

r
NO]
HC03"
HCO;:
CH 3CO Z
SO~-

Fe(CN)~Fe(CN):10
CI0
BrO;
HS0

4
4

(Q-I'cm2/equiv)
197.6
76.34
78.3
78.8
71.44
41.5
54.6
40.9
80
101
III
54.38
67.32
55.78
50

Dj x 105 (cm 2/s)

5.260
2.032
2.084
2.044
1.902
1.105
1.454
1.089
1.065
0.896
0.739
1.448
1.792
1.485
1.33

"From Newman [328J. Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.

zinc at infinite dilution are listed along with those of other ions [328]. In KOH solutions
the diffusion coefficient of zincate ions varies with KOH concentration and temperature
as shown in Fig. 2.8 [221]. The diffusion coefficient of Zn 2+ in 0.05M ZnCl 2 + 1M KCl
solution has been found to be 0.89 x 10-5 cm 2/s [1120].
The mobility of zinc ions is generally lower than that of most uncomplexed anions,
as a result of the much larger hydration shell of the zinc ions. For example, the transport
number of zinc in Zn(Cl0 4)2 solution is 0.44 in a solution at infinite dilution and 0.335
12.----------------------------------,
x

E 8

c
Q)

'(3

~o
o 4
c:

'00

:l

'I:

is

o
o

L -_ _ _ _ _---'--_ _ _ _ _---"_ _ _ _ _ _- '

Molarity of KOH, M

12

FIGURE 2.8. Variation of diffusion


coefficient of Zn(OH)~- (3 x 10-3M)
in KOH solutions with temperature
and KOH concentration. After
McBreen and Cairns [22IJ.

27

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

100

E
E
.c
0

;3-

80
60

:~

ti

:l
"0

40

c:
0

()

20
0

10

12

14

ZnCI 2 Concentration , M

FIGURE 2.9. Conductivity of aqueous zinc chloride solutions at 298 K. After Thomas and Fray [8971.

in a molar solution [327]. In general, for a salt with a cation transport number t+ [t+ =
A)(},+ + },J J less than the anion transport number t_, t+ will decrease with increasing
concentration. The conductivity of a zinc salt solution increases sharply with concentration at low concentrations but decreases at high concentrations due to increased viscosity.
Figure 2.9 shows the changes in conductivity as a function of zinc chloride concentration
[897].
2.4. DOUBLE-LAYER PROPERTIES
The values of the double-layer capacitance of a zinc electrode near its reversible
potential in aqueous solutions range from 16 to 20/lF/cm2 [3, 180, 532, 782]. The
double-layer capacitance is a function of potential and solution concentration. Figure 2.1 0
shows typical capacitance-potential curves of a zinc electrode in two different sodium
sulfate solutions [1291]. Determination of capacitance at potentials anodic to the Zn rest
potential is difficult because of the rapid anodic dissolution.
The capacitance curves measured on single-crystal zinc electrodes are, in general,
similar to those meas ured on polycrystalline electrodes [782]. Values of capacitance much
larger than about 20 /lF/cm 2 are generally associated with faradaic and specific adsorption
processes. The pH of a solution has a significant effect on the capacitance of a zinc
electrode because a strong interaction between OH- and the electrode occurs at pH> 3.4
[810]. Capacitance values of 40 to 600 /lF/cm 2 have been observed in KOH solutions due
to OW adsorption [12, 763, 786]. Also, OW adsorption appears to be associated with
surface inhomogenei ty. It has been reported that in strong KOH solutions, OH- adsorption
occurs on the zinc electrode but not on amalgamated zinc since the latter may have less
surface inhomogeneity [478]. Sustained immersion in neutral or alkaline solutions may
cause an increase of the electrode capacitance with time because of the gradual formation
of a solid film on the surface [70 I] .

28

CHAPTER 2
30

25

~ 15

<5

10

1 0.5 N Na 2SO., pH=2.87


2 0.01 N Na2SO., pH=2.96

OL---~------~~-------L--

-1.5

-2

-1

______ ______
-0.5
~

E, Vshe

FIGURE 2.10. Relationship between capacitance and potential for a zinc electrode in Na2S04 solutions. From
Keifets and Krasikov [1291].

Measurements in O.lM NaCl04, NH4Cl0 4, and NH4Cl solutions of different pH


values indicate that the specific adsorption of Cl- has a negligible effect on the capacitance
of the zinc electrode in the potential range between -1.11 and -1.65 VSCE [810]. In acidic
solutions, where OH- adsorption is absent, cation interactions may affect the capacitance
values.
The double-layer capacitance of a zinc electrode in an aerated 0.5M NaCl solution
has been found to decrease with increasing thickness of the electrolyte layer as shown in
Fig. 2.11 [608]. It is largely independent of the layer thickness in deaerated solution.
The potential of zero charge (PZC) can be obtained from the minimum of the
capacitance-voltage curve. Experimental investigations typically locate the PZC of zinc
around-0.85 to -1 VSCE [180, 782, 532]. It has been found that the PZC varies with crystal

50,--------------------------------------,

Eo

40

~30
ai

ffi

.~ 20
Q.

ro

<.)

10

O~------~------~-------L------~

-2

-1

______~
3

Layer thickness, mm (Log)

FIGURE 2.11. Interfacial capacitance C as a function of electrolyte layer thickness. After Keddam et al. [608].

29

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS


TABLE 2.4.

Electrochemical Techniques Used to Study Zinc Electrode Kinetics


References

Technique
Dynamic polarization
Steady-state polarization
Measurement of i-t
transients
Measurement of V-t
transients
Impedance measurements
Rotating electrode

Corrosion

10, 33, 110, 106,


III. 112
106,703,790
114,790

Dissolution!
deposition

H+ and 02
reduction

22, 176,796,800, 10,110,797


1251
61,181, 700, 763, 113,683
786, 789
12,324

118,790

12,785,789,887, 7,9
903
113,700,701
61,683,763,786, 110,683, 1251
894
106,113,116,128 405,786,789,
106, 116, 128,
796,888
445,797, 1251

Passivation

526,702,794,890
526,904
18,681,904
18,127,886,889
526, 702
702

orientation [180]. Since the potential of a zinc electrode in a solution is nonnally negative
to E pze , the electrical double layer is populated primarily with cations in the inner
Helmholtz plane.
2.5. KINETICS OF ELEMENTAL REACTIONS
The kinetics of the zinc electrochemical reaction processes such as dissolution,
deposition, hydrogen evolution, oxygen reduction, passivation, and surface film formation, has been the subject of many electrochemical studies. Table 2.4 lists the different
electrochemical techniques used in these studies.

2.5.1. Dissolution
The dissolution of zinc has been extensively studied [12. 785, 786, 790]. Zinc
dissolves readily near its equilibrium potential, with the formation of divalent zinc ions.
In acidic solutions the dissolution product is simply Zn 2+. In alkaline solutions the
predominant zinc species has been identified to be tetrahedral Zn(OH)~-. The charge
density for dissolution of one monolayer of solid zinc can be calculated using the lattice
constants of the zinc crystal to be 522 jJ.C/cm2 [3].
Table 2.5 lists the exchange current densities and Tafel slopes for zinc dissolution in
various solutions. The exchange current density and Tafel slope are a function of many
factors. Figures 2.12 and 2.13 show the exchange current density of zinc in KOH solutions
as a function of KOH concentration and the concentration of zincate ions [887]. The
exchange current density increases with KOH concentration and reaches a maximum at
a KOH concentration of about 8M. It changes only slightly with zincate concentration.
According to Dirkse [1132], the decrease in the exchange current density at high KOH
concentrations shown in Fig. 2.12 is related to the ionic strength, which influences the
mobility of the reacting species and modifies the exchange current density. Johnson et al.
[181] studied the anodic dissolution of zinc in solutions containing CI-, Br-, 1-,

30

CHAPTER 2

TABLE 2.5 . Tafel Slopes and Exchange Current Densities for Zn Dissolution in Various
Solutions
b(mV)

Solution
IMHCI
1M H ZS0 4

25
42

O.IM NaZS04' pH = 1-5


0.06MHCI

30
90
94.4

1M ZnCl z, pH

=2

1M NaCI , pH = 3.8

25

1M NaZS04' pH = 3.8
O.IM NaCI , pH = 5.3

38
30
20

O.OIM ZnCl z + 0.32M


K ZS0 4

Reference(s)
14
14
0.0008 a

1.75

110
1I0
14,118
181

120

IMZnS04
O.IMKOH
O.4MKOH

200

13

24

887
794

40
28

IMKOH
IMKOH
IMKOH
I. 8M KOH + ZnO
7MKOH

445
15
1120

93

311
887

786

33-95
50
27

8-370
209

12
785

"Zn2+ concentration is between 10-5 and I0-4M.

SO~-, NO;, and acetate. They found that the Tafel slopes for zinc dissolution tend to be
60-85 m V/decade for solutions containing NO; and 15-40 mV/decade for those without
NO;.
Dissolution may occur on a bare surface or on a surface covered by a solid film. Solid
films formed during Zn dissolution may have different compositions and various degrees

0 .5

SS C
.., 40 C
+ 2S' C

0 .4

E
~
.-

O C

...

0 .3
0 .2
0 .1

KOH Concentration, M

FIGURE 2.12. Variation of the exchange current density for the zinc electrode in KOH solutions with
temperature and KOH concentration. After Dirkse [1132] .

31

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

KOH Cone ,

0 ,3

03 M

+ 7M

$"

)!(

10 M

12 M

:;

Q)

FIGURE 2.13, Exchange current


density in KOH solutions as a function
of zinc concentration, Reprinted from
Dirkse and Hampson 18871. with kind
permission from Elsevier Science Ltd,
The Boulevard, Langford Lane,
Kidlington OX5 1GB, United Kingdom,

C>

'"

,J;;

"
w

0,1 -

)(

0.4

0,6

1.2

1.6

Molarity of Zincate. M

of compactness and thus may significantly affect the dissolution process. In solutions
containing no species with which zinc can form insoluble salts (e.g., NaCI and Na 2S04
solutions), the zinc electrode maintains a bare surface during dissolution at a pH below
3.8 [110). At a pH value of 5.8 in 3 M NaCI or Na2S04 , according to Baugh [110], the
zinc electrode is coated with an oxide film: this film may affect both the anodic and the
cathodic process but is not passivating, In O.lM NaCI solution, the dissolution of zinc at
low pH values (up to 3) is kinetically controlled whereas at high pH values (> II) it is
controlled by diffusion of ZnO~- or HZn02 away from the surface [ 1161.
The anodic dissolution of zinc may be greatly inhibited by the formation of solid
surface films in solutions of carbonates [3, 127, 196], nitrates [591, phosphates [481, 800,
797), and chromate, molybdate, and tungstate [57,59,98,199). In phosphate solutions,
according to De Pauli et al. [800), the dissolution of zinc occurs through a solid surface
film which changes composition and structure with pH [481,800]. At pH > 12 a
dissolution-precipitation mechanism operates whereas at pH < 12 a solid-phase process
dominates. The presence of PO!- ions promotes zinc dissolution whereas the presence of
HPO;- ions inhibits it. Kuznetsov and Podgornova [101) suggested that zinc dissolves to
form [ZnH 2P0 4 at pH 4.5 and [Zn(HP0 4 hf- at pH 9.5. In saturated solutions of zinc
sulfate, a thin pore-free salt layer forms, and zinc dissolution occurs through a direct
reaction of zinc with sulfate ions to form solid zinc sulfate [888). The presence of Pb and
Sn in alkaline solutions has been found to inhibit zinc dissolution as a result of the
formation of a smooth inert surface film [27). Silicate was found to prevent the precipitation of a solid film on the zinc electrode in alkaline solutions and therefore maintains a
high dissolution rate [7871.
Davydov et al. [405] measured the limiting anodic current density of the zinc
electrode in KOH solutions of different concentrations. They found that the anodic
limiting current density, im , increases with KOH concentration up to about 6M, as shown
in Fig. 2.14. The larger values of i for higher KOH concentrations are explained in terms
of a higher solubility of the anodic reaction products. Because of the dissolution, the
concentration of zincate ions at the electrode surface at the i can be much higher than
the solubility in the bulk, and thus the solution near the surface is oversaturated. Owing

ll1

ll1

32

CHAPTER 2

.------------------------------------, 10

:i:

o
c

1
'E

'"u
C

o
u

'"u
.!'!

:;

(j)

OL---------------~~--------------~

Concentration of KOH

FIGURE 2.14. Limiting current density, i lll of dissolution of a zinc disk electrode (l11 = 750 rpm) in KOH and
calculated concentrations of hydroxyl ions near the surface of the electrode, c?W, as a function of bulk solution
concentration, c~H-. After Davydov et al. [405].

to the formation of Zn(OH)~- ions, which consumes OW ions, the concentration of OW


at the surface is considerably lower than that in the bulk, as shown in Fig. 2.14.
The surface area, morphology, and other properties of the electrode surface may
change as a result of dissolution. For examples, Yamashita [112] found that the surface
of Zn becomes very smooth after 10 days' immersion in 1M ZnS04 ; Powers reported that
anodic dissolution starts preferentially at grain boundaries in 7M KOH solution [27] and
that hexagonal pits are produced on single crystals during low-potential dissolution in 7M
KOH solution [29]; and Menzies et al. [488] observed that the anodic dissolution of Zn
in sulfamic acid-formamide solution leads to polishing of the surface.
2.5.1.1. Dissolution Efficiency. In most solutions, the dissolution of zinc occurs with
a valency of2 and with almost 100% faradaic efficiency [181,532]. However, the apparent
valence may be lower than 2 in certain solutions. Figure 2.15 shows that in the absence
of NO), the anodic dissolution of zinc occurs with a valency of 2 [181]. In the presence
of NO), the dissolution is accompanied by the reduction of NOj', which leads to surface
disintegration. Because the disintegration is outside the faradaic circuit, it results in an
apparent valence lower than 2. There are two possible explanations for the lower apparent
valence values: (1) stepwise oxidation: Zn ~ Zn+ + e- (at anode) followed by Zn+ +
oxidant = Zn 2+ + reductant (in solution); and (2) anodic disintegration, generating fine
metallic zinc particles. These particles are very active and are oxidized rapidly by the
oxidant in the solution [182]. Nonfaradaic dissolution is also found in NaBr03 [1119], in
NaCIO z [34], and in ZnBrz solutions containing Brz [706].
2.5.1.2. Mechanism. The mechanism of dissolution is different for acidic and
alkaline solutions and for complexing and noncomplexing solutions. In acidic and
noncomplexing neutral solutions, the ZnJZn z+ electrode reaction appears to occur in two
consecutive charge-transfer steps [110, 786, 789]:
(2.9)

33

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

..
Q)

g
Q)

1.9

+ 0 .10 /0.30 M

c:

0.
0.

'"
~

0 .0/0.333 M
... 0.05/0.317 M

Iii
>

KN0 3 /K 2 S0 4

Concentrations

0.30 / 0.2 33 M
X O.SO /0.167 M

1.8

1.7

0.70 / 0.10 M
" 1.0

I 0.0

X 2.0/0.0 M

1.6 L---------------------~------~

0 .04

0.08

i, A I em"

FIGURE 2. I 5. Apparent valency of zinc dissolving anodically in different KNO r K2S04 solutions at 25C.
Reprinted from Johnson el al. [18 I j, with kind permission from Elsevier Science Ltd, The Boulevard. Langford
Lane, Kidlington OXS 1GB, United Kingdom.

(r.d.s. )

(2.10)

with the reaction in Eq. (2.1 0) as the rate-determining step (r.d.s.) and Zn+ as an adsorbed
and/or a solution-soluble intermediate. This dissolution mechanism gives a Tafel slope of
2.3 x 2RTl3F (40 mV).1t can be seen in Table 2.5 that the dissolution in many solutions
may follow this simple mechanism. At low overpotentials, the concentration of adsorbed
intermediate is small, and the reaction can be treated as a pseudo-one-step reaction.
However, at higher potentials the contribution from the adsorbed species becomes
significant [786]. This contribution has been measured by many investigators on solid
zinc electrodes as well as on amalgamated zinc electrodes [532].
This simple reaction scheme is also reported to occur in other electrolytes where zinc
complexes form. Hurlen and Fischer [789J found that the Zn+/Zn 2+ charge-transfer step
in concentrated acidic chloride solutions occurs between the couple
ZnCI 2 (Hpr/ZnCI 2(H 20)" but species with one or no chloride ligand take over as the
main electroactive species at chloride concentrations below 1M. The reaction scheme
represented by Eqs. (2.9) and (2.10) was also proposed by Armstrong and Bell [786] to
describe the dissolution of zinc in alkaline solutions, in which hydroxo-zinc complexes
generally form. However, the dissolutIOn mechanism becomes more complicated when
complexes are formed because more reaction steps are required to account for the
formation of these complexes.
Figure 2.16 shows that the rate-determining step for zinc dissolution depends on the
type of anions in the electrolyte. The dissolution of zinc in NaCI solution, although
following Eqs. (2.9) and (2.10), is diffusion-limited, probably because of the diffusion of
chi oro-zinc species (e.g., ZnCI~-) away from the electrode surface [110]. Armstrong and
Bell [786] found that the concentration of adsorbed intermediate species and the diffusion
of zincate ions away from the surface are important parts of the dissolution process in 1M
KOH since the dissolution current of zinc depends on the rotation rate of the electrode.

34

CHAPTER 2
0 .13 r - - - - -- - - - - - - -- - - -- - - ,
-NaCI0 4

0.12

10<.>

Na 2S0 4
..J..NaCI

0 .11

..: 0.1

~========:::;;::;:::!:::::::::::::::!:==::::::1:==r=::::I===;

0 .09

0 . 08L---------------~----------------~----~

0.05

0.1

W " i SH2

FIGURE 2.16. Rotation speed dependence of the anodic dissolution current for zinc in different molar solutions
at pH 3.0. Reprinted from Baugh rIlOl. with kind permission from Elsevier Science Ltd. The Boulevard.
Langford Lane. Kidlington OX5 1GB, United Kingdom.

Johnson and co-workers [181,182] proposed the reactions in Eqs. (2.11)-(2.13) to


explain the dissolution mechanism in neutral solutions containing various anion species
that are not reducible by zinc. In this scheme, the desorption of ZnO ads is the ratedetermining step and the Tafel slope is 2.3RT!2F (30 m V). When the solutions contain
also NO) ions, the Tafel slope appears to be 2.3RTIF (60mV), which is associated with a
cathodic reduction from NO) to NO;.
(2.11 )
(2.12)
(2.13)
Cachet and Wiart [849] proposed a reaction scheme [Eqs. (2.14)-(2.16)] for the
dissolution of zinc in de aerated ZnCl 2 and NH 4 CI solutions in which zinc complexes
form. The dissolution involves two parallel paths. The major path, Eq. (2.15), is
catalyzed by Zn;ds' The minor path, Eq. (2.14), is much more dependent on the
diffusion of the chi oro-zinc species than the major one. Both reaction paths are stimulated
by chloride anions. The formation of ZnOHadS' is a side reaction and is caused by the
chemical oxidation of zinc by the electrolyte. Deslouis et al. [700] confirmed the validity
of such a reaction scheme for zinc dissolution in aerated sulfate solutions.

I~
Zn

Zn~ds

Znads
2+ ----+ Zn 2 + + sol e

(2.14 )

Zn~;l + Zn~ds

(2.15)

+ e-

I+Zn

L--=:...

+ 2e-

(2.16)

35

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

In alkaline zincate solutions, Cachet et ai. [763, 1195] measured four loops on the
complex-plane impedance plots with decreasing frequency: (i) a capacitance loop generally highly depressed in connection with the current penetration within pores of a surface
film; (ii) an inductive loop corresponding to the presence of a monovalent intermediate
Zn+ in the reactive interface; (iii) a capacitive loop resulting from the precipitation and
escape of Zn 2+ ions by diffusion from the pore bases; and (iv) an inductive loop consequent
on the slow decrease of the pore length with increasing anodic polarization. They
postulated that there are at least four adsorbed species at the electrode surface and that
the dissolution of zinc does not occur by a series reaction. Some adsorbed species are
formed and consumed by slow reactions taking place in parallel with the main reaction
path, similar to the reactions in Eqs. (2.14) and (2.16). The active dissolution takes place
essentially at the base of pores in a layer of oxidation products whose degradation by the
anodic current can be depicted as a slowly decreasing layer thickness with increasing
anodic polarization. The zincate ions formed at the pore bases partially precipitate inside
the pores but mostly escape from the pores by a diffusion process whose rate increases
with decreasing pore length. In addition, these authors suggested that the much faster
charge transfer observed for dissolution than for deposition is due to a drastic change III
the kinetics occurring within a small potential domain passing through the equilibrium
potential.
Bockris et ai. [12] proposed a multistep reaction mechanism [Eqs. (2.17)-(2.20)]
for the dissolution of zinc in alkaline solutions, with the rate-determining step being
reaction in Eq. (2.19). A similar scheme was proposed by Muralidharan and Rajagopalan
[790], who, however, pointed out that the mechanism described by Eqs. (2.17)-(2.20) is
only valid under transient conditions. Under steady-state conditions, the rate-determining
step becomes the diffusion of zincate away from the surface rather than the charge-transfer
reaction (Eq. 2.19). Furthermore, under steady-conditions, zinc redeposition occurs
owing to the slow diffusion of zincate away from the surface.
Zn + ow ;:::::::= Zn(OH) + e-

(2.17)
(2.18)

Zn(OH) + OI-t;:::::::= Zn(OH);:


Zn(OH);: + OW

Zn(OH):; + e-

Zn(OH)3 + OH-;:::::::= Zn(OH)~-

(r.d.s.)

(2.19)
(2.20)

According to the reaction scheme proposed by Hampson et ai. [785] [Eqs. (2.21)(2.24)], the adatom surface diffusion step, Eq. (2.21), is the rate-determining step in
alkaline solutions. It is considered that the adsorbed atoms are stabilized by OH-, which
is extensively adsorbed at the zinc electrode. This mechanism implies that the reaction is
insensitive to the zinc ate concentration in the solution. The charge-transfer coefficients
appear to be about 0.9 for the anodic reaction and about 0.1 for the cathodic reaction,
indicating that the anodic and cathodic reactions are not the same. This mechanism can
be used to explain the different dissolution rates observed on zinc electrodes with different
crystallographic orientations, which have different densities of kink sites [893].

36

CHAPTER 2

Zn kink + OW ~

Zn(OH)~ds

(r.d.s~

(2.21 )
(2.22)
(2.23)
(2.24)

The exchange current density in concentrated KOH solutions (Fig. 2.13) depends
only slightly on the zincate ion concentration. This seems to be in agreement with either
the reaction mechanism given by Eqs. (2.17)-(2.20) or that given by Eqs. (2.21 )-(2.24).
The decrease of the exchange current density, io, with increasing KOH concentration from
7M to 12M can be attributed to (a) the formation of different anodic dissolution products
in the electrochemical reaction at KOH concentrations below and above 8M and (b) a
shortage of water molecules for the hydration process when the KOH concentration is
high.
It appears that zinc dissolution can follow different mechanisms depending on the
electrolyte and experimental conditions. The differences between the various proposed
mechanisms arise essentially from the differences in the final dissolution products and
their properties, which include the type and number of intermediates, their mobility, and
their state of adsorption and solvation. Each mechanism can be characterized by a set of
distinctive kinetic parameters such as the Tafel slope, reaction orders, etc. Table 2.6 lists
the theoretical values for several different reaction schemes, and Table 2.7 compares the
values obtained from different studies [790].
2.5.2. Deposition
In most cases, zinc deposition plays a negligible role in zinc corrosion for two
reasons: (a) corrosion generally occurs at a potential anodic to the reversible potential of
zinc, where the deposition is insignificant compared to the dissolution, and (b) corrosion
is usually encountered in solutions containing very little ionic zinc. Thus, zinc deposition
is only discussed here in a rather general fashion. However, much of the information
presented above (Section 2.5.1) regarding dissolution can be applied also to deposition.
More information on the subject of zinc deposition can be found in the literature [61, 683,
1158].
Zinc deposition occurs at potentials negative to the Zn reversible potential. In
aqueous solution, both zinc deposition and hydrogen evolution may occur at potentials
negative to the zinc reversible potential. Thermodynamically, hydrogen evolution is a
more favorable reaction at a cathodic potential because of its more positive reversible
potential. However, the cathodic reactions on zinc near the zinc reversible potential are
dominated by Zn deposition when the zinc concentration is higher than 10-4M. This is
attributed to the small io and a large Tafel slope for the hydrogen reaction on zinc. For
example, in alkaline solutions at potentials at which substantial zinc deposition occurs,
the hydrogen evolution current is very small, less than 10 j1A/cm 2 [12].

0.5

Znad + 20H- ~ Zn(OHh + 2eZn(OHh + 20W ~ Zn(OH)~-

Znad + OH- ~ ZnOH ad + eZnOHad + OH- ~ Zn(OHh + e-

0.75

o
40

120

120

60

Anodic

120

40

40

60

Cathodic

Tafel slope (mV/decade)

Anodic

-3

-3

-2

Cathodic

Reaction order with respect to


OW

Anodic

Cathodic

Reaction order with respect to


zincate

"Reprinted from Muralidharan and Rajagopalan [7901. with kind permission from Elsevier Science-NL. Sara Burgerhartstraat 25. 1055 KV Amsterdam. The Netherlands.

mr
mr

Zn(OH) + OW ~ Zn(OH);
Zn(OH)2" +
~ Zn(OH)3" + eZn(OH)3" +
~ Zn(OH)~-

Zn(OH) + e-

Zn + OW

0.25

Zn + OW ~ ZnOlf,;"d
ZnOlf,;"d ~ ZnOHad + e ZnOHad + OH- ~ Zn(OHh + eZn(OHh + 20W ~ Zn(OH)~-

Zn(OHh + 20W ~ Zn(OH)~-

d log CZn(OH)~-

dlogaoH-

Mechanism

0.25

dlog io

dlogi o

TABLE 2.6. Possible Mechanisms for Dissolution/Deposition of Zinca

~
Z

"
~

CI:l

::r:

\.l

38

CHAPTER 2

TABLE 2.7.

Comparison of the Results from Different Studies on the Mechanism of


DissolutionlDeposition of Zinc in Alkaline Solutions"
Muralidharan and Rajagopaland

Parameter
[

d log

io]

dpH

Hampson et al. b Bockris et al. e

Steady state

Current step

Potential step

0.2

0.14

0.10

0.66

0.2

0.67

0.65

0.3

0.27

49 12

50 10

9020

9020

113 30

175 20

20020

200 20

0.72 to 1.05

0.53 to 0.7

-0.04 to -0.21

0.0

-0.75 to -1.96

0.7 to 1.3

0.76 to 0.8

0.31 to 0.53

0.06

0.1

0.23 to 0.39

0.04 to 0.07

2.56

0.5

0.91

0.34 to 0.47

2-

cZn(OH)4

dlogio
d log CZn(OH)~-

Anodic Tafel slope


(mY/decade)
Cathodic Tafel slope
(mY/decade)
Cathodic reaction order
w.r.t. zincate
Cathodic reaction order
w.r.t. OHAnodic reaction order
w.r. t. zinc ate
Anodic reaction order
w.r.t. OH-

62 10e
320f
55 8e
28040f

to

1.06

"Reprinted from Muralidharan and Rajagopalan [790J, with kind permission from Elsevier Science-NL. Sara Burgerhartstraat
25, 1055 KY Amsterdam, The Netherlands.
h

Ref. 785.

Ref. 12.

Ref. 790.

Low overpotential.

! High overpotential.

In solutions in which Zn 2+ exists as a complex, the electrode reaction must begin with
the formation of a tetrahedral solution complex, which then undergoes consecutive
dissociations until zinc metal forms, which requires two charge-transfer steps [532], e.g.,

The reaction mechanisms for deposition of zinc near the reversible potential are generally
considered to be the opposite of those proposed for Zn dissolution [12, 786, 789, 790,
894]. In certain cases the rate-determining step for anodic dissolution is considered to be
different from that for cathodic deposition [763, 785, 887]. The Tafel slopes for deposition
in alkaline solutions are in general larger than those for dissolution, i.e., 120-300 m V
versus 20-40 mV [12, 785, 790].
Kim and lome [1120] found that the deposition of zinc in ZnCl 2 solution is the
reverse of the two-step reaction given by Eqs. (2.9) and (2.10), with a reaction order of
about one with respect to the Zn 2+ concentration. Epelboin et al. [61] postulated that
deposition of zinc in acidic sulfate, LecIanche cell, and alkaline zincate solutions depends
on the presence of Hads , Zn;ds, and other adsorbed anions and may have an autocatalytic

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

39

step, Zn 2+ + Zn;ds + e- = 2Zn;ds, in which a monovalent intermediate is involved (the


adsorption of Hads acting primarily as an inhibitor for zinc deposition). On the other hand,
according to Cachet and Wiart [683], zinc deposition in highly acidic sulfate electrolytes
is associated with an inhibition of hydrogen evolution, possibly due to the formation of
the intermediate Zn;ds. The presence of Ne+ ions has been found to destabilize the zinc
deposition process in sulfate solutions by stimulating hydrogen evolution [1030, 1251].
The morphology of the surface deposit varies with overpotential, current density,
and Zn2+concentration [62,221,324]. Smooth, dark gray porous, or dendritic Zn deposits
are formed as the overpotential changes from low to high values [62]. In alkaline solutions,
smooth deposits occur at low overpotentials with vigorous stirring; dark gray porous
deposits occur at low overpotentials 70 mV); and dendritic deposits occur at high
overpotentials (>75 m V) [221]. The transition from moss to dendrites corresponds to the
onset of mass-transport control since dendritic growth is a diffusion-controlled process
and is influenced by flow of solution, especially at lower concentrations. Dendrites initiate
at places where the local current density is high [1120]. They originate from the tips of
pyramids arising as a result of rotation of a screw dislocation. As a pyramid grows, its
radius of curvature decreases, and eventually the tip becomes a point for a spherical
diffusion [221].
Cathodic potential oscillations during zinc deposition in alkaline solutions containing zinc ions have been found to occur in the current range of 0.5 mA/cm 2 to 0.17 A/cm 2
[132]. The phenomenon has been explained as a result of the balancing effect between
deposition and diffusion of zinc ions and the competing effect of the zinc deposition and
hydrogen evolution reactions at the electrode surface. Impurities (Ni, Co, Cu, Cd, Sb, Ge,
As, Bi) in electrowinning solutions induce instability in zinc deposition and alter the
deposit morphology [1253].

2.5.3. Hydrogen Evolution


2.5.3.1. Potential of Hydrogen Electrode. The standard potential of the hydrogen
electrode, defined by the reaction in Eq. (2.25), is conventionally taken as E~ = 0 [11].
The reversible hydrogen potential in aqueous solutions depends on the hydrogen gas
pressure, PH,' and the activity of hydrogen ions, a H\ as expressed by Eq. (2.26).

W+e-~~H2(gs)
EH =

E~

- RT I2FlogpH, + RT IF-log aH>

(2.25)
(2.26)

In concentrated alkaline solutions the reversible potential at 25C can be calculated from
the equation

EH = 0.0296 log aH,O


where aH,o is the activity of water in the solution [7]. The reversible hydrogen potential
cannot be measured on a zinc electrode in aqueous solutions owing to the active nature
of zinc, which has a reversible potential much lower than that of the hydrogen electrode.
Hydrogen gas has a very low solubility in water; under a hydrogen pressure of 1 atm,
aqueous solutions contain approximately 0.8 x 1O-3MH 2 [II]. The solubility of hydrogen
is greatly decreased in concentrated electrolytes because of salting-out effects, as shown

40

CHAPTER 2
20
H2 SO4
" KOH

::: 15

.9-

E'"u

;10
:0
::>
(5

(J)

12

10

16

14

Concentration (N)

FIGURE 2.17. Hydrogen solubility as a function of electrolyte normality at 30C. After Riietschi [1146].

in Fig. 2.17 [1146]. Ions with large hydration shells are particularly effective in saltingout. Hydrogen solubility is, therefore, lower in KOH than in H2S04 or NH4 Cl. The
diffusion coefficient of hydrogen, detennined from limiting currents to a rotating platinum disk electrode, greatly decreases with increasing concentration of the electrolytes as
shown in Fig. 2.18 [1146].
2.5.3.2. Exchange Current Density and Tafel Slope. The hydrogen overpotential, Yf,
is related to the exchange current density, io, and Tafel slope, b, through the Tafel equation:
Yf

=b log ilio,

(2.27)

b=RTlaF

where a is the charge-transfer coefficient. Figure 2.19 shows the Tafel plots, measured
by Lee [7], on a zinc electrode in 6NKOH solution. A clear linear relation between current

5.------------------------------------,
4

~3
()

'"
CIl

oL-__ ____
o
2
4
~

L __ _

____

__

____

10

12

__

~L-

14

__

16

Concentrat ion (N)

FIGURE 2.18. Dependence of the hydrogen diffusion coefficient on electrolyte normality at 30C. After
Riietschi [1146].

41

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

L10

1.00
~
0

>

0.90

.J

<
;::: 0.80
z
w

~a::

070

>
0

0.60
0.50
10

0.1

0 .01

CURRENT DENSITY (ma fern l

1000

100
)

FIGURE 2.19. Hydrogen overpotential on Zn in 6N KOH. After Lee [71.

and overpotential is observed. The values of the exchange current density and Tafel slope
determined in various solutions are listed in Table 2.8.
As can be seen in Table 2.8, in most cases the Tafel slope for hydrogen evolution on
the zinc electrode has a value of about 120 m V/decade, which is also the value found for

TABLE 2.8.

Tafel Slopes and Exchange Current Densities for Hydrogen Reduction on Zinc in
Aqueous Solutions

Solution
INHCI
IN H 2 S04
INH 2 S04
H 2S04, 0.05-2N

O.IM NazS04' pH = 1-8


1M NaCI , pH = 5.8

= 5.8
=6
1M NH4c!' pH = 6
1M NH 4 CI, pH

1M (NH 4lzS0 4, pH
IN LiOH
IN NaOH
INKOH
5NKOH
6NKOH
9NKOH

9NKOH

b(mVj

232
124
120
120
120
200
120
125
174
150
120
140
160
124
145
124

log io

-10.8
-10 to -10.8
-8.9

Reference
14
14
6
II
445
110
III
33
33
311
311
311
311

-9.1
-8.2

7
311
7

01

. _0

~
<l

-10

-5

Cr

Fe

Ni

II
,

Zn Ge

30
I

V Mn Co Cu Go As

I , , , ,

25
<~

Zr

45
, ,
I

50
I

Nb Tc Rh AQ In Sb
Me Ru Pd Cd Sn Te

I,

40

I 2
5p

I 2 34

To

75

/~.

80

N
6p

Re Ir Au TI 8i
Os PI HQ PI> Po

1 23

222222112222

23456791010101010

HI

FIGURE 2.20. Values of hydrogen exchange current density, log iQ, for various metals in acid solutions. After Kita [6).

4p

I 2 45 2 2 I 2 2 2 2 I 2 2 2

,,

.:

,~

H-."'T""'1"""T'"+'-,.,-r-T""T-+-......--,-,,-

55211211012222265

Ti

Pi
v. I

(I)

(b

l
.,.i ~ ,l

4d 2 4 5 5 7810101010101010 5d

Si

Alomic number

3p

AI

51 '

fA
f

..

.;.'

.,.l-'\'
.
..

3s 2 2 2 3d 2 3 5 5 6 7 8 1010 10 10

MQ

13

~I

./ .

\ ..

~\:

~r
I
.

I.

-10

-5

.2

01

.!?

<t

(')

;>:l

ITI

-i

"'0

:c

'"

...

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

43

hydrogen reduction on many metal electrodes. The presence of chloride ions seems to result
in larger b values. The high overpotential for hydrogen reduction on zinc, compared to that
on other metals, is mainly due to the low exchange current density as shown in Fig. 2.20.
According to Brodd and Leger [532], the low values of the hydrogen exchange current density
are a result of the weak interaction between zinc and hydrogen. Because of this weak
interaction, zinc is essentially free of a chemisorbed layer of atomic hydrogen.
The periodic variation of io with the atomic number for each long period (Fig. 2.20)
indicates that the reaction kinetics for hydrogen evolution are essentially determined by
the intrinsic properties of the electrode materials. The variations of io and b with changes
in the experimental conditions reflect mainly the influence of surface condition, solution
chemistry, etc.
According to Brodd and Leger [532], who collected data from several studies, the
exchange current density of the hydrogen reaction on zinc is largely independent of pH except
in concentrated acid or alkaline solutions. as shown in Figs. 2.21 and 2.22 r116,445,532].
2.5.3.3. Reaction Mechanism. The hydrogen evolution reaction in acidic solution
can be expressed by the following equation:
(2.28)
whereas the following equation describes the reaction in alkaline solution:
(2.29)

..

-11

Ref. 135
Ref. 136
Ref. 137
'Y

-10
0

O"l

-'

_O.s'

e--e

'"

-0.4

e--I

e-

0.01

0.1

10

Acid concentration (N)

FIGURE 2.21. Values of exchange current density, io, and charge-transfer coefficient, a, for hydrogen
overpotential in sulfuric acid as a function of concentration. The reference numbers in the figure are from Bradd
and Leger [532]. Reprinted by courtesy of Marcel Dekker, Inc.

44

CHAPTER 2

---11,----------------------,

-10
Ref. 138
~ Ref. 139
_ Ref. 140

Ol

-'

-9

--0.5

.. ...
~

-.-.--.-------=-.:..-~--

- - 0.4':--:-:---------::'-:---------'::-----=-=--:--:-':-_-_ _ _,
0.01
0.1
10
Alkaline concentration (N)

FIGURE 2.22. Values of io and a for hydrogen overpotential in alkaline solutions as a function of concentration.
The reference numbers in the figure are from Brodd and Leger [532]. Reprinted by courtesy of Marcel Dekker,
Inc.

The consistency of the a value in acidic and alkaline solutions (Figs. 2.21 and 2.22)
suggests that the electric field has the same effect on the electron-transfer reactions
involving H30+ and H20. A Tafel slope of 120 mV/decade and a charge-transfer coefficient of 0.5 may indicate that the discharge reaction is the rate-determining step in
accordance with the elementary steps generally proposed for hydrogen evolution [11,
532]:
Acid

(2.30)

Alkaline

(2.31 )
(2.32)

The rate equations for hydrogen evolution on a zinc electrode can be expressed by the
Butler-Volmer equation by neglecting the back reactions because on zinc the reaction
occurs at a potential far from its equilibrium value [7]:
Acid

(2.33)

Alkaline

(2.34)

where l/J is the potential between the metal electrode and the bulk of the solution, and Kj
is the rate constant for the forward reaction. The Nernst equation for the reactions in Eqs.
(2.28) and (2.29) can be expressed by Eqs. (2.35) and (2.36), respectively.

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

45

Acid

(2.35)

Alkaline

(2.36)

The overpotentiall] is
I]

= J - Jeq

Thus,

2RT I Fin i

(2.37)

- 2RT IFln i

(2.38)

Acid

I]

= const. + RT IF In( aH,o+aH,o) -

Alkaline

I]

= const. + RT IFln(aH,oao H)

assuming a = 0.5. At a constant temperature and for a given electrolyte, Eqs. (2.37) and
(2.38) yield the Tafel equation, i.e., Eq. (2.27). A plot of I] versus log i at 25C for Eqs.
(2.37) and (2.38) displays a slope of 120 mV/decade.
Catino [311] postulated that in alkaline solutions of concentrations lower than 5N,
water reduction is controlled by the recombination of adsorbed hydrogen atoms lEq.
(2.32) rather than Eq. (2.31 )j. In more concentrated alkaline solutions (7-9N), hydrogen
reduction is promoted by the alkali metal cation, which acts as an electron bridge at low
overpotentials:
(2.39)
(r.d.s.)

(2.40)

Catino suggested that hydrogen reduction at high overpotentials follows the alkali metal
penetration mechanism described by Eqs. (2.41) and (2.42). In very concentrated solutions, the alkali metal cations may become partially dehydrated in the inner portion of the
double layer and thus force the reactant water molecule to move away from the zinc
surface [532].
(r.d.s.)

(2.41 )
(2.42)

Different processes may be involved in different electrolytes at different overpotentials. In solutions with pH values between 3 and 12, hydrogen evolution may be controlled
by nonactivation steps. For example, diffusion of protons to the surface has been found
to be the rate-determining step at low overpotentials in slightly acid solutions in the pH
range 3.5-6 [116, 110,445]. In near-neutral and slightly alkaline solutions, hydrogen
reduction is found to be affected by the formation of a surface oxide film [110, 116,445].
Powers [27, 29] found that formation of an anodic film catalyzes hydrogen evolution in
alkaline solution. Baugh [110] reported that in acidic solution of pH 3.8, hydrogen
evolution occurs via proton reduction at low overpotentials and water reduction at high

46

CHAPTER 2
1.000 r---------------------------------~--_.

100

NaCI0 4

10

... Na 254
+ NaCI

1~----~----~------~----~--~~----~

-1.6

- 1.5

-1.4

-1. 2

-1. 1

-1

FIGURE 2.23. Effect of anions on the polarization curves for zinc in molar solutions at pH 3.8. Reprinted from
Baugh [III], with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington OX5
1GB, United Kingdom.

overpotentials. Figure 2.23 shows the polarization curves measured in three different
electrolytes, indicating the effect of electrolyte composition on hydrogen evolution.
2.5.3.4. Effect of Solution and Electrode Composition. The overpotential for hydrogen reduction is strongly affected by the ions present in the solution owing to their specific
adsorption on the zinc electrodes and their interaction with water [II]. As shown in Fig.
2.23, hydrogen evolution is affected by the presence of anions, especially at low overpotentials. NH; in weak acid solutions has been found to affect hydrogen reduction by direct
reduction (Eq. 2.43) and by changing the concentration of H30+ near the surface (Eq.
2.44) [IIIJ.
(2.43)
(2.44)
The presence of Fe 2+, Cu 2+, N?+, As 3+, Sn 2+, and Sb3+ ions promotes hydrogen
evolution on zinc [10, 115, 683, 1251, 1252]. These elements have more positive
reversible potentials and lower hydrogen overpotentials than zinc, and the precipitation
of these elements on a zinc surface causes an increase of rate in hydrogen evolution. On
the other hand, Pb 2+ ions inhibit hydrogen evolution as shown in Fig. 2.24 [10, 115].
H4 PO; has a catalytic effect on hydrogen reduction [943]. The presence of oxyanions
through formation of cathodic films has been found to affect hydrogen reduction [199,
597J.
The presence of Zn 2+ in acid solutions and Zn(OH)~- ions in alkaline solutions
generally results in a reduction of the hydrogen reduction rate [10, 683, 1252]. Figure
2.25 shows the effect of Zn 2+ concentration on the hydrogen evolution rate [1252]. The
rise in current at potentials more negative than -1.04 V seE is due to zinc deposition. The
decrease in the rate of hydrogen reduction before the current rise at more negative
potentials is attributed to the adsorption of zinc ions at potentials less cathodic than

47

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS


-0.4
-0.5
-0.6

"1

"i.
~

-0.7

Cu"
... Sn"

C
Q)

-0.8

a.

-0.9

+ base curve
PO"

(5

-1
-1 .1
0 .01

0.1

10

100

(rnA I em')

FIGURE 2.24. Polarization curves for zinc in 6N KOH at 25C in the presence of 1Q-3 M Zn"+. Cu 2+. Sn 2+. or
Pb 2+. After Mansfeld and Gilman [IOj.

required for zinc deposition [683]. In alkaline solutions, the inhibition of hydrogen
evolution with the addition of ZnO is due to the formation of zincate ions. Zincate ions
lower the activity of water according to the following reaction [10]:
(2.45)

Polycrystalline and single-crystal zinc surfaces exhibit nearly the same hydrogen
overpotential characteristics [532]. However, impurities present in the zinc electrode may
change the kinetics of hydrogen reduction [9, 11,438,891]. In acid solutions the presence
of a trace amount of lead in zinc results in a significant decrease of the hydrogen evolution
[1250], Lee [9, 1121] reported (Table 2.9) that Hg in Zn decreases the exchange current

[Zn"]. gil

100

5
20
60

()

..
~

100

10

'0;

Q)

"0

:l

FIGURE 2.25. Effect of Zn 2+


concentration on voltammograms
in 200-gll H 2 S04 solutions. After
Wang et al. [1252].

0.1 L-_ _--'--_ _ _- - l L -_ _ _....J...._ _ _----J


-0.8
-1
-0.9
-1 .1

Potential (V seE)

48

CHAPTER 2

TABLE 2.9. Tafel Slopes, Charge-Transfer Coefficients. and


Exchange Current Densities for the Hydrogen Evolution Reaction
on Surfaces of Various Zn Alloys in 9N KOH"
Surface

Tafel slope

io (Alcm 2 )

Zn
Zn-2%Hg
Zn-4%Hg
Zn-8% Hg
Zn-4%Cd
Zn-8%Cd
Zn-0.2%Pb
Zn-0.8% Pb
Zn-2%Pb
Zn-0.05%Mn
Zn-0.5%Mn
Zn-2% Hg-0.2% Pb
Zn-0.8% Pb-0.05% Fe

0.124
0.116
0.098
0.086
0.158
0.154
0.137
0.134
0.172

0.48
0.51
0.60
0.69
0.37
0.37
0.43
0.44
0.34
0.43
0.42
0.48
0.48

1.5 x 10-9
2.7 x 10- 10
8x 10- 11
6 x 10- 12
7 x 10-8
1.5 x 10-8
2 x 10-9
1.3 x 10-9

0.138
0.140
0.125
0.125

6.2 x 10-8
5.1 x 10-8
7.5 x 10-8
6 x 10- 10
9 x 10- 10

"Ref. 9.

density io for hydrogen reduction in alkaline solutions and reduces the Tafel slope. Pb
causes a significant increase in the Tafel slope while having a varying effect on i o. Mn and
Cd increase the Tafel slope slightly and increase io significantly. According to Lee [9, II],
the small amounts of impurities in zinc lower the hydrogen overpotential but do not
change the mechanism ofthe hydrogen evolution processes in alkaline solutions. Alloying
with noble metals generally facilitates hydrogen evolution.

2.5.4. Oxygen Reduction


2.5.4.1. Solubility and Diffusivity. The solubility of oxygen decreases significantly
with increasing temperature as shown in Table 2.10. It is also affected by dissolved salts
in water. In water with dissolved salts up to 1000 ppm, the oxygen solubility is basically
constant [558], but it decreases significantly in concentrated solutions as shown in Fig.
2.26. The process of dissolving oxygen gas in water is, however, not efficient. The water
surface acts as a barrier to the incoming oxygen molecules. A water surface at 25C admits
only one in 6,000,000 impinging molecules of oxygen [403].
The diffusivity of O 2 in water at 25C is about 1.9 x 10-5 cm2/s [496]. It decreases
with the amount of salts dissolved in the water. For example, it is 1.24 x 10-5 cm2/s in
0.5MNazS04 solution [113]. The diffusivity of Oz in KOH solutions as a function ofKOH
concentration is shown in Fig. 2.27 [496].
2.5.4.2. Reaction Kinetics. Oxygen reduction is, apart from hydrogen evolution, the
most important cathodic reaction in the metal corrosion process. The electrode reaction
for oxygen reduction in acid solutions is [1, 1122, 1123]

Eo =1.229 V SHE

(2.46)

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

TABLE 2.10.

49

Solubilities of Air and Oxygen in Water"


Air
Percent oxygen in
dis sol ved air

a (10 3)"

Temperature (0C)

34.29
34.69
34.47
34.25
34.03
33.82
33.6

29.18
25.58
22.84
20.55
18.68
17.08
15.64
14.18
12.97
12.16
11.26
11.05

0
5
10
15
20
25
30
40
50
60
80
100

Oxygen

"

0.0489
0.0429
0.0380
0.0342
0.0310
0.0283
0.0261
0.0231
0.0209
0.0195
0.0176
0.0170

0.00695
0.00607
0.00537
0.00480
0.00434
0.00393
O.O035,!
0.()()30l
0.00266
0.()()227
0.00138
0.00000

" Data from Ref. 495.


/, Volume of gas. in milliliters, measured at DoC and 760 mm, dissolved in I ml of water when the pressure of the gas (without
the contribution of the water vapor) is 760 mm.
" Weight of gas. in grams, dissolved in 100 g of water when the pressure of the gas plus that of the water vapor is 760 mm.

and in alkaline solutions is

Eo = 0.401 V SHE

(2.47)

The reversible oxygen potential cannot be determined on a zinc electrode owing to the
active nature of zinc. The O2 reversible electrode potential and Tafel parameters can be
measured on a Pt electrode [1122]. Figure 2.28 shows that the Tafel slopes for Pt in
02-saturated 1M H2S04 solution are 93 and 126 mV/decade for reduction and oxidation,

'0
E

.!

0 .6

]fOA
:0
~
'0
Ul

0.2

KOH

ooL-------2~0--------4~O--~----6~O--------8LO-------l~OO
We ight % electro lyte

FIGURE 2.26. Solubility of 02 in KOH, H 2S04 , and H 3 P04 solutions at 25C. After Drane [558].

so

CHAPTER 2

2 .-------------------------------------~

u'"

1.5

Q)

0 .5

20

10

30

40

50

Weight % KOH

FIGURE 2.27. Oxygen diffusivity in KOH solutions at 25C. After Gubbins and Walker [4961.

1).. ------------

log apparent current density (A/cm 2)

FIGURE 2.28. Plot of the anodic and cathodic overvoltage obtained galvanostatically on a bright Pt electrode
in an Oz-saturated. 2N H 2S0 4 solution. From Hoare [1122]. Reprinted by permission of John Wiley & Sons,
Inc.

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

51

respectively, and the exchange current density is l.3 x 10-9 A/cm". The detailed reaction
mechanism for oxygen reduction is complex as reduction of one oxygen molecule
involves a four-electron charge transfer. In-depth reviews on the subject can be found in
the literature [1122,1123].
As on most other metals, the reduction of oxygen on zinc occurs in two well-defined
steps. In the first step, H20 2 is generated according to Eq. (2.48), and it is then reduced
in the second step (Eq. 2.49) [113,128,1123].
(2.48)
(2.49)
These two reactions depend on the type and concentration of anions in the solution. The
half-wave potentials for the oxygen reduction in a number of solutions have been
represented by an equation of the form [1139]
EI/2

=a -

b log C

(2.50)

in which C is the concentration, and a and b are constants. The ease of reduction of oxygen
decreases in the order sot> cr > Br- > ClOt > NO] > 1- in the concentration range
10- 2-I M. The effect of anions on oxygen reduction is attributed to surface adsorption to
act as an electron bridge for available ions [1139].
As pointed out by Tarasevich et al. [1123], the large negative free energy changes
associated with the decomposition of H"02 and HO; suggest that these species should be
very unstable in both acid and alkaline solutions. However, in the absence of impurities,
the decomposition of hydrogen peroxide is very slow in aqueous solutions. Therefore,
reduction of the peroxide (Eq. 2.49), is not complete, and various amounts of peroxide
may be produced as a result of the whole reaction process.
Boto and Williams [128] studied the electrode behavior of zinc in oxygen-saturated
sulfate solutions in the pH range between 4 and 11. They found that the reduction of
oxygen produces a mixture of H20 2 and hydroxide formed via a two-electron reduction
of the peroxide. Depending on solution composition and pH, the average number of
electrons for the oxygen reduction varies between 2.4 and 3.9. Also, according to Boto
and Williams, the reduction is controlled by different processes in different pH ranges:
In a low pH range (between 4 and 6), both reactions (2.48) and (2.49) proceed on
the zinc surface. In a higher pH range (up to 11), only reaction (2.48) proceeds
on the zinc surface because the buildup of corrosion products at high pH prevents
reaction (2.49) from occurring on the surface.
Within the pH range 4-6 at the corrosion potentials, E,o,," oxygen reduction on
zinc is diffusion-controlled. In the pH range 6-11, on the other hand, it is
controlled by the processes inside the passive film.
Wroblowa and Qaderi [797] investigated the mechanism of oxygen reduction on zinc
in O.IM K3P0 4 solutions in the pH range between 10.5 and 12.25 using a ring-disk
electrode. Figure 2.29a shows that above --0.6 VNHE the surface is covered with passive
surface films, and below -0.65 VNHE the zinc surface is bare. On the anodic potential

52

CHAPTER 2

- 0.4

0.4

0 .8

Disc potential , VNHE


FIGURE 2.29. Zinc disk and gold ring currents as a function of disk potential. Electrolyte: 1M borate buffer
+ O.IM K3P04 ; sweep rate: 0.01 Vis. (a) Disk background currents in deaerated electrolyte; (b) pre reduced
(anodic positive sweep) disk currents in oxygen-saturated electrolyte; rotation rates (rpm): (I) 700, (2) 1000,
(3) 1500, (4) 2500, (5) 3600; (c) ring currents corresponding to curves in (b); (d) disk currents; cathodic sweep;
rotation rates as in curves b; (e) ring currents corresponding to curves in (d); ring potential set at 1.08 V NHE
Reprinted from Wroblowa and Qaderi [797J, with kind permission from Elsevier Science Inc., 655 Avenue of
the Americas, New York.

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

53

sweep (Fig. 2.29b,c) the ring current is only a very small fraction of the disk current,
indicating that very few peroxyl ions are produced on the bare zinc surfaces and the
reaction proceeds primarily by a direct four-electron reduction to hydroxyl ions, i.e., via
the reaction in Eq. (2.51). On the other hand, peroxyl ions are produced on the passive
surface, both in the passive and prepassive regions (Fig. 2.29d,e).
(2.51a)
H02+2e+W~OW

(2.51b)

In oxygenated solutions of NaHS0 3, Rosales and Granese [943] observed that the
reduction rate of oxygen on a zinc electrode increases with increasing NaHS0 3 concentration. In neutral 0.5M Na2S04 solutions, according to Deslouis et ai. [113], the oxygen
reduction is diffusion-controlled through a layer of corrosion products in the vicinity of
the corrosion potential but occurs on an active surface at higher overpotentials. The film
of zinc corrosion products acts as a barrier to the diffusion of oxygen but does not directly
alter the reaction steps in the oxygen reduction.
When the thickness of the electrolyte on the electrode surface is close to or smaller
than that of the diffusion layer, the oxygen reduction rate increases significantly. Figure
2.30 shows that the reduction current greatly increases with decreasing electrolyte
thickness at thicknesses less than 100 J1m [156]. According to Rosenfeld [336] the
increased reduction current density in thinner electrolytes is due not only to the reduction
of the diffusion-layer thickness but also to the self-mixing effect in thin electrolytes
induced by evaporation and variation in surface tension and temperature. The dependence
of oxygen reduction on the thickness of the electrolyte is very important with respect to
atmospheric corrosion processes.

FIGURE 2.30. Dependence of the rate of oxygen reduction on Pt on the thickness of the electrolyte layer. 1M
Na2S04' E = -0.65 VSHE' Reprinted from Stratmann et al. [156], with kind permission from Elsevier Science
Ltd, The Boulevard, Langford Lane, Kidlington OX5 1GB, United Kingdom.

54

CHAPTER 2

2.6. CORROSION PROCESSES

2.6.1. General Considerations


Corrosion is an electrochemical process in which the surface of a metal in contact
with an electrolyte is oxidized with the simultaneous reduction of some species in the
electrolyte on the metal surface and which, over time, results in the deterioration of the
metal. Generally, a corrosion process can proceed in one of three modes, depending on
the compactness and stability of the corrosion products as shown in Fig. 2.31: (a) direct
dissolution without hindrance from corrosion products; (b) direct dissolution with hindrance from corrosion products; and (c) indirect dissolution through the formation of
passive films. A corrosion process in a given environment can involve one, two, or all
three modes and can change with time from one mode to another.
A corrosion process can be studied with various electrochemical techniques as shown
in Table 2.4. Compared to such corrosion testing methods as weight loss measurements,
electrochemical techniques are fast and can be used to obtain instantaneous information
on a corrosion process, which cannot be provided by weight loss measurements.
Among the electrochemical techniques, AC impedance technique is a particularly
useful method for studying electrode kinetics at the corrosion potential. Also, impedance
techniques, along with the linear polarization technique, are the most commonly used
methods for determining corrosion rates. In this section, the corrosion information
obtained with impedance technique is presented. The linear polarization technique will
be discussed in Chapter 5.

2.6.2. Impedance o/Corroding Electrodes


2.6.2.1. Impedance Techniques. The impedance of an electrode is one of the most
important quantities that can be measured in electrochemistry. When the impedance is
sampled over an infinite bandwidth, the impedance data contain all the information that
can be obtained from the system by purely electrical means [137]. Some of the advantages
of impedance techniques are (1) the use of very small signals which do not disturb the

Zn(OH),

2H+ - Zn"

2H,O

Zn - Zn 2 + + 2e

porous film
(a)

(b)

(e)

FIGURE 2.3\. Schematic illustration of different modes of corrosion: (a) direct dissolution without hindrance
from corrosion products; (b) direct dissolution with hindrance from corrosion products; (c) indirect dissolution
through the formation of passive films.

ELECTROCHEMICAL THERMODYNAMICS AND KINETICS

55

electrode processes being studied, (2) the possibility of studying corrosion reactions and
measuring corrosion rates in low-conductivity media, where traditional DC methods fail.
such as corrosion inside concrete or under paint, and (3) the fact that polarization
resistance as well as double-layer capacitance data can be obtained from the same
measurement.
A metal/electrolyte interface undergoing simple reduction or oxidation reactions can
be simplistically described by an electric circuit as shown in Fig. 2.32, in which R Q is the
resistance of the electrolyte, Cd is the double-layer capacitance, Ret is the charge-transfer
resistance, and Zw is the Warburg impedance, which is related to diffusion processes [133,
137]. By passing a sine-wave potential signal of small amplitude across the electrode and
measuring the AC current, one can obtain an AC impedance Z, expressed as

where j = ~ -1, and ZRe and Z,m are frequency-dependent real numbers. When Zim is plotted
against ZRe for different frequencies, one obtains the complex plane of impedance (called
a Nyquist plot). Alternatively, log IZ I and (/J can be plotted versus log w (Bode plot),
where IZ I = (Z~e + Z;m)! 12, tan (/J = -ZR/Z,nl' and w is the frequency of the AC signal.
Each plot has its advantages. The complex plane frequently is more useful for mechanistic
analysis. On the other hand, the Bode plot directly employs frequency as the independent
variable, so that a more precise comparison between experimental and calculated impedance can be made.
Figure 2.33 schematically shows the complex plane of impedance for a simple
system such as that in Fig. 2.32. From this plot. the values of the elements in the circuit
of Fig. 2.32 can be obtained, such as the solution resistance between the surface and the
reference electrode, the polarization resistance, R,), or the charge-transfer resistance, Ret.
In the case of a simple system such as that shown in Fig. 2.32, R" = Ret. The polanzation
resistance can be used to calculate corrosion current, as discussed in Chapter 5.
In cases in which pseudoinductance is measured in the low-frequency range, R" may
not be equal to Rei' The selection of R,! or RCI for use in corrosion rate calculations depends
on the circumstances since there are a number of possible sources for the pseudoinductance [719]. The polarization resistances measured by an AC impedance technique have
been found to generally agree with those obtained by a DC linear polarization technique
for many systems [718].

FIGURE 2.32. Equivalent electric circuit for a simple-charge transfer reaction at a planar electrode
surface.

56

CHAPTER 2

z""

Mass

Kinetic
control

transfer
control

/ 1

\
R + R

- - - Z-,..

FIGURE 2.33. Impedance plot for an electrochemical system. Regions of mass transfer and kinetic control are
found at low and high frequencies, respectively.

In real corrosion systems, the reaction processes are often more complicated than
that described with the circuit in Fig. 2.32, as the electrode surface may be porous or
covered with a surface coating. Different equivalent circuits are used to describe the
impedance data obtained from these systems. For example, for filmed or coated electrodes
the impedance can be expressed by a diagram of the type shown in Fig. 2.34, where Cd
is the capacitance of the intact coating layers and R, is the resistance inside pores of the
coating [134, 135]. The theoretical analysis for the impedance of different electrode/electrolyte systems can be found in the literature [133-137].
2.6.2.2. Impedance of Zinc Electrodes. The impedance techniques have been used
in a number of studies on the corrosion of zinc. Table 2.11 presents some impedance

c.

~t
R,

FIGURE 2.34. Equivalent circuit for an electrode surface covered with a solid surface film.

Dissolution and precipitation of


hydroxide

Diffusion in solid surface film

Simple dissolution

1M Na2S04' r = 2000 rpm,


pH = 10, deaerated

1M Na2S04 + 0.2M Na2P04'


r = 2000 rpm, pH = 10, deaerated

O.SM Na2S04 + O.OIM acetate,


pH =4.7, deaerated

Revealed Process

6t

1m

10

20

(ncm')

2000

4000

(nem')

Zim

(n em')

1m

60

100

10

2000

..... 0.6

20

4000

600

Spectra

0.'"

10

O.OU

(n

em')

Rc (n em')

ZR,

o.m

Rc (n em')

Impedance Spectra of the Zinc Electrode Measured at or near the Corrosion Potential in Various Solutions

Solution and conditions

TABLE 2.11.

93

702

702

(continued)

Reference

-..J

Ul

t/.l

Pi

;.::

:..
3:
Pi
t/.l
:..
Z
o

Slz

3:

~
l'

~g

=4.7, Zn2P20rtreated, de aerated

1M NaCl, pH =3.8, de aerated

O.SM Na2S04, r =600 rpm

pH

O.SM Na2S04 + O.OIM acetate,

Solution and conditions

Dissolution with diffusioncontrolled proton reduction

Dissolution and diffusion mixed


control

Dissolution under passivation

Revealed Process

TABLE 2.11.

.
10

100

200

300

0.002

100

Spectra

It

100

200

0.1

300

l00~

(n em')

Ii

In',:'
Z.

100

ZI",
(nem')

(Continued)

ZR, (n em')

ZR< (n em')

0.002

110

700

93

Reference

()

tv

;J>

:I:

=8
Multistep formation of oxide and
carbonate surface film

Surface adsorption and surface


diffusion controlled reactions

= 2000 rpm

7M KOH + O.IMZnO, r

350 ppm NaHC0 3, pH

Charge-transfer-controlled
dissolution

= 2000 rpm

3M KOH + O.IM ZnO, r

Zim

1m

...l

(n em')

(ncm'1 ~

Z""

(ncm')

20

0.1

11

em')

(n em ')

Re (n

'7Rc

.
0.1

ZRc (n em'l

(continued)

704

147

147

'"

""Zttl
-l
n

Z
'0

'";J>

;J>

-<
Z

=:::
0
'0

:;.:l

:c:
ttl

t"'
-l

;J>

ttl

:c:

:;.:l

-l

ttl

ttl

t"'

Charge-transfer-conlrolled reaclions

NH; adsorplion and reduclion +


dissoluli on

O.IM (NH4}zS04' pH = 5.9

Revealed Process

3.5% NaCi. pH '" 6.4

Solution and condilions

TABLE 2.11.

(Oem')

Zim

1000

(Oem')

1m

1000

2000

ZRc (Oeml)

ZIte (Oem')

SpeClra

l /"

(Continued)

427

70 1

70 1

Reference

tv

>

(")

:I:

Charge-transfer-controlled reactions

Charge-transfer-controlled reactions

\0% NH4C\, deaerated

1M Na2S04 + 1.5M ZnS04 + 0.0 1M NBu4Br

1000

Zim
(Oem')

1000

Zim
(Oem')

t-

1000

2000

21

ZRe (Oem')

ZRe (Oem')

843

ttl

0....

::l
n
Vl

ttl

>
Z

Vl

(5

::

>

0
-<
Z

::0

ttl

::a

>
r

(5

:r:

b;
n
;d

62

CHAPTER 2

spectra reported in the literature for the zinc electrode at or near the corrosion potential.
It can be noted that these spectra differ greatly not only in shape but also in the numerical
values, indicating that different electrochemical processes may occur on the zinc surface
depending on the nature of the electrolytes.
The nature of a corrosion process can often be revealed by an impedance spectrum.
Deslouis et al. [700], based on the impedance spectra of zinc in deaerated sodium sulfate
solutions, that the corrosion resistance is determined by a dissolution and a diffusioncontrolled process. In anodic dissolution, the first step yields an intermediate Zn;ds' The
further oxidation of Zn;ds follows two parallel paths: a major path is to form Zn;~ in the
solution, and a minor one involves the formation of Zn;;s on the surface. The overall
corrosion is a dissolution through formation of zinc hydroxide with an accompanying
diffusion-controlled cathodic process.
Cachet et al. [702] found that the presence of HPO~- in Na2S04 solution increases
the impedance of a zinc electrode. A strong inhibition of zinc dissolution occurs owing
to the formation of a protective surface layer. A Warburg impedance is measured at low
frequencies, indicating that the corrosion process is controlled by the diffusion of ions
through the phosphate passivating layer [702]. Similar diffusion-controlled processes
through a carbonate passive film have been proposed for the corrosion of a zinc electrode
in bicarbonate solution [704].
In 1M NaCl at pH 3.8, Baugh [110] found that the corrosion of a zinc electrode is a
simple charge-transfer dissolution limited by a dissolution-controlled proton reduction,
since at high frequencies the Nyquist impedance plot reduces to a semicircle and at low
frequencies a Warburg impedance develops, having a slope of 45 in the complex plane.
He also proposed that formation of an oxide film may be involved in the corrosion
processes since the double-layer capacitance is considerably smaller around the corrosion
potential than in the cathodic region.
Deslouis et al. [113] proposed an equivalent circuit for a zinc electrode in 0.5M
Na2 S04 as shown in Fig. 2.35, where Ret is the dissolution charge-transfer resistance, R Q
is the electrolyte resistance, W is the Warburg impedance, Cd is the double-layer capaci-

FIGURE 2.35. Equivalent circuit for zinc/O.5M Na2S04 interface. Reprinted from Deslouis et al. [113],
with kind permission from Elsevier Science-NL. Sara Burgerhartstraat 25, 1055 KV Amsterdam, The
Netherlands.

63

ELECTROCHEMICAL THERMODYNAMICS AND KlNETICS

7oo.------------------------------------, 7o

Eu

600

60

500

50

400

40

13
l.L

~300

30 ~

200

20

100

10

OL---~~~----~------~------~~.--J O

-1.9

-1.8

-1,7

-1 .6

-1,5

- 1.4

FIGURE 2.36, Variation of current density J, double-layer capacity C,b and surface film capacity Cr as a
function of potential in 0.5M Na2S04' Reprinted from Deslouis et ai, [1131, with kind permission from Elsevier
Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.

tance, and Cris the capacitance of the corrosion product film. Figure 2.36 shows the values
of the elements in Fig. 2.35 as a function of potential. The lower Cd value and the definite
values of Cf near the corrosion potential indicate the presence of a surface film.
The corrosion of zinc in sulfate or chloride solutions seems to involve a charge-transfercontrolled dissolution process, with a formation of a corrosion product film on the surface,
and a diffusion process through the film [113, 700, 702, 704]. When the zinc surface is
free of corrosion products or the corrosion product film is of a porous nature, the corrosion
process is controlled by charge-transfer-controlled dissolution and/or the diffusion of the
reactants for the cathodic reaction. When the surface is covered with a passive film, the
corrosion process may be controlled by a diffusion process through the film. Compared
to the dissolution at an anodic potential, at which diffusion of the dissolution products
such as Zn 2+ or Zn(OH)~- may be the rate-determining process at a large dissolution
current, the dissolution rate at the corrosion potential is seldom controlled by the diffusion
of the dissolution products because the oxidation/dissolution rate is usually very small at
the corrosion potential.

3
Passivation and Surface Film
Formation
3.1. INTRODUCTION
Passivation is a process in which the metal surface transforms from an active state
to an inactive state owing to the formation of a barrier layer. The passivation of zinc has
been the subject of numerous studies as shown in Table 3.1. It should be noted that there
is a clear difference between the studies made in strong alkaline solutions and those made
in other solutions. In general, the studies made in strong alkaline solutions are related to
battery applications, and the focus is on the maximum current prior to passivation and
the time to passivation. This focus arises because passivation is a problem in alkaline
batteries under a high discharge rate. On the other hand, studies made in neutral and
slightly alkaline solutions are generally related to corrosion, and the focus is on the
conditions and processes of passivation as well as on the stability of the passive films.
Accordingly, the material presented in this chapter is organized in two main sections;
dealing with passivation in alkaline solutions and passivation in other solutions. Prior to
these two sections, a description of the conditions and characteristics of passivation is
provided. A later section is devoted to anodization, an anodic process used to produce a
solid surface film which generally passivates the surface. The last section discusses the
stability of passivation and passivation breakdown.
3.2. CHARACTERISTICS AND CONDITIONS
Passivation can be simplistically characterized by an anodic polarization curve, as
shown in Fig. 3.1 [8, 1126]. In the active state, the metal electrode dissolves according to
the reaction Me = Me'+ + ze-, and the dissolution current increases sharply with increasing
potential. At a certain potential value, E p' the passivation potential, the current stops
increasing and starts decreasing rapidly to much lower values, marking the onset of
passivity. The current on the passivated surface, called the passivation current, iI" can be
several orders of magnitude smaller than that on an active surface at the same potential.
With further increase of the potential beyond a certain value, E b , the current may start to
sharply increase, and the electrode is said to be in a transpassive state. This sharp increase
65

66

CHAPTER 3

TABLE 3.1. Passivation Overpotentia!, 1] p' breakdown potentia!, E h , and Passivation Current
Density, ip , of a Zinc Electrode in Various Solutions
Solution
0.5M NaH2P04
0.5M NaH 2P04
0.5M NaH 2P04
O.IM Na3P04 + O.IM Na zH4
350 ppm NaHC0 3
O.OIM NaHC0 3
0.15M Na2B407 + 0.3M H3B0 3
O.IM Na2HAs04
O.IM Na2Cr04
O.IMNaCI
1M NaN0 3
0.IMNaMo04
0.2M H 3B03 + O.IM NaOH
H3B04 + NaOH
0.2M Na2HP04
1M Na2S04 + 0.2M Na2HP04
O.OOIMKOH
O.OIM NaHC0 3
1M Na2C03
3M NaCI
O.IM Na3P04 + O.IM Na2HP04
O.OIMKOH
O.OIMNaOH
H3B04 + NaOH
O.IMNaOH
0.3MNaOH
0.5MKOH
0.5MKOH
IMKOH
IMKOH
4MKOH
4MKOH
5MKOH
7MKOH
7M KOH + 0.25M ZnO
7M KOH + 0.25M ZnO

pH
4.5
6.2
6.5
7.1
8
8.1
8.4
8.9
9
9
9
9
9.2
9.2
10
10
II
11.5
11.5
11.7
12
12
12.3
12.9
13.5

14

IJ ea (V)

0
0.2
0
0.15
0.2
0.5
0.15
0.2
0
0
0
0
0.22
0.18
0.25
0
0
0.22
0
0
0.23
0
0.2
0.3
0.26
0.3
0.3
0.32
0.39
0.36
0.43
0.37
0.38
0.36
0.27
0.3

Eh (V SCE)

0.9

ie (j.1Ncm z)
0.05
200
0.2
30
28
5

2.0
1.2
-0.75
-0.76
1.5
2.1
1.2
0.6
-0.6
2.8
0.3
1.6
0.8
1.6
l.l
1.4

IO
100
I
0.2
200
200
0.8
300
I
I
50
300
200
2
I
9
2

5
100
500
600
17,000
15,000
18.000
3,000
20,000
5,000
5,000
2,500

Reference
IOI
603
IOI
481
704
194
526
21
98
45
45
98
16
355
698
702
46
127
3
3
526
46
37
355
526
19
422
1128
24
1128
794
681
1128
27
29
26

aDifference between passivation and corrosion potentials.

in current is either associated with the breakdown of the passive film, leading to a severe
dissolution of the electrode, usually localized, or with the onset of another reaction such
as oxygen evolution. When it is associated with the breakdown of the passive film, Eb is
called the breakdown potential. It is also termed the pitting potential since localized
corrosion, such as pitting, generally occurs above the breakdown potential.
Generally, passivation occurs when the dissolution of a metal produces a situation
in which the solubility of a salt or hydroxide in the electrolyte near the electrode surface
is exceeded and a compact solid film forms [1126, 1127]. As a result of the film formation,
ions must move from the metal phase into the surface film in order for further dissolution

PASSIVATION AND SURFACE FILM FORMATION

67

passive

active-passive transition

FIGURE 3.1. Schematic plot of a typical current-potential curve


showing the transition from the active to the passive state of a
metal.

current

of the metal to take place. At least three processes are involved in the dissolution on the
passivated electrode: (i) transfer of metal ions from the metal phase into the surface film;
(ii) transfer of ionic species from the solution phase into the surface film; and (iii) transfer
and hydration of metal ions across the film/solution interface. This last process is the
dissolution of the film and determines, in general, the corrosion rate of the metal in the
passive state. When there is no other reaction, such as oxidation of water, the passivation
current, ip ' equals the net corrosion rate of the metal in the passive state.
The occurrence of passivation on zinc surfaces is determined by the thennodynamic
and kinetic conditions for formation of a stable and compact solid surface film. According
to the potential-pH diagram shown in Fig. 2.2 in Chapter 2, passivation of a zinc surface
does not occur in acidic solutions without the presence of film-forming agents. In slightly
alkaline solutions containing no complexing agents with which zinc can form soluble
salts, passivation of zinc is thermodynamically possible through the formation of zinc
oxides or hydroxides. In the presence of ionic species, the possibility of passivation may
either increase as a result of the formation of a solid zinc salt layer or decrease as a result
of the formation of more soluble zinc compounds in the solution. For example, the
presence of carbonate promotes the formation of zinc carbonate in near-neutral or neutral
solutions and thus extends the pH range in which passivation is possible to lower values
compared to that for carbonate-free solutions (Fig. 2.5).
The appropriate thermodynamic conditions do not necessarily guarantee the occurrence of passivation. The actual occurrence of passivation depends also on kinetic
conditions. While thermodynamic conditions determine whether formation of stable zinc
salts is possible as a result of zinc dissolution, kinetic conditions determine the chemistry
near the electrode surface and the nature of the surface film formed. The stability,
continuity, and compactness of the film eventually determine the degree of passivation.
Depending on the conditions, passivation may occur instantly in some cases while it may
take days or months in others.
Table 3.1 presents the solution compositions and potential ranges in which the
passivation of zinc is observed. It may be noted that in some solutions passivation occurs

68

CHAPTER 3

it the open-circuit potentials whereas in others an overpotential is needed. The corrosion


potential of an electrode can be used as an indication of the state of passivation. A
corrosion potential that is much more positive than the reversible potential usually
indicates the passivation of the electrode surface. On the other hand, the occurrence of
passivation mayor may not result in a corrosion potential that is significantly more
positive than the reversible potential.
3.3. ALKALINE SOLUTIONS
Studies on passivation of zinc electrodes in alkaline solutions are mostly related to
zinc alkaline batteries [18, 24, 681, 794,889,903]. The utilization of zinc electrodes in
alkaline batteries depends on the ability of the electrode to remain active during the anodic
dissolution process. The occurrence of passivation prevents their maximum utilization.
Due to this special interest the parameters obtained from these studies are often the peak
current density before passivation and the time to passivation at a given current density.
The peak current density is generally determined from a dynamic potential-current curve
whereas the time to passivation is most often obtained from a potential-time curve.
3.3.1. i-V Curves
Figure 3.2 shows a typical anodic current-potential curve for a zinc electrode in an
alkaline solution [24]. The curve can be divided into four regions: an active dissolution
region (I), a first linear region (II), a second linear region (III), and a passive region (IV).
The current values and the limits of the regions vary with hydroxide concentration,
temperature, and hydrodynamic conditions.
The characteristics of the i- V curves depend on the potential sweep rate and
convective conditions in the electrolyte, except in region I the i- V relation is essentially

100

"""' 80
Na
u

:;;:

a 60
.

III

II

'-'

IV

<Il

i::
<l)

E
<l)

1::
;::l

40
20
0

-1.5

-1.4

-1.3

-1.2

-1.l

-1.0

Potential (Vsce)
FIGURE 3.2. Current-potential curve measured on a zinc rotating disk electrode. Conditions: IN KOH, 300
rpm, 4-m VIs sweep rate, 25C. Regions: I, initial dissolution; II, first linear region; III, second linear region;
IV, passive region. After Chang and Prentice [24].

69

PASSIVATION AND SURFACE FILM FORMATION

20 r---------------------------------,-1
First peak cuuenl

... Second peak curr.nl

-1.1

Firs. peak poteontJa]

X Second

peak potential

<- 10
.s

-1.2

~
:s!

c:
OJ

-1.3

-1.4

O L-----------------~------------~

(Sweep rate)'" (mV/s) 'A

FIGURE 3.3. EtTect of sweep rate on a stationary disk. Conditions: IN KOH, 25C. After Chang and Prentice
[24].

unaffected by the potential sweep rate. Figure 3.3 shows that the potentials of the first
and second peaks are not affected by sweep rate. However, the peak potentials are found
to be dependent on the sweep rate in some situations [19,1128].
Figure 3.3 also shows that the current densities at the peaks increase with increasing
sweep rate, the relationship becoming almost linear at higher sweep rates [24]. The lack
of a fully linear relationship between the peak currents and the square root of the sweep
rate indicates that the reaction is not totally controlled by diffusion in the electrolyte. This
is in accordance with the dependence of the current peak on the rate of electrode rotation,
shown in Fig. 3.4 [1128]. On a rotating electrode, the peak current densities are less
dependent on the potential sweep rate. At rotation rates below 600 rpm, a straight line is
200

150

E<>

.s..: 100

50

,/

(Revolution

I s)

10

'12

FIGURE 3.4. Peak current at a rotating zinc disk electrode in IN KOH as a function of the square root of the
rotation rate. After Hull et al. [1128].

70

CHAPTER 3

0 .----------------------------------,---,

0- 0 . 5
en

;f.

en

:I:
V>

:>

>

-1

-1.5

20

10

T i me , min

FIGURE 3.5. Typical potential-time curve for a zinc anode in a 7.S4M KOH solution containing 0.5M
dissolved ZnO. Current density is 40 mNcm 2. After Sato [1137].

observed, which is indicative of a mass-transfer control. With increasing rotation rate,


diffusion control in the overall reaction becomes increasingly less important.

3.3.2. Passivation Time


The time for passivation is most often determined by the use of galvanostatic
techniques [18, 25, 889, 903, 904], by which the potential is measured as a function of
time under a constant current density. Figure 3.5 shows a typical E-t curve. where the
time at which the potential rises rapidly is taken as the time to passivation, ~, [1137]. There

2,---------------------------------------,
7.24 M

"'6

T 4.98 M

1.5

...::

+ 2.92 M
. 0.784 M

;E-

....

...

Vi
c:

Q)

"0

C
~

8 0 .5

.. .. ...
0.5

1.5

-111

2.5

, S l7

FIGURE 3.6. Current density vs. reciprocal of the square root of the passivation time for upward-facing zinc
anodes in KOH solutions of various concentrations. After Liu et al. [IS].

PASSIVATION AND SURFACE FILM FORMATION

NE
u

71

1.0 M

,. 2 .0 M

...::
~

'(i;

3 .5 M

4.5 M

c:

'"

X 5 .0 M

. 7.0 M

:;

"12.8 M
2

20

10

30

40

50

FIGURE 3.7. Anodic current density vs. reciprocal of the square root of the passivation time in KOH solutions
of different concentrations at 23C. Reprinted from Dirkse and Hampson [889[. with kind permission from
Elsevier Science Ltd. The Boulevard. Langford Lane. Kidlington OX5 1GB. United Kingdom.

is generally a linear relation between current density and the reciprocal of the square root
of the passivation time, (", for current densities up to 1.5 A/cm2 , as shown in Fig. 3.6 [18].
Figure 3.7 shows that the i_t~l /2 curves are also linear for higher current densities [889].
Extrapolating the curves in Figs. 3.6 and 3.7 to the current axis, one notes that the
maximum current density attainable without passivation is rather low.
It is generally found that for a wide range of conditions, equations of the type

(i - i o)tp112 = k

(3.1 )

can be used to describe the relationship between i, the applied current density, and tp ' the
time required for passivation. In this equation, io and k are constants whose values depend
on electrolyte concentration, temperature, and convective conditions.
The form of Eq. (3.1) indicates the important role of diffusion in the passivation of
the zinc electrode [889]. In semilinear diffusion, if the original concentration in the bulk
solution is Co, after a time f the concentration, c, of the dissolution product at the electrode
surface is
C

Co

+ (2i1F)(tlnD) I 12

(3.2)

with D the diffusion coefficient and F the Faraday constant. Passivation occurs when the
solution in the vicinity of the electrode reaches its capacity limit for the dissolution
product. If C"" is the critical concentration required to cause passivation, Eq. (3.2)
becomes
(3.3)
Equation (3.3) is valid when diffusion in solution is the only mode of mass transport.
When a certain amount of convection is taken into account, Eq. (3.3) can be modified to
an equation of the form of Eq. (3.1) in which io represents the mass transport by the

72

CHAPTER 3

processes other than diffusion. Different modifications of Eq. (3.1), taking into account
oxide film growth or chemical reaction steps in addition to diffusion and convection, have
been used to describe the relationship between current density and passivation time [18,
25,889].
Increase in temperature appears to prolong the passivation time. Measurement of
dissolution rate as a function of temperature indicates that the passivation process on zinc
electrodes in alkaline solutions is under mixed control of diffusion and activation
processes [889]. The effective activation energy was reported to be in the range of
27.2-41.9 kllmol, which is higher than the typical activation energy for a diffusioncontrolled process, about 12.6 kllmol, and is lower than that for an activation-controlled
process, about 41.9 kllmol [24].
The orientation of the electrode surface is found to affect the passivation time.
Different passivation times are measured for upward-facing, downward-facing, and
vertical electrodes [18, 26, 1130]. Longer passivation times are required for vertical
electrodes than for horizontal ones owing to the onset of natural convection [26]. The
thickness of the diffusion layer as a function of electrode orientation has been studied by
O'Brien et aZ. [1138].
Dirkse and co-workers [886, 902] studied the effect of ionic strength on passivation
time. They found that for binary KOH-HzO mixtures the passivation time at a given
current density increases with increasing KOH concentration at a relatively low ionic
concentration, while it decreases with increasing KOH concentration at a high ionic
concentration. As shown in Fig. 3.8, the slope of the i versus t;1 IZ curve increases with
KOH concentration up to 7-8M and then decreases with further increase of KOH
concentration [889]. This phenomenon was explained in terms of loss of the unbound
water as the ionic strength of the solution increases.
The presence of zincate ions in the solution seems to shorten the passivation time
[18]. Figure 3.9 shows that the slope of the i versus t;112 curve decreases with increasing
zincate concentration [1131]. According to Hampson et aZ. [1131], passivation occurs
when the concentration of zincate ions at the anode is equivalent to that of OH-.
The presence of other ionic species also affects passivation time. In one study,
passivation time was found to decrease linearly with increasing carbonate concentration

-",

O.B

:J

:;
"00.4
Q)

a.
o
iii

oOL-..- -2-'---.......I..4 -

-...l.6--....Ja'--- -,l..o- -'.L.2---.J'4

Concentration of KOH (M)

FIGURE 3.8. Slopes of the plots of


current density vs. reciprocal of the
square root of the passivation time for
KOH solutions of different concentrations. Reprinted from Dirkse and
Hampson [889], with kind permission
from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington OX5
1GB, United Kingdom.

PASSIVATION AND SURFACE FILM FORMATION

73

1.35

0.95L---------------~--------------~------~

0.5

Zn 2 + Concentration, M

FIGURE 3.9. Variation of k. the slope of the i vs. t;1 12 curve, with Zn 2+ concentration in 7M KOH for horizontal
zinc anodes at 23C. After Hampson [1131 J.

up to 2M in 7M KOH solution; this decrease was attributed to the increase in the viscosity
of the solution [1137]. In another study, it was found that Pb and Sn added to the
electrolyte give rise to a smooth and compact metallic film over the Zn surface and thus
increase the degree of passivation [27].

3.3.3. Characteristics
The amount of surface film needed for passivation to occur can be very small. Hull
et al. [1128] measured i- V curves using a rotating ring-disk electrode. They found that,
as shown in Fig. 3.10, the i- V curves observed with both ring and disk electrodes are, in

b
-to

+4.0

...

...~

:l
U

:l

Vl

Ci

-o.~

bll

c:

+2.0

C2

-1.0

Disc potential, V (Hg/HgO)

-1.0

Disc potential, V (Hg/HgO)

FIGURE 3.10. (a) Current-voltage curves recorded on the disk of a graphite-zinc ring-disk electrode at
different rotation rates. Scan rate = 10 mV/s; IN KOH; electrode area = 0.035 cm 2 . (b) Reduction current
observed at the graphite ring held at a fixed potential of -1.45 V (Hg-HgO) for the current-voltage curves
shown in (a). After Hull et al. [1128].

74

CHAPTER 3

600.0

Thick silver
colored film
Film darkens
Brigh.ly
e.chod

200.0

Film beeomes
G brigh.

BI.ckfilm_

0.0 L....,6"--_'---_ _ _...I...-_ _ _-'J'--'-H'--_~


- 1.40

-1.28

-1.16

- 1.04

Electrode Potential vs Hg-HgO (Volt)

-0 .92

FIGURE 3.11. Current-voltage


curve of a zinc wire electrode and the
corresponding changes of the electrode surface in unstirred 5N KOH at
a potential scan rate of 1.1 mVIs. After
Hull et at. [1128}.

general, of almost identical forms. The collection efficiency on the ring electrode is
constant for all the regions of the i- V curves, indicating that the fraction of the current
that is utilized in the formation of the surface films on the zinc disk at either the peak
potential or during the onset of passivity is so small that there is no detectable change in
the amount of dissolved species arriving at the ring electrode. Thus, the majority of the
current produced at the disk for all potential regions is utilized in direct electrochemical
oxidation of the zinc to soluble products.
The physical appearance of the zinc electrode surface during anodic polarization in
alkaline solutions varies with the potential. Figure 3.11 summarizes the changes in the
appearance of the zinc electrode surface visually observed by Hull et al. [1128] during
cyclic anodic polarization in an alkaline solution. At potentials less anodic than that at
point B, the surface of the electrode remains bright and the development of etch patterns
is clearly visible. At point B, a milky white film forms, which, in the absence of stirring,
slowly flakes from the electrode surface in the region C-D. At point D, as the current
begins to fall, the color of the surface beneath this film can be observed to darken until a
region of intense current oscillation is reached. The passivated surface at F is very black;
however, the color rapidly lightens as the potential increases further, leaving the surface
covered with a white crystalline passivating layer. The oscillation at the onset of passivation is attributed to the changes in the pH of an electrolyte layer adjacent to the surface
as a result of the oxide formation and diffusion [1128]. The oscillation has also been
explained as a result of the distribution between IR potential drop in the electrolyte and
the potential across the double layer [904].
Aurian-BIajeni and Tomkiewicz [20] studied the impedance spectra of zinc electrodes in alkaline solutions and concluded that the anodic passive films are composite
layers of oxides and solution. The growth of the layers and the passivity are dictated by
the diffusion of the electrolyte across these layers. The passive layer becomes more and
more compact, with grains approaching a spherical shape. The layer thickness starts to
increase after a certain porosity and conductivity are reached. It was found that the
dielectric properties of the oxide-solution composite layer were not the linear combination of the dielectric characteristics of the oxide and the solution [1134].

PASSIVATION AND SURFACE FILM FORMATION

75

The passive films formed on zinc electrodes in alkaline solutions have often been
found to have semiconducting properties [422,484,526,687]. For example, Scholl et al.
[422] studied the photocurrent spectra of the films formed in 0.5 and 1M KOH solutions
under various conditions and found that the films exhibited properties characteristic of
an amorphous semiconductor.
It is important to note that the oxide films formed in alkaline solutions are not "truly"
passive for there is a significant current in the passive potential region. For example as
reported by Dirkse [904], the steady-state dissolution current density in the passive region
in 10-40% KOH is in the range of 19-33 mA/cm 2. Also, as can be noted in Table 3.1,
the passivation current densities in concentrated KOH solutions, e.g. 1-7M, can range
from 2.5 to 20 mA/cm 2.

3.3.4. Mechanisms of Formation of Passive Films


The detailed mechanisms of zinc passivation in alkaline solution under specific
conditions are complex and are still not fully understood. However, in simplistic terms,
as Dirkse [904] has summarized, the passivation generally proceeds by the following
steps:
1. Zinc is oxidized to form ZnO or Zn(OH)2'
2. They dissolve in the electrolyte to form Zn(OH):1 or Zn(OH)~-.
3. When the electrolyte can no longer dissolve the ZnO or Zn(OHh that is produced
by the charge-transfer reaction, a solid film forms and passivates the surface.
Powers and Breiter [26, 27, 29] studied the change in the appearance of the surface
of zinc electrodes in 7M KOH under a microscope during anodic polarization. Two
different films, both zinc oxides, were found to form under different conditions. Type I
is white and porous and forms by precipitation from a supersaturated layer of electrolyte
covering the electrode. Type II forms directly on the electrode surface and ranges in color
from light gray to black. The dark color of the type II film is due to the excess of zinc in
the film. Type I film precipitates near the potential of the first peak of the i- Vcurve. Type
II film is more coherent and skinlike than Type I film and forms directly on the surface
beneath the type I film at a slightly higher potential than that of the first peak. The
characteristics of the two types of films are summarized in Fig. 3.12. Absence of
convection seems to be important for the formation of type I film; in the presence of
convection, only type II film is observed to form [27].
The formation of type II film is considered to be responsible for the transition
from the active to the passive state of zinc in alkaline solutions. Cobweblike structures
are formed when type II film dissolves due to the gathering together of the excess
zinc in this film [29]. The cobwebs are electrically conductive and can be further
oxidized. Type II film also appears to serve as a catalyst for hydrogen evolution at
potentials anodic to the zinc/zinc oxide equilibrium value. The formation of hydrogen
bubbles can mechanically dislodge the passivating film and cause reactivation of the
electrode [27].
The duplex nature of the oxide film formed on a zinc electrode surface was also
reported by Szpak and Gabriel [28]. Depending on the conditions, the anodic oxides, on
the micron scale, have either a carpetlike, a boulderlike, or a thistlelike structure. Figure

76

CHAPTER 3

With convection nearly absent

1.

While. Oocculem, IYpe 1 film.


formed by precipitalion from a
. / supersaluraled layer of eleclrolyle, ZnO

Coherenl, strongly lighlabsorbing,


direcllyformed lype II fiLm, ZnO
with excess of Zn
Zn eleclrode

2.

In presence of convection
~_ _

Type II film
Zn electrode

FIGURE 3.12. Schematic illustration of the films formed on zinc electrodes in 7M KOH under different
convection conditions. After Powers [27].

(b)

(a)

Adsorption

(c)

Adsorption

Transport

QznloHIV

Adsorption

Transport

iZn'OH'/

ad.y

,
/II
~Zn!oHll
l~'
znloHr;

OZnIOHI;:

_-Q0H

M~o..:
I

I :

Distance

Polymerization

(d)

(f)

(e)

Adsorption

Transport

/1

Adsorption

Transport

I
I

I
I

!\
I

Nucleation and growth

I
I
I
I

I
I
I

Oxide densification

Oxide film folding

FIGURE 3.13. Schematic representation of the stages in the development of an anodic ZnO film: (a)
quasiequilibrium state; (b) development of transport region; (c) formation of polymerization region by trapping
of monomers; (d) nucleation and growth region ; (e) oxide densification; (f) oxide film folding. After Szpak and
Gabriel [28].

77

PASSIVATION AND SURFACE FILM FORMATION


-0.80 ,--------,.---------------r-,....-,

Zn + 4 ORZn(OH)/ + 2e

~ -1.00-

00

::c:

Zn + 40R -> Zn(OH)/ + 2e


Zn(OH)/ -> ZnO + 20R + H20
Growth of type I ZnO

is -1.20
.!:!

8<1.l
-g
l:l

Zn + 20H-

->

ZnO + H2 0 + 2e :

Growth of type II ZnO

--r :
I

Lilil -1.40 L - -_ _ _--'-_ _ _ _ _ _ _ _ _ _ _ _.....L---Jw

r- t.

- - 1 - - - - - tb

-----+-1 tel

Time

FIGURE 3.14. Proposed scheme for the processes associated with the anodic passivation of zinc in alkaline
solutions. After Liu et at. 1181.

3.13 schematically illustrates the stages in the formation of the oxide film. The nucleation
is considered to be associated with the saturation of Zn(OH)3 monomers and to be
completed within milliseconds. The growth of the nuclei to an observable size, e.g., to a
radius of 10-5 cm, may take a much longer time depending on the conditions.
Liu et al. [18] proposed a multistep reaction process for the passivation of zinc in
alkaline solutions as shown in Fig. 3.14. In the first step, the anodic dissolution proceeds
for a time, t", producing zincate ions which accumulate near the surface. When a critical
concentration, Ceril , which may be several times the solubility of ZnO in KOH solution,
is reached, type I ZnO begins to precipitate. The anodic dissolution continues through the
porous oxide film up to a time th at which the rate of mass transfer of hydroxide ions
through the film falls below that required for the formation of zincate ions and formation
of type II zinc oxide is initiated. After an additional time, te , the whole surface is covered
with type II oxide and becomes passivated.
Cabot et al. [681,794] described the formation of a passive film under a potentiostatic
condition as consisting of two stages. In the first stage, the process is controlled by
diffusion of zincate ions in the solution near the surface. In the second stage, during the
film growth the process is controlled by ion migration through the pores in the film. The
growth tends to be two-dimensional at low solution concentrations and three-dimensional
at high concentrations. According to Cabot et al., a porous solid precipitate, possibly
Zn(OH)2' might already form in the linear region before the current peak on an anodic
i- V curve, because a linear region on an anodic i-V curve implies solution resistance
control in the pores of the hydroxide film.
3.4. OTHER SOLUTIONS

3.4.1. Slightly Alkaline and Carbonate Solutions


The typical behavior of a zinc electrode in carbonate solution can be seen in Fig.
3.15, which shows a potential-time curve obtained by Kaesche [127] for a zinc electrode

78

CHAPTER 3
- charging curve
- - -discharging curve

c
........... oxygen evolution (?)
I
I

--- growth of oxide layer

I
..

F
oxide reduction
\,--- --.(, G

paSSivatIOn
_ hydrogen evolution

Time

hydrogen evolution
..... __ [ __ _

FIGURE 3.15. Potential-time curve for zinc in 1M


Na2C03 solution, 60C, during galvanostatic anodic
charging and cathodic discharging. Reprinted from
Kaesche [1271. with kind permission from Elsevier
Science Ltd. The Boulevard. Langford Lane.
Kidlington OX5 1GB. United Kingdom.

in 1M Na2C03 solution at 60C. At point A, the anodic charging is switched on from a


preset cathodic potential where hydrogen evolution has been occurring. In the time
between points A and B, the electrode is passivated, followed by oxide growth between
points Band C, with superimposed oxygen evolution beyond point C. If the current is
reversed during charging, then between points F and G the oxide is reduced, and the
potential finally returns to the exclusively hydrogen-evolution potential. The passivation
time, tp ' corresponding to point B, is found to be a logarithmic linear function of the
charging current density, i a , with a slope of about -1.6. This relation indicates that the
passivation in the carbonate solution has a constant value of the product iJ;12 , differing
significantly from that in concentrated alkaline solutions, where ii ~ 12 is generally a
constant.
In Kaesche's experiments in 1M Na2C03, the passivation time increases with
increasing temperature from about 0.1 s at 20C to about 0.5 s at 80C at a charging current
density of 10 mA/cm2 [127]. For a tp value of about 0.1 s at room temperature, the product
of ii p is about 1 mC/cm 2, corresponding to approximately an oxide monolayer. Apparently, the thickness of the passive film at passivation increases with increasing temperature. The current efficiency in 1M Na2C03 at room temperature is nearly 100% since the
ratio icfJiia, with ic and tr the current density and the time for the reduction of the oxide
film, is near unity. At a higher temperature the ratio is lower than 1, as shown, for example,
in Fig. 3.16 for a temperature of 60C [127]. The decrease in current efficiency observed
with longer charging times is attributed to oxygen evolution.
The passivation characteristics vary with the concentrations of carbonate and bicarbonate in the solution. In 1M Na2C03 solution, the passivation is fast and brought about
by a small amount of oxide, about a monolayer [127]. On the other hand, in O.OIM
NaHC0 3 solution, no passivation occurs after 30 min of polarization at 25C, but
passivation occurs readily at higher temperatures. In O.IM NaHC0 3, the passivation is
very slow and is associated with the formation of a thick oxide film. At higher temperatures, this film becomes thinner and has a larger number of nuclei and thus requires a
shorter time to reach passivation. According to Kaesche [127], the passivation process in
O.lM NaHC0 3 is quite different from that in 1M Na2C03 in the following respects: (a)
longer passivation time by an order of magnitude; (b) less than 20% current efficiency;
(c) formation of white oxide flakes on the surface; (d) formation of a larger amount of
oxide, about 300 mC/cm2, at passivation; (e) reduction of passivation time with increasing
temperature, from 24 min at 40C to less than 1 min at 90C; and (f) larger oxide nuclei.

79

PASSIVATION AND SURFACE FILM FORMATION

..... . .

::0.8

::>;

0.6

...

.~

o
c

Q)

0.4

Q)

'E
Q)

5 0.2

oL--------------L--------------~--------~

10

0.1

Amount of charge, me I cm 2

FIGURE 3.16. The ratio i,t"!Va as a function of the anodic charge ie/a' measured for zinc in 1M Na2C01 at
60C. I i" I = I ic I = 10 mAlcm 2 Reprinted from Kaesche r127]. with kind permission from Elsevier Science
Ltd. The Boulevard. Langford Lane. Kidlington OX5 1GB. United Kingdom.

According to Muralidharan and Rajagopalan [206], in O.OIM NaOH solution zinc can be
passivated with the formation of three monolayers of Zn(OHh at the active centers. The
formation of these monolayers is two-dimensional with an instantaneous nucleation.
D' Alkaine and da Cunha [23] found that the peak current density of the anodic i- V
curve is linearly related to the peak potential in carbonate solutions, indicating an
ohmic-controlled process in the oxide film:
(3.4 )

where p is the ionic resistivity, and II' is the film thickness at the peak potential. This
resistance is found to decrease with increasing carbonate concentration as shown in Fig.
100,-----------------------------------------,

a.

0..

oL---------------~--------------

__

_ L _ _~

Na 2C0 3 Concentration (M)


FIGURE 3.17. Representation of oxide film resistance (pip) versus Na2C03 concentration. After D' Alkaine
and da Cunha [231.

80

CHAPTER 3

3.17. D' Alkaine and da Cunha reasoned that since the film thickness changes little with
concentration, the decrease in the resistance with increasing concentration shown in Fig.
3.17 may be attributed to changes in the resistivity of the film.
Kannangara and Conway [3] conducted a detailed study of the passivation of zinc
in alkaline and carbonate solutions. Figure 3.1S shows the cyclic voltammograms for Zn
in 1M Na2C03 and in 3M NaCI solutions at pH 11.5. The passive potential range in the
carbonate solution is significantly larger than that in the chloride solution. The major
features of the i- V curves are identified as follows:
Peak Al consists of the dissolution current and formation current for ZnO or
Zn(OH)2'
Peak Al consists of Ala and Alb' with Ala due to the dissolution of surface defects
(grain boundaries, etc.), which is diffusion-controlled, and Alb due to the fonnation of a compact layer of ZnO or Zn(OH)2, which is controlled by a surface
mechanism. The fonnation of this compact oxide layer, corresponding to only
one or two monolayers, passivates the Zn surface.
Peak C I is the reduction current for ZnO or Zn(OHh.
The slope of the current density of peak Al versus the square root of the potential
sweep rate, about 0.7, was explained by Kannangara and Conway [3] as being due to a
mixed process of diffusion-controlled dissolution and film formation. The dissolution
process (with a peak current density proportional to the square root of the sweep rate)
proceeds simultaneously with the film fonnation process (with a peak current density
proportional to the sweep rate) until the surface is completely covered by the passive oxide
film. The amount of dissolution during the passivation is described by qJ qc, the ratio of
the charge under peak Al to that under peak C I . This ratio decreases with decreasing
solubility of the oxide and with increasing potential sweep rate as shown in Fig. 3.19.
Similar parallel dissolution and film fonnation processes were proposed by Rangel
and Cruz [704] for a 350 ppm bicarbonate solution of pH S.O. The dissolution of zinc
produces an adherent film with OH- ions diffusing from the bulk into the porous film
against a flux of zinc species. They measured the current peak as a function of temperature
and obtained an apparent activation energy of IS.4 kJ/mol, which is in the range for a
diffusion-controlled process. The dissolution mechanism was considered to follow the
reactions given by Eqs. (2.14)-(2.16) in Chapter 2.
Using X-ray diffraction, Huber [S02] identified the anodic films formed in hydroxide
and carbonate solutions. The oxide film formed in Na2CO, is thin and light in color and
is insulating, whereas that fonned in NaOH is of y-zinc hydroxide with a small amount
of oxide and is thick and dark. The dark color is due to the presence of an excess of metallic
zinc. ZnO has also been identified in other studies as the passive film in carbonate
solutions [3]. The ionic resistance of the passive film has been found to decrease with
increasing concentration of CO~- [23].

3.4.2. Phosphate Solutions


The passivation of zinc in phosphate solutions is of particular importance in the
surface treatment of zinc and its alloys. Phosphating, an immersion or spray process by

-8

-4

-1

c,

o
2

0, evolution

A:-/

E (Vscd

-4

-2

- 0.8

C,

11--

- 0.6

reduction of ZnO

reduction of Zn + +

dissolution and
film formation

- 0.4

FIGURE 3.18. Typical cyclic voltammograms for polarization of polycrystalline Zn in 1M Na2COj (a) and 3M NaCI solution (b) at pH
11.5. Potential sweep rate = 50 mY/s oThe electrode is polycrystalline, of apparent area 0.07 cm 2, etched in HC!. After Kannangara and
Conway 131.

(,)

......

ME

00

(5

!::

'"

.."

!::

m
::::::
r

'";;;;
n

[/)

>Z
o

(5

~'-l

[/)
[/)

82

CHAPTER 3

15

pH 13
T

pH 12.7

+ pH

11.5

200

300

Sweep rate (mV s" )

FIGURE 3.19. Values of q,,/qc from cyclic voltammograms for Zn oxidation and redeposition, as a function
of potential sweep rate , S, for 1M Na2S04 solutions. After Kannangara and Conway [3J.

which an insulating zinc phosphate coating is formed on the zinc surface, is commonly
used in the steel industry to enhance paint adhesion on zinc-coated steels.
Passivation of zinc in phosphate solutions can occur at the open-circuit potential in
a wide pH range [101,481, 784]. The open-circuit electrode potentials were found to shift
to more positive values, a sign for the occurrence of passivation [784].
De Pauli et al. [481] investigated the i- V characteristics of the zinc electrode in O.lM
Na3P04 + O.lM Na2HP04 in the pH range between 6.5 and 13.1. Figure 3.20 shows that
the shape of the i- V curve strongly depends on pH. The peak current density before the
passivation region is a function of potential sweep rate and, when plotted against the pH.
has a minimum around pH 9,5, as shown in Fig. 3.2l. In the higher pH range. the peak
current density has a linear relation with the square root of potential sweeprate, S. which
is similar to the behavior in alkaline solutions, In the lower pH range the peak current
density has no linear dependence on 5 112 or 5, indicating a different passivation mechanism, Also, it was suggested that at pH > 12 a dissolution-precipitation mechanism
operates whereas at pH < 12 a solid-phase process prevails,
Furthermore, different mechanisms may operate in different temperature ranges . At
a temperature near ooe, the peak current is independent of the electrode rotation speed,
whereas at higher temperatures it depends linearly on the square root of the rotation speed
[603]. According to De Pauli et al. [481, 698], PO~- ions promote zinc dissolution since
the peak current density, i lll , increases with increasing PO~- concentration; the capacity
ofPO~- ions to provide OH- ions at the interface was suggested as a probable explanation
for this effect.
The passive film can be a monolayer or a multilayer film, depending on the
concentration of phosphate and the temperature of the solution. Awad and Kamel [784]
attributed the passivation in phosphate solutions to the formation of a highly polymerized
zinc phosphate layer on the electrode surface. According to De Pauli et ai, [603], the
passivation operates through a dissolution-precipitation mechanism with the participa-

83

PASSIVATION AND SURFACE FILM FORMATION

pH
- - - -13.1

- - 9.1
0.9 -'-7.1

.,

0.6

f .

C
.,

0.3

0.6

1.2

\.

\
\

J. ".\ I.

.1

/ ,'i!

./ \

r.
/

-0.9

I,

0.3

12<

./

\ I

\ r

./

J'

-Ul

Y
.1.5

1.2

0,9

Potential (V"J
FIGURE 3.20. Current-voltage relationship for a stationary Zn electrode in buffered phosphate solutions of
different pHs. Potential sweep rate 0.05 Vis. Reprinted from De Pauli et al. [481], with kind permission from
Elsevier Science-NL Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands .

3 .5

0.1 VIs
y 0,05 VIs

+
:;c
,3.

0.025 VIs

Ol

.2

2.5

24L-----~6------~8------~
10-------12------~14~~

pH

FIGURE 3.21. Dependence of log it" on pH. [pol-] = 0.2M. Reprinted from De Pauli et al. [4811. with kind
permission from Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.

84

CHAPTER 3

,.-...

"'E

.Q
.;:;:;

.,

;:

-1

0>

I:
=>

-2

- 1.8

- 1.5

- 1.2

-0.9

-0.6

-0.3

Potential (V ...)
FIGURE 3.22. Effect of potential reversal at different anodic potentials on the reduction peaks. S = 0.05 Vis.
rpO~-l =0.2M, pH = lO.9. After De Pauli et al. [4811.

tion of phosphate species in the solution and Zn 2+, which diffuses through the thin
nonpassivating film, according to the following scheme:
(3 .5)
(3.6)

Spherical nodules with high phosphorus content can be formed in the multilayer films .
The formation of the nodules was considered to be a result of repassivation at the places
where breakdown of the passive film has occurred [603, 702].
Passive films may have different phases at different potentials. Figure 3.22 shows
that the positions and the number of cathodic peaks change when the potential range of
anodic polarization is increased, indicating that phase transformations of the passive film
may be involved during the anodic polarization [481]. This phase transformation in the
passive region was postulated by De Pauli et al. [481] to involve a slow chemical reaction
of the form
(3.7)

The dissolution in the passive region appears to be associated with the diffusion of
ions through the passive layer [702].

3.4.3. Miscellaneous Solutions


Boron compounds are often added as a pH buffering agent in the solutions used for
passivation studies on zinc electrodes. However, it has been found that, in addition to their
buffering effect, these compounds also participate directly in the passivation process. Zinc
becomes passivated very easily in boric acid-sodium hydroxide solutions in the pH range
of 9.2-12.3 [355], and the amount of charge needed for the passivation is many times less

85

PASSIVATION AND SURFACE FILM FORMATION

1.0

.'I
\

,,

_ . - 1M crO,"
-IMMoO/
- - -- I \VO.'"

--1.0
0.01

- -

- - - - -

________

- -- - - - -

________

- -

- -- ,

________
-w

0.1

10

Current density, rnA/em'


FIGURE 3.23. Effect of oxyanions on passivity in aerated 1M solutions at pH 9 and 40C. After Bijimi and
Gabe [98].

than that needed in alkaline solutions of similar pH values [907]. The passivation may be
attributed to the formation at nonsoluble zinc borate salts [16, 17] or to the buffering effect
of borate ions on the electrolyte near the electrode surface [45].
Pimat et ai. [482] studied the passivation of zinc in chromate solutions at pH 1.5.
Zinc exhibits a passive behavior in the potential range from -0.9 to 0.05 VseE' Passivation
cannot occur when sulfate is also present because of the competitive adsorption of sulfate
on the electrode surface. Bijimi and Gabe [98] found that the passivation current density
of zinc in a chromate solution increases with temperature. Compared to other oxyanions,
chromate is the most effective in passivating the zinc surface, as shown in Fig. 3.23.
Aeration or deaeration has no effect on the passivation of zinc in solutions of oxyanions.
According to Macias et ai. [175,202], the passivation of zinc in Ca(OHh-containing
solutions is due to the complete coverage of the surface with a compact layer of
Ca(Zn(OHh)22H20, the formation of which is determined by the concentration of Ca2+
ions. The passivation results in a decrease in corrosion current to about 0.5-1 /lA/cm"
and a shift of the corrosion potential to about -0.5 VSCE'
De Pinto et ai. [21] studied the passivation of zinc in arsenate solutions and found
that arsenate ions increase the dissolution rate in the active potential region but tend to
form an insoluble compound with zinc ions to passivate the zinc surface. Thick anodic
films can be formed in arsenate solutions. Cracks were found to develop in the solid
surface film, more in the solutions with dissolved 02' Passivation can also occur in
chloride, sulfate, and other solutions under anodic conditions within certain pH ranges,
generally due to the surface saturation of the zinc salts [532].
3.5. ANODIZATION
Anodization is a process used to produce a solid surface film of a certain thickness
and properties. Anodic coatings of various colors, from white to gray to black, can be
produced in aqueous solutions of sodium hydroxide and sodium carbonate. Figure 3.24
shows that, depending on the current density, a white or a black film can be formed on a
zinc surface through anodization in a solution of NaOH and Na 2C0 3 [494].

86

CHAPTER 3

>

200 mA I cm 2

:g

cCl>

" 100 mA I cm 2

+ 50 mA l cm 2

(5

0Cl>
"'0

30 mA I cm 2

- - - Wh ite films
- - - - - Black films
_ . __

_ __ _

2 3 4
Time , m in

FIGURE 3.24. Potential-time curves of zinc electrodes anodized at various current densities in O.146N NaOH
and O.054N Na2C03' After Whitaker and Fry [494].

Black or dark-colored coatings can be produced by anodization in NaOH solutions


with pH values greater than 13.3 at current densities of 70-140 mA/cm 2 Gray to white
oxide coatings can be produced in alkaline solutions having a pH equivalent to that of
O.OOl-O.IN NaOH solutions. White or light-colored coatings can also be produced in
Na 2C0 3 solutions. The coatings produced in Na2C03 solutions are less porous and are 10
to 100 times thinner than those produced in NaOH [493]. The conditions fortheformation
of a black oxide film in NaOH solution are shown in Fig. 3.25. The black color is attributed
to metallic zinc particles dispersed in the film, resulting from the reactions
(3.8)
(3.9)
When zinc is anodized in NaOH solutions, first a thin film composed of very tine
crystals of oxide (50-150 A in size) is formed. Upon prolonged anodization, this primary
film adjacent to the metal becomes thicker, and a porous layer is formed [404]. The anodic
films formed in alkaline and carbonate solutions have been found to consist primarily of
zinc oxide. The oxide is converted to zinc hydroxide with difficulty and is practically
insoluble in pure water. However, if some carbon dioxide is added to the water, the oxide
layer is rapidly converted to basic carbonate [493].
The black oxide films prepared from anodization in NaOH and Na2C03 solutions
appeared to have a high absorption for wavelengths shorter than 2.2 pm and a high
transmission for wavelengths longer than 2.2 pm [1133]. The anodic coatings obtained
in Na3P0 4 solutions primarily consist of zinc phosphates [96]. MuItilayers are formed on
zinc electrodes in slightly acidic solutions in the presence of NaH 2P0 4 . The films are
passivating and contain ZnO and inclusions of phosphate; the amount of inclusions
increases with temperature [603]. The anodically produced films in chromate-containing

87

PASSIVATION AND SURFACE FILM FORMATION

Compact black oxide layer


1-

_ _ _ __

Thin layer of porous


black oxide
UJ

:t

>'"
W

-0.5
Black oxide layer

Thin passive layer

-1.0

0.1

10

Current density, rnA/ern2


FIGURE 3.25. Anodic behavior of zinc in O.SM NaOH. After Bianchi et (//.13591.

solutions vary in color from clear to slightly iridescent to yellow to black [493]. The clear
film consists essentially of CrPl with some water.
3.6. STABILITY OF PASSIVATION
3.6.1. Type of Passivation

Stability of passivation refers to the ability of the electrode surface to maintain its
state of passivation. It can be characterized by the potentials at which passivation occurs
and ends and by the passivation current density in the passive region. Depending upon
the state of polarization, zinc passivation in various solutions can be divided into two
types: that which occurs only under a certain anodic polarization (type A) and that which
occurs also at the open-circuit potentia!, i.e., the nonpolarized condition (type B). In many
cases, passivation is associated with the formation of a solid film that is stable only at
certain anodic potentials. Sometimes, the solid film is not stable at all potentials, and
passivation is achieved only under certain anodic current densities to maintain a metastable solid film. In alkaline solutions the passivation, due to the formation of Zn(OHh,
disappears when the anodic polarization is removed as shown in Fig. 3.26, because the
metastable passive hydroxide film quickly dissolves in the solution [904].
In general, the passivation of zinc electrodes in concentrated alkaline solutions is of
type A since it is necessary to impose an anodic overpotential on the zinc electrode. On
the other hand, the passivation in many slightly alkaline solutions is of type B. In the case
of type B passivation, the oxide or salt films are relatively stable, and the electrode surface

88

CHAPTER 3
2
1.6

1.2
~

a
0 .8

-t

0.4

Time, seconds

FIGURE 3.26. Voltage decay at 25C in 30% KOH saturated with ZnO polarized at 1.9 Vzinc for 3 min. After
Dirkse [904J.

can maintain the passive state without an external polarization, for example, in slightly
alkaline, carbonate, and phosphate solutions.
Stability of passivation can be further characterized by the passivation current density
in the passive region. Generally, the current in the passive region consists of two parts. A
part of the current is due to film growth to maintain the barrier thickness which chemically
dissolves. Another part of the current is due to the direct dissolution of zinc through the
pores or defects of the film. If the passive film is stable, then the current is mainly from
the dissolution through defects in the film. On the other hand, if the film is compact but
not stable and dissolves, the current is mainly used for maintaining the film thickness.
Table 3.1 shows that the passivation current, iI" can vary by orders of magnitude from one
case to another. There appears to be a clear distinction in the data shown in Table 3.1
between concentrated alkaline solutions and other solutions. Generally, in alkaline
solutions the passivation current densities are higher than I mA/cm 2, while in other
solutions they are much lower than 0.1 mA/cm 2
The current density in the passive region in alkaline solutions as a function of
concentration and potential sweep rate is shown in Fig. 3.27 [794]. In situations in which
stable salt films can form, the passivation current is determined primarily by the compactness of the films. Generally, the solid films formed through a direct oxidation process are
more compact than those formed through a dissolution-precipitation process, the latter
tending to result in porous films [607].
The means of transport for charge, reactants, and dissolution products depend on the
structure of the surface film, electrolyte composition, and test condition. Mass transport
by diffusion, capillary force, and convection are important to the reactions on the
passivated zinc surface in alkaline solutions. The passive films are generally porous [26,
27]. Ionic diffusion through the oxide film tends to be the rate-limiting process for zinc
in phosphate solutions [702]. The passive oxide films formed in certain borate, borax, and
NaOH solutions are found to be electronically conductive since oxygen evolution can

PASSIVATION AND SURFACE FILM FORMATION

80

89

. 3.0 M KOH
'" 2.0 M KOH

60

+ 1.0 M KOH

20

+
o L-----~------~------~--------------~~

0 2 3 4 5
s '/2 (mV/2 s ,')

FIGURE 3.27. Passivation current (Ip) VS. SII2 plots for KOH solutions of different concentrations. Reprinted
from Cabot et al. [7941. with kind permission from Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV
Amsterdam, The Netherlands.

proceed on the oxide films [355,907, 1129]. The passive film formed in carbonate
solutions seems to be an insulator, as the passive region can extend to potentials more
positive than 3.0 V SCE without significant increase in the anodic current ]3, 907]. The
passive films formed in various alkaline, phosphate, and borate solutions have been found
in many studies to conduct current under certain conditions through a semiconducting
mechanism [422, 484, 526, 687].

3.6.2. Passivation Breakdown


Passive films on a metal electrode tend to break down at certain anodic potentials.
Thus, the stability of passivation can also be characterized by the potential above which
breakdown of the passive film occurs. Generally, the higher the breakdown potential, the
more stable the passive film is. In practice, breakdown can be determined with an anodic
polarization curve as shown in Fig. 3.1.
The metal dissolution at passivation breakdown is usually localized, leading usually
to the formation of pits, and the breakdown potential may be taken as the pitting potential.
The rate of pit growth can be very rapid at the breakdown potential because of the large
driving force.
The value of the breakdown potential is very sensitive to solution chemistry. As
shown in Fig. 3.18, breakdown does not occur for an anodic polarization up to 3.0 V SHE
in 1M Na 2C0 3 at pH 11.5, while it occurs at about -0.1 VSCE in 3M NaCI solution at the
same pH. Depending on the composition of the base solution, the addition of a very small
amount of chemical species can greatly affect the value of breakdown potential. The
addition of as little as 150 ppm of chloride ions in a borate solution has been found to
reduce the breakdown potential by about 0.5 V [355]. Generally, Ct, Br-, r-, F, CIO:;,
SO~-, and CH 3CO:; have been found to reduce the breakdown potential [16, 17,46,355],
while OW, NUl' HPO~-, CrO~-, CO~-, WO~-, MoO~-, and BOi- have been found to
increase the breakdown potential [37,45,98, 1091. Examples of the variation of the

90

CHAPTER 3

20 ,----------------------------------------,
NO ~

T CH,CO',
+ CIO;
~

.~

CIO ;

10

Q)

X F'

Sr-

6S0: '

:s

X I'

v CI .

-1

-0.9

-0.8

-0. 7

Potential, V (Hg/HgO)

FIGURE 3.28. Influence of different anions on anodic behavior of zinc. Potential sweep rate = 25 mV/min;
pH = 9.2; [H 3B0 3] = O.2M; [NaOH] = O.IM; [anion] =O.IM. Reprinted from Augustynski era!. [16], with kind
permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington. OX5 1GB, United Kingdom.

breakdown potential with the addition of some ions are shown in Fig. 3.28 [16, 17]. Figure
3.29 shows that the breakdown potential in slightly alkaline borate solutions decreases
with increasing concentration of CIO; and Cl- [16]. Sergi et al. [174] found that
passivation breakdown is more sensitive to variations in pH than to changes in chloride
concentration.
Breakdown can arise from a variety of effects, which can be generally divided into
physical effects and chemical effects. Physical effects include changes in field strength,
dielectric properties, and mechanical failure of the barrier film caused by internal stress
or volume changes resulting from transitions in crystal structure. The chemical effects
relevant to oxide breakdown originate at the film/electrolyte interface. These include
-0.4 .----------------------------------------,

0- 0 .6
en
J:
0;
~

>

-0 .8

OJ

~
Q)
(5

n.

-1

_1 .2L-______

-2

__________

____________L __ _ _ __ _

-1

Log anion concentration (M)

FIGURE 3.29. Variation of breakdown potential of zinc with concentration ofCIO; and Cl-. pH = 9.2; [H 3B0 3 ]

=O.2M; [NaOH] = O.IM. Reprinted from Augustynski er al. [16], with kind permission from Elsevier Science
Ltd. The Boulevard, Langford Lane. Kidlington. OX5 1GB, United Kingdom.

PASSIVATION AND SURFACE FILM FORMATION

91

nonuniform dissolution of the film and defects introduced by ion adsorption on the surface
or ion incorporation in the film. In many cases, these physical and chemical effects cannot
be separated. Chemical changes can lead to mechanical failure; on the other hand,
mechanical failure may result in localized attack. It is the interdependent effects of the
physical and chemical processes that control the breakdown of passivity.
The mechanism for breakdown of passivation on zinc surfaces has been the subject
of several studies. Galvele and co-workers [45. 652J postulated that passivation hreakdown is caused by OH- ion depletion at the zinc/electrolyte interface. The formatIon of
hydroxides at the surface produces protons, which reduce the pH near the surface. Since
zinc oxide or hydroxide is only stable in slightly alkaline solutions. the decrease in pH
may prevent the formation of a stable film. On the other hand. since the dissolution rate
of zinc exponentially increases with potential, the dissolution at a higher potential will
result in a lower pH value near the surface. When the pH in localized areas near the surface
is outside the range in which the oxide film is stable, the dissolution of zinc in these local
areas will not produce nor maintain passivation at these areas by a passive film. The
breakdown potential is therefore the minimum potential at which an acidified solution
can be produced and maintained in contact with an active dissolving metal. Thus, whether
breakdown occurs under a given potential is essentially dictated by the thermodynamic
equilihrium at the local areas. In the model put forward by Galvele and co-workers. any
anion that does not interfere with the zinc dissolution and acid consumption at the
interface should not affect the breakdown potential. Anions that can reduce surface
acidification. for example. by a buffering action, will generally enhance the stability of a
passive film on a zinc surface.
On the other hand, according to Augustynski et al. [16, 17], who studied the effects
of many anions on the breakdown potential of zinc in borate solutions, the depassivation
effect of the anions is due to the ability of these anions to form more soluble salts with
zinc. Through specific adsorption on the electrode surface, these anions locally prevent
the formation of a passive film. The difference in the effects of the various anions on the
breakdown potentials is largely due to the difference in the solubilities of the corresponding zinc salts and in the extent to which the anions are specifically adsorbed.
Although the models of Galvele and Augustynski differ in the proposed role of the
anions. both models are essentially based on the solubility of the passive film at localized
areas. Galvele's model emphasizes the effect of local acidification, which increases the
solubility of passive films, while Augustynski's stresses the effect of anion adsorption
and accumulation at localized points, which cause changes in the solubility of zinc in the
electrolyte near the surface. The combination of the two views perhaps gives a more
complete picture concerning the phenomenon of passivation breakdown of a zinc electrode. It can be generalized that anions which enhance surface acidification or form zinc
salts more soluble than zinc oxide and hydroxides tend to reduce the stability of
passivation while those which reduce surface acidification or form less soluble salts tend
to enhance the stability of passivation.
In essence, the occurrence of breakdown is determined by the ability to repassivate
the local surface area once activated by any physical and chemical inhomogeneities and
fluctuations, which are always present within the system. If the condition is such that
repassivation is not possible, the active dissolution at these areas will intensify and
expand, leading to the breakdown of passivation.

4
Electrochemistry of Zinc Oxide
4.1. INTRODUCTION
Zinc oxide is a semiconducting material and is commonly found in the corrosion
products of zinc and its alloys. Many passive films formed on zinc electrodes in various
electrolytes have been shown to have some of the semiconducting properties ofZnO, and
the semiconducting properties have been found to play an important role in the corrosion
behavior of zinc in many situations. However, correlations between the electrochemical
behavior of zinc oxide and the corrosion of zinc are still lacking. Also, zinc oxide is
notable as one of the most frequently used materials in studies of semiconductor
electrochemical and photoelectrochemical phenomena. Much of the early understanding
of semiconductor electrochemistry was actually obtained from these studies. It is thus felt
that a systematic overview of the electrochemical properties of zinc oxide would be useful
not only for a deeper understanding of many zinc corrosion phenomena but also for further
research on the semiconducting behavior of zinc oxide itself.
4.2. BASIC PROPERTIES
4.2. J. Physical Properties

Zinc oxide exhibits many useful optical and thermal properties and is widely used
in the production of rubbers and paints. As a semiconductor, zinc oxide possesses a set
of unique electronic and photoelectronic properties and has been used in a number of
applications such as varistors and photocopying products. Selected properties of zinc
oxide are shown in Table 4.1 [570].
Zinc oxide has a wurtzite structure in which the oxygen atoms are arranged in a
hexagonal close-packed lattice with zinc ions occupying half the tetrahedral sites, as
shown in Fig. 4.1 [860]. The two types of ions, Zn 2+ and 0 2-, are tetrahedrally coordinated
and are therefore positionally equivalent. Due to their marked difference in size, these
ions fill only about 44% of the volume in a zinc oxide crystal, leaving some relatively
large (O.095-nm radius) open spaces [570].
The natural color of zinc oxide powder is white, but it displays pronounced changes
in color when heated or when certain impurities are incorporated into the crystals [11791.
In the visible region ofthe spectrum, zinc oxide powder has good hiding power (the ability
93

94

CHAPTER 4

TABLE 4.1. Properties of Zinc Oxide"


Molecular weight
Lattice
Lattice constants
Density
Dielectric constant
Refractive index
Energy band gap
Enthalpy of formation
Melting point
Specific heat
Solubility in H 20

Zn: 65.38; 0: 16.00; ZnO: 81.38


Hexagonal. wurtzite
a =0.324 nm, C =0.519 nm, cia = 1.60
5.78 glcm 3 or 4.21 x 1022 ZnO molecules/cm 3
8.54
2.008
3.2eV I
Zn(s) + Z02(g) ~ ZnO(s) -83.17 kcallmol
Vaporizes at -1700 D C at normal atmospheric pressure; melts at
197YC under pressure
9.66 call(mol-K)
1.6 x 10-6 g per gram of H 20 at 2YC

"Reprinted from Van [570]. with kind permission from Elsevier Science Ltd. The Boulevard, Langford Lane.
Kidlington OX5 lOB, United Kingdom.

to prevent light transmission) and tinting strength, depending on the refractive index and
particle size. The white color of zinc oxide powder, composed of transparent and colorless
microcrystals, is a result of a series of optical processes: surface reflection, transmission
through crystals, refraction, and scattering of light rays. Reflection of light at an oxide/air
interface is low, about 11 %, the rest of the light being transmitted through the crystal.
Oxide particles with sizes that approach the wavelength of the incident light are highly
effective in scattering those rays, thereby reducing the degree of light penetration into the
powder layer. The hiding power varies with particle size to a maximum of 0.25 pm as
shown in Fig. 4.2 [1277]. Zinc oxide with particle size smaller than 0.06 pm attenuates
(i.e., scatters and absorbs) ultraviolet radiation most effectively [451].
Zinc oxide possesses a set of unique thermal and optical properties, as summarized
by Brown [1179]. It changes from reflector to absorber abruptly at a wavelength of 0.385
pm, close to the border between the UV and visible regions, as shown in Fig. 4.3 [1179].
With dopant additions and proper heat treatment, zinc oxide can be a versatile phosphor
that converts ultraviolet light and X-ray radiation into light of various colors [570]. Zinc
,

"

FIGURE 4.1. Crystal wurtzite structure of zinc oxide. 0, zinc; e, oxygen. After Addison [860].

ELECTROCHEMISTRY OF ZINC OXIDE

95

ZnO

in
en

E
en

:c

g>
Q)

>

(j)

a:

0.4

0.2

0.6

0.8

Particle size in microns

FIGURE 4.2. Relative light transmission of zinc oxide as a function of particle size. Zinc oxide is availahle in
a wide range of particle sizes and provides a broad spectrum of hiding power since the relative light transmission
of the oxide is a U-shaped function of the particle size. Optimum hiding power is ohtaincd with particles of
O.25-flm size. After Stutz 11277].

oxide also shows significant photoconductivity in the ultraviolet region and throughout
most of the visible region of the spectrum.
4.2.2. Electronic Properties

Semiconductors are substances with electronic conductivity between that of metals


(10 6 _10 4 Q-Icm-I) and dielectrics 10- 10 Q-Icm-I). Zinc oxide is intrinsically an n-type
semiconductor due to electrons excited from ionized zinc interstitials existing in the zinc
oxide crystal lattice. At 25C, the typical electronic conductivity is 1 Q-Icm-I [11791.

100

~
Q)

()

c
cO
t5
Q)

50

'lii
a:

OL-______
300

400

______L -______L -_ _ _ _ _ _
500

600

700

____

800

Wavelength (nm)

FIGURE 4.3. Percentage of light reflected by zinc oxide as a function of wavelength. Zinc oxide exhibits a
pronounced absorption edge in the ultraviolet range (low reflectance) at 385 nm. After Brown [1179].

96

CHAPTER 4

,',""
Conduction Band
Donor -"'--'-

Valence Band
./

./

;'

;'

FIGURE 4.4. The effect of donor and acceptor


defects on electron transport across the band gap .

The quantum theory of solids presents a complete and rigorous description of the
nature of current carriers in semiconductors. According to quantum theory, the energy
spectrum of electrons in an ideal crystal consists of energy bands filled with energy levels
(allowed bands) and with no energy levels (band gaps). The width of a band gap and the
distribution of electrons in the allowed bands determines the electronic nature of a crystal
(i.e., metal, semiconductor, or dielectric). For a semiconductor, the upper, unfilled band
is called the conduction band while the lower, almost filled band is called the valence
band, as shown in Fig. 4.4. The width of the band gap, E~ = E,. - En which is the most
important electronic characteristic of a crystal, depends on the strength of the chemical
bonds. For ZnO, Ex = 3.2 eV.
The electronic conductivity of semiconductors, as expected from the band structure,
can be generated by electrons of atoms of the basic substance in the crystal (intrinsic
conductivity) as well as by electrons of impurity atoms or by the presence of defects
(extrinsic conductivity). In intrinsic semiconductors at T> 0 K, the generation of current
carriers occurs as a result of the thermal excitation of some electrons from the valence
band to the conduction band, with the corresponding thermal rupture of some chemical
bonds. Simultaneously, an equal number of positively charged holes are created in the
valence band. In an electric field, these holes behave like particles possessing a positive
charge equal in absolute value to the charge of the electron. For extrinsic semiconductors,
impurities and defects (which have energy levels located in the band gap) are classified
as either donors or acceptors as shown in Fig. 4.4. Donors, usually located at energy levels
slightly below the conduction band, give up excess electrons to the conduction band,
thereby creating electron conductivity (n-type semiconductors). Acceptors, located at
energy levels slightly above the valence band, capture valence electrons from atoms of
the basic substance, producing hole conductivity (p-type semiconductors).
An important concept in the description of semiconducting properties is that of the
Fermi level, EF, which is defined as the energy level for which the probability of being
occupied by an electron is ~. For an intrinsic semiconductor at room temperature, EF lies
essentially midway between the conduction band and the valence band within the band
gap. For a doped material, the location of EF depends on the type and concentration of
the dopant. For moderately or heavily doped n-type solids, EF lies slightly below the
conduction band. Similarly, for moderately or heavily doped p-type materials, EF lies just
above the valence band.
The conductivity of zinc oxide samples has been observed to be in the range of 10- 17
to 103 Q-Icm-I, depending principally upon the method of sample preparation [1179]. A

ELECTROCHEMISTRY OF ZINC OXIDE


Zn2+ O~
Zn'+ O~
eZn+
Zn2+
Zn2+ o~
Zn2+ O~
Zn
e
O~
Zn2+ O~
Zn'+ O~
Zn2+
Zn2+
Zn2+ O~
Zn2+ O~
eO~
Zn2+ O~
Zn2+ O~

O~

(a)

97
Zn'+

O~

Inl+

O~

Zn2+

O~

Zn'+

O~

Zn2+

O~

O~

In 3 +

Zn2+

O~

O~

e-

Zn2+ 0"
e
O~
Inl+
Zn 2 + 0"
Zn'+

O~

Zn'+
e
O~

(b)

FIGURE 4.5. Schematic lattice structure of zinc oxide (al and doped zinc oxide (b).

single crystal of pure zinc oxide has very low conductivity and is an insulator. The
semiconducting property of zinc oxide depends on the presence of defects in the zinc
oxide lattice. Two types of semiconducting zinc oxides are distinguished according to the
types of defects; one contains interstitial and the other substituted zinc atoms, as shown
in Fig. 4.5. In the interstitial type, zinc oxide is partially reduced by reaction with agents
such as carbon monoxide or hydrogen at elevated temperatures (400-900C). Each atom
of oxygen removed releases an atom of zinc and two electrons. The zinc atom moves to
the void space to become an interstitial atom, which may be in the form of Zn, Zn+, or
Zn 2+, depending mainly on temperature. The substitutional type of zinc oxide is produced
in the presence of metallic vapor or salts at elevated temperature. A portion of the zinc
atoms in the zinc oxide crystals are replaced by the foreign metallic atoms. The zinc atoms,
upon release from their lattice positions, diffuse to the crystal surface, where they
vaporize. Depending on the type of metallic atoms, the conductivity of zinc oxide can be
either increased or decreased.
The electronic structure and surface characteristics of zinc oxide are found to
increase the rate of many chemical reactions [570]. Zinc oxide has great absorptivity for
Hb CO, and CO 2 after being cleaned of absorbed HP and CO 2 by vacuum heating. The
catalytic activity of ZnO is generally increased with increase in conductivity. Upon
exposure of zinc oxide to air or to oxygen, some of the electrons near the surface are
spontaneously trapped by physically adsorbed oxygen to form negative ions on the
surface [1179]. These ions, formed by transfer of electrons from the interior of ZnO to
the surface, create an upward bending of the bands. The adsorbed ions can be desorbed
either by heating or by generation of holes with light. The holes are driven to the surface
to neutralize the adsorbed ions, and the photoelectrons in the conduction band compensate
the positive space charge of the ionized donors, resulting in a flatband condition.
4.3. SEMICONDUCTOR ELECTROCHEMICAL BEHAVIOR

4.3.1. Basic Theories


When zinc oxide is immersed in an aqueous solution, protons, hydroxyl ions, and
other ions adsorb on the surface. In the simplest case of pure water, OH- ions are attracted

98

CHAPTER 4
liquid

H
0

H
0

H
0

Zn --- 0 --- Zn --- 0 ___ Zn ___ 0 --- Zn ___ 0

solid

FIGURE 4.6. Schematic lattice structure of zinc


oxide surface in water.

to the zinc sites and H+ ions are attracted to the oxygen sites on the ZnO surface as shown
in Fig. 4.6. The surface generally adsorbs an excess of one species and becomes charged
either positively or negatively depending on the reactions at equilibrium:
(4.1 )

(4.2)
When a semiconductor is brought into contact with a solution containing a redox
couple (e.g., Zn 2+/Zn or H+/H2)' if electrostatic equilibrium is attained, the Fermi levels
in the two phases must become equal (the electrochemical potentials must become equal).
In the case shown in Fig. 4.7 for an n-type semiconductor, where EF of the semiconductor
is higher than that in solution, electrons will flow from the semiconductor to the solution
phase. The excess charge in the semiconductor does not reside at the surface, as it would
in a metal, but instead is distributed in a region near the surface, called the space charge
region. The resulting electric field that forms in the space charge region is shown by a
bending of the bands. In the case of Fig. 4.7b, where the semiconductor is positively
charged with respect to the solution, the bands are bent upward (with respect to the level
in the bulk semiconductor), and the degree of band bending is measured by V,.
When the semiconductor has no excess charge, there is no space charge region and
no electric field and the bands are not bent. The electrode potential under this condition
is called the flatband potential, Etb . The flatband potential is a very important quantity for
a semiconductor electrode as it connects the parameters that can be experimentally
determined to the parameters derived from semiconductor/electrolyte interface physics.
The value of the flatband potential of a metal oxide semiconductor is found to be
quantitatively related to its electron affinity [1180].
When the interface is irradiated with light of energy greater than the band gap, E~,
photons are absorbed and electron-hole pairs are generated (Fig. 4.7c). Some of these
electrons and holes, especially those formed in the bulk semiconductor beyond the space
charge region, recombine with the evolution of heat. However, the space charge field
causes the separation of electrons and holes. Thus, in the case of Fig. 4.7c, the holes arrive
at the surface at an effective potential equivalent to the valence band edge and cause the
oxidation of the redox species in the solution from R to 0 while the electrons move into
the external circuit through the semiconductor electrode lead. The flow of holes and
electrons in opposite directions can be measured as current (photocurrent). The larger the
band bending, the more holes are driven to the surface and the larger is the photocurrent.
Thus, the onset of the photocurrent is near E!b, at which the band bending is zero.

E,

(a)

solution

O/R

vsT

(b)

interface

O/R

(c)

EEl'

--------....-r I
- - .-

/R

I hv > E,

O/R: (a) Before contact in the dark; (b) after contact (in the dark) and electrostatic equilibration; (c) junction under irradiation.

FIGURE 4.7. Representation of the formation of the junction between a semiconductor and a solution containing a redox couple

semiconductor

Ev

EF

Ec

""
'C

Z
\)
o
><
6tT1

o"T1

-I

Vl

s::

tT1

:r:

o\)

;>:J

-I

tT1
\)

tT1

100

CHAPTER 4

Similarly, for a p-type semiconductor the bands are usually bent downward and the
electrons generated by irradiation are moved by the field in the space charge region toward
the surface, causing reduction of 0 to R.
The electrode potential of a semiconductor can be changed by an extemal power
source so that the degree of band bending is altered. Depending on the type of semiconductor, electron or hole current can be generated by bending the bands in one direction.
In the dark under reverse bias (increasing band bending), there is essentially no current
flow because for an n-type semiconductor there are few holes and for a p-type semiconductor there are few electrons available in the semiconductor to participate in the
reactions. On the other hand, under a forward bias (decreased band bending), there are
more electrons (n-type semiconductor) or more holes (p-type semiconductor) at the
semiconductor surface. More thorough descriptions of fundamental semiconductor electrochemistry can be found in the literature [1177, 1183].

4.3.2. Flatband Potential


The flatband potential of a semiconductor electrode can be experimentally determined by measuring the capacitance as a function of potential. The capacitance of the
space charge layer, Csc. the degree of band bending, V, = E - Efb and the dopant concentration, N D, are related and can be described by the Mott-Schottky equation [1177]:

lIC;c = (2IeeeoND)(-V, - kTIe)

(4.3)

Thus. a plot of lIC~ versus potential E is linear. In this plot the potential at which the
line intersects the potential axis yields the value of Efb, and the slope can be used to
calculate the doping level ND . Figure 4.8 shows a typical Mott-Schottky plot reported by
Dewald [514] for single-crystalline ZnO in 1M KCl at pH 8.5. The flatband potential can
also be estimated by determining the onset potential for photocurrent [423].
1,200

,,

1,000
N

fu

800

u2-

600

400

'-0.59 ohm" em"

theoretical SlOPer",

200
0
-0.5

0.5

1.5

Pote ntial (V seE)

FIGURE 4.8. Mott-Schottky plots for two crystals under exhaustion conditions. The dashed lines represent
the theoretical slopes. The intercepts of the linear plots give the values of the flatband potentials of the crystals.
After Dewald [514).

101

ELECTROCHEMISTRY OF ZINC OXIDE

TABLE 4.2. Dopant Concentrations, N D' Slopes of Logarithmic Current vs. Potential (in
Millivolts), and Flatband Potentials, En" of ZnO in Various Solutions
Solution

pH

104_ IO -z M Fe(CN)~-

3
6
8.8
12
8.5
3.8

Acetonitrile
O.IMNaOH

13

IMKCI
O.IM NaZS04
1M KCI + borate + 10-3M Fe(CN)~1M KCI + O.OlM KOH + 0.5M K 3 Fe(CN)n
1M KCI + borate

ND (cm- 3 )

1020
3 x 10 18
3.3 x 10 17
5 x 10 17
2.6 x 10 10
2 x IO IR
2 x 10 17
3 x 10 18

log i-V
slope

605
65 5
65 5

Efb (SCE)

-0.46
-0.52
-0.45
-0.65
-0.47
-0.2
-0.75
-0.82

Reference

455
474
950
1033
514
951
919
474

The value of the flatband potential is determined by two factors: variation of the bulk
Fermi level and interaction of surface states with the electrolyte. The first is related to
variations in dopant concentration because the Fermi level with respect to the conduction
band edge, Ec, is equal to kTleln(NJND ) for a nondegenerated semiconductor, where
Ne is the effective density of states in the conduction band. The flatband potentials of ZnO
determined in various solutions are shown in Table 4.2. They vary from 60 to 65 m V per
decade in N D, in agreement with the theory. The second factor is subject to variations in
surface treatment and the nature of the electrolyte. These variations are manifested by
changes in VH , which is the voltage drop across the Helmholtz double layer and depends
on the excess surface charge. It is generally found that for non degenerated semiconductors
VII is primarily determined by adsorption/desorption processes between the surface and
the electrolyte. The contribution from electron transfer between the surface and the bulk
of the semiconductor is negligible. This is because the amount of charge stored in the
semiconductor associated with this transfer is on the order of 10 12lcm 2 or less, which is
very small in comparison with the amount of charge (on the order of 10 15lcm 2 ) associated
with the adsorption/desorption processes [1177].
In solutions in the absence of specific adsorption of other ionic species, adsorption/desorption of H+ or OH- is responsible for the excess charge stored on the surface.
As reported by Morrison [951], the flatband potential of ZnO in aqueous solutions of
various pH values is essentially independent of the presence of different redox couples.
The Helmholtz potential of ZnO is primarily determined by H+ and OH- ions and is little
affected by the presence of other species.
In the reaction

where H30+ is a hydronium ion in solution, the free energy of the reaction varies with the
double-layer potential Vfj, as the proton must acquire the potential energy (eVH ) to become
adsorbed [1177]. Therefore,
(4.4)

102

CHAPTER 4

-0.2
-0.4

-0.6

+ Rest dark potential

-0.8

Rest potential

ft Flatband potential

-1

-1.2

10

12

14

pH

FIGURE 4.9. pH dependencies of the potentials of the ZnO electrode: the flatband potential, the rest
photopotential, and the rest dark potential. After Matsumoto el al. [474].

where A is a constant. The double-layer potential is proportional to the charge adsorbed.


Assuming that CH , the capacitance of the Helmholtz double layer, is independent of VH ,
the relation between the charge adsorbed and the double-layer potential can be described
as
(4.5)
Since [W] only varies slowly with [H30+], according to Eqs. (4.4) and (4.5) and as an
approximation one obtains
VH = B + kT/e . In[H30+] = B - 0.059pH

(4.6)

(where B is a constant) which indicates that the Helmholtz potential decreases about 59
m V per pH unit. The flatband potential then also varies 59 m V per unit pH since it is
expressed as
(4.7)
The approximately 60-m V/decade variation of the flatband potential with pH on a ZnO
electrode was first confirmed by Lohmann [1196]. This relationship has also been
reported in several studies as shown in Fig. 4.9 [474, 514]. As illustrated in Fig. 4.18,
there is also a linear dependence of the flatband potential on pH for passive films formed
on a zinc electrode [526].
The flatband potential depends on the crystallographic plane and on the surface
condition of ZnO [737]. Dewald [514] found that the flatband potential for a ZnO crystal
etched in H 3P04 solution was 130 mV more positive than that for one etched in KOH
solution. The effect of etching was explained by Dewald on the basis of acid-base
equilibrium. On an ideal {1120} surface, there are equal numbers of zinc and oxygen atoms.
Each of the surface oxygen atoms has an unshared pair of electrons, and each zinc atom

ELECTROCHEMISTRY OF ZINC OXIDE

103

- 0 . 4 8 , - - - - - - - - - - - - -- -- - - - - - - - ,
First run

-0.5
Ul

'I'

:;:

G.
~

Second run

+ Third

-0.52

run

'" -0.54
-0
a.

-0

c:

1i

-a

u:

.f

-0 .56

+
-0.58

---'J..2- - - ,'-6- - - - -2'0----=-'24

-0. 6 0L ----'4---..J.
8

Time after etching, hours

FIGURE 4.10. Variation of the flatband potential with time after etching in H,P0 4 . The different symbols
correspond to three successive runs on the same crystal. After Dewald [514].

has an empty pair of orbitals. These oxygen electrons are shared, and the zinc orbitals are
filled by bonding with the ionic species in the solution, which at equilibrium determine
the flatband potential. Different etchants may result in different ratios of surface zinc
atoms to oxygen atoms and thus change the excess charge on the surface.
Figure 4.1 0 shows that the flatband potential tends to change with time in the solution
[514]. This gradual change is attributed to slow dissolution or corrosion of the surface
leading to a new surface condition which is independent of the initial surface treatment.
According to Dewald [514], the direction of the change depends on the nature of the
surface treatment, while the final limiting value is independent of the nature of the surface
treatment.
The frequency used for measuring the capacitance for the Mott-Schottky plot has
been found to have varied effects on the value of the flatband potential. Dewald [5141
reported that the Mott-Schottky plot measured in 1M KCl at pH 8.5 is essentially
independent of frequency from 100 to 10,000 Hz. On the other hand, Vanden Berghe et
al. [1033] found that the flatband potential of ZnO in 1M KCl containing 0.5M
Fe(CN)~- at pH 12 varies by 70 m V over the frequency range 130-10,000 Hz. The surface
states, energetically distributed over the whole range of the band gap, are often considered
to be responsible for the frequency-dependent flatband potential.
4.3.3. Band Structure
The energetic positions of conduction and valence bands of a semiconductor at the
surface depend on the interaction between the semiconductor and electrolyte through
ionic and electronic exchange at the interface. They are a function of the semiconductor
material and of the composition of the electrolyte. The conduction band edge in the bulk
for an n-type semiconductor, the valence band edge for a p-type semiconductor, and the
Fermi level differ by a fixed amount of energy determined by the doping concentration (e.g.,
about 0.1 V for moderately doped semiconductors). The position of the band edges at the

104

CHAPTER 4
E (NHE)

....L..--r------~./.-----

E, ____-,__________.-~---

3.2 V

E,

ZnO
2

rI

Ev ----~-----------------

semiconductor

En.

L,
4

ohp

electrolyte

interface
FIGURE 4.11. Band structure of moderately doped ZnO in the dark at pH 7. ohp. Outer Helmholtz plane.

surface can be determined through the flatband potential using the Mott-Schottky
equation (Eq. 4.3).
The band structure can be illustrated for given values of the flatband potential, doping
concentration, and rest potential. Figure 4.11 shows the band structure of moderately
doped ZnO in the dark at pH 7. At this pH value the potential drop in the Helmholtz double
layer, VH , is estimated to be about 0.1 V using the pH value of 8.8 at the point of zero
charge, where VH = 0 [1181]. The point of zero charge, according to Blok and De Bruyn
[1181], is mainly influenced by the presence of impurities in the oxide and the nature of
the electrolyte. Several other factors may also influence the value of the point of zero
charge, as reviewed by Parks [1185].
The band bending for the ZnO semiconductor in Fig. 4.11 is 0.53 V. Assuming a
dopant concentration of 1017/cm3, the width of the space charge layer is calculated to be
about 0.07 ,urn according to Ls= (2V,cocfeND)112 [1177]. Accordingly, the amount of
immobile charge in the space charge layer is Qs = eNoL, = 1.1 x 10-7 Cfcm 2, and the field
strength at the surface is Cs = eNDL/ceo = 1.5 x 105 V fcm.
This band structure is a function of the flatband and rest potentials. Figure 4.9 shows
that the rest dark potential ofZnO in 0.IMNa 2S0 4 is essentially independent of pH [474].
This means that the degree of band bending, V" also increases with increasing pH since
the rest dark potential Vrd is expressed as
(4.8)
The variation of V, is caused by the charge transfer between the surface and the bulk
semiconductor, which has little effect on the magnitude of VH for nondegenerated
semiconductors [l177]. The reason V, varies with pH, according to Matsumoto et at.

105

ELECTROCHEMISTRY OF ZINC OXIDE

[474], is that the protons and hydroxyl ions not only adsorb on the ZnO surface but also
react with the metal ions on the surface. This leads to a change of the valence of the metal
ions so that the potential drop across the space charge layer changes with pH according
to

v, = const. -

(4.9)

0.059pH

Combining Eqs. (4.6)-(4.9), one obtains


Vrrl

= const.

(4.10)

4.3.4. Electrode Kinetics in the Dark

The typical semiconductor electrode is characterized by the phenomenon that in the


dark current can only flow in one direction, depending on the type of semiconductor. For
an n-type semiconductor at anodic potentials (reverse bias), the current in the dark, being
limited by the availability of holes, is close to zero and is essentially independent of the
potential. At cathodic potentials (forward bias), the current is not limited, and it increases
with decreasing potential bias because electrons, the majority carrier, are responsible for
the current flow. The magnitude of the cathodic current at a given potential depends on
the concentration of oxidizing agents in solution while the anodic current is essentially
independent of the presence of reducing agents.
The electrode kinetics of ZnO have been extensively investigated by Freund and
Morrison [892, 921,950,9511. The anodic current in the dark is found to be very low,
typical for an n-type semiconductor. For example, the dark anodic current density on ZnO
in 1M KCl is less than 5 nA/cm" up to a potential as high as 10 V seE [9511. Under a forward
bias in a solution containing Fe(CN)~-, the cathodic current is proportional to the
concentration of Fe(CN)~-. A plot of the logarithmic current versus potential over the
current range 0.02-20 j1A/cm" gives a 60 m V change in the surface barrier per decade of
current change. This behavior is consistent with the simple theoretical model for a
semiconductor electrode under a forward bias. The rate of reaction is first-order in the
density of electrons at the surface, n, = ND exp( -e V/kT), and first-order in the density of
unfilled states at the solid/electrolyte interface (adsorbed Fe(CN)~- ions), no" and can be
expressed as
(4.11 )
is the number of
where J is the cathodic current density, h' is the rate constant,
Fe(CN)~- ions per cubic centimeter in the electrolyte, and 11., is the number of conduction
band electrons per cubic centimeter at the semiconductor surface. The linear dependence
of current on Fe(CN)~- concentration indicates that the adsorption of ferricyanide ion on
a zinc electrode follows a linear isotherm. It is also found that the current is independent
of the presence of Fe(CN)t, indicating that the reaction is irreversible; i.e., electrons are
not transferred from adsorbed Fe(CN)t to the conduction band. This effect is an
indication that the filled energy level is far below the conduction band.
The cathodic current-potential relationship described by Eq. (4.11) has been observed in solutions containing various redox couples, as shown in Fig. 4.12 1951]. All the
110X

106

CHAPTER 4

curves give slopes in the range of 65 5 mV/decade. The I/C2 versus potential plots give
straight lines, agreeing well with Eq. (4.3) [950]. The 60-m V slope for log i versus Vand
the linear Mott-Schottky plots suggest that the rate-limiting step in the reduction process
is the transfer of electrons within the space charge layer. Thus, the zinc electrode behaves
like a typical semiconductor: the Helmholtz potential, VH , does not change with the
applied potential, and the potential change occurs only across the space charge layer, V,.
Figure 4.12 also shows that for a given current density the surface barrier required
in different solutions is different. The different barrier heights are related to the electrode
capacitance and can be determined from capacitance measurements. Table 4.3 shows that
at a given cathodic current density the capacitance of a ZnO electrode in 1M KC1 solution
10.3r-----.----.-----.-----.-----.-----.----.-----,

o
o

10 ' CeIHSO.I . wH IS . 1000 11 lace


IQ J KMnO . p H 87 100011

>

IO ' 2M KMnO . pH ~ 5 1000' 1

..,

1O' 2M INH.1 2 I<CI6 38 100011


_
1O. 2M 1'. II I Cvan.de. pH 38 100011

102M F. II I

"

1O. 2 M A9INH.12NOJ pH 12 (00011

IO 3M Fe tIl Fhenaolhfoh~ln pH 1.5 (0001]

Cv.n.de.

pH 12.0 100011

10 ,5

'"E

i-

0
0

'Vi
C

....C

10.6

'1:::1

10'7

10 ~L_

__

~~

__

____

____

____- L____- L____ __


06
07
08
~

Surface barrier Vs, V


FIGURE 4.12. Variation of the cathodic current with the surface barrier, for various oneequivalent oxidizing
agents. Reprinted from Morrison [951], with kind permission from Elsevier Science-NL, Sara Burgerhartstraat
25, 1055 KV Amsterdam, The Netherlands.

ELECTROCHEMISTRY OF ZINC OXIDE

107

TABLE 4.3. Effect of Different Oxidizing Agents on the


Capacity of a ZnO Electrodea Measured during Cathodic
Reactions in 1M KCl Solutions b
Substance added

Capacity (nF)

O.IMHCOO-

02 + O.IM HCOO0.IMH 20 2
O.IM H20 2 + 0.2M HCOO-

600
580
113
108
60

62

"Geometric area = 0.2 cm': dark current = -5.0 0.1 nA.


hRef. 921.

varies with the addition of different oxidizing agents [921]. With no oxidizing agent
present, the capacitance is high, corresponding to a large surface barrier (large VJ, Table
4.3 also shows that both H20 2 and O 2 are very active oxidizing agents on ZnO while the
formate ion is not active. According to Morrison [951], since the Helmholtz potential of
the system is independent of the kind of redox species in the solution, this difference in
surface barrier must be due to the difference in electron reactivity, O'[X], with 0' the cross
section of the surface state for electron capture and [X] the concentration of the unfilled
surface states. For different oxidizing agents, both 0' and [X] can be different. While the
amount of unfilled surface states is determined by adsorption of the oxidizing agent, the
cross section depends on the extent of overlap between the energy levels in the conduction
band on the surface and the energy levels in the adsorbed oxidizing agents. Because
different agents have different redox potentials, overlapping of the energy levels in the
solution and in the bands of the semiconductor varies.
ZnO electrode behavior in the dark has been studied by different techniques. Kohl
and Bard [919] measured the cyclic voltammogram of a ZnO electrode in a nonaqueous
(acetonitrile) solution containing various redox couples. They compared the reduction/oxidation potentials of the couples obtained for ZnO with those obtained on a Pt
electrode. Gomes and Cardon [939] used electrochemical noise measurements to study
the oxidation reactions of 1- and aliphatic alcohols on a ZnO electrode. Bindra et al. [477]
studied lead deposition on zinc oxide in 1M KN0 3 solution. Eger et al. [806] found that
a very strong accumulation layer with a surface electron density approaching 1014/cm2 is
achieved on a ZnO surface in 2M KOH solution using a constant-pulse technique. The
charge density and thickness of accumulation layer was found to be a function of the
barrier height.
Breakdown. Breakdown of a semiconductor electrode made from a nondegenerated
semiconductor occurs when the near-zero current at reverse bias sharply increases with
increasing reverse potential bias. At breakdown the electrode loses its "insulating"
character and becomes "conductive." Breakdown is observed on the zinc oxide electrode
when itis anodically polarized to high potentials as shown in Fig. 4.13 [946]. The potential
at breakdown generally decreases with increasing dopant concentration. As a result of the
breakdown, both dissolution of zinc oxide and formation of oxygen from water oxidation
occur. The ratio of zinc oxide dissolution to oxygen formation is found to decrease with

108

CHAPTER 4

12,------------------------------------.---,

Eu
<t:

2,

>-

10
8

'iiic

CD
-0

::;
<.)

No = 10 '6 cm3

2
OL---__
1

__

__

_____ L_ _ _ _
10

~~

____

____

50

100

Electrode potential E (VscEl

FIGURE 4.13. Potential dependence of breakdown currents at ZnO electrodes with different doping concentration N D Electrolyte: 1M KCI. pH 6; sweep rate: 50 mV/s. After Pettinger et al. [946).

increasing current. This ratio varies with the crystallographic orientation of the zinc oxide
electrode, indicating the specific catalytic effect of surface structure on different reactions.
Breakdown of a ZnO electrode at high anodic potentials in aqueous solution is
associated with an electron tunneling process. According to Pettinger et al. [455], by
applying high voltages to a moderately or highly doped semiconductor electrode, the
thickness of the potential barrier for energy levels in the upper part of the forbidden zone
is reduced so that electrons can tunnel through the barrier from the redox couple in the
solution into the conduction band, as illustrated in Fig. 4.14. The probability for an
electron to tunnel increases with decrease of the thickness of the space charge layer
[1197]. Thus, the tunneling current increases with increasing dopant concentration since
the thickness of the space charge barrier is a function of dopant concentration according
to L, = ( 2Vl4;/eND) I 12. This equation also indicates the dependence of the current on
potential at breakdown. Since the majority of the applied voltage is across the space charge
layer and its thickness is a function of the voltage, the tunneling current can be changed
by several orders of magnitude by varying the potential.
According to the model illustrated in Fig. 4.14, the tunneling current also depends
on the energy levels of the reducing agents and their concentrations in the solution. Thus,
at breakdown, at a given potential the anodic current may vary depending on the kind of
reducing species in the solution.
In addition to the tunneling process from a redox level in the solution into the
conduction band, band-to-band tunneling has also been considered possible [946].
However, at a band bending less than 3.2 V, electron tunneling from the upper edge of the
valence band to the conduction band is not possible for ZnO, which has a band gap of
3.2 eV
Electron tunneling does not appear to be only an anodic process on zinc oxide.
Pettinger et al. [455] also observed a cathodic current on a highly doped ZnO electrode
at a potential as high as 0.7 VseE. at which considerable band bending occurs. Since the
electron concentration at the surface is very low at such a potential, cathodic reaction can

ELECTROCHEMISTRY OF ZINC OXIDE

109

electron
donor

3.2 V

ZnO
electrolyte

Ev----~--------~

FIGURE 4.14. Model representation of


electron tunneling from the donor in the
electrolyte into the conduction band. After Pettinger et al. [4551.

semiconductor
interface

only occur with the participation of bulk electrons. Furthermore, for a highly doped ZnO
electrode, the barrier thickness can be sufficiently thin for tunneling to occur at a relatively
small band bending. The cathodic current observed at a rather positive potential can then
be attributed to the electron tunneling from the conduction band to the oxidizing agent in
the solution. Therefore, it appears that electrons can tunnel through the space charge
barrier of highly doped ZnO electrodes in both directions when proper redox couples are
present in the electrolyte to act as the electron donor and acceptor.
4.3.5. Photo electrochemical Kinetics

When a ZnO electrode/electrolyte interface is under illumination, a photopotential


or a photocurrent can be measured depending on whether or not the electrode potential
is controlled. The photopotential and photocurrent are essentially the result of the change
in the surface concentration of electrons and holes induced by absorption of light. The
concentrations of electrons and holes are determined by

n, = no exp(eV/kT)

(4.12 )

p, =Po exp(-eV/kT)

(4.13)

Under a moderate illumination intensity, the concentration of the majority carrier in the
bulk of an n-type semiconductor, no, will change very little because the number of
electrons generated by moderate illumination is very small compared to no. However, the

110

CHAPTER 4

5
3

2
E (V",)

FIGURE 4.15. Current-voltage curves for ZnO in HzS04 (pH 3) under illumination of intensities
I L .) and I LZ ' I dark = 10-6 _10- 7 Ncm 2. After Gerischer [129].

minority carrier, Po, can be drastically changed, by orders of magnitude, under illumination. Thus, according to Eq. (4.13), the surface hole concentration is greatly increased as
a result of illumination.
Figure 4.15 shows an example of thephotocurrent observed on a ZnO electrode in
H 2S04 solution [129]. The photocurrent increases with increasing anodic potential since
the concentration of surface holes increases with increasing surface potential barrier V,.
Above a certain potential, the photocurrent reaches a constant value, the saturation
current, when all the photogenerated holes available at the surface are exhausted by the
anodic reactions. The saturation current depends only on bulk properties of the semiconductor for a given light intensity [793].
Theoretically, photocurrent is directly proportional to the light intensity 10 and is a
function of the space charge layer thickness L" the hole diffusion length L,,, and the
absorption coefficient a, as described by Eq. (4.14) [1183,1187].
i photo oc

eIo{ 1 - [exp(-aL,)/(l + aLp)]}

(4.14 )

The depth of light penetration in a semiconductor depends on the wavelength of the light.
For example, with 370-nm-wavelength light the penetration in ZnO is about 0.2 Ilm
[1188].
The net current and the quantum yield are a function of a number of competing
processes as described by Gerischer [793] in Fig. 4.16. For an n-type semiconductor, the
externally measurable current i is the difference between the photocurrent and the forward
current of electrons. The majority current is decreased to zero at a certain anodic bias.
The flux of holes to the surface, defined as idf' is exclusively controlled by the solid-state
properties, while all the other reaction steps depend on the surface properties of the
semiconductor. The holes arriving at the surface can either be transferred to the electrolyte
by a reaction with a redox couple, by an oxidative or reductive decomposition process,
or recombine via surface recombination centers. The measured photocurrent, ipho,o, can
therefore largely deviate from idf' Consequently, photocurrent-voltage curves of different

1II

ELECTROCHEMISTRY OF ZINC OXIDE


Electrolyte

n-type semiconductor

Conduction band

~---''''----~'f-~~=-''::....l..;oL...::-=-~- Ee

-E F
Surface
recombination

Difussion current

hv

Light
absorption

Recombination

Valence band

FIGURE 4.16. Kinetic processes controlling the photocurrent yield. Reprinted from Gerischer [793], with kind
permission from Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.

semiconductor samples can have very different shapes and depend differently on changes
in the experimental conditions.
Thus, in the absence of surface recombination and with a fast rate of electron transfer,
the photocurrent increases steeply when the depletion layer starts to form, and a saturation
current is quickly reached. On the other hand, with sufficient surface recombination or in
cases of slow electron-transfer reactions, the apparent onset of the photocurrent is shifted
to higher bias, and the saturation current is only reached at a larger band bending. The
crystal orientation, surface structure, and surface treatment have been demonstrated to
affect the surface recombination processes [793]. It is found that mechanically induced
defects near the surface of ZnO single crystals act as hole traps. which are efficient
recombination centers [1188].
A ZnO electrode under illumination exhibits a number of phenomenon including
current doubling, photocatalysis. and electroluminescence. Also. single-crystal ZnO is
found to be photosensitive to light with energy less than the band gap in a solution
containing agents, such as ferrous ion, which act as a dye sensitizer [949]. The sensitization is attributed to electron injection from the photoexcited adsorbed-dye surface states.
Masuda et al. [792] detected an acoustic signal during photoelectrochemical reactions at
a ZnO electrode in O.2M KN0 3 + 0.0 1M Pb 2+. The acoustic signal was found to associate
with the deposition of an oxide layer of Pb0 2 and to be a function of the light wavelength.
4.3.5.1. Current Doubling. Current doubling is a phenomenon observed in the
anodic reaction of a ZnO electrode under illumination in an electrolyte containing two or
more equivalent reducing agents. In current doubling an anodic current is produced,
which is approximately double the original limiting hole current generated by illumination [921,1032]. A two-step mechanism has been suggested for this process
(4.15)
(4.16)

112

CHAPTER 4

TABLE 4.4. Anodic Reactions on Illuminated ZnO in O.IM KCl


at 2 VSCE with and without the Addition of Various Substances"
Added substance(s)
None
HCOOHCOO- +0 2
HCOO- +H 20 2
H 20 2
O2

Current (nA)

H20 2 produced
(molecules/hole)

200
350
200
200
200
200

0.1
0
0.1
0.1

"Reprinted from Morrison and Freund [8921. with kind permission from Elsevier Science Ltd. The Boulevard. Langford Lane. Kidlington OX5 1GB. United Kingdom.

where A represents the two equivalent reducing species at the surface, which can be As 3+,
CW, HCOO-, methanol, or ethanol. A is oxidized by a hole to become a radical A+.
Because of its closer position to the conduction band edge, A+ is further oxidized by
injecting an electron into the conduction band to generate the current doubling effect.
The occurrence of current doubling on a ZnO electrode depends also on the presence
of other oxidizing chemical species. Table 4.4 shows the anodic currents measured under
illumination at 2 V SCE in O.IM KCl solution, with or without an oxidizing agent [892]. As
seen in the first row, hydrogen peroxide is produced when no active chemicals are present,
and the yield is one molecule per 10 holes. With the addition of O.IM formate ions, a
current doubling agent H20 2 is no longer produced and the anodic current is almost
doubled. The third and fourth rows show that in the presence of oxidizing agents the
current doubling effect offormate disappeared. According to Morrison and Freund [892]
with no active chemicals present in the electrolyte, the hole p+ oxidizes the lattice oxygen
of the ZnO according to
(4.17)

When formate is added, the holes now react with formate ions:
(4.18)

The radical HCOO is a highly reactive form, and it spontaneously becomes oxidized by
injecting an electron into the conduction band of the ZnO:
(4.19)
Therefore, for each hole reacted, an electron is injected, leading to current doubling. If
an oxidizing agent is introduced into the system (third row in Table 4.4), the free radical
reacts with the oxidizing agent rather than injecting an electron into the conduction band:
(4.20)

ELECTROCHEMISTRY OF ZINC OXIDE

113

Fujishima et al. [1171 proposed a different mechanism for the current doubling
process, with the dissolution of ZnO as the first step. They observed the formation of zinc
ions during current doubling on ZnO in O.4M KN0 3 solution containing HCOONa due
to the dissolution of the ZnO electrode.
The lifetime of the radicals generated during current doubling is short. As was
investigated by Cardon and Gomes [1032], the lifetime of the intermediate radicals
formed on ZnO during the oxidation reactions is not longer than 10-5 s.
The efficiency of current doubling is affected by the presence of nonreactive anions.
Micka and Gerischer [942) studied the effect of the presence of various anions on the
current doubling of zinc oxide in O.OlM HCI0 4 + 0.00 1M HCOOH. They found that many
anions have a suppressing effect on the current doubling reactions on ZnO. The suppressing effect on the photocurrent was found to decrease in the order 1- ::c: Br- > CI- >
H 2P04 > NO) > SO~- ::c: ClO:;. According to Micka and Gerischer, the results indicate the
importance of anion adsorption at the surface of a ZnO crystal in the reactions with holes.
Those anions which have a strong tendency to form complexes with Zn 2+ cations are
adsorbed preferentially and react with holes, preventing the oxidation of formic acid or
methanol. In strong alkaline solutions, current doubling disappears, also indicating a
competitive process of hole capture by OH- ions.
4.3.5.2. Photocatalysis. When a semiconductor is illuminated and electrons and
holes move to the surface, the electrons will reduce chemical species at the surface; the
holes will oxidize the species at the surface. The chemical changes will occur in the
medium in which the semiconductor is immersed. The process of producing these
chemical changes is termed photocatalysis [1177). For an n-type semiconductor the hole
current flowing to the surface is proportional to light intensity and is dependent on V, as
described by Eq. (4.14). The electrons and holes arriving at the surface may recombine
with no net chemical change in the solution. On the other hand, catalysis occurs when a
net chemical change results from the reactions of the electrons and holes wi th the chemical
species in the solution.
The photocatalytic effect ofZnO on chemical reactions in aqueous solution has been
extensively investigated by Morrison and Freund [921,1177). They found, as an example,
that formate ions only react with O 2 when a ZnO electrode is present. The electrons and
holes reaching the ZnO surface cause the reduction of oxygen and oxidation of formate
ions, respectively, according to the following sequence of reactions:
(4.21 )
(4.22)
(4.23)
(4.24)
The quantum efficiency of ZnO-catalyzed reactions depends on many factors.
Cunningham and Zainal [807) found quantum efficiencies ranging from 10- 3 to 0.29 for
reactions produced by UV illumination of zinc oxides suspended in solutions containing
agents such as NaN0 3 and KMn0 4 Hada et al. [944] reported quantum yields of

114

CHAPTER 4

0.24-0.39 for the reduction of Ag+ by the irradiation with 365-nm light of a ZnO surface
in aqueous solutions containing 5 x 10-5-1 x 1O-3M AgCI04 The electrons generated by
irradiation reduce the Ag+ ions adsorbed on the ZnO surface. Bernas [1188] found that
the quantum yield of H2 0 2 production in a zinc oxide suspension under illumination is
increased when appreciable quantities of interstitial zinc atoms are present in the oxide.

4.3.6. Electroluminescence
Electroluminescence is a process in which light of a certain wavelength is emitted
from a semiconductor electrode as a result of the electrochemical reactions. Electroluminescence has been observed on zinc oxide under both forward and reverse bias [798,
1038]. Under reverse bias, the light originates from the recombination of electron-hole
pairs either by direct band-to-band recombination, when the corresponding luminescence
is centered at 390 nm, or by recombination at self-activated centers intrinsic to ZnO, when
the light emission is broad and centered around 550 nm. Under forward bias, in addition
to the luminescence from the self-activated centers, a light emission is observed at 728
nm which is attributed to the centers activated by electron impact.
Luminescence may occur during reduction reactions on a zinc oxide electrode
[1035]. The light emission is attributed to homogeneous electron transfer between radical
cations and anions which are generated during reduction on the zinc oxide electrode.
According to Pettinger et al. [1036], these radical species have very high electron affinity
are normally required in order to inject holes into the deep-lying valence band of ZnO.
Although not stable in aqueous solution, short-lived species such as S04- and OR are

- R
4

I
I

I
I3
I

v, > E,

hv
I

ZnO

E, ----'-- - - semiconductor

e lectrolyte

AGURE 4.17. Band structure of a highly


doped semiconductor/electrolyte interface.
(I) Band-to-band tunneling under the condition V,2: Eg . (2) Reaction of holes with constituents of the electrolyte. (3) Generation of
radicals by the process. (4) Electron injection from the radical. (5) Recombination of
electrons and holes. After Pettinger et al.
[946].

115

ELECTROCHEMISTRY OF ZINC OXIDE

TABLE 4.5. Flatband Potentials, Efb , of Non-Single-Crystal ZnO Electrodes in Various


Solutions
Solution
INKOH
INKOH
O.IMNaOH
O.IM Na3P04 + O.IM Na2HP04
0.02MNaOH

Fonn of zinc oxide


Anodic film
Polycrystalline
Anodic film
Anodic film
Particulate films

Efb

(SCE)

-0.86
-0.86
-0.77
-0.69
-0.6

Reference
423
423
484
526
580

found to have a long enough lifetime to inject holes into the valence band of ZnO when
they are generated by an electrochemical reaction.
Anodic luminescence is associated with breakdown such that an anodic reaction
becomes possible in the dark [798, 946]. The emission intensity increases with current,
and the maximum emission intensity lies at a wavelength of 390 nm, corresponding to
the energy of the band gap. According to Pettinger et al. [946], as shown in Fig. 4.17, the
anodic breakdown of zinc oxide at high anodic potential is due to two tunneling processes:
one involves a band-to-band tunneling from the valence band to the conduction band, and
the other involves an interface tunneling from an electron donor in solution to the
conduction band of the semiconductor. The holes generated by the band-to-band tunneling are consumed in the anodic dissolution of the zinc oxide. The photoemission occurs
only when the two tunneling processes proceed simultaneously because the emission of
photons is only possible if both kinds of charge carriers are available for the recombination
to occur.
4.4. THIN ZnO FILMS
Thin zinc oxide films can be formed by anodization, by heating of metallic zinc, or
by physical and chemical deposition [113,423,788,948]. Also, the passive films formed
on a zinc surface under many conditions are essentially thin zinc oxide films. Compared
to the physical characteristics of bulk single-crystalline ZnO, those of thin-film ZnO,
being polycrystalline in general, vary greatly. For example, it has been reported that the
conductivity of a thin oxide film formed by thermo-oxidation depends strongly on the
environment because of its large surface-to-bulk ratio [788].
The tlatband potential and band gap of thin ZnO films are generally found to be
similar to those of a single crystal of ZnO. Table 4.5 shows that the tlatband potentials
determined for zinc oxide films and other non-single-crystal zinc oxides are close to that
of bulk ZnO (see Table 4.2). Similar results are shown in Fig. 4.18, reported by Bothe et
al. [526], for the tlatband potentials at anodic passive films formed in solutions with pH
values ranging from 8.4 to 12.9. An almost 60 mY/decade pH dependance indicates that
the potential drop in the Helmholtz double layer is primarily due to the adsorption/desorption of W or OW ions. Figure 4.19, reported by Burleigh [423], illustrates that the
variation of quantum yield with photon energy for anodic films formed in 1M KOH
solution depends on the formation potential. The film formed at -0.5 Y has a similar band
edge to that of polycrystalline zinc oxide. Photocurrent has been observed for a very thin

116

CHAPTER 4

- 0 . 4 , - - - - - - - - - - - -- - - - - - - - - - - ,
Zinc oxide film
. . ZnO single crystal

'""UJ -0.6

~
re

w -0.7

-0.8
-0.9

L..-_ _--'-_ _ _- ' -_ _ _- ' -_ _ _L...._ _......1._....1

11

10

12

13

pH

FIGURE 4.18. Dependence of the flatband potential Etb on the solution pH for ZnO film and ZnO single crystal
obtained from the extrapolation of the corresponding Mott-Schottky plots. Reprinted from Bothe eT al. [526],
with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington OX5 1GB, United
Kingdom.

(10-20 A) passive film fonned on a zinc surface in 1M NaOH + O.IM borate solution of
pH lO.5 [1184].
Scholl and Prentice [354,422] measured the photocurrent spectra of the passive films
formed on a zinc electrode in KOH solutions. They found that the photocurrent spectra
at the rest potential exhibit a strong ultraviolet response with a shoulder tailing into the
visible region as illustrated in Fig. 4.20. The shoulder is attributed to the existence of
interband energy levels characteristic of an amorphous film. There are clear differences
in the photospectra of the passive film formed in different potential regions, indicating
differences in composition and structure of the films.

- ZnO polycryslal (1 x)

0.4

.Zn: o.so Vsce (100x)

(5

.s=
a.
C;;

+ Zn: 1.00 Vsce (1oox)

0.3

Q)

~
"C

a;

0.2

';;'

E
:::I

a'"

0.1

:::I

Photon energy (eV)

FIGURE 4.19. Photospectra for anodic oxide films on zinc formed at - 1.00 VSCE (4 hand id =5.8 mAlcm 2 )
and -0.50 VSCE (2.5 hand id =0.02 mAlcm 2) and polycrystalline ZnO. From Burleigh [423]. Copyright by
NACE International. All Rights Reserved by NACE; reprinted with permission.

117

ELECTROCHEMISTRY OF ZINC OXIDE

g.--------------------------------------,

c~o _ :)

N. t':

... C~t _ O N.

.....

OIl

-t ."IO Y, P.C .

-1.40e v, P.C .
C'I .O H. E -1.<00 V. s.c.
~

t::
<!)

.....
.....
;:\
u

80

..
0

..c::

0..
"0
<!)

~
0

oa
. t:::l

8.....
0

100

.00

~oo

0
0

0
N

eoo

700

800

Wavelength (run)
Fl(JURE 4.20. Photocurrent spectra at the rest potential (Ere") in stagnant O.SN and I.ON KOH. T = 293 K, for
polycrystallinc WC) and single-crystal (S.C) zinc. The solid line is a B-spine fit of the circles. After Scholl ~t
ul14221

According to Scholl and Prentice [3541, the photocurrent measured on zinc passive
films in alkaline solution is linearly dependent on the illumination power over two orders
of magnitude. The photocurrent spectra show that the band edge corresponds closely to
that of the bulk material, which suggests that the passive film is largely composed ofZnO.
Hotchandani and Kamat 1580J investigated the photoelectrochemical behavior of ZnO
particulate films. They found that the photoelectrochemical properties are very similar to
those of hulk ZnO. At wavelengths shorter than 320 nm, most of the incident photons are
absorbed by the film. The efficiency of the incident photo-to-current conversion at this
wavelength was found to be 15%.
The electrochemical properties of thin zinc oxide films vary greatly because of the
variations of composition, structure, and morphology in films formed under different
conditions. Novak and Szucs 1796] found that the solid layer formed on the surface of
the zinc electrode during anodic polishing in an organic acid exhibits the characteristics
of an n-type semiconductor. According to Deslouis et at. [113], the photoelectrochemical
effects induce the dissolution of ZnO or the dissociation of H20 hut do not directly alter
oxygen reduction on the ZnO film. Sengupta and Chatterjee [948] found that a zinc oxide
film electrochemically deposited on a zinc surface has a very high surface state density,
which causes a pinning of barrier light. Fruhwirth et al. [847] found a linear relation with
a slope of unity between the photopotential and the electrode potential for a single-crystal
ZnO electrode in O.IM Na 2 B4 0 7 lOH1 0. By comparison to the single crystal, the zinc
oxide film formed on a corroded zinc surface yields much smaller photopotentials, and
the potential dependence is not linear. The smaller photopotentials of a corrosion product
layer is attributed to the high doping and thinness of the layer.

118

CHAPTER 4

TABLE 4.6. Donor Concentrations, ND, and Flatband


Potentials, E fb' Determined from the Linear Portion of
Mott-Schottky Plots a

ND

E fb
(V)

(I020/cm 3)

Sample
Zn
Zn-O.4%Co
Zn-0.6%Co
Zn-1.0%Co
Zn-I.2%Co
Zn-0.2%Ni
Zn-1.2%Ni
Zn-6.0%Ni
Zn-12.0%Ni

2.8
3.3
4.1
5.0
5.7
3.9
5.1
8.7

-0.77
-0.76
-0.75
-0.74
-0.73
-0.75
-0.69
-0.70
(-0.70?)

"Ref. 484.

Passive films formed on Zn-Co and Zn-Ni alloys are found to exhibit similar
properties to ZnO films [373,483,484]. These films behave like n-type semiconductors.
Table 4.6 shows the flatbandpotentials and dopant concentrations determined from
Mott-Schottky measurements for various Zn-Co and Zn-Ni alloys. The donor concentration of the passive films is very high and appears to increase with increasing Co and
Ni concentrations. The band gap of the passive films was found to be a function of the
electrode potential as shown in Fig. 4.21 [484]. The physical structure and the degree of
surface inhomogeneity of the passive films on Zn-Ni and Zn-Co alloys were investigated
by Juttner and Lorenz [373,483] using electrochemical impedance spectroscopy.

3 .35 ,---------------------------------------,

3.3 f-

>
~3.25
w'"

3 .2

"

Zn l .2% Co
... Zn l .2% Ni

" l",

+ Zn

."''''

. ZnO single crysta l

"''''",
'" '"
... tt++ "'", "'''''''
+ +
+ +
,

. ............ . ... . ... .......... '+'-'1-" '+'"

3 . 15 L-----------~-------------L------------~

-1

(Vsce)

FIGURE 4.21. Band gap energy, Eg , as a function of the electrode potential of a ZnO single crystal and of
passive layers on Zn, Zn-1.2% Co, and Zn-1.2% Ni. After ViJche et al. [484).

ELECTROCHEMISTRY OF ZINC OXIDE

119

4.5. STABILITY
4.5.1. Conditions of Stability and Decomposition Reactions

The stability of ZnO in aqueous solution is a function of pH according to the


pH-potential diagram (Fig. 2.2). ZnO is thermodynamically stable in the pH range
between 6 and 12. In solutions with low or high pH values, ZnO is not stable and dissolves
owing to the relatively high solubilities of zinc in these pH ranges. In acidic solution the
dissolution of zinc oxide in the dark is attributed to both chemical and electrochemical
reactions [1187]. The chemical dissolution is due to direct reaction with hydrogen ions:
(4.25)
and the electrochemical reaction is
(4.26)
This reaction is electrochemical because it depends on the electrode potential to regulate
the adsorption of protons on the surface of the zinc oxide. In solutions other than acidic
or strongly alkaline ones, zinc oxide is usually stable under anodic polarization but
dissolves when breakdown occurs under high anodic potentials.
Zinc oxide may not be stable under illumination and may decompose according to
(4.27)

leading to dissolution of the semiconductor.


At the open-circuit potential, the photoinduced decomposition is accompanied by a
reduction reaction such as 2H+ + 2e- ~ H2 through a conduction band process. The rate
of decomposition can be increased with the addition of oxidizing agents to increase the
rate of electron consumption. The decomposition rate increases with anodic potential as
the concentration of photogene rated holes increases with the potential (Fig. 4.15).
According to Gerischer [129], the overall reaction involved in the photodecomposition of ZnO consists of the following steps:
ZnO + 4hv

ZnO + 4e- + 4p+

Hole + electron generation


First hole trapped on surface (slow)
Second hole trapped (slow)
Formation of oxygen molecule (fast)
Zn 2+ leaving surface (fast)

The whole reaction:


2ZnO + 4hv ~ 2Zn;~ + O 2 + 4e-

(4.28)

120

CHAPTER 4
E(NHE)

-I

ZnO + 2H- + 2e- -Zn + H, O

2H' + 20'" H,
ZoO + OH- + 2h- - Zn(OH)" + 1120,
H,O + 2h ' '" 2 H" + 1120 ,

3_2 V

ZnO
2

E.----~----------~--

rsem iconductor

L,
electrolyte

interface
FIGURE 4.22. The relative positions of various redox couples with respect to the edges of bands at the ZnO
surface_

The instability of a ZnO electrode under illumination in aqueous solutions is


essentially due to the fact that the decomposition potential of ZnO by holes is more
negative than the edge of the valence band E,,, as shown in Fig. 4.22 [941, 1176].
Depending on the decomposition product, the decomposition potential may vary according to the reactions:
(4.29)
1

ZnO + OW + 2p+ ~ Zn(OHt + 202

E~ = 0.52 VseE

(4.30)
(4.31 )

Figure 4.22 shows that the anodic decomposition potential is lower than the redox
potential for water oxidation and thus the decomposition is a thermodynamically more
favorable reaction. ZnO decomposition is normally the predominant reaction for hole
consumption at low current densities. However, water oxidation occurs as a parallel
reaction at high current densities [946].
According to Fig. 4.22, ZnO is stable against cathodic decomposition by electron
reduction because the position of the decomposition potential is above the conduction
band edge. Cathodic decomposition can only occur at high cathodic polarizations when
the conduction band edge is shifted to above the position for the cathodic decomposition
reaction.

121

ELECTROCHEMISTRY OF llNC OXIDE


50 ~---------------------------------,0.1

Eu
..:

::L

;!-

-50

'c;;
c:
(J)

-0

!----.......----

o ~----__~------------~====~

Dissolution fate

c
'N

0.05 "0
(J)

"

-100

c:

.2
"S

:s

0-150

"0
<J)
<J)

6
-200L-----~----~------~----~------~0

-2

-1

Potential (V seE)

FIGURE 4.23. Correlation of zinc dissolution rates with the current-potential relation for an 8.5-Qcm lnO
single crystal in the dark in l.OM KOH. After Justice and Hurd [475J.

4.5.2. Rate of Decomposition


The dissolution rate ofZnO depends on many factors. Justice and Hurd [475] found
that the dark dissolution rate of ZnO in 1M KOH is relatively low and is essentially
independent of anodic potential up to 3.0 VSCE as shown in Fig. 4.23. The change of the
cathodic current has no effect on the dissolution rate. On the other hand, under illumination, as shown in Fig. 4.24, the dissolution rate is much higher and increases proportionally to decreasing cathodic current or increasing anodic current. In 0.5M KOH solution
the dissolution rate is lower compared to that in 1M KOH but the photocurrent is larger,

100,------------------------------ - - , 0.15

1:

i~

"'E

Cl

O. I

E
0

c:
'N

"0
(J)

A Di:ssoh.nion ralt

0.05

Poh~"iz.ation cutvl

"
c:

.2

"S

"0
(f)

6'"

_200L-~--~----~----~------L-----~ 0

-2

-1

Potential (V SCE )

FIGURE 4.24. Correlation of zinc dissolution rates under illumination with the current-potential relation for
an 8.5-Qcm lnO single crystal in l.OM KOH. The open-circuit potential is -650 mY SCE' After Justice and
Hurd [4751.

122

CHAPTER 4

indicating that the photocurrent is not fully generated from the ZnO dissolution process.
The dissolution in the dark is primarily a chemical process which depends only on the
concentration of OH- ions:
ZnO + 20H- + HP ~ Zn(OH)~-

(4.32)

The dissolution under illumination according to Justice and Hurd [475], the amount of
zinc actually measured in the solutions, is always greater than the amount calculated from
the anodic currents. This indicates that under illumination a photo-assisted dissolution
occurs in addition to the chemical dissolution process. Thus, the dissolution rate of ZnO
is a function of both OH- and hole concentrations at the surface, which are a function of
both potential and light intensity.
The photo carriers generated under illumination are consumed either for ZnO
decomposition or for water oxidation in the absence of other redox couples. According
to Fruhwirth et al. [1187], the contribution of each reaction is a function of pH. At a pH
of about 9.2, water oxidation is the predominant reaction because the saturation concentration of Zn 2+ ions is about 10-7 molll and will be reached immediately after immersion
of the ZnO electrode. With decreasing pH, the solubility of Zn 2+ ions is increased, and
the contribution of ZnO decomposition to the photocurrent increases. Fruhwirth et al.
[1187] found that the photocurrent displays a linear dependence on pH in a nonsaturated
solution whereas it is independent of pH in a Zn 2+-saturated solution. Erbse et al. [1278]
found a linear relation between the dark dissolution rate and pH in the pH range between
4 and 6.
Gerischer and Sorg [1186] reported that the dissolution rate of zinc oxide in 0.5M
KCl solution decreases with increasing pH in the acidic range. In the alkaline pH range,
the presence of NH3 leads to a pronounced increase of the rate, which is followed by a
decrease at pH 10-13. The dissolution rate depends on the type of anion in the solution
and increases proportionally to the square root of the electrode rotation speed. The latter
dependence is attributed to the diffusion of H+ ions. Gerischer and Sorg also found that
the dissolution rate is independent of pH (independent of band bending) in the presence
of acetic acid as the proton donor. It was thus concluded that the dissolution rate of ZnO
is controlled mainly by chemical processes and that the electric forces at the interface
play only a limited role in the reaction kinetics. According to Gerischer and Sorg, the
rate-determining step in the dissolution of ZnO in the dark is hydrolytic bond splitting
between Zn atoms in kink sites and 0 atoms in lattice sites, catalyzed by H+ ion donors,
on the one hand, and by molecules or anions that coordinate to the Zn 2+ions, on the other.
The dependence of ZnO decomposition on crystal orientation was interpreted by
Morrison [1177] in terms of hole capture, mainly by different hydroxyl groups on the
different crystalline faces. Because of the structure of the wurtzite crystal, only Zn ions
are exposed on a (0001) surface; any residual oxygen over the surface of the zinc plane
is weakly bonded and will be dissolved into the solution. Then a preferred adsorption of
OH- is expected for this plane because the surface sites are all cationic. Holes coming to
the (0001) surface may interact and oxidize these adsorbed OH- ions, but lattice ions will
not be oxidized. On a (OOOT) face, however, surface oxygen ions are exposed. For this
plane, preferred adsorption of H+ is expected, giving OLH- or O~- groups, where OL is a
lattice oxygen ion. Holes coming to the surface will then be able to oxidize the lattice

ELECTROCHEMISTRY OF ZINC OXIDE

123

100

+ Single

crystal

.. Sinter

'0
<::
o

"fii
"'C
. 50
4>

0;

a.
E
o

(.)

o~----~--------------------~~--~

0.0001

0.001

0 .01

Concentrat ion of I' (mol - dm-')

FIGURE 4.25. Percentage of competitive oxidation of ron ZnO electrodes as a function of 1- concentration.
Arter Kobayashi et al. [9181.

oxygen to a -1 valence or to the zero valent (0 2) state, enabling lattice oxygen to evolve,
followed immediately by the now weakly bonded underlying zinc ions. Thus, the (0001)
plane corrodes much more readily than the (0001) plane.
The anodic photodecomposition of zinc oxide can be quenched by competitive hole
capture with the addition of a redox couple that has a potential more negative than the
ZnO decomposition potential [1186]. Kobayashi et al. [918] reported that the percentage
of hole capture by anodic decomposition of ZnO is reduced by addition of 1- in the
solution, as shown in Fig. 4.25. The ratio of hole capture through r reduction versus that
through ZnO decomposition increases with r- concentration but decreases with photocurrent (Fig. 4.26). A moderate effect is observed with the addition of Br- ions, and a very

100

~
.:....

'0
<::

Electrolyte

"'C

'x
0

concenlr,

50

CD

.~

(mol o dm 3)

+10- 2
.. 10 .3

'~

a.
E

+ 3.16 X 10-4

(.)

10-4

0
1 E-06

Disk photocurrent, A

FIGURE 4.26. Percentage of competitive oxidation of 1- on a sintered ZnO electrode as a function of the disk
photocurrent and C concentration. After Kobayashi et al. [9181.

124

CHAPTER 4

small effect with the addition of Cl- ions. The effectiveness of 1- in quenching the
decomposition of ZnO in aqueous solution is due to its more negative redox potential
compared with the anodic decomposition potential for ZnO. On the other hand, the redox
couples Br-/Br and Cl-ICl are located at potentials more positive than the decomposition
potential and therefore are not effective in competing for hole capture.

5
Corrosion Potential and
Corrosion Current
5.1. INTRODUCTION
Corrosion potential is a mixed potential (also an open-circuit potential or rest
potential) at which the rate of anodic dissolution of the electrode equals the rate of
cathodic reactions and there is no net current flowing in or out of the electrode. Corrosion
current is the dissolution current at the corrosion potential.
Corrosion potential and corrosion current are two important parameters which
connect the fundamental electrochemistry and the practical corrosion behavior of metals.
The value of the corrosion potential indicates the state of a corroding metal while that of
the corrosion current reflects the instantaneous corrosion rate at the time of measurement.
In this chapter, the fundamental relationship between corrosion potential and corrosion current is described, and the various methods for measuring the corrosion current
and their limitations are briefly discussed. The corrosion potentials and currents are then
presented according to the specific effects of the various material, solution, and measurement factors. Lastly, the effect of time and the correlation of corrosion current to real
corrosion loss data are discussed, and the errors which may be generated in determining
corrosion currents are analyzed. Since in most cases oxidation and/or dissolution is the
dominant anodic reaction on a zinc electrode and in most studies the corrosion potentials
are determined by measuring the open-circuit potentials, no distinction is made in the
discussion regarding the corrosion potential versus the open-circuit potential or the rest
potential.
5.2. RELATION BETWEEN CORROSION POTENTIAL AND CORROSION
CURRENT
The basic theory on the relationship between corrosion rate and electrochemical
polarization resistance was formulated in the 1950s [693,694]. According to the theory,
for a corroding metal there are two coexisting electrochemical reactions: the dissolutiondeposition of the metal, M r + + re- ~ M, and a reduction-oxidation of a species in the
electrolyte, Z"+ + ne- ~ Z. Each of these reactions has its own exchange current and Tafel
125

126

CHAPTER 5

-0.1

#
~

>
0

-0.3
ACTIVE
0.01

LO

0.1

Current density

10

100

{J.IA/cm 2)

FIGURE 5.1. Relationship between overvoltage and current for a corroding electrode system consisting of two
coexisting electrochemical reactions. After Stern and Geary [694].

slopes so that the corrosion of the metal occurs at a potential, the corrosion potential, at
which the total rate of oxidation equals the total rate of reduction:
+-

(5.1)

<--

iz + i", = i: + i",
f--

---?

f-

where i", is the rate of reduction of Mr+ and i", is the rate of oxidation of M, and iz and iz
are the rates of reduction and oxidation of species Z, respectively. The corrosion current
Ls th!; difference between the dissolution and deposition currents of M, namely, icorr =
i", - im . Figure 5.1 illustrates the potential-current relationships for a mixed electrode
system. The rates of each reaction under activation control are:

~ = i~ exp(-2.31J /fJJ

(5.2)

~ = i~ exp(2.31J /fJ)

(5.3)

-->

(5.4)

+-

(5.5)

im = i~, exp(-2.3IJm1fJm)
ill! = i ~ exp(2.31J ml Pm)

CORROSION POTENTIAL AND CORROSION CURRENT

127

where i~ and i~, are the exchange currents and", and" '" are the overpotentials for the
reactions Z"+ IZ andM r +1M, respectively, and /J, and /J", are Tafel slopes for the two
reactions, assuming the same values for the forward and backward directions. When the
corrosion Eotential, Ecoff ' is sufficiently removed from the equilibrium potential of the
reactions, i, and 7,,, become insignificant in comparison to 7, and 7,,,, and the corrosion
current can then be expressed as
<-

-->

(orr = im = i:

At any potential E deviating from Eco", the anodic dissolution current can be described
by an expression analogous to the rate expression for a single redox couple:
<--

=im -

->

iz

= icorr[exp(2.3" I/J",) - exp(-2.3" I/J,)J

(5.6)

with" = E - ECOIT"

5.2.1. Polarization Resistance and Corrosion Current


For small values of", exp(2.3" I {Jill) and exp(-2.3" I /J,) may be approximated by I
+ 2.3" I/J", and I - 2.3" I/J,. Equation (5.6) thus reduces to
(5.7)
Differentiating" with respect to i in Eq. (5.7), one obtains
(5.8)

or
icnrr = BIR"

(5.9)

with B =2.3(/J, + /J",)/(/Jfl",), and R" =d" Idi'l~o.


RI' is the polarization resistance, and B is called the Stern-Geary constant. It can be
seen in Eq. (5.9) that the corrosion current for a corroding metal can be determined by
measuring the polarization resistance, RI" at the corrosion potential when B is known. The
polarization resistance is most commonly determined by the linear polarization technique.
In linear polarization the slope of the polarization curve, the polarization resistance,
is determined within an overpotential range close to the corrosion potential. The constant
B is determined from the anodic and cathodic polarization curves. Figure 5.2 shows
examples of linear polarization curves of zinc in three different solutions [1101.
The polarization resistance can also be measured by impedance techniques. The
determination of polarization resistance by impedance techniques is described in Chapter
2. Compared to the linear polarization technique, the advantages of the impedance
techniques are that mechanistic information is obtained, that the measurement is independent of the solution conductivity, and that electrode capacitance is also obtained along

128

CHAPTER 5

Eu

"NaCI

+ NaCI04

bNa2S04

2-

-3
-4

-5

-4

-3

-1

-2

E (mV SCE )
FIGURE 5.2. Linear polarization plots in the vicinity of the corrosion potential for zinc in molar Na2S04,
NaCI04, and NaCI solutions at pH 5.8. Reprinted from Baugh [110], with kind permission from Elsevier Science
Ltd, The Boulevard, Langford Lane, Kidlington OX5 1GB, United Kingdom.

with Rp [718]. The disadvantages are that impedance techniques are more complicated
and that the Tafel slopes cannot be obtained from an impedance spectrum.
Other electrochemical techniques are sometimes used for measuring polarization
resistance. Boto and Williams [114] used a differential pulse method to determine the
polarization resistance of zinc in aerated NaCl solutions. In this method a short potential
pulse of several millivolts is applied, and the current response is measured. 11ll1i is then
taken as the polarization resistance Rp. It was reasoned that because the polarizing pulse
is applied for only a very short time, less than 60 ms, the perturbation to the corrosion
system is minimized.

10,000

... 80 2-

1,000

CI-

NE
u

"-

100

.-

10

1~----~----~----~------~~--~~--~

1.04

1. 06

1.08

1.1

1. 12

Potential (Vsce )

~.1

1 .16

E corr

FIGURE 5.3. Anodic polarization of zinc in 1M NH4CI and 1M (NH4hS04 solutions. After Dattilo [33].

CORROSION POTENTIAL AND CORROSION CURRENT

129

In addition to the methods for determination of polarization resistance, corrosion


current can also be determined by Tafel line extrapolation of the anodic or cathodic i- V
curves to the corrosion potential as illustrated in Fig. 5.3 [33]. The underlying methodology for this method is straightforward. The anodic dissolution is described by
ia = icorrexp(2.31J / Pm), and thus at the corrosion potential IJ = 0 and i" = iCOJr There is
generally good agreement between the corrosion rates determined by the extrapolation
method and by the polarization resistance method for zinc in various electrolytes [33,
110,534].
Corrosion current is meaningful only for evaluating certain corrosion situations
where the corrosion is essentially uniform over the whole surface area. It may result in
large errors if it is used for evaluating localized corrosion, such as pitting corrosion or
intergranular corrosion. The corrosion loss caused by localized corrosion must be
evaluated with other parameters. Localized corrosion is discussed in Chapter 7.

5.2.2. Conversion Factors


In practice, corrosion rates are most often measured with methods other than
electrochemical techniques and are expressed in terms of weight loss or thickness loss in
units of milligrams per square decimeter per day or microns per year, etc. The conversions
between corrosion current density and corrosion rate expressed in terms of weight loss
and thickness loss rates are
i=-W

nF
m

(5.10)

pnF
m

(5.11 )

and
l=--T

in which Wand Tare the corrosion rates in weight loss rate (b/cm"s) and thickness loss
rate (cmls), respectively, p is the density of the metal, n is the charge per atom, m is the
molar mass weight, and F is the Faraday constant. For zinc, p = 7.3 g/cm 3, n = 2, and m
= 65.4 g/mol; thus,
1 j1A/cm"

=4.64 x 10-7 j1m/s = 14.5 j1m1yr = 2.92 mg/(dm2day) (mdd)

or
1 j1m1yr = 0.07 j1A1cm 2
In the case in which the cathodic reaction is mainly due to hydrogen evolution, the
extent of corrosion can also be measured by the amount of hydrogen collected during the
corrosion process. For zinc at room temperature and at atmospheric pressure, the conversion between the corrosion current and the volume of hydrogen is [311]:
I j1A = 0.4 ttl/hr

130

CHAPTER 5

In North America, weight and thickness for zinc coating are commonly expressed in
English units. The conversions between the different weight and thickness units are:
I ozlfe =305 g/m2 =30.5 mg/cm 2 = 1.70 mil

=43.2 ,urn

From a practical point of view, 1 ,uNcm2 is a rather large corrosion rate. The typical
corrosion rate of zinc in most atmospheric environments is no more than a few microns
per year, which is only a small fraction of 1 ,uNcm2
5.3. CORROSION POTENTIAL AND REACTION KINETICS
As can be noted in Fig. 5.1, corrosion potential is the potential at which the total
anodic dissolution current equals the total cathodic current. Any changes in conditions,
such as surface preparation, solution composition and concentration, pH, temperature,
time, convection, aeration, etc., that affect the anodic and/or the cathodic reactions will
affect the value of the corrosion potential. Thus, the corrosion potential as well as the
corrosion current of a metal can vary greatly, depending on the specific conditions. Table
5.1 lists the corrosion potentials and corrosion currents of zinc (or galvanized steel) in
various environments.
In solutions in which passivation does not occur or there is no oxidizing agent, the
corrosion potential of zinc generally lies in the vicinity of its equilibrium value. This is
primarily due to the high exchange current density for zinc dissolution and the relatively
much smaller exchange current density for the cathodic reactions on zinc. The equilibrium
electrode potential for zinc in acidic or near-neutral noncomplexing solutions is -1.122
VseE, assuming that the concentration of Zn 2+ ions in the solution is 10-4M, which is a
likely concentration level in an aqueous solution containing a zinc electrode without the
presence of Zn 2+ initially. From Table 5.1 it can be seen that most corrosion potentials
measured in various solutions are close to this value. The small differences in the values
among solutions of similar compositions and pH may be attributed to differences in
aeration and measurement procedures. Aeration usually affects the cathodic reaction rate,
whereas the measurement procedure may affect the surface condition of the electrode.
For example, time of immersion in the electrolyte may affect the surface concentration
of Zn 2+ ions and therefore the electrode potential of zinc. Large differences can occur
between the corrosion potential and the equilibrium value when the surface of zinc is
passivated or there are oxidizing agents in the solution.
A change in the corrosion potential in either the anodic or the cathodic direction may
correspond to a decrease or an increase in the corrosion current. The relative values of the
corrosion potential and their relation to the corrosion currents under various conditions
can be summarized using the schematic polarization curves in Fig. 5.4. In this figure,
the corrosion potential of an active zinc electrode in a solution is E~o", and
E ~oIT' E~o", E~o", and E~orr are the corrosion potentials under various conditions.
The corrosion potential of an active zinc surface is E~orr with I" and Ie as the anodic
and cathodic curve respectively. In the case in which the anodic dissolution is inhibited,
for example, by surface adsorption of a chemical species, the anodic curve becomes 2".
This will result in a more positive corrosion potential (from E~orr to E ~orr) if the cathodic
reaction remains unchanged. In such a situation, the corrosion current is reduced upon a

CORROSION POTENTIAL AND CORROSION CURRENT

131

TABLE 5.1. Corrosion Potentials and Corrosion Currents of Zinc in Various Solutions
Material"
G90

Galvanized
steel

Solution
INHCl h
INH 2S0 4
0.12MCr03
O.IMHCI

O.IMNaCI"
O.IMNaCl h
0.05M H2SO 4"
0.5MNH4C1"

G90

IMKCI
1M NaCI"
1M Na2S0/
IN NaClf
100 ppm NO"]
0.IMNa2S0/
0.IMNa2S0/
2.7MNaCI
IN NaCl h
0.IMNa2Cr04
O.IM Na2Mo04

0.IMNa2W04
6MNH 4C1
O.IM succinic acid
1M NaCI"
1M Na2S0/
O.IM Na2S04 + O.IM (NH 4)oSO/
I
O.IM (NH4)2S04'
O.IM (NH4)2S0/
0.5NNaCl t
1M (NH 4lzS0i
1M NH4C1

0.IMNa3P04
O.IMNaCI
3.5% NaCl d
3.5% NaCl h
1.0NNaCI"

pH
-0

I
2
2
3
3
3.8
3.8
4
4
4
4
4.8
5
5
5
5
5.3
5.6
5.8
5.8
5.8
5.9
5.9
6
6
6
6.2
6.3
6.4
6.4

1.0NNaCIf

3% Na2S0l
5% NaClf
0.IMNa2Cr04
99.2% Zn

(V SCE )

-1.00
-0.98
+0.05
-1.02
-1.057
-1.085
-1.07
-1.06
-1.03
-I.I
-1.09
-1.04
-0.65
-1.102
-1.125
-1.12
-1.04
-0.93
-0.88
-1.07
-1.17
-1.095
-1.09
-1.13
-1.07
-1.056
-1.143
-1.03
-1.15
-1.133
-0.65
-1.098
-1.187
-1.14
-1.07

120 p~m HCO;- + 10 ppm


S04-' NO;-

8.8

-1.07
-0.55
-0.89
-1.06
-1.11
-0.99
-0.995
-1.0
-1.12
-0.79

0.IMNa2Cr04
O.IM Na2Mo04

9
9

-0.57
-0.73

7.3

O.IMKNO/
O.IMKCI

0.IMNa2S0/
IMZnSO/

IMZnSO/
0.IMNa2S0/
Zn (0001)

Ecorr

1M (NH 4 lzS0 4 h

icorr "

(mA/cm")

Reference

400'
3'

14
250
59
1.8'
703
1129
1129
0.91'
64
0.053'
63
75
0.014,e
110
0.05 cc
110
110
196
1129
1129
0.009""
534
45
199
199
199
0.028"<"
534
0.8'"
114
0.009""
110
0.004""
110
128
701
701
5
0.017"'"
33
0.011c.C
33
710
0.08S'"
114
701
701
1.8 X 10-4 <" 176
0.0014<"
176
0.05'"
41
41
710
38
38
38
0.002'"
112
0.001<"
112
128
0.009'"
446
410
199
199
(continued)

132

CHAPTER 5

TABLE 5.1. (Continued)


Material a

Galvanized
steel

Zn (0001)
Galvanized
steel

Solution

pH

0.IMNa2W04
0.IMNa2S0/
0.IMNa2S0/
100pprnNO:J
NaClO/
O.OIMNaOH
O.OOIMKOH

9
9
9
9.5
II
11
II

0.IMNa2S0/
0.1 gil Ca(OH)2
0.IMNa2S0/
Sat. Ca(OH)z
0.05M NaOH + 0.5M H 20 2
0.IMNa2S04
O.3M KOH + Zn(OH)~-"
0.5MNaOH
Sat. Ca(OH)z + 0.5N NaOH

II
11.1
12
12.6
12.7
13

0.5M NaOH + 0.08M H20 2


IN LiOH"
IN NaOH b
IN NaOH b
IN KOH"
3NNaOHb

13.5
13.6
13.7

>14

30% KOH"
5NKOH
6NKOH b

8NNaOH
9NKOHb

>14
>14
>14
>14

Ecorr (V seE)
-0.65
-1.077
-1.168
-0.78
-0.74
-0.11
-0.79
-1.141
-0.45
-1.171
-0.43
+0.01
-1.393
-1.43
-1.47

-0.6
-1.47
-1.46
-1.52
-1.47
-1.54
-1.49
-1.55
-0.73
-1.57
-1.61

icorr 2

(rnA/ern)

O.Ole

0.Dl5"

6 X 1O-4e

<0.04
0.06
0.02"

O.Ole

0.018 c
0.14'
0.012 e
0.2'
0.016
0.55 e

0.015 c

Reference
199
1129
1129
196
34
710
104
1129
197
1129
202
359
1129
12
446
202
359
311
311
790
311
790
25
311
10
790
311

" Unless otherwise specified. the material is of pure zinc.


b Deaerated.
C

Extrapolation from i- V curves.


Aerated.

, From polarization resistance.


Nondeaerated.

shift to a more positive potential. On the other hand, if the anodic dissolution kinetics
remain unchanged but the cathodic reactions is represented by curve 2c instead of curve
Ie (due to, for example, lowering of the pH), the potential also becomes more positive
(from E~orr to E~orr)' However, in this case the corrosion current is increased upon a shift
to a more positive potential.
The anodic curve becomes 30 when the surface is passivated. If the cathodic reaction
is unchanged, the corrosion potential of the electrode becomes E~orr' more positive than
E~om and the corrosion current is generally much smaller than that at E~orr"
Values of the corrosion potential that are much more positive than E~orr are usually
associated with both the passivation of the zinc surface and the presence of oxidizing

133

CORROSION POTENTIAL AND CORROSION CURRENT

_ .-2,
I,

---'
FIGURE 5.4. Schematic polarization curves illustrating Eeorr and icorr under various conditions
(see text).

rt-r-T1'--._

E' "'"

E'ool'T
J30"",

I,
i,

i.

agents (e.g., chromate ions or hydrogen peroxide) in the solution. The cathodic and anodic
polarization curves in such a solution are illustrated by 3c and 3", the coupling of which
yields the corrosion potential E~()ff" The corrosion rate in this situation is usually very low.
However, if the surface is not passivated by the presence of an oxidizing agent, the
corrosion rate can be very high. This can be appreciated by coupling curves la and3cIn a similar manner, a decrease in potential can be caused by either faster anodic
dissolution kinetics or a slower cathodic reaction. For example, a decrease in corrosion
potential due to deaeration of the solution is usually observed for a zinc electrode in a
neutral solution. Deaeration removes the dissolved oxygen and thus reduces the cathodic
reaction rate.

5.4. Ecorr AND

icorr

UNDER VARIOUS CONDITIONS

5.4.1. Effect ()f Zinc Ions

The electrode potential of zinc is a function of the zinc concentration in the solution.
Thennodynamically, as illustrated by the pH-potential diagram in Chapter 2 (Fig. 2.2),
the effect of zinc concentration on the zinc potential depends strongly on pH. It has been
found, as shown in Fig. 5.5, that the corrosion potential of zinc in sulfate solutions is
almost independent of the concentration of Zn 2+ at low pH values, but at pH > 5 the
corrosion potential starts to show a linear dependence on the logarithm of Zn 2+ concentration. The slope of this line is 2.3 x RTl2F, corresponding to the dependence of the
equilibrium potential of zinc on the concentration of Zn 2+, according to the Nernst
equation [4451. The insensitivity of the corrosion potential to changes in the bulk Zn 2+
concentration at low pH values is probably due to a high surface concentration of Zn 2+
resulting from the dissolution of zinc, which occurs rapidly at low pH.

134

CHAPTER 5

-1.-----------------------------------------,

- 1.05
pH 6108

pH 5.4

pH 4.45

pH 3 .3

pH 2 .7
'l

- 1.15

pH 2 .2

~------~--------~------~--------~
-0.5
-3 .5
-2.5
-1.5
-4.5

Log [ZnSO.l

FIGURE 5.5. Zinc electrode potential in O.IM sulfate solutions as a function of zinc ion concentration and pH
of the solution at a disk velocity of 9 rev/s and a temperature of 2Ye. After Gmytryk and Sedzimir [445].

In alkaline solutions the zinc potential is found to be a logarithmically linear function


of Zn(OH)~- concentration, as shown in Fig. 5.6 [131]. The deviation caused by the
hydrogen reaction on the zinc electrode is less than 1 mV owing to the fast rate of the
zinc reaction and the slow rate of the hydrogen reaction. The slopes of the lines in Fig.
5.6 are between 26.9 and 28.6 mY/decade for KOH concentrations between 3.1 and
1O.8M, which is very close to the slope predicted from the equilibrium equation
(5.12)

E= 0.441 - 0.118pH + 0_02951og aZn(OH)~

Bockris et al. [I2] found that the zinc potentials experimentally measured in 0.3-3M
KOH solutions containing various amounts of zincate are 20 to 30 mV more positive than
those calculated according to the equilibrium described by Eq. (5.12)_
-1 _34

r-----------------------------------------,

-1.36

oOl -1.38

:r:

3.1 M OH
K

-1.4

'" 4 .8 M KOH

w -1.42

7.7 M KOH
y

10.8 M KOH

-1.44

-1 . 46~--------~---------L----------~------~

0 .001

0.01

0.1

10

K.Zn(OH), Concentration (M)

FIGURE 5.6. Zinc electrode rest potentials as a function ofZn 2+ concentration in KOH solutions at 25e. After
Isaacson et al. [131].

CORROSION POTENTIAL AND CORROSION CURRENT

135

-0.75,--- -- - - - - - - - - -- - -- -- - ,

-0.85

c:QI

(5
c.

-0.95

Sr -

QI

... CI-

+ NOj

'0

QI

-1. 05

" 1-

X 50 2 4
CI0 .j

10"8

10

.,

10

FIGURE 5.7. Potential-log C curves for zinc in solutions of corrosive anions. Reprinted from Gouda et al.
l 1139], with kind permission from Elsevier Science Ltd. The Boulevard, Langford Lane. Kidlington OX5 1GB,
United Kingdom.

The addition of zinc ions to a solution generally reduces the corrosion current of the
zinc electrode. For example, it has been found that the corrosion current of zinc in 1,8M
KOH solution is 45 I1A/cm 2 , but the addition of 0.03-0, 1M zincate ions to this solution
decreases the corrosion rate to below ll1A!cm 2 [12],

5.4.2. Effect of Anions and Cations


According to Gouda et ai, [1139], anions can be classified into two main groups
based on the mode of variation of the steady-state corrosion potential of zinc with anion
concentration, In the first group, to which SO~-, CI-, 1-, Br-, CIO:!, and NO) belong, an
increase in anion concentration is accompanied by a decrease in the potential of the zinc
electrode as shown in Fig. 5.7. The anion concentration and the corrosion potential follow
the relationship
Eeorr = a - b log C

(5.13)

with C the concentration, and a and b constants. For SO~-, Cl-, Be 1-, and CIO:!, b is 33
mY/decade; for NO), it is about 15 mY/decade. The lower slope for NO) is probably
related to the slight oxidizing character of the nitrate ion. The reduction in the corrosion
potential with anion concentration was suggested by Gouda et al. to be the result of an
altered balance between anodic and cathodic areas set up by the competing influence of
oxide film formation and destruction by the anions. According to Awad and Kamel [784],
the adsorption of the anions on the bare anodic areas of the metal surface accelerates the
ionization of the metal, perhaps through reduction of the activation energy of the process.
The potential decreases with an increase in the anion concentration in order to increase
the cathodic reaction rate.
In the second group of anions, to which CrO~-, H2PO;;, WO~-, HPO~-, and NO;
belong, the corrosion potential depends on the type of anion but not on concentration at
low concentrations; it increases with concentration at high concentrations as shown in

136

CHAPTER 5
-0.45 , - - - --

- - - - - -- - --

- - -- - - - - ,

H2 PO';
Q;
u

G -0.65
~

HPOt

c:

X N0 2

'"

(5

a.

'"

'0

<:5

CrOt

+ wot

If)

-0.85

'"
UJ

.,

_' . 05L----L----~----L----L----~----~--~--~
~3

10

10

-2

Log C (M)

10

10

10'

FIGURE 5.8. Potential-log C curves for zinc in solutions of corrosion-inhibiting anions. Reprinted from Gouda

et al. [1139], with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington OX5
1GB, United Kingdom.

Fig. 5.8 [1139]. The more positive corrosion potentials in these solutions are due to the
surface passivation taking place at certain concentrations, depending on the type of anion.
The corrosion potential of a passivated zinc electrode is also sensitive to the presence of
other ionic species as shown in Fig. 5.9 [59].
The corrosion potential in phosphate solutions varies with the concentrations of the
primary, secondary, and tertiary phosphate ions. Figure 5.10 shows the corrosion potential
as a function of the concentrations of NaH 2P0 4, Na2HP04, and Na3P04' Generally, a shift
of the potential to more positive values takes place at a certain phosphate concentration
which depends on the solution pH. According to Awad and Kamel [784], this potential
0.2.---------------------------~

--0J,!', ,;:06L.------------l NaN03

______

~~

_ -0.2

0 .3

Q)

If)

G -0.4
n;

'E

'"

-0.6

(5

c.. -0.8 ~--_----------70i-,0~i-6------------------.:

-1.2

j NaCI
JNa"SO,

0,3

-1 ----

0,03
0.3

L...._ _ _ _...L..._ _ _--'-_ _ _ _ _ _ _ _-'--_ _ _ _ _- ' -_ _- '

10

20

30

40

Time, hours

FIGURE 5.9. The effects of several anions on the potential-time curves for zinc immersed in 0.12-molll Cr03
solutions at 20C. The concentrations of the anions, in moles per liter, are labeled on the plot. Reprinted from
Williams [59], with kind permission from Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam,
The Netherlands.

137

CORROSION POTENTIAL AND CORROSION CURRENT

-O.5r-----------------------------------------,
Na,HPO.

-0.7

+ NaH,PO. Na,HPO. mix


Na,PO.

~ -O.9

LoU

- 1. 1

_1.3L-------~-------L------~--------L-------~

-4

-3

-2

-1

Log C (M)

FIGURE 5. I O. Effect of concentration of sodium primary phosphate (NaH 2P0 4). primary-secondary phosphate mixture. secondary phosphate (Na2HP04)' and ternary phosphate (Na3P04) on the potential of the zinc
electrode. Reprinted from Awad and Kamel [7841. with kind permission from Elsevier Scicnce-NL. Sara
Burgerhartstraat 25. 1055 KY Amsterdam. The Netherlands.

shift is attributed to the fonnation of a protective serniglassy interface resulting from the
adsorption of phosphate ions in a complex fonn on the surface of a zinc phosphate layer.
After reaching a maximum value, the potential decreases upon further increase of
concentration owing to adsorption of phosphate ions on the bare cathodic areas of the
metal, which decelerates the reduction of oxygen. In a Na 3PO c NaOH mixture, both OHand PO~- ions take part in forming the serniglassy interface. At low concentrations,
adsorption of H2PO:; results in a reduction of the anodic dissolution whereas at high
concentrations H4 PO; ions exist and catalyze hydrogen reduction through preferential
discharge of hydrogen bonded to the ions.
In Ca(OHh solutions of concentrations in the range 0.1-3 gil (pH 11.1-12.55), zinc
shows a corrosion potential of about -0.5 VseE after 2-4 days of immersion [175, 197,
202J. The ennoblement of the zinc surface is due to the formation of a compact calcium
hydroxyzincatc. In HSO:; solutions, when the concentration is above 5 x 1O- 3M, the
potential is independent of concentration. This is associated with competitive adsorption
between S02 and HSO:; [943]. The presence of NaCI0 2 in aqueous solution leads to more
positive Ecorr and larger icorr values due to the effect of ClO:; reduction, similar to the effect
of NO:; reduction [34]. The corrosion current in aqueous solutions is significantly reduced
by adding rare-earth salts such as CeCl 3 [605].
Many organic species affect the corrosion potential or corrosIOn current of zinc in
aqueous solutions, including acrylic type anions [39], dimethyl sulfoxide, N-dimethylformamide, and acetonitrile [463], pyrazole [15], oxalate ions [40], phosphines [64],
benzene thiols [164], phenothiazine [63], zinc gluconate [100], and n-decylamine [207].
The corrosion potential and current of zinc and zinc alloys have also been detennined in
real environments such as concrete [468], soils [357], and natural waters [709, 565].
5.4.3. Effect afpH

Figure 5.11 shows the corrosion potentials experimentally detennined in various


noncomplexing and nonoxidizing solutions for zinc or zinc coatings as a function of pH

138

CHAPTER 5

0r-----------------------------.

1,

a-0 .5

UJ"

"iii

"E.,
15

Q..

c:

i ...

.9

e'"

(;

1.5

.~ ',...

FIGURE 5.11 . Corrosion potentials


in various solutions (data from Table
5.1). The solid line indicates the reversible potential calculated from the
Nernst equ ation, assuming IO-4M
Zn 2+ in the solution.

~~

10

12

14

pH

(data from Table 5.1). The solid line indicates the reversible potential calculated from the
Nemst equation assuming 1O-4M Zn 2+ in the solution. It can be seen that the corrosion
potentials in the pH range 4-8 are close to the calculated values. However, the corrosion
potentials in acidic and alkaline solutions are somewhat higher than the reversible
potential values. This may indicate that the concentrations of Zn 2+ in these solutions are
higher than lO-4M, at least near the surface, where the zinc ions from the dissolution may
accumulate. The larger discrepancies between the calculated values and the measured
values in the pH range 8-12 are due to the formation of solid oxide or hydroxide films,
resulting in various degrees of passivation.
Gmytryk and Sedzimir [445] measured the corrosion potential of zinc in deaerated
O.IM Na2S04 solutions in the pH range between 0 and 9. Figure 5.12 shows that the
corrosion potential changes the most between pH 4 and 6. The more positive values in
-1

-1.05

QI

1 .1

-1.15

-1.2

pH

FIGURE 5.12. Corrosion potential ofZn as a function of pH in O.IM Na2S04' Reprinted from Gmytryk and
Sedzimir [445], with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington
OX5 1GB, United Kingdom.

139

CORROSION POTENTIAL AND CORROSION CURRENT


0.8 02 M PO'"

'4

A O.2M PO ~'w ilh roln .

.,.. NaOH

+ 0.2M

PO~ .

1 " NaOH

..

X NaOH wilh rotation

ell

-1.2 L---:"'------~-

(;
<)

1.4

_1.6 L-__ __ ____ __ ____L __ __ L_ _ _ _L __ __ L_ _


13
14
11
12
10
6
7
8
9
~

pH

FIGURE 5.13. Dependence of Zn electrode corrosion potential, Eeorp on pH for different types of solutions:
0.2M [pol-I; 0.2M [pol-] with electrode rotation; NaOH + O.2M [pol-I; NaOH; and NaOH with electrode
rotation. Reprinted from De Pauli et al. [481], with kind permission from Elsevier Science-NL, Sara
Burgerharstraat 25, 1055 KV Amsterdam. The Netherlands.

the lower pH range are attributed to the anodic polarization required for the relatively
larger dissolution rate of zinc. In the pH range between 4 and 6, the cathodic part of the
corrosion process is controlled by diffusion of hydrogen to the surface.
Figure 5.13 shows the dependence of Eeorr on pH in solutions containing various
amounts of phosphate [481]. Within the pH range 7-10 an almost constant potential is
found, whereas within the pH range 11.5-14 a slope of 60 mV per unit of pH is obtained.
The constant corrosion potential value from pH 7 to 10 is attributed to the reaction Zn +
HPO~- = HZnP04 + 2e-. Electrode stirring is found to have an effect on the corrosion
potential between pH 10.5 and 11.5, where the corrosion potential values are more
positive and scattered, but has no effect at other pH values.
-3 r---~----------------------------------------,

NE -4
u

.a

._u

~.5

b
C

" d
Xe
-6

12

pH

FIGURE 5.14. Logarithm of current density for zinc in deaerated O.IM NaCI solutions as a function of pH of
the solution. (a) and (d) Reaction control; (b) intermediate control; (c) two simultaneous reactions; (e) diffusion
control regime. Reprinted from Zembura and Burzynska [116], with kind permission from Elsevier Science
Ltd, The Boulevard, Langford Lane, Kidlington OX5 1GB, United Kingdom.

140

CHAPTER 5
3

'f2

+
+

<:

.s

(;

u1

+
0

10

11

pH

FIGURE 5.15. Corrosion current density of zinc as a function of pH in unbuffered O.IM Na2S04 solutions and
oxygen-saturated solutions: acetate buffer, pH 4.7; succinate buffer, pH 5.6; phosphate buffer + 0.1 mol KClIl.
pH 7.0; sodium acetate (0.1 molll), pH 8.0; sodium acetate (0.1 molll) + EDTA (0.1 mol/I) + NaOH to pH 9.
Reprinted from Boto and Williams [128], with kind permission from Elsevier Science-NL, Sara Burgerharstraat
25, J055 KV Amsterdam, The Netherlands.

In contrast to the corrosion potential, which varies only slightly with pH in acidic
solutions, the corrosion current is a strong function of pH. Figure 5.14 shows the corrosion
current of zinc in deaerated O.lM NaCI solution as a function of pH [116]. The corrosion
current is relatively high in acidic and alkaline solutions and is lowest at pH 9. This
U-shaped dependence of the corrosion current on pH agrees well with weight loss data
obtained from immersion tests (see Chapter 9, Fig. 9.13).
Boto and Williams [128] found that in oxygenated sulfate solutions in the pH range
4-6 the corrosion of zinc is controlled by oxygen diffusion as shown in Fig. 5.l5. At
higher pH values, zinc hydroxide builds up on the surface, and the reaction is no longer
oxygen-diffusion-controlled. As a result, the corrosion current decreases to much smaller
values at pH > 6.

5.4.4. Effect a/Temperature


The equilibrium potential of zinc changes by only a few millivolts for a change in
temperature of several tens of degrees, as can be seen in Fig. 2.1 in Chapter 2. Figure 5.16
also shows that the zinc potential changes little as a function of temperature in alkaline
solutions containing various amounts of zinc ions [131].
Significant changes in the corrosion potential as a result of changing temperature
can, however, arise from changing the surface state of the electrode from active to passive.
It is a well-known phenomenon that the polarity reversal of a zinc/steel galvanic couple
in hot water or aqueous solutions is primarily due to the change of the zinc surface from
an active state to a passive state. In distilled water, potential can significantly change with
temperature variations in the presence of small amounts of ionic species in the solutions.
Figure 5.17 shows that the corrosion potential of zinc increases with increasing temperature in waters containing trace amounts of HCO), SO~-, and NO) [410]. Generally,
NO), HCO), and cOj- ions cause an increase in corrosion potential with increasing
temperature, while SO~-, Ct, SiO~-, and Ca2+ cause a decrease [196,410]. More infor-

CORROSION POTENTIAL AND CORROSION CURRENT

141

-, .34 r - - - - - - - -- -- - -- - - - - - - - - - - ,
+ 0.0132M 0.0476M "0.0966M
. O.I 54M
-1.36

0.35M

Q)

'"
>
~ -1.38
w

-1 .4

'0

40

30

20

Temperature

60

50

(Oel

FIGURE 5.16. Effect of temperature on Zn potential in 4.SM KOH containing Zn 2+ at different concentrations.
After Isaacson et al. [13 I].

mation on the variation of corrosion potential in hot water with respect to changing
temperature are presented in Chapter 7 (Section 7.2.4).
There are very limited data on the effect of temperature on the corrosion current of
zinc. Yamashita [112] found that the corrosion current of zinc in 1M ZnS0 4 solution may
increase with temperature in the presence of oxygen but decreases with temperature in
the absence of oxygen. The difference was explained as probably due to the formation of
a surface corrosion product.

5.4.5. Effect of Aeration and Convection


Aeration or deaeration changes the concentration of dissolved oxygen in an electrolyte. Figure 5.18 shows that aeration has a significant effect on the corrosion potential of

-, , - - - - - - - - - - - - - - - - - - - - - - ,
-0.9

30'C

....,-0.8
u

'"

2:,.-0.7
40' C

n3

'g-0.6
Q)

<5
Q.-

0.5

- 0.4

70' C

-0.3~---L---~--~L---~---~--~

Time, hou rs

FIGURE 5.17. Effect of temperature on zinc potential. Solution composition: 115-140 ppm HCOl 10 ppm
SO~-, and 10 ppm NO;. From Hoxeng and Prutton [410]. Copyright by NACE International. All Rights
Reserved by NACE; reprinted with permission.

142

CHAPTER 5

-1,----------------------------------,

c: -1 .2
Ul
u

-a.rated

C
Q)

-0-

OIitBe-rated

c.

c:: -1.4

';;;

(j

-'.6

FIGURE 5.18. Effect of aeration on


the corrosion potential of zinc in O.IM
Na2S04 solutions of different pH values. After Zhang and Hwang r1129J.

12

pH

zinc, particularly in near-neutral or slightly alkaline solutions [1129]. The more negative
potential values in the oxygen-free solutions are due to the reduced total cathodic reaction
rate, as curve 2e changes to curve Ie in Fig. 5.4. Dissolved oxygen has less of an effect on
the corrosion potential in acidic or alkaline solutions, where hydrogen evolution is the
predominant cathodic reaction.
In addition to the aeration conditions, the form of the electrolyte affects the rate of
oxygen reduction. Figure 5.19 shows that the corrosion potential in aerated 0.5M NaCl
solution decreases with decreasing solution thickness from 200 mm to 0.1 mm whereas
in deaerated solution the corrosion potential increases with decreasing thickness [608].
The decrease in the aerated solution is attributed to the depletion of dissolved oxygen,
since the amount of dissolved oxygen decreases with decreaSing volume of electrolyte

-1.06 , - - - - - - - - - - - - - - - - - - - -- - - -- - - - - - -- - ----,

Q)
()

G'"

-1.1

Aerated

iij

~
Q)

De.aerated

(5

Cl.

.~ -1.14

e
o

-,

-1.'8 '--_ _ _-'-_ _ _--'-_ _ _---'-_ __ ----'_ _ _---.J


-2
o
2
3
Layer thickness, mm (Log)

FIGURE 5.19. Corrosion potential Ecorr of zinc in aerated and de aerated O.SM NaCI solutions as a function of
electrolyte layer thickness. After Keddam et al. [608].

143

CORROSION POTENTIAL AND CORROSION CURRENT

TABLE 5.2. Effects of Grit Size of Grinding Paper on Corrosion and Surface Parameters of
Zinc in O.IM Na2S04 at 25C"
Grit size

Eeorr

(V)

-1.053
-1.051
-1.062
-1.077

120
240
320
400

icarr

IJiA/cm 2 )

1187
819.2
483.1
184.2

icath

!JINcm 2 )

623.9
304.4
206.7
178.6

Rp (Qcm 2 )

fI" (V/decade) fie (V/decade)

0.161
0.158
0.153
0.79

37.51
63.24
97.54
143.2

0.262
0.487
0.373
0.263

._--

"Ref. 383.

on the surface. On the other hand, the increase in the deaerated solution is attributed (0
the slower anodic reaction kinetics due to the buildup of zinc species in the solution.
Convection, which enhances the transport of the species in an electrolyte, has little
effect on the corrosion current of zinc in acidic solutions because the corrosion in acidic
solutions is, in general, controlled by an activation process. Convection usually affects
the corrosion current in nonacidic solutions, in which the corrosion process may be
associated with different diffusion-controlled reactions depending on aeration and pH
[128, 445). Bo(O and Williams [128] found that the corrosion current in aerated sulfate
solution at pH 5.8 increases linearly with the square root of the rotation rate of the
electrode [128]. Gmytryk and Sedzimir [445] reported that the corrosion current III
deaerated O.IM NaCl solution from pH 4 to 6 depends on the convective condition.
5.4.6. Effect of Surface Condition
Surface conditions are very important in electrochemical corrosion measurements.
The condition of the surface is perhaps one of the most nonreproducible factors and may
account for the remarkable differences in the corrosion data reported in different studies.
Depending on the surface finishing condition, the degree of cleanness and roughness can
vary greatly. The effective surface area can be many times greater than the apparent
surface area, depending on the polishing or grinding material. For example, Table 5.2
shows that the grit size of the grinding paper has a significant effect on the corrosion
potential and corrosion current density [383]. For a change in grit size from 400 to 120,
the apparent current density increases by as much as a factor of 6.
For a single-crystalline zinc sample, the corrosion current may be a function of the
crystal orientation. Table 5.3 shows the Eeorr and polarization characteristics of single
TABLE 5.3. Electrochemical Parameters for Zinc (0001), (1010), and (1120) Surfaces in
Deaerated 1M (NH4)2S04 Solutions"

____Eeorr (V seE) ___b" (mV)


_____be (mV) ___ Rp(Qcm-)

Exposed surface

(0001)
( 10TO)
( 1120)

=~~~

-1.116
-1.132
-1.132

31
31
32

~~

124
132
127

_L~_~_

1052
1447
1499

"Reprinted from Abayarathna et al. [446], with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane,
Kidlington OXS 1GB. United Kingdom.

144

CHAPTER 5

10,--------------------------------------.
6-hour immersion

E0

A 3.hour immersion

... ...

3~
~

0
0

0 . 1~----~----~-----L----~

048

12

____~~____~

16

20

24

Drying time (hours)

FIGURE 5.20. Effect of drying time in air after immersion in 5% NaCI solution on corrosion current density
of electrodeposited Zn-7% Fe alloy. Reprinted from Sagiyama et ai. [611], with kind permission from The Iron
and Steel Institute of Japan, Chiyodaku, Tokyo, Japan.

crystals in 1M (NH4)2S04 solution [446]. It appears that the polarization resistance is


slightly lower for the (0001) orientation than the others. The opposite relationship is
observed in alkaline solutions.
The individual grains of a polycrystalline surface may show different potentials.
Tsuru et at. [705], using a scanning microelectrode, measured differences in the corrosion
potential (up to 15 m V) among individual grains of a polycrystalline zinc electrode.
Other surface factors may affect the values of EcoIT and icorr' Chiu and Selman [706]
found that the corrosion potential and current varies across a curved zinc surface in a
flowing electrolyte owing to the transport and ohmic effects of the electrolyte. The
corrosion potential of a chromate-treated zinc surface is usually similar to that without
the treatment, but the corrosion current is generally much smaller with the chromatetreated surface because of the passive film formed on the zinc surface [75, 701]. Sagiyama
et at. [611] found that drying in air after immersion in the solution drastically reduces the
corrosion current of a zinc alloy as shown in Fig. 5.20 and attributed this decrease to the
formation of a more compact corrosion product film as a result of drying. The corrosion
current of electroplated zinc coatings in neutral chloride and sulfate solutions was found
by Dattilo [33] to be similar to that of pure Zn.
5.5. ZINC ALLOYS
The corrosion potential of a zinc alloy depends on both the equilibrium potential of
zinc and that of the alloying element. Since zinc has a low position in the emf series,
alloying with most elements will result in a more positive corrosion potentiaL In general,
addition of small amounts of other elements, i.e., a few percent, does not significantly
alter the corrosion potential from that of pure zinc. The potential only becomes significantly different from that of pure zinc when a certain level of the alloying element is
present. Compared to the corrosion potential, the corrosion current of zinc alloys varies
with the alloying element in a more complicated manner, as it is affected not only by the

CORROSION POTENTIAL AND CORROSION CURRENT

145

nature and quantity of the alloying element but also by the fonnation of intermetallic
phases and the microstructure of the alloy.
Zinc has been alloyed with many elements, such as aluminum, iron, nickel, titanium,
and copper. Table SA lists the corrosion potentials and currents of some zinc alloys in
various solutions.
Figure 5.21, reported by Selvam and Guruviah [330], shows the corrosion potentials
and currents of Zn-Mn alloys in 3% NaCI solution. The corrosion potentials of the alloys
with Mn contents below 50% are close to that of pure zinc, while the corrosion current
is the lowest for alloys with Mn contents between 10 and 30%. For zinc-copper alloys,
Fig. 5.22 shows that the corrosion potential is close to that of pure zinc when the Cu
concentration is below 30% and is close to that of pure copper when it is above 40% [361.

TABLE 5.4. Corrosion Potentials and Corrosion Currents of Some Alloying Elements and Zinc
Alloys in Various Solutions
Alloy

Solution

Cu
10%Cu
AI

O.IM Na2S04
5% NaCI"
INHCI"
IN NaCI
INHCI"
SS%AI
INH 2S04
INHCI"
5%AI
S5%AI
LON NaCl a
LON NaCi
94% AI
3% NaCI
sr/e AI
5% NaCI
55%AI
5% NaCI
INNaCI
S%AI
INNaCI
INHCI"
Fe
Steel
INNaCI
5% NaCI"
10% Fe
10% Mg
S% NaCl"
5% NaCl"
10%Ti
5% NaCI"
10%Cr
10% Ni
5% NaCl"
12%Ni
5% NaCl a
5% NaCl d
1O- 3M Cl0.19%Ni
Mn
3% NaCI
14.3% Mn
3% NaCI
INNaOH
2%Pb
O.IMHCI
2%Cd
O.IMHCI
1% Ti + 0.5% Cu O.OIN NaOH"
----.--~-

-------~--

/1

Dcaerateo.
Extrapolation from i- V curves.

From polarization resistance.

Aerated.

II

pH

0
4

6.5

4
4
()

12

Ecorr (V SCE)

-0.05
-0.843
-0.76
-0.64
-1.01
-0.96
-1.05
-1.05
-1.17
-1.00
-1.09
-1.01
-0.99
--1.04
-0.48
-0.69
-0.915
-1.016
-1.003
-0.942
-0.9
-1.11
-0.97
-1.02
-1.267
-1.043
-0.5
-1.01
-1.02
-1.26
..

-----~.--

icorr

(mAlcm 2 )

Reference
]g

Sh
2h
0.3 h

O.Sh
2. I x 10-4 ,
2.3 x 10- 5,

0.4"

0.026'
0.59"
0.36"

1241
14
14
14
250
14
176
176
42
]47
34R
14
14
14
14
1241
1241
1241
1241
1241
287
287
32
330
330
330
703
703
37
_0. _ _ _ -

146

CHAPTER 5
-1r-----------------------------~

170

wu

Corrosion potential

>'"
:::- -1 .1

o
~

.~

120

Cll

<5

..:;
Ql

iii
c
o

a:

Cl.

c
o

'0;

.~ -1 .

70

(;
()

(;

()

-1.3 ' - - - - ' - -- - - ' - -- -...L..- ---'---...J20

00

100

%Mn

FIGURE 5.21. Corrosion potential and current of Zn-Mn alloys


in 3% neutral NaCI as a function
of manganese content. After Selvam and Guruviah [330].

Figure 5.23 shows the changes in corrosion potential after 35 days' immersion in 5%
NaCl solution for Zn-Ni and Zn-Co alloy coatings [44]. The coatings containing less
than 10% Ni and 7.5% Co have potential values close to that of zinc and maintain them
after 35 days of immersion. The corrosion current as a function of Co and Ni concentration
is shown in Fig. 5.24. According to Short et al. [44], the ennoblement of the potential
with increasing Ni or Co content is associated with an increase in the anodic Tafel slope.
They also found that for the Zn-Ni and Zn-Co alloy coatings more positive corrosion
potentials are associated with lower corrosion currents. Similarly, Hosny et al. [606] found
that the addition of 8 gil cobalt in the zinc plating bath resulted in 25% reduction of
corrosion current compared to pure zinc coating. Darwish [32] reported that, compared
to the corrosion potential of pure zinc, the corrosion potentials of Zn-Ni alloys vary more
with pH in acidic and near-neutral solutions.
-0.2

r---- -- - - - -- - - - - - -- ----,

Q; -0.4 to
r/)

~ -0 .6
c:

oa.
Q)

-0,8

'iii

e(;

()

-'~---~~
-, .2 :------::-':-______---'-________-'--_____--'-__-.J
o
20
40
60
80
Concentration of Cu ('Yo)

FIGURE 5.22. Potential of Zn-Cu alloys in air-saturated 3.5% sodium chloride solution at 25C as a function
of copper content. From Budinski and Wilde [36]. Copyright by NACE International. All Rights Reserved
by NACE; reprinted with permission.

147

CORROSION POTENTIAL AND CORROSION CURRENT

-O.6 r----------------------------------------,
.... initial

35 days immersion

Q)

2:.'" -0 .8
z'"
c

Q)

oa.
c

'iii
~

-1

..... ... ... .. . ..... . .. . .........

_ 1.2 L-----~-------L------~------~----~----~

12

16

20

Concentration of Ni or Co (%)
FIGURE 5.23. Variation in corrosion potential with time for Zn-Ni and Zn-Co coatings in 5% NaCI solution.
Data are taken from Ref. 44.

Figure 5.25 shows the corrosion potential as a function of Fe content in electroplated


coatings in 5% NaCI solution at three different pH values [612]. There are roughly three
distinct regions of behavior with respect to Fe content. The corrosion potential of Zn - Fe
alloys with Fe contents less than a few percent is similar to that of pure zinc. Alloys with
10-50% Fe show a relatively constant corrosion potential 200 m V higher than that of
pure zinc. For alloys with Fe content higher than 70%. the corrosion potential is close to
that of pure iron.

100

"*'Ni

Ec.>

+Co

..::

3.
C
~
10
:;
c.>

c
0

'in
~

<:;

12

16

20

24

Concentration of N i or Co (%)

FIGURE 5.24. Effect of Ni and Co concentration on the average corrosion rate , between 20 and 40 days. of
zinc alloy coatings in 5% NaCI solution. After Short el af. [44].

148

CHAPTER 5
-0.4

pH 9,5
... pH 11.7

-0 ,6

+ pH

13.6

-;;; -0,8
u

'"

-1

C
Q)

~ -1.2
-1.4

40

20

100

80

60

Fe content in coating (wt 'Yo)

FIGURE 5.25. Relationships between corrosion potential of Zn-Fe alloy coatings measured in alkaline 5%
NaCl solutions and Fe content in the coating. Reprinted from Sagiyama et al. [6121, with kind permission from
The Iron and Steel Institute of Japan, Chiyoda-ku, Tokyo, Japan,

According to Sagiyamaetal. [611], Zn-Fealloys containing 4-27% Fe exhibit much


lower corrosion currents than of pure zinc and alloys containing more than 30% Fe when
tested in neutral 5% NaCI solution. The dependence of the corrosion current on iron
content is a function of the pH of the testing solution. At pH 9.5 and 11.7, the dependence
of corrosion current on Fe content is similar to that observed in the neutral solutions as
shown in Fig 5.26 [612]. The same figure shows that at pH 13.6 the corrosion current
increases with Fe content up to about 50% Fe and then decreases with further increase in
the Fe content. At pH 12, the corrosion current is found to decrease with increasing Fe
content over the whole composition range. In another study, Chang and Wei [365] found
that electrodeposited zinc-iron alloys with 20-40% Fe have the lowest corrosion current
in de aerated O.IM NaCI solution.

100

E'-'

.3- 10

pH 9.5

... pH 11 .7

+ pH 13.6
1

20

40

60

80

100

Fe content in coating (wt 'Yo)

FIGURE 5.26, Relationships between corrosion current density obtained from polarization curves for Zn-Fe
alloy coatings measured in alkaline 5% NaCI solution and Fe content in the coating, Reprinted from Sagiyama
et ai, [612], with kind permission from The Iron and Steel Institute of Japan, Chiyoda-ku, Tokyo, Japan.

CORROSION POTENTIAL AND CORROSION CURRENT

149

The corrosion of most zinc alloys is associated with a dezincification process, in


which the zinc is preferentially dissolved. As a result of dezincification, the surface
concentrations of the alloying elements increase. Thus, the corrosion potential and current
tend to vary much more with time than in the case of pure zinc.
5.6. EFFECT OF TIME
The time at which a measurement is taken is one of the most important factors in
determining the corrosion potential and current of a metal in a corrosion system. It is
perhaps also a major factor in the differences among the results reported in various studies
of similar corrosion systems.
Time invariably brings two basic changes to a corrosion system: (I) a change of the
physical structure and chemical composition of the corroding metal surface and (2) a
change in the composition of the solution, particularly in the vicinity of the surface.
Specific changes that may occur include changes in surface area and roughness, adsorption of species, formation of passive films, saturation of dissolution products, precipitation of a solid layer loosely attached to the surface, and exhaustion of reactants.
Mechanistically, these changes may lead to alterations in the equilibrium potentials, the
type of reactions involved, the rate-controlling process, etc. As a result, the corrosion
potential and current may vary drastically depending on the nature and extent of these
changes. The potential and current may not reach a constant value if the surface and
solution change continuously. For example. in the case of zinc alloy coatings, such as
batch hot-dipped zinc coatings, the composition varies from the surface ro the coating/steel interface. Thus, the corrosion potential will change with time as the coating
gradually dissolves.
Depending on whether the surface is active or passive, the corrosion potential of a
zinc electrode may reach different values at a steady state. As shown in Fig. 5.1 I, the
corrosion potential of an active zinc surface in neutral nonoxidizing and noncomplexing
salt solutions is about -1.1 Vso' near the equilibrium value. In solutions in which passive
corrosion products can form, the corrosion potential can be much more positive than the
equilibrium value.

TABLE 5.5. Time to Reach a Steady Zinc Corrosion Potential (V seE) under Various Conditions
-.---.-~.----

Solution
I gil Ca(OH)2' pH 12.3
Distilled water, 40"C
100 ppm cr. 40C. pH 4
10 ppm NO:;. 80C, pH 8
30 ppm NaHC0 3 , 4()OC
IMZnSO.j,2YC
IMZnS04,50C
0.1 wI. % NaCI, 25C
O.5N Na2S04, pH 4
O.IM Na2S04, pH 6
O.IMNa2S04'
pH 13
-------------- - - - -

Time

Initial Ecorr

.------.

3 days
240 min
150min
100 min
50 min
120 min
200 min
20 min
10 min
10 min
2 min

-1.39
-0.69
-0.90
-0.81
-1.0
-1.01
-1.01
-0.99
-1.12
-1.09
-1.41

Steady Eeorr
-0.45
-0.93
-0.97
-0.48
-0.84
-0.99
-1.00
-1.05
-1.11
-1.06
-1.42

Reference

------

197
196
196
196
196
112
112
365
14
1129
1129

150

CHAPTER 5
TABLE 5.6. Effect of Time on Corrosion Currents of Zinc
Initial
Solution

Tap water, pH 8
30 gil NaCI, pH 5.5
30 gil NaCI, pH 7
0.5M Na2S04 a
0.5M Na2S04a
IMZnS04
5% NaCI, pH 6.3
O.OIMKOH
8 gil Ca(OH)z
0.2M C 2Hz{COONa)z, pH 5.6
O.IM NaCI, pH 5.3
5% NaCI," pH 5.6
IMNaCI

icoIT

Later iCOIT

(j1Ncm 2)

(j1Ncm 2 )

38.8
200
80
46
66
2.38
25
3
9
640
60
0.9
8

1.1
42
38
18
26
0.1
12
0.1
0.1
830
250
8
13

Duration

Reference

97 days
5 hours
5 hours
23 hours
23 hours
10 days
35 days
30 days
30 days
7 hours
40 hours
I day
4 days

697
664
664
700
700
112
44
104
197
114
118
611
607

"Sample rotated at 600 rpm.


hZn-7% Fe.

The time required for reaching a steady-state value varies with the test conditions.
As shown in Table 5.5, the time for the corrosion potential to reach a steady-value varies
from a few minutes to several days. Table 5.6 shows the change in corrosion current with
immersion time in various solutions. It is noted that the corrosion current decreases with
time in some solutions whereas it increases in others. It generally decreases with time in
solutions in which some kind of inhibitive corrosion product film forms. On the other
hand, the corrosion current increases with time as the surface becomes more activated or
roughens in solutions in which such films do not form.
Walter [118] investigated the effect of time on the corrosion potential and current of
zinc in a O.IMNaCI solution of pH 5.3 buffered with phthalic acid. As shown in Fig. 5.27.
although the corrosion potential remains relatively constant, the corrosion current increases with time by a factor of 3. The value of the equilibrium potential also increases
due to the buildup of zinc ions in the solution.
Macias and Andrade [104] measured the corrosion potential and current of galvanized steel as a function of time in alkaline solutions of various pH values. Figure 5.28
shows that the corrosion potential of the zinc coating decreases with increasing pH after
the first hour of immersion. However, after 33 days of immersion the corrosion potentials
shift to more positive values. It was found that the formation of corrosion products during
the course of the immersion passivates the surface and is responsible for the change in
potential to a more positive value. Figure 5.28 also shows that corrosion currents at pH>
11 after 33 days of immersion decrease significantly as compared to the one-hour values
as a result of passivation. A similar effect is observed in Ca(OHh solutions at pH 12-13.8
[175].
Yamashita [112] measured the corrosion current of zinc in 1M zinc sulfate solution
and found that it decreased drastically after three days of immersion owing to the
formation of corrosion products on the surface as shown in Fig. 5.29. He noted that even

151

CORROSION POTENTIAL AND CORROSION CURRENT

-1.05,------------------------------------------,
50

>
..
40

ba

.0.

+
+
30

200

,:;-

<.J

c:

10

Immersion time, hours

FIGURE 5.27. Effect of immersion time on electrochemical parameters measured for zinc corroding in aerated
phthalate-buffered O.IM NaCI solution, pH 5.3. ieof" Corrosion current density; io, metal exchange current
density; RIP polarization resistance; Eeof" corrosion potential; Eo, metal equilibrium potential; hI" anodic Tafel
slope for the metal; Me electrochemically derived mass loss; M, solution-analysis-derived mass loss. Reprinted
from Walter [ 118 [, with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington
OX5 1GB, United Kingdom.

152

CHAPTER 5
o .---------------------------------------~

-0.2
-.:J
~

-0.4

-0.6

C
Q)

(5

-0.8

oc:

_1

a.

'iii

e
o -1.2

(J

-1.4

...

+
+

+ 1-hour immersion

33-day immersion

-* +

-1f--# ++

-1 .6 L-________---'__________--'-__________-'-____....l
10.5

11.5

12.5

13.5

pH

100

+ +

10

4"

#"

1 1-

0 .1 I-

I-hour immersion
33-day immersion

0 . 01~----------~----------~----------~----~

10.5

11.5

12.5

13 .5

pH

FIGURE 5.28. Corrosion potential (a) and current density (b) of galvanized bars as a function of pH after I
hour and 33 day of immersion in test solutions. Data are taken from Ref. 104.

10

E
u
(;

._0

0 2 present

*0 2 absent

0.1L---------~~--------~~--------~~~~

0 .01

0.1

10

Duration (days)

FIGURE 5.29. Influence of immersion time on the apparent exchange current density of the zinc electrode in
I .OM ZnS04 solution at 25 C. After Yamashita [112).

CORROSION POTENTIAL AND CORROSION CURRENT

153

with intense stirring of the solution a corrosion product film could still form. Deslouis et
al. [700] reported that the corrosion rate of zinc in neutral sulfate solution decreases within
one day to less than half of the initial value because of the accumulation of corrosion
products on the surface.
5.7. CORRELATION BETWEEN CORROSION CURRENT AND WEIGHT LOSS
RATE

It is always important for the corrosion rates determined with electrochemical


techniques to correlate with the corrosion data derived from the more realistic corrosion
measurements such as weight loss measurements. The correlation between the corrosion
current and the weight loss rate of zinc in aqueous solutions has been investigated in
several studies [104, 114, 118, 790].
Macias and Andrade [104] found that by proper! y choosing the Tafel slopes in
accordance with the specific anodic and cathodic reactions, good correlation could be
obtained between the corrosion current and the corrosion rate determined by weight loss
measurement for galvanized steel in alkaline solutions as shown in Fig. 5.30. Similar
correlation was found in solutions containing calcium hydroxide [197]. In a study by
Chang and Wei [365], a good agreement was found between the corrosion rates determined by weight loss measurement and by an electrochemical technique for zinc-iron
alloys in deaerated O.IM NaCI solution, as shown in Fig. 5.31.
Muralidharan and Rajagopalan [790] compared the corrosion rates derived from
various measurement techniques and found that in concentrated alkaline solutions the
corrosion rate derived from steady-state Tafel extrapolation is in close agreement with
that determined by weight loss measurement, as shown in Fig. 5.32. Compared to
steady-state data, the corrosion currents obtained from transient methods are much higher.
The discrepancy between the values obtained by transient and steady-state techniques is
attributed to the slow diffusion of zincate ions away from the electrode surface, which
Sample

E
<.>
0>

100

'6

vi

tfl

.c0>
'0;

15,

10

"

,g

'0

"2

"

14

7- - 8
3 - 6
5

Qj

_ 3

'5

11

0. 1
0 .1

2
3
4
5
6
7
8
9
10

10

100

E lectrochem ical wei ght loss , mg / cm'

12
13
14
15
16
17
18

pH
10.95
11.95
12.62
12.79
12.68
12.76
13.00
12.97
13.17
13.43
13.58
13.81
12.59
13.02
13.40
13.52
13.56
13.74

FIGURE 5.30. Comparison of gravimetric and electrochemical weight loss results for galvanized bars.
Numbers refer to sample designations in table at right. After Macias and Andrade 1104].

154

CHAPTER 5
10
-

'*

8
>a.
E
ai

l'!

Weight loss mel hod


Electrochemical mel hod

c:
0

'0;

20

Zn

40

60

Fe (wt%) in Fe-Zn coating

80

100

Steel

FIGURE 5.31. Corrosion rates, in mils per year, at 25C for steel, Fe-Zn alloys of different Fe contents, and
zinc. Reprinted from Chang and Wei [365], with kind permission from Elsevier Science Ltd, The Boulevard,
Langford Lane, Kidlington OX5 1GB, United Kingdom.

later causes redeposition of zinc. The transient data extrapolated to time zero give the
rates without the effect of redeposition.
Walter [118] studied the correlation between the corrosion rates obtained by linear
polarization and by solution analysis in a neutral O.lM NaCl solution. The rate calculated
from the polarization resistance was found to be typically 50-60% higher than that from
solution analysis. Walter attributed the error in the polarization data to neglect of the metal
ion deposition when the corrosion potential is near the equilibrium electrode potential.
In a study by Boto and Williams [114], the corrosion currents determined by a
transient polarization technique were correlated with the weight loss rates determined by

5 ,-------------------------------------~
- Weight loss

+ Steady sta te

NaOH concentration, M

FIGURE 5.32. Comparison of corrosion rates obtained by different methods in NaOH solutions at 40C. After
Muralidharan alld Rajagopalan [790J.

CORROSION POTENTIAL AND CORROSION CURRENT

155

chemical analysis. The differences between the results obtained by the two techniques
were generally less than 15%. The errors in the corrosion current values were considered
to be related to the number of cathodic reactions involved in the corrosion process, a better
correlation between corrosion current and weight loss being obtained when only one
cathodic reaction was involved.
Deslouis et al. [700] found the corrosion current measured by the impedance
technique in an aerated sulfate solution to be 10 to 30% lower than that determined by
atomic absorption, which was explained by the fact that the electrochemical measurement
is instantaneous while the atomic absorption method is time-averaged.
The differences between the electrochemically determined corrosion rate and the
gravimetrically determined corrosion rate can arise from many causes, which generally
belong to one of two types: (1) time effects and (2) assumptions made in the determination
of the corrosion current. The corrosion currents determined by electrochemical methods
are intrinsically different from the corrosion rates determined by gravimetric methods.
The corrosion current is an instantaneous parameter and is a function of time as discussed
in the previous section. Depending on the conditions, the corrosion current may decrease
or increase with time. On the other hand, the corrosion rates measured by gravimetric
methods are time-averaged. Assuming the equivalent corrosion current for the gravimetric
corrosion rate is igt , the relation between the corrosion current, icoIT' and igt can be expressed
as
(5.14)
It is evident from Eq. (5.14) that the value of icorr is close to that of igt when they are
measured at a time that is significantly longer than the time required for the system to
reach a steady state. On the other hand, the values of i gt and icorr can be very different when
they are measured during the time when icorr changes significantly. In such a case, the
time-averaged corrosion current will have a better correlation with that of the equivalent
corrosion current from gravimetric measurements. It has been reported that the time-averaged corrosion current, compared to a single instantaneous current value, showed a
better agreement with the corrosion rate obtained by atomic absorption analysis for zinc
in a neutral sulfate solution [700J.
For the same reason, corrosion current and gravimetric corrosion rate provide
different information about a corrosion system. Corrosion current, when measured
properly, reflects the corrosion rate at the time of measurement but provides little
information on the total corrosion loss. On the other hand, the time-averaged corrosion
rate determined by gravimetric methods, although reflecting the total corrosion loss up
to the time of the measurement, provides no information on the corrosion activity at the
time of measurement. A unique advantage of the electrochemical techniques is that they
can be used to measure the initial corrosion activity of a freshly exposed metal surface,
which usually cannot be assessed through the use of a gravimetric method.
Another common origin for disagreement between icorr and igt lies within the
determination of the corrosion current itself. Depending on the corrosion system and
experimental conditions, errors in corrosion current values can be caused by (i) lineari-

156

CHAPTER 5

zation of the current-potential equation; (ii) incorrect estimation of the Tafel slopes; (iii)
neglect of reverse partial reactions; (iv) neglect of solution resistance; (v) use of a high
potential scanning rate; and (vi) neglect of mass transport. In particular, significant errors
are obtained when icorriit (it being the limiting diffusion current) is close to 1, a high
potential scanning rate is employed in a low conductive electrolyte, or RQIRp is large
(R Q and Rp are the solution resistance and polarization resistance, respectively). The
theoretical analysis of these issues can be found in a number of studies [118, 694, 717,
718,790].
Although it is often difficult to obtain a one-to-one correlation between values of
corrosion current and gravimetric corrosion rate because of the many possible error
sources, experimental results indicate that a certain proportionality exists between the two
values [104, 197,365,790]. For example, the curve for the corrosion current as a function
of pH shown in Fig. 5.14 agrees very well in shape and position with that of the thickness
loss rates shown in Fig. 9.13 in Chapter 9. Under a given set of test conditions, the relation
between icorr and igl can perhaps be simplified to igl = aicoIT' with a being a constant. The
value of a depends on the specific corrosion system and the measurement techniques and
procedures. In situations in which the objective is to compare the relative corrosion rates,
it is often not necessary to know the exact value of a.

6
Corrosion Products
6.1. INTRODUCTION
The corrosion products discussed in this chapter are the solid materials formed on
the surface of a corroding metal. Being a layer between the metal and the environment,
corrosion products greatly affect the corrosion behavior of the metal. In general, corrosion
products differ in composition, structure, morphology, and properties depending on the
specific conditions under which the corrosion process occurs. Based on their effect on
corrosion, corrosion products can be roughly divided into two major groups: (i) those
having an effect of blocking the anodic and/or cathodic reactions and thus drastically
reducing the corrosion rate, and (ii) those having little inhibiting effect or even having an
effect of enhancing the corrosive reactions. The corrosion products in group ii are
generally thicker, bulkier, and more porous than those in group i.
Characterization of corrosion products is essential to the understanding of a corrosion process. Zinc corrosion products have been characterized by various analytical
techniques such as X-ray diffraction (XRD), Fourier transform infrared (FTIR) spectroscopy, thermogravimetry (TG), glow discharge optical spectroscopy (GDOS), scanning
electron microscopy with X-ray microanalysis (SEM-EDS), X-ray fluorescence spectroscopy (XFS), X-ray photoelectron spectroscopy (XPS), secondary ion mass spectroscopy
(SIMS), and ion chromatography (IC) [253,462, 1164]. Each technique has its unique
advantage in obtaining information on certain aspects of a corrosion product. In a recent
study [1164] the following choice of techniques has been suggested: (a) XRD for
determination of crystalline phases; (b) GDOS for qualitative analysis of in-depth
distribution of major and minor elements except for Nand CI; (c) SEM-EDS for
examination of morphology and quantitative analysis of lateral distribution of elements
with atomic numbers of 6 and higher; and (d) IC for quantitative analysis of waterdissolved ionic species.
This chapter compiles the information that has been obtained on the composition,
structure, morphology, and properties of zinc corrosion products according to the specific
environments in which they form. The corrosion environments are divided into four
groups: (a) atmospheric environments, (b) waters, (c) aqueous solutions, and (d) environments other than those included in (a)-(c). The fourth group consists mostly of environments used in accelerated tests such as the salt spray test and cyclic tests. A large part of
this chapter deals with the characteristics of the corrosion products formed in atmospheric
157

158

CHAPTER 6

environments since they have been the most extensively investigated. Also, the focus in
this chapter is on the nonelectrochemical aspects; the electrochemical behavior of zinc
surfaces covered with a passive film or corrosion products have been discussed in Chapter 3.
6.2. IN ATMOSPHERIC ENVIRONMENTS

6.2.1. Composition and Structure


The composition of zinc corrosion products formed in atmospheric environments
has been the subject of many studies [173, 331, 868, 1163]. Table 6.1 lists the zinc
compounds that have been identified in the corrosion products formed in various atmospheres. It should be noted that although atmospheres are conventionally defined according
to four general types, i.e., rural, industrial, urban, and marine, each atmosphere as a
corrosion environment is, in essence, unique. That is, each atmosphere is characterized
by its own particular combination of specific factors such as air composition, temperature
variation, seasonal climate changes, and type, amount, and frequency of rain. Thus, it is
not unusual to find very different kinds of corrosion products formed in atmospheres of
the same type.
Many zinc compounds can form in each type of atmosphere. However, for a specific
atmosphere, only certain compounds dominate. Generally, among the zinc compounds,
oxides, hydroxides, and carbonates are most often found in corrosion products [173, 331 J.
Zinc sulfate, ZnS04 nH 20, and basic zinc sulfate, Zn4S0iOHknH 20. are also frequently
found [868, 1259]. In some industrial atmospheres the amount of zinc sulfates in corrosion
products can be as high as 50% [555]. In coastal areas, zinc hydroxychloride,
Zn5(OH)sCI2H20, and zinc chlorohydroxysulfate, NaZn4Cl(OH)6S046H20, are common compounds [173, 974, 1165]; however, the relative amount of chloride in zinc
corrosion products is usually low owing to the high solubility of the chloride-containing
compounds [297].
The composition of zinc corrosion products formed in a particular atmosphere
generally changes with time. Friel [173] reported that the corrosion products formed on
a zinc coating in industrial and marine environments after a 9-year exposure are mainly
Zn 5(C0 3MOH)6 with some ZnO, ZnS04, and Zn 5(OH}gCI 2. During the exposure, ZnO
becomes increasingly abundant while the amount ofZnS04 decreases. Biestek et al. [868]
found that in rural, urban, industrial, and marine atmospheres hydrated zinc oxide, basic
zinc carbonate, and zinc sulfate of various compositions are formed in the first two years.
However, after 10 years of exposure, the corrosion products contain only basic and neutral
zinc sulfates.
The distribution of the chemical compounds identified in the corrosion product layer
on zinc may not be uniform with depth. Flinn et al. [1171] found O/Zn ratios consistent
with ZnO in the outer 15 nm of the surface layer and with ZnC0 3 or Zn(OH)2 deeper into
the film. The extreme outer 1 nm is rich in surface contaminants such as chlorine and
sulfur. The soluble salt ions may also penetrate into the interior of the surface through
cracks and defects [331].
Many zinc corrosion products formed under atmospheric conditions are found to be
crystalline [173, 1163]. Data on the crystal structures of the various zinc compounds
identified in zinc corrosion products, as recently summarized by Odnevall and Leygraf

CORROSION PRODUCTS

159

TABLE 6,1, Zinc Compounds Detected in Corrosion Products of Zinc Formed in Various
Atmospheric Environments
Atmosphere
Rural

Compounds

------------~--------

ZnO
Zn(OHh
ZnC0 3
Zn5(C03MOH)6
Zn4S04(OH)6,nH20
(Zn, CU)4S04(OH)6AH20
Zn4CI2(OH)4S04,5H20

Urban

Zn(OHh
ZnC01
ZnS04,nH 2O
Zns(C03MOH)6
Zn4S04(OH)6,nH20
(Zn, CU)4S04(OH)6AH20
Zn4CI2(OH)4S04,5H20
NaZn 4CI(OH)6S04,6HP

Industrial

iI

II

ZnO
Zn(OH)2
ZnS04
ZnCO J
Zn4C03(OHkH20
Zns(C03lz(OH)6
Zns(OH)sCI2H2O
Zn4S04(OH)6,nH20
NaZn4C](OH)6S04,6H20

Reference( s)

325
297," 1163," 1171"
325, 1171"
297," 511," 868,,,,1> 1164,"./)
Cracked, uniform layer"
1259"
297," 868,,,1> 1163," 125917
Spherical particles ( 1-5 pm)' 1164"
974"
Uniform, nodular, and finegrainedd
Platelet islands"
Platelet islands"

ZnO
Zn(OH)2
Zns(C03lz(OH)6
ZnS0 4,nH 2O
Zn4S04(OH)6,nH20
Zn4CI2(OH)4S04,5H20
Zns(OH)sCI2,H2O

Marine

Structure and morphology

Platelet islands!

Platelet islands g
Platelet islandsg

297," 555,,,,1> 1171"


1171"
868,"./' 1163, 1259 b
297," 1164"./'
297," 868,"./' 1163, 1259"./'
1163"
1163"./'
1163"./'
173"
297," 555",1>
173," 297," 511"
173," 555,"./' 868," 1163,
1259"
297," 868,"1> 1163
1163,"./) 1259",h
1163"./'
173," 297," 1163, 1259"
297," 868"
173"
297," 325
325
173," 511," 868,"./) 999"./'
173," 297," 999"./'
297," 868"
999, ",I> 1259",1>

Under an unsheltered condition.


Under a sheltered condition.

'Ref. 1164,
"Ref. 1171.
"Ref. 1165,
I Ref. 1163,
g

Ref. 999,

[1166], are presented in Table 6,2, According to Odnevall and Leygraf [ 1166], there is a
structural resemblance between hydroxycarbonate, hydroxychloride, hydroxysulfate, and
sodium zinc chlorohydroxysulfate, as illustrated in Fig, 6,1. These compounds have
layered structures with sheets of Zn 2+ in octahedral and tetrahedral coordination, and the
main difference is the chemical content and bonding between the sheets, The structural

160

CHAPTER 6

TABLE 6.2. Crystal Structures of Zinc Compounds Found in the Corrosion Products of Zinc in
Atmospheric Environments"
Compound
Oxide,ZnO
Hydroxide, p-Zn(OHh

Hydroxycarbonate,
Zns(C03h<H)6

Cell structure

Phase structure

Hexagonal, a = 3.25 A, c = 5.21 A


Orthorhombic, a = 8.49 A, b = 5.16 A, Tetrahedrally coordinated zinc atoms
c=4.92A
forming a three-dimensional
network with "hydroxyl ions
Monoclinic, a = 13.58 A, b = 6.28 A, Sheets with octahedrally and
c = 5.41 A, P= 95.6
tetrahedrally coordinated zinc
atoms, held together by carbonate
ions

Sulfate
ZnS04H20

Monoclinic, a = 7.51 A, b = 7.59 A,


c = 6.94 A, P= 116.25
Monoclinic, a = 5.95 A, b = 13.60 A,
c = 7.95 A, P= 90.4
Monoclinic, a = 9.98 A, b = 7.25 A,
c = 24.28 A, P= 98.45
Hydroxysulfate,
Triclinic, a = 8.36 A, b = 8.37 A,
Sheets with octahedrally and
c = 20.68 A, a = 90.06, P= 89.93,
tetrahedrally coordinated zinc
Zn4S04(OH)64H20
r = 120.11
atoms; sulfate groups connect on
either side of the sheets, which are
held together by hydrogen bonding
Hydroxychloride,
Sheets with octahedrally and
Hexagonal, a = 6.34 A, c = 23.64 A
tetrahedrally coordinated zinc
Zns(OH)gCI2 H20
atoms, held together by weak
O-HCI bonds
Chlorohydroxysulfate,
Hexagonal, a = 8.37 A, c = 13.05 A
Sheets with octahedrally and
tetrahedrally coordinated zinc atoms
NaZn4CI(OH)6S046H20
are coordinated with sulfate groups;
sodium atoms are coordinated
between the sheets
Chlorosulfate,
Monoclinic, a = 10.92 A, b = 4.14 A,
c=7.ISA,p= 102.62
Zn4CI2(OH)4S045H20
"Data from Ref. 1166.

resemblance between these compounds may facilitate the transformation from one phase
into another under the proper environmental conditions.
The zinc compounds found in atmospheric corrosion products may be different under
rain-sheltered and unsheltered conditions, as shown in Table 6.3, which contains data
reported by Johansson and Gullman [1259]. For example, the latter authors found that
zinc hydroxycarbonate, Zns(C0 3MOH)6, forms only under an unsheltered condition after
exposure for 1 to 5 years, while zinc hydroxysulfate, Zn4S0iOHk4H20, forms only
under a sheltered condition in a rural environment.
The wet/dry pattern is an important factor affecting the specific composition of a
corrosion product. In particular, periodic drying has an important effect on the formation
of zinc salts such as hydroxysulfate and hydroxychloride. Precipitation of these salts
under atmospheric conditions may occur during every drying period of the wet/dry cycle.
However, the situation is very different in a fully immersed condition, where the
precipitation of these salts does not necessarily occur because they have a much higher

CORROSION PRODUCTS

161

(b)

(a)

(c)

(d)

Tetrahedrally-,

."

Sulfate group,

Octahedrally coordinated Zn

Carbonate group,

Hydrated Na

FIGURE 6. I. Crystal structures of zinc hydroxycarbonate (a), zinc hydroxychloride (b), zinc hydroxysulfate
(c), and sodium zinc chlorohydroxysulfate (d). Structural details are given in Table 6.2. After Odnevall and
Lcygraf [11661.

solubility than oxide or hydroxide. As will be seen in later sectIons, these salts are not
usually found on zinc surfaces that are fully immersed in solutions.
In indoor environments, zinc surfaces may be covered mainly with contaminants
from the air. Munier et ai. [406] determined the amount and composition of water-soluble
chemical compounds accumulated on a zinc surface that had been exposed to an indoor
environment for 40 years. They found that the average contamination rate was 1.2
,ug/(cm2yr) for chloride, 1.8 ,ug/(cm2yr) for sulfate, and 0.33 ,ug/(cm2'yr) for nitrate, The
chloride is present mainly as zinc chloride because zinc is a good scavenger for chloridecontaining species. The sulfate and nitrate are present as a mixture of ammonium, sodium,
calcium, and zinc salts.

162

CHAPTER 6

TABLE 6.3. Zinc Compounds Detected in the Corrosion Products Formed under Sheltered and
Unsheltered Conditions in Four Types of Atmospheric Environments"
Duration of exposure
Atmosphere Condition
Rural

U
S

Urban

S
Industrial

Marine

3 months

Zn4S04(OH)6nH20
ZnO
Zn4S04(OH)6,nH20

I year

5 years

Zns(C03h(OH)6
Zn4S04(OH)6nH20

Zns(C03MOH)6
Zn4S04(OH)6nH20

Zn4S04(OH)6,nH20

Zn4S04(OH)6,nHzO

Zn4S04(OHknH20
ZnS04-H2O

Zn4S0iOH)6,nH20
ZnS04,H 2O

Probably
Zn4CI2(OH)4S04,5H20
ZnS04,H 2O
Zn4CI2(OH)4S04,5H20

ZnS04,H 2O
Zn4CI2(OH)4S04,5H20

ZnS04,H2O
Zns(OH)gCI2, H2O
Zn4C12(OH)4S04,5H20

Zns(C03lz(OH)6
NaZn4CI(OH)6S04,6H20
Zns(C03lz(OH)6
NaZn4Cl(OH)6S04,6HzO

ZnO
Zns(C03lz(OH)6
NaZn4Cl(OH)6S04,6H20 NaZn4Cl(OH)6S04,6H20
Zns(OH)gCI2, H 2O
NaZn4Cl(OH)6S04,6HzO
NaZn4Cl(OH)6S04,6H20

"Ref, 1259,
"S, Sheltered; U. unsheltered,

A very thin surface oxide film fonns on zinc even in dry air. Leroy and Schmitz [253]
investigated the surface oxide films of various zinc- and zinc-aluminum-alloy-coated
steels. Based on scanning Auger microprobe (SAM) analysis, they concluded that:
(a) The surface of electroplated galvanized steel in air is oxidized to a depth of
more than 100 A.
(b) The surface of hot-dip zinc coating (0.15% AI) is covered with a thin aluminum
oxide film less than 50 A in thickness. This film is composed of 18% Zn, 22%
0, and 60% AI. The thickness of the oxide film is related to the cooling rate of
the coating during the coating process because the aluminum oxide film is
generated by Al diffusing from the bulk to the surface and reacting with oxygen
in the air.
(c) The aluminum oxide thickness on a 5% Al zinc alloy coating (Galfan) is less
than 50 A, and it is less than 20 Aon a 55% Al coating (Galvalume). The thinner
oxide on Galvalume could be due to a rapid passivation of the surface, generally
observed on pure aluminum.
Friel [173] found that amorphous AI-Zn sulfate is the principal corrosion product
formed on Zn-55% AI-coated steel in industrial and marine environments. AI(OHh and
AlzeOHkHzO were also found in the marine atmosphere. An increase in the exposure
time in the marine atmosphere from 3 to 9 years results in a decrease of the zinc
concentration in the AI-Zn sulfate and an increase in the AI(OHh concentration. The
lower amount of the zinc corrosion product fonned in the longer exposure time was
attributed to its more soluble nature, compared to the aluminum compounds on the

CORROSION PRODUCTS

163

surface. In another study, aluminum sulfate and hydroxide were found to be the main
corrosion products of Zn-55% AI-coated steel wire in rural and marine environments
[456].

6.2.2. Quantity and Morphology


The amount of solid corrosion products formed on a zinc surface depends on many
factors, among which rain is particularly important. As illustrated in Fig. 6.2 [1264], the
amount of corrosion product washed away by rain is comparable to that remaining on the
surface. It is also notable in Fig. 6.2 that the relative amount of zinc washed away by rain
depends on the time of the year.
Schikorr and Schikorr [555] determined the amount of adhering solid corrosion
products and the relative amounts of different zinc compounds in the corrosion products
formed in several atmospheric environments. As shown in Table 6.4, the amount of
corrosion products remaining on the surface increases with exposure time. However, the
percentage of zinc compounds in the corrosion products decreases with time, indicating
an accumulation of contaminants from the atmosphere. These authors also found that
although there is less corrosion under a sheltered condition, the amount of corrosion
products is greater than under an unsheltered condition, indicating the washing effect of
rain. Because of the lack of rain washing under a sheltered condition, the amount of
corrosion products can actually be greater than the weight loss.
Flinn et al. [1171] investigated the relation between the corrosion weight loss and
the amount of corrosion product retained on the surface of zinc samples exposed for
various lengths of time in four different locations in the United States. Figure 6.3 shows

o
'Q

Weigh, loss
Zinc content in runoff rain \vater
Zinc content in corrosion product

0.2

01)

'"
'"0

..J

.;::
ell
.;;;

0. 15

~
-0

'"

0. 1

;g

u
<>
.s

0.05

ov.

Jan .

Mar.

May

July

Sep.

Month of the Year


FIGURE 6.2. Variation of corrosion rate and zinc content of corrosion product and runoff rain water with the
time of year. After Slunder and Boyd 1217].

CHAPTER 6

164

TABLE 6.4. Weight Loss (WL), Amount of Corrosion Products (CP), and the Percentage of
Zinc Compounds in the Corrosion Products (ZP) on Zinc Sheet Samples Exposed in Several
Atmospheric Environments"
Test site
Berlin-Dahlem,
unsheltered

I month
6 months
2 years
I month
I year
I month
1 year
2 years

Berlin-Dahlem,
sheltered
Beside engine shed

CP (g/m2)

WL(g/m 2)

Duration of exposure

2.5
8
15
3.3
20.2
4.4
13.3
58.8

2.2
20
52
1.4
15
7.6
55
133

ZP(%)
90
87
77
77
62
74
72
50

aRef. 555.

that the amount of corrosion products retained on the surface increases with the weight
loss. It also shows that a large part of the corroded zinc is washed away by the rain since
the weight of the corrosion products, containing also oxygen, water, and other contaminants, should be several times greater than the zinc weight loss if all the corroded zinc is
retained on the surface. The runoff of the corrosion products was found by Flinn et al.
[1171] to be mainly due to the effect of hydrogen concentration in the rainwater.
The microscopic morphologies of corrosion products formed in many different
environments appear to be rather similar [999,1163]. According to Odnevall [1163], small
islands with thin platelets form initially, and larger islands or patches with thicker, larger,
and more rounded platelets form afterward. The islands and patches grow and merge,
gradually covering the whole surface. In a rural environment, small spherical platelet
islands less than 1 /lm in diameter were found after 14 days of exposure under a sheltered
condition. They grow in size to form larger islands. Under unsheltered conditions,
networks of very thin sheets are formed within a few days of exposure. In a marine
environment, small islands (20-30 /lm) with layered structures are randomly formed on
the zinc surface within 24 hours of exposure [999]. Within the islands, thin hexagonal
100

North Carolmill

0,

~ DIstrict of Columbia

80

.J:.
C>

@ 40
c

e<;

N9W York

Slope = 1.645

60

'0;
~

'iii

ew Jersev
Q

.~

t.

~.,

".,,,, _

Slop.= 0. 575

40

60

80

Weight lo ss, mgldm'

100

120

FIGURE 6.3. Amount of corrosion


products retained on 191 zinc samples
after exposure for I, 3, and 14 months
at four different sites in the United
States. After Zhang and Tran [1117].

CORROSION PRODUCTS

165

platelet (0.5 f1m x 3 ,11m x 2 Jim) crystals of zinc hydroxychloride, Zn s(OH)sCI 2Hp,
form. These corrosion products are not uniform but vary in thickness. They have an
average thickness of 0.7 f1m after 14 days, 1.2 f1m after 30 days, and 1.7 f1m after 90 days.

6.2.3. Formation Processes


The formation of corrosion products in an atmospheric environment is a complex
and continuously changing process. The degree of complexity and the rate of change
depend on the type of atmosphere and the various factors involved.
Feitknecht [404] investigated the formation of oxide films on a cleavage surface of
a zinc single crystal in air. He found that first an amorphous film was formed. After four
weeks, the film thickness was -100 A, and the film was amorphous at the surface but
crystalline in the interior. The very small crystals were oriented with their a and c axes
parallel to the a and c axes of the zinc. The growth rate on anodically polished surface
was faster, the film thickness reaching a few hundred angstroms in several days.
In a review article on atmospheric corrosion mechanisms, Graedel [331 J proposed
the formation reactions for the zinc compounds commonly found in atmospheric corrosion products based on the stabilities of various zinc compounds. In general, the first step
in the process is the formation of oxides and hydroxides followed by the formation of
zinc carbonates as the system reaches equilibrium with CO 2 in the air. In S02 or
CI--containing environments, solid zinc sulfate or zinc chlorides may form. Figure 6.4

I
I
I
I
I

y(OW.I:
H2 0 I

r:---.1,..,
'Zn( HC 02)2'
L _____

FIGURE 6.4. Schematic representation of the processes involved in the formation of carbonate and organic
components during the atmospheric corrosion of zinc. After Graedel133 I I.

166

CHAPTER 6

schematically illustrates the processes postulated by Graedel for the formation of zinc
carbonate. The reactions involved in the formation of the various zinc compounds are
shown below.
For zinc hydroxycarbonate:

For zinc hYdroxychloride:


Zn(OHh(s) + 4Zn 2+ + 6(OW) + 2Cr -7 ZI1s(OH)gCI 2
For zinc sulfate and zinc hydroxysulfates:
(x = 4, 6, or 7)

Zn(OHh(s) + (y - l)Zn 2+ + (2y - 2)(OW) + SO~- + 4H20 -7


(y =

3 or 6)

Odnevall and co-workers [974, 999, 1163-1166] investigated the sequence of


formation of various zinc compounds in the corrosion products formed in various types
of atmospheric environments. In a rural environment, zinc hydroxycarbonate,
Zns(C03MOH)6, forms initially; this product is also found at later times under an
unsheltered condition but gradually transforms to Zn4SOiOHk4H20 under a sheltered
condition [999]. Similar results were reported by Johansson and Gullman [1259]. In
marine environments, the initial step consists of the formation of basic zinc carbonate,
Zns(C03MOH)6, which is subsequently transformed into zinc hydroxychloride,
Zns(OH)gCI 2H20. Later, sodium zinc chorohydroxysulfate, NaZn4Cl(OH)6S046H20. is

( Zn(OH)2)

FIGURE 6.5. General reaction scheme for the formation of the major corrosion products of zinc under sheltered
conditions. After Odnevall [1163].

~::--..~':''t~~H)6nH20

ZnO
Zn,(CO,),(OH),
Zn,(OH),Cl,.H,O
Zn.SO.(OH),.nH,o
NaZn4Cl(OH)6S04.6H,o

Major compounds

Zn.S04(OH) . nH 20
Zn4Cl,(OH)4S04.5H,O

1 ZnS04nH,O

Zn,(COl),(OH),

Zn(OH),
ZnS04nH,O
I Zn,(COl),(OH),
Zn.SO.(OH),.nH,O
Zn.Cl,(OH)4S045H,O

---0>-------+---------11--:1

1 year

FIGURE 6.6. Sequence of formation of the major zinc compounds found in the corrosion products formed in four different types of atmospheric
environments under a sheltered condition. The circle below the compounds indicate the earliest detection of the compounds in the corrosion
products.

'0

1 month

NaZn.Cl(OH),S046H,O

1 week

ZnS04nH,O
Zn.SO.(OH) . nH,O
Zn.Cl,(OH)4S04.5H,o

Zn,(OH}gCl,.H,O

1 day

--l

'"

(/l

--l

;0

'"tl

(5
Z

(/l

;0

o;0

168

CHAPTER 6

formed [999]. In urban and industrial environments, the initial step is the formation of
Zns(C0 3MOH)6, followed by the formation of hydroxysulfate, Zn4 SOiOHknH zO, and
eventually by the formation of zinc chlorohydroxysulfate, Zn4CliOH)4S045H20 [1165].
The final step can also be the formation of zinc hydroxychloride under some conditions.
The sequence of formation of various zinc compounds and the transformations among
them depend on the chemical species present in the atmosphere and also on the climatic
conditions. Odnevall [1163] proposed a general reaction scheme, shown in Fig. 6.5, for
the formation of several major corrosion products formed in different environments.
The sequences of formation of the major zinc compounds found in the corrosion
products in four typical types of environments are summarized in Fig. 6.6 based on the
data in Table 6.1 and the results of Odnevall and Leygraf [1163-1166]. Initially, the zinc
surface is covered quickly with zinc hydroxide, which is gradually converted into zinc
carbonate. Within one month of exposure, almost all major zinc compounds can be
detected in the corrosion products. In the more severe environments, such as marine and
industrial environments, the formation of chloride and sulfate compounds can be very
fast, occurring within one day. As corrosion continues, the various zinc compounds
generally increase in quantity, but some may also disappear as a result of their transformation into other compounds, depending on the specific environmental factors. The
sequences of formation of the various corrosion products are generally similar under
sheltered and unsheltered conditions, although in most cases the process is faster under
an unsheltered condition [974, 999, 1165].
6.3. IN WATERS

6.3.1. Fresh Waters


In fresh waters, as shown in Table 6.5, the most common corrosion products are ZnO
and Zn(OH)2' The hydroxide found in corrosion products may have different structures
depending on water conditions, but p-Zn(OH)2 is found to occur most frequently [404,
462]. Zinc carbonate may be the major constituent in the corrosion products when the
water is saturated with air or carbon dioxide [462, 688]. The corrosion products formed
in water are not always crystalline. For example, Gilbert and Hadden [437] reported that
the corrosion product formed on a zinc coating immersed in cold supply water is
completely amorphous.
The relative amounts of zinc oxide and hydroxide in corrosion products formed in
water are a function of many factors, such as temperature, aeration condition, and time
of immersion. According to Kotnik [462], ZnO is generally the initial corrosion product
in water in the temperature range O-lOOC. The stable corrosion products in the lower
temperature region (O-30C) are mainly zinc hydroxides. However, between 30 and 90C
the products are mainly ZnO whereas at the boiling point they are a mixture of zinc oxide
and the hydroxides. Similar results were reported by Gilbert [458]: zinc hydroxide is
usually produced in cold water whereas in hot water the corrosion product is usually zinc
oxide.
Terada [1048] studied the dissolution of zinc oxide in water charged with CO 2, When
CO 2 is passed into water in which ZnO is suspended, the oxide dissolves quickly on brisk
stirring and forms a metastable solution of Zn(HC0 3h. The zinc concentration in the

CORROSION PRODUCTS

169

TABLE 6.5. Composition of Corrosion Products on Zinc in Different Waters


Reference( s)

Water

Composition of corrosion product

Deionized water

ZnS(C03)2(OH)6.a ZnO"
ZnO." Zn(OH)/,
ZnO
ZnO with some Zn(OHh

Cold water

Zn(OHh
Amorphous Zn(OHh

Hot water

ZnO

462,458

Seawater

ZnO
ZnCI 2
ZnCI 2-4Zn(OH)2
Zn4(C03)(OHkH20
ZnS'

179.209
209
179.209
179.209
209

Water saturated with air.

/>

Water saturated with CO 2 -free air.

Seawater containing sulfide.

654.688
458,654
462
402
462
437

solution increases to a maximum and then begins to decrease, reaching a constant value
within 4 h at 30e. No solid carbonate is present in the solid phase before the maximum
concentration is reached, but afterward the solid phase changes in composition to
5ZnO2C0 2 AH 20 and slowly approaches the composition of ZnC0 3 At 40 and 50C,
the zinc concentration in the solution attains a constant value within 40 min, and the solid
phase attains also a final composition of 5ZnO2C02AH 20.
Water temperature affects not only the composition but also the compactness of the
corrosion products. Table 6.6 shows the characteristics of corrosion products formed in
distilled water at various temperatures [412]. At low temperatures the corrosion products
exist as a gelatinous and adherent substance. An increase in temperature to 55C is
accompanied by a definite change from the gelatinous form to a granular, nonadherent
form. At 65C the corrosion film is completely granular and somewhat nonadherent and

TABLE 6.6. Characteristics of Corrosion Products Formed in Water at Various Temperatures"


Temperature (0C)
20
50

55
65
75
95
100
"Ref. 412.

Tenacity

Appearance
Definitely gelatinou~
Slightly less gelatinous
Mostly granular
Decidedly granular, becoming flaky, and
compact
Decidedly granular, flaky, and compact
Compact, dense, and flaky specimen and
fracturing scale
Varies from grayish white to black, very
dense, resembling enamel

Very adherent
Adherent
Nonadherent
Nonadhercnt
Nonadherent
Adherent, removed by bending
Very adherent and difficult to remove
by mechanical means

170

CHAPTER 6

appears to be more compact. Increasing the temperature further is accompanied by an


increase in the compactness and tenacity of the film.
Temperature also determines the stability of the corrosion products. Zinc oxide was
found to be stable in both cold and hot water [458]. Zinc hydroxide transforms to zinc
oxide in dry air, in C0z-free air. or in hot water. According to Gilbert and Hadden [437],
ZnOH 20, once dehydrated, cannot take up water to form zinc hydroxide.
Kotnik [462] found that Zn(OH)z, whether of the f3 or the e form, and whether present
as a layer on zinc or as a powder, is converted to zinc oxide in the presence of water at
temperatures near 90C. Neither low pressure (less than 1 mm Hg) nor storage in a
desiccator causes conversion of zinc hydroxide to the oxide at temperatures of 25-30C.
Zinc hydroxide forms when zinc oxide powder is kept in water at low temperature
(0-10C) for extended periods of time. Kotnik also determined the conductivity of the
corrosion products formed in water at different temperatures. It was found that the
conductivity of the products formed at 88C is greater than that of those formed at lower
temperatures by a factor of 103. The conductivity of the corrosion products, which consist
mainly of ZnO, does not remain constant but decreases with time.
The structure and composition of zinc corrosion products formed in water varies with
time, especially at an early stage. For example. Kotnik [462] reported that a short exposure
to water at temperatures below 10C results in the formation of corrosion products
containing a considerable amount of zinc oxide. A longer exposure results in the formation
of products composed mainly of hydroxides, with only a small amount of zinc oxide. As
another example, according to Feitknecht [404], in distilled water on a cleavage surface
of a single crystal, an amorphous film is formed initially. After a few minutes, starting
from randomly distributed centers underneath the film, a layer of oriented crystalline
oxide begins to grow, extending sideways and finally covering the whole surface. With
time. about one day, the crystals gradually lose their orientation, and an uneven thickening
of the film occurs, resulting in the initiation of local anodes.

6.3.2. Seawater
The zinc compounds that have been identified in the corrosion products formed in
seawater are listed in Table 6.5. Calcium and magnesium compounds are also found in
the corrosion products, and their relative amount increases with increasing temperature
[209]. Also, in seawater containing sulfide at pH > 7.2, the corrosion products were found
to be mainly ZnS, which has the lowest solubility in this pH range [209].
The composition of the corrosion products in seawater also depends on the time of
exposure. According to Mor and Beccaria [179] when a zinc sample is immersed in
seawater, zinc oxide is formed within the first 8 h. At longer times, up to 72 h,
ZnCI 2-4Zn(OHh becomes the main corrosion product, and thereafter the main products
are ZnCI 2-4Zn(OH)2 and ZniC03)(OH)6H20. Increasing temperature was found to favor
the formation of more compact corrosion products [209].
The corrosion products formed on zinc anodes (i.e., dissolution products as they are
formed under an anodic current), which are often used in seawater applications, have been
studied by Perkins and co-workers [464, 1169]. They found that the corrosion products
formed in seawater on a zinc anode sample, which was galvanically coupled to a steel
sample, were mainly zinc oxide. The corrosion film was a porous three-dimensional

CORROSION PRODUCTS

171

network of discrete single-crystal plates of ZnO with a size of 10-100 ,urn. Arrays of
plates had a population density on the order of 106 plates/cm2. Individual plates grew to
about 30 ,urn in diameter and several microns in thickness after one week of exposure.
Side plates nucleated at specific crystallographic orientations to the primary plates. The
arrays of plates exhibited a preferred orientation with fast growth directions normal to
the basal surface.
The effects of anodic current density and the flow rate of seawater over the surface
of a zinc anode on the formation of the dissolution products of zinc were investigated by
Perkins et al. [1169]. They found that increasing flow velocity favors two-dimensional
growth of the corrosion products. At certain velocities, the dissolution product was thin,
flat, and compact but cracked. The toughest, most compact films were formed at low
current density and with long time exposure under static conditions.
6.4. IN SOLUTIONS
Zinc oxide is a common corrosion product in diluted solutions of salts such as NaC!
or Na 2S04 with zinc hydroxide present in various amounts as a minor component. In
concentrated solutions ofNaCI, Na 2S04, and other salts, the corresponding zinc salts may
form in addition to zinc oxide or hydroxide, or both. The presence of solutes, in general,
favors the transformation of zinc oxide to zinc hydroxide, which is the main corrosion
product in water at room temperature [462]. In solutions of certain salts such as H 3 P0 4
or CrO" with which zinc can form compounds of low solubility, the corrosion products
can be concentrated with the corresponding zinc compounds. The formation of zinc salt
films such as phosphate films and zinc chromate films has an effect of passivating the
zinc surface and has a range of industrial applications. The corrosion products formed in
various solutions are shown in Table 6.7.
6.4.1. Effect of pH

The formation of zinc corrosion products in solutions is primarily determined by the


pH of the solution in the absence of species with which zinc may form compounds of low
solubility. In acidic solutions, zinc has high solubility and dissolves with formation of
Zn 2+ ions. Since the solubility of zinc decreases with increasing pH in acidic solutions,
precipitation ofZn(OH)2 occurs when a certain pH value is reached. In alkaline solutions,
with pH > 9, the solubility of zinc increases with increasing pH, and in the high-pH range,
zinc oxide and hydroxides tend to dissolve with the formation of zincate ions.
According to Roetheli et al. [497], who investigated the corrosion of zinc in HCI and
NaOH solutions of different pH values, no corrosion films are formed on zinc samples
held for up to 30 days in solutions with pH lower than 5 or higher than 13.5. Corrosion
films may form as a result of corrosion in the pH range 5-13.5, but the films formed at
pH < 6 and pH > 12.5 are rather loose.
Baugh [II OJ found, in his study of electrochemical corrosion of zinc in slightly acidic
solutions of NaCI0 4, NaC!, and Na2S04' that the formation of an oxide film may occur
in the pH range 3.8-5.8. The oxide film was found to be porous and not passivating.
Macias and Andrade [105] investigated the zinc corrosion products in alkaline
solutions. Table 6.8 shows the composition of corrosion products formed on galvanized

172

CHAPTER 6

TABLE 6.7. Zinc Corrosion Products Formed in Various Solutions


Solution

Corrosion product(s)

Ca(OHh
0.1 gIl
0.8 gIl
Saturated
O.I-I.SMKOH
O.SMNaCI
CaC0 3, saturated, 60C, 10 ppm SO~O.IM NaC!, 70C, O 2 bubbling
O.IM NaI, 70C, 02 bubbling
O.OOSM ZnS04
0.OO2M ZnCI 2
1O-4-2MKCI
0.SMNa2S04
IMZnS04
0.01-IMCoCI2
0.INNa2S
Na2Cr20rH20 + H 2SO4
O.OIM NaHC0 3
0.IMH 3P20 7
H3P04
NaOH+ H 20 2

Reference( s)

ZnO
Ca[Zn(OHhh 2H2O
Ca[Zn(OHhh2H20 (pH < 13.3).
ZnO (pH> 13.3)
ZnO, Zn(OH)2
Zn(OH}z, ZnCI 2nZn(OHlz
ZnO, Zn(OHh, 4ZnOC0 2AH 20 (s)
ZnO
ZnO
ZnO + Zn(OHh + ZnS04
Zn(OHh + ZnCI4Zn(OHlz
ZnO
ZnO+ Zn(OHh
Zn(OHh
ZnO + CoO
ZnS
ZnO + Cr203 + ZnCr04
Zn4C03(OH)6H20
Zn2P207
Zn3(P04hAH20
ZnO + Zn particles

197
197
202
105
404
462
462
462
462
462
402
113
112
97
831
57,66
194
93
1175
359

"Room temperature unless otherwise specified.

steel in KOH solutions of various concentrations. In solutions of various KOH concentrations, the corrosion product formed on the zinc surface is mainly ZnO initially, but it
is mainly Zn(OH)2 after about 30 days of immersion. The rate of transformation from
ZnO to Zn(OH)2 increases with increasing pH. The morphology of the corrosion products
formed in alkaline solutions was found to greatly depend on pH. The zinc oxide formed
at pH 11-12 is porous and nonadherent whereas that formed at pH 12-12.8 is thin and

TABLE 6.8. Corrosion Products Identified on Galvanized Steel Bars during Immersion Tests in
KOH Solutions"
Corrosion product( s)
KOH concentration

After 24 hours

Intermediate stage

I.SM

ZnO

ZnO (9 days)

0.6M

ZnO

0.2M
O.IM

ZnO
Possibly ZnO

Zn(OHh, traces of ZnO


(8 days)
Zn(OHh (14 days)
Zn(OHh (18 days)

"Ref. 105.

End oftest
ZnOZnFe204, traces of
Zn(OH)2 (36 days)
Zn(OHh, traces of ZnO
(35 days)
Zn(OHh (28 days)
Zn(OHh (33 days)

CORROSION PRODUCTS

173

2o,-----------------------------------,
EG956, Dip "EG952, Spray + CRS958. Dip CRS952. Spray
~ 15

(5

"

.c.
coo

-. 10

*E

rn
'iii

3:

2~~--~4------~6------~8------~10------~12~~~~
14
pH

FIGURE 6.7. Dissolution of phosphate coatings obtained by different methods after immersion in stirred HCI
or NaOH solutions of different pH values for 30 min at 2SOC. After van Ooij and Sabata [982\.

compact. At pH 12.8-13.4 the product is a layer of well-packed Zn(OHh crystals. At pH

> 13.4 it is a continuous porous ZnO layer.


The dissolution products formed in phosphate solutions, mainly zinc phosphates,
vary greatly in composition, morphology, structure, and properties depending on the
conditions of their formation [93-95, 256, 578, 581]. Zinc phosphate films are stable
within the pH range between 3 and 12 as shown in Fig. 6.7 [982,993]. At low pH they
dissolve completely, whereas at high pH they are converted to zinc hydroxide, which then
decomposes to zinc oxide or dissolves as zincate ions.
The possibility of formation of solid corrosion products is determined by the surface
pH value, which often differs from the bulk value. Baugh [111] found that the oxide film
forms at higher bulk pH values in ammonium salt solutions than in sodium salt solutions,
due to the dissociation ofNH; ions within the double layer, which lowers the pH near the
electrode surface. Boto and Williams [128] noted that a corrosion product forms on the
zinc surface in unbuffered sulfate solution in the pH range 5.6-6.2 whereas it docs not
occur in buffered solution in the same pH range. Because of the buffering effect, the
surface pH is less likely to be different from that of the bulk, and a higher bulk pH value
is therefore required for the formation of hydroxide in the buffered solution compared to
the unbuffered solution.
6.4.2. Formation Processes

The formation and transformation of zinc corrosion products in solutions has been
systematically investigated by Feitknecht [404]. He found that amorphous hydroxide is
precipitated on adding hydroxyl ions to a dilute solution of zinc chloride, and it changes
with time to oxide or hydroxides through the following reactions:
Zn 2+ + 20H- ---7 am. Zn(OHh

CHAPTER 6

174

am. Zn(OHh

ZnO

pH=7-9
pH = 7-9
pH =11-12

If the Cl- concentration is higher than O.OIM and the pH is below 7, two different
hydroxide chlorides (I and II) may be formed by the following reactions:

7Zn2+ + 120W + 2Cl- ~ 6Zn(OHhZnCI2 (II)

Similarly, basic zinc carbonates can be obtained by precipitating a solution of a zinc salt
with a mixture of sodium hydroxide and sodium carbonate. They can also be formed by
bubbling air containing CO 2 through a suspension of amorphous zinc hydroxide.
The compositon of corrosion products formed on zinc surfaces may not be
uniformly distributed. According to Sergi et al. [174], Zn(OH)2 completely covers the
surface in 1M KOH for 70 days, but only a fraction of it is in the form of well-defined
crystals; the coverage is not complete in the presence ofCl-. Leidheiser and Suzuki [103]
found that ZnO formed on single crystals of different zinc alloys by immersion in 3%
NaCI solution has different thicknesses. Feitknecht [404] reported that a zinc sample
immersed in O.SM NaCI solution for several days is covered by very loose corrosion
products consisting of spindlelike ZnO and platelets of p-Zn(OH)2. Craters, up to 1 mm
in dimension, are randomly distributed underneath the loose oxide and hydroxide. In the
middle of the crater, there is a pit filled with hydroxide chloride. After prolonged exposure,
it is covered with a layer of p-Zn(OH)z. As shown in Fig. 6.8, the inner part of the wall
around the crater is also composed of hydroxide chloride; then follows p-Zn(OH)2
covered partly by ZnO, and finally a zone with p-Zn(OH)2. Based on the distribution of
the corrosion products across a crater, Feitknecht proposed the pH conditions for the
formation of corrosion products. The pH rises from about 6.S at the crater to nearly 12 at
the zones away from the crater where p-Zn(OH)z is formed. Feitknecht also noted that if
a zinc specimen is immersed in a sodium chloride solution through which air is bubbling,
it remains bright for several months owing to the formation of a thin layer of zinc
carbonate, which is oriented with the c axis perpendicular to the surface of the specimen.
The formation process of the corrosion products in alkaline Ca(OH)2 solutions has
been studied by Macias, Andrade, and co-workers [197,202, 1214]. According to their
results, the initial corrosion product in solutions of all concentrations is invariably ZnO.
At a later stage, Ca[Zn(OH)3hH20 becomes the main corrosion product when the
solution contains more than 0.8 g of Ca(OH)2 per liter. The corrosion products are
mixtures of ZnO and Zn(OH)2 when the Ca(OH)z content is lower than O.S gil. The

CORROSION PRODUCTS

17S

4 Zn{OHh. ZnCI,

[[[[[J

6 Zn(OH),. ZnCI,

B Zn(OH),
ZnO layer

ZnO disperse

B Zn(OH),

FIGURE 6.S. Schematic illustration of the distribution of the


corrosion products on zinc around an active center in O.SM
NaCI solution. After Feitknecht [404[.

morphology of the corrosion products varies greatly with changes in the pH of the
solution. When the pH is around 12.6, the surface is totally covered by the crystallized
corrosion products during the first one or two days. As the pH increases, the size of these
crystals increases, and they seal the surface almost completely, leaving only small zones
of the metal surface uncovered. At just above the threshold pH value of 13.3, the corrosion
products appear as isolated crystals.
The corrosion products in chromate solutions form through a dissolution-precipitation mechanism [67]. Zinc dissolves as tetravalent chromium is reduced. The reduced
trivalent chromium precipitates as a complex chrome gel on the metal surface. The
chromate films can be represented by the general formula Cr203-Cr03xH20 [66]. The
thickness and color of the chromate film depend on the conditions employed [57,651.
The color of a film reflects the amount of hexavalent chromium in the film. For example,
a yellow film typically has a thickness of 0.1-0.6 flm [57]. The resistance of a yellow
film is in the range of 100 to 1000 flQ [65]. Black chromate films can be obtained by
adding Cu 2+ or Ag+ to chromate solution [811.
Formation of corrosion products that are hygroscopic can cause water condensation
at relative humidities much lower than 100% [406,557]. The zinc compounds, particularly zinc chloride, are very hygroscopic and tend to have a higher rate of moisture pickup
than zinc metal [406]. Similarly, it was found that the corrosion products formed initially
have the effect of enhancing the ability of the surface to retain more corrosion products
afterward [1094].
The properties of corrosion products are a function of various material and environmental factors and thus essentially vary from situation to situation. The properties that
are important in relation to the corrosion of zinc and its alloys are: (a) electrical
conductivity; (b) compactness; (c) adherence to the substrate surface; Cd) hardness; and

176

CHAPTER 6

(e) resistance to dissolution and decomposition. These properties determine the electrochemical activity, permeability, physical stability, and chemical stability of the corrosion products.

6.4.3. Zinc Alloys


There is very little systematic information about the corrosion products formed on
zinc alloys although the chemical compounds in the corrosion products are mostly
reported as supplemental data in many corrosion studies.
It has been reported that the IJ phase in a hot-dipped zinc coating was covered by
uniaxial blocks of Zn(OHh crystals after immersion in 1M NaOH for 70 days and by a
layer of clusters of needle-shaped crystals of ZnO after immersion in chloride solution
[174]. The amount of crystal blocks is much smaller for the ( phase compared to the IJ
phase. Suzuki and Enjuzi [692] noted that the compactness of corrosion products of
Zn-Fe alloy coatings is affected by the presence of metallic ions in the test solutions.
Macias and Andrade [105] found the presence of ZnFe204 in the corrosion products of
galvanized steel in alkaline solutions. Cheng et al. [393] identified the corrosion products
on 5% AI and 55% Al zinc alloys immersed in Ca(OH}z-saturated solution of pH 12.5 to
be mainly calcium aluminum hydroxide hydrate Ca2Al(OH)73H20. Satoh et al. [339]
reported that the corrosion product formed on Zn-Ni alloys is more concentrated with
ZnCI-4Zn(OH)2 than that formed on galvanized steel [339]. In another study by Short et
al. [44], the corrosion product of a Zn-Ni coating formed during an immersion test in
5% NaCl was identified to be mainly Zn 5(OH)gCl z, and the coating after removal of the
corrosion products was found to be rich in Ni.
6.5. IN OTHER ENVIRONMENTS
Table 6.9 shows the corrosion products formed in environments other than natural
atmospheric environments, water, and aqueous solutions. These environments consist
mostly of those associated with laboratory simulated or accelerated test conditions.
In general, it appears that either ZnO or Zn(OHh tends to be the corrosion product
when the form of moisture is such that a concentrated salt solution does not form on the
surface of the zinc. Other zinc compounds may form when a concentrated solution is
generated on the surface. In situations in which only a thin layer of moisture is involved,
such as in a humidity chamber or under water spray, a small amount of contaminant can
produce a concentrated electrolyte, because of the small quantity of moisture. For
example, a thin layer of moisture in open air is usually saturated with carbon dioxide,
thus providing a favorable condition for the formation of corrosion products enriched
with zinc carbonates.
It is generally found that in clean damp air the corrosion products formed on zinc
surfaces consist mainly of basic carbonate, 2ZnC03 3Zn(OHh [437,688]. The relative
amount of zinc carbonate in the corrosion product is a function of the concentration of
carbon dioxide in the air. In situations in which the access of air is restricted, for example,
in the interior of coils of wire, the ratio of zinc oxide to basic zinc carbonate tends to
increase with increasing restriction of air. When zinc corrosion occurs in air free of carbon
dioxide, the corrosion product, after drying in air for a day or two, consists of zinc oxide
only, indicating that dry zinc oxide does not transform into zinc carbonate in dry air [437].

177

CORROSION PRODUCTS

TABLE 6.9. Corrosion Products on Zinc Surfaces Exposed in Environments Other than
Atmospheric Environments, Waters, and Aqueous Solutions"
Corrosion product( s)

Test

Medium

Reference
---~---

Watcr
Water
Watcr
Damp air
Air (90% RH) at 70C
Air (100% RH) containing
1.5% CO 2
Air ( 100% RH) contai ning no
CO 2
Air containing HCI
Mix of S02 and water vapor
Water containing S02
0.1 % solution of 0.88
ammonia
20 wI. % HCI
5% NaCI
1O-3M Na2S04
CI- -laden concrcte
0.05% (NH.jhS04 + 0.35%
NaCI
Acetic acid, pH 3.8
Alcohols or acetone. etc., +
O.OSMHCI

Water sprinkled between


two sheets
Water trapped between a
bundle of wires
Fog
Exposed
Flowing air

ZnO

688

ZnO

437

52

Exposure

4ZnOC02'4H ZO
ZnC0 3
ZnO + Zn(OH)2 + ZnC0 3
2ZnC0 3 3Zn(OHh
2ZnC0 3 3Zn(OHh

Exposure

ZnO

437

Exposure
Exposure
Exposure in a sealed tank"
Exposure in a sealed tank"

ZnCl 2-4Zn(OHh, ZnCl z


S + ZnS + ZnS04
ZnS04H20 + ZnS04 6H20
2ZnC0 3 3Zn(OHh

811
499
437
437

Exposure in a sealed tank"


Spray
Cyclic wetting and drying

Zinc oxychloride
ZnO
Zn(OHh
Zn5(OH)8CI2-H20
ZnCI 2-4Zn(OHh + ZnO

437
610
1094
1173
213

2ZnC0 3 3Zn(OHh
ZnO + amorphous zinc
compounds

437
500

Cyclic wetting and drying


Spray
Immersion

6X~

1170
437

--------~

"Reprinted from Helwig [52]. with kind permission from Elsevier Science Inc., 655 Avenue of the Americas. New York.
hHung above the solution.

Helwig [52J identified the corrosion products formed under different types of moist
conditions in various accelerated tests; he found that, in short-term tests, abundant air is
needed for the formation of zinc carbonate, as shown in Table 6.10. The results of longer
TABLE 6.10. Appearance and Identity of Corrosion Products Formed in Various Accelerated
Tests"
Test
7 cycles (30 min each) of steam
pressure
4 days in water film (under glass
slide)

Appearance

Identity of corrosion product(s)

White water-spot marks

ZnO
ZnO

3 days in condensation cabinet

Moderatel y heavy white powder


with dark areas underneath
Very heavy, bulky white powder

4 days in water fog

Very heavy. bulky white powder

ZnO and 2ZnC0 33Zn(OH)2

1 day in salt spray

Very heavy, bulky white powder

ZnO, ZnC0 3, and ZnCI 2-4Zn(OH)2

4 days on outdoor rack

Spotty gray-whitc powder

4ZnOC0 2 -4H 20

"Ref. 52.

ZnO and 4ZnOC0 2-4H 20

178

CHAPTER 6

term tests (e.g., water fog, condensation, and water film for 6 months) reveal that the
predominant corrosion product is 4ZnOC0 24H20.
Gilbert and Hadden [437] investigated the effect of the presence of acid vapors in air
on the formation of corrosion products. In the presence of vapors of organic acids, zinc
salts of fatty acids have been detected in the corrosion products. In moist air containing
sulfur dioxide vapor, the corrosion products consists mainly of zinc sulfate. In the
presence of hydrochloric acid vapor, the corrosion products are largely zinc oxychloride.
Lyon et at. [213] found that the corrosion products formed in a cyclic test with a
wetting solution of 0.35% NaCI and 0.05% (NH4)zS04 were a mixture of
ZnCI 24Zn(OH)2, ZndOH)ls(S04)3Cl3(H20)s, and ZnO after 340 hours of test but consisted of only ZnO after 1020 hours. McLeod and Rogers [499] reported that the corrosion
product on zinc exposed to a moving mixture of air and sulfurous acid (sulfur dioxide +
water vapor) contained sulfur, zinc sulfite, and zinc sulfate.
Hronsky [500] reported that the corrosion of zinc in some organic solvents, such as
alcohols and acetone, containing 0.05-0.5M HCI resulted in the formation of a poorly
adhering, permeable, black layer consisting of amorphous zinc compounds and zinc
oxide. The corrosion products formed in organic solvents such as alcohols, acetone, and
benzene generally consist of zinc oxide and amorphous zinc compounds.
Mabuchi et at. [287] investigated the corrosion products formed under various
organic paints. They found that generally under a hydrophilic coating, after a salt spray
test, the only corrosion product is ZnCl r 4Zn(OH)z; while under a hydrophobic coating,
ZnO is also found.
In concrete containing no free chloride, Ca[Zn(OHhh2H20 is the usual corrosion
product formed on a corroding zinc surface [1214]. ZnO and Zn(OHh have also been
identified in corrosion products formed on zinc surfaces in concrete. The process for
formation of Ca[Zn(OH)3lz2H20 can be generally expressed as

6.6. EFFECT OF CORROSION PRODUCTS ON ZINC CORROSION


Generally, a corrosion product layer that is insulating, compact, adherent, and
insoluble serves as a good barrier against a corroding environment. It has been established
that the low corrosion rate of zinc in clean atmospheric environments is largely due to the
formation of a thin, compact basic zinc carbonate film that has a low solubility in water
and dissolves only very slowly. The compact nature of the film prevents the permeation
and retention of oxygen and water. The corrosion rate is increased by rain, to an extent
that depends on its acidity, and airborne contaminants that promote the dissolution or
decomposition of the carbonate film. As a result of such decomposition in the presence
of contaminants, other zinc salts may form. Thus, depending on the relative proportions
of zinc carbonate and other zinc salts in the corrosion products, further corrosion may
also be determined by the stability and compactness of the salt film.

CORROSION PRODUCTS

179

In water and aqueous solutions, the corrosion rate of zinc changes with temperature
owing to the different properties of the corrosion products in different temperature ranges.
The relatively higher corrosion rate in the 60-90C range (Fig. 9.3 in Chapter 9) was
attributed by Cox [412] to the formation of non adherent corrosion products. However,
according to Kotnik [462], the formation of a nonadherent oxide film is not the reason
for the relatively higher corrosion rate in this temperature range. He believed that the
effect is due to the formation of a zinc oxide film which has a better electrical conductivity
than the zinc hydroxide formed at lower temperatures.
The zinc/steel galvanic couple exhibits a polarity reversal in hot water that is directly
related to the physical and chemical properties of the corrosion product film. The
ennoblement of the zinc electrode potential in hot water has been attributed in a number
of studies [458,462] to (a) passivation of the zinc surface by zinc oxide, which is the main
corrosion product in hot water, and (b) cathodic depolarization due to the semiconducting
nature of the oxide, the electrical conductivity of which increases with temperature.
Polarity reversal occurs when this ennoblement reaches a certain potential which is more
positive than that of the steel.
The dependence of the zinc corrosion rate on pH (Fig. 9.13 in Chapter 9) is also
determined by the formation and nature ofthe corrosion products. Generally, at pH values
below 5, there is no corrosion product formation, and thus corrosion depends solely on
the rate of cathodic reactions, which are usually rate-determining on an active zinc
surface. In the pH range between 5 and 9, the corrosion rate is hindered to a varying extent
by the formation of corrosion products, which are usually bulky and porous. In the pH
range between 9 and 12, the corrosion products are thin and compact and are thus passive;
the corrosion rate of zinc is the lowest in this pH range. At pH values higher than 13, the
corrosion rate becomes high again because the high solubility of zinc prevents the
formation of a stable corrosion product film on the surface. Figure 6.9 illustrates the effect
of the corrosion products formed on galvanized steel in a Ca(OH)2 solution as a function
of pH in the pH range 11-14 [1214].
I

I
no H, evolution
ZnO

HJ

H,

CaHZ

isolated
crystals -Zn(OH),

ZnO
I Zn(OH),
I isol. CaHZ
I

localized
corrosion
0
11

passivation

corrosion

12 12.2
13
13.3
Alkaline solutions containing Ca'+ ions

14

FIGURE 6.9. Corrosion behavior of galvanized steel immersed in Ca(OH)2 solutions in the pH range 11-14.
CaHZ = Ca[Zn(OHh h2H20. Reprinted from Andrade and Macias [1214 J, with kind permission from Elsevier
Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.

180

CHAPTER 6

The formation of compact salt films, such as chromate or phosphate films, can greatly
reduce the corrosion rate in a wider pH range. For example, it is reported that in distilled
water a chromate film on galvanized steel delayed the formation of white rust from I day
to 37 days [593]. The protective effect of chromate films on zinc is due to the stability of
Cr20 3 and, more importantly, to the presence of hexavalent chromium in the film. The
hexavalent chromium leaches out when in contact with water, thus providing a local
supply of chromate ions to react with zinc and passivate an exposed surface area. Thus,
generally, the more hexavalent chromium the film contains, the longer the protection
effect. Compared to chromate films, phosphate films are generally thicker and more
insulating but porous, allowing zinc dissolution through pores [993]. However, depending
on the pH range of the solution, formation of the dissolution product may seal the pores
and prevent further dissolution within the pores.
Corrosion product films formed by other surface treatments are found to have various
hindering effects on zinc corrosion. In one study it was reported that the corrosion of Zn
in 3% NaCI is slightly inhibited by dipping the sample in a solution containing a low
concentration of cobalt [73, 296]. In another study, it was found that the corrosion rate of
zinc in aqueous solutions can be significantly reduced by adding rare-earth salts, such as
CeCI 3, to the solution [605]. The inhibition is attributed to the formation of a passive
cerium oxide film.
It has been found in several studies [287,339,351,610] that certain zinc compounds
are more protective than others. Figure 6.10 shows the corrosion resistances of several
zinc alloys as a function of the relative amount of ZnCl z-4Zn(OHh in the corrosion
products [610]. It appears that a high percent of ZnCl z-4Zn( OH)2 in the corrosion products
indicates a higher corrosion resistance. Among the alloys tested, Zn-Mg has the highest
content of ZnCI 2-4Zn(OH)z and thus the highest corrosion resistance. It has also been
reported that an Alz03-dispersed Zn-Co-Cr (0.5-2:0.3-1:0.1-2%) coating was more
100

~
~

Q)

50

zc

a..

O~~----------------L-----------------~
o
500
1,000
Time to red rust occurrence (h)

FIGURE 6.10. Relation between peak percentage of ZnCl z4Zn(OHlz in the corrosion products of various
vapor-deposited zinc alloy coatings and the time to red rust occurrence in the salt spray test. ZA, Zn-lO.7%
AI; ZC, Zn-IO.0% Cr; ZM, Zn-IO.O% Mg; ZN, Zn-9.7% Ni; ZT, Zn-9.8% Ti; EG, electroplating. Reprinted
from Kawafuku et al. [610], with kind permission from The Iron and Steel Institute of Japan, Chiyoda-ku,
Tokyo, Japan.

181

CORROSION PRODUCTS

1000

ZnCI 2'4Zn(OH)1

....

.... , , /

500

,/

Zn-Co-Cr-AllO]

--.,---,.------0-

~o-o

ci

,./'

ZnO

()

2' 1000
<II

C
<IJ
C

Zn-Co

".. ..

500'.......
,
j ..0-, ......... - - - - ....

,r ,,

-,----------0-

~
~ 1000

Zn

,~

FIGURE 6.11. Amounts of ZnO and


ZnCI 24Zn(OHh detected by X-ray diffraction in
Zn, Zn-Co, and Zn-Co-Cr-AI 20 J coatings after
exposure in a salt spray test (SST) for various
amounts of time. Reprinted from Yasuda et al.
[351], with permission from ASM International.

500

rl '\. _---------0-

Ijf"-

0---"1:

''-.., . . . . .

'=

._---'_...J
02345
6
SST Time (Oay)

corrosion resistant than a pure zinc coating, which was attributed to a higher content of
ZnCI 2 AZn(OHh in the corrosion products of the Al 20 3 coating as shown in Fig. 6.11
[351].
Several explanations for the apparent corrosion-inhibiting nature of ZnCI 2AZn(OHh
have been proposed. Kawafuku et al. [610] considered that ZnCI 2AZn(OH)2 is a more
compact corrosion product than ZnO because the c axis of the hexagonal close-packed
structure of ZnCI 2 AZn(OHh is parallel to the direction of the film growth. Others have
suggested that the inhibitive effect of the corrosion products composed mainly of
ZnCI 2AZn(OH)2 is probably due to a decrease of the oxygen reduction rate on
ZnCI 2 AZn(OH)2 [287, 339].

7
Corrosion Forms
7.1. INTRODUCTION
The form of corrosion can be defined according to the nature of the corrosion, e.g.,
galvanic corrosion and crevice corrosion, or according to the effect of the corrosion on
surface morphology, e.g., general corrosion and pitting corrosion, or according to its effect
on bulk properties, e.g., intergranular corrosion and stress corrosion cracking. The
corrosion forms that are commonly found to occur on zinc are general corrosion, galvanic
corrosion, pitting corrosion, and intergranular corrosion.
Galvanic corrosion is a particularly important form of corrosion for zinc applications,
whether as a coating, an anode, or a zinc-dust paint. For zinc, unlike many other metals,
in most situations galvanic corrosion is desirable because it is required for galvanically
protecting the coupled metal, usually steel. Pitting corrosion usually occurs under certain
conditions when the zinc surface is covered with a passive film, or when the surface is
inhomogeneous owing to the presence of impurities, or when the surrounding environment, for example, soil, is inhomogeneous. Intergranular corrosion is found to occur on
certain zinc alloys, particularly zinc-aluminum alloys, when used in warm or hot, moist
environments. It is an important form of corrosion for the applications of zinc alloys
because aluminum is the most used alloying element for die casting and coatings.
The galvanic corrosion, pitting corrosion, and intergranular corrosion of zinc are
considered below in three separate sections that include detailed data and discussion of the
corrosion mechanisms. General corrosion, although it is the most common form of corrosion
for zinc, is not specifically considered in this chapter because it is dealt with in almost all the
other chapters. Other forms of corrosion, such as stress corrosion cracking and hydrogen
embrittlement, are not common in zinc applications. and the limited amount of information
available is only briet1y mentioned. On the other hand, a special subsection is devoted to wet
storage stain, which, although not a form of corrosion defined in corrosion text books,
commonly occurs on galvanized products during periods of storage and transportation.
7.2. GALVANIC CORROSION
7.2.1. lnlroduclion
The galvanic corrosion of zinc coupled to other metals, particularly to steels, has been
the subject of numerous studies. Table 7.1 lists the coupled metals, the environments, and the
specific aspects investigated in a number of studies.
183

184

CHAPTER 7

TABLE 7.1. Studies on the Galvanic Corrosion of Zinc Coupled to Other Metals in Various
Electrolytesa
Metal

Measurement(s)b

Electrolyte

Effect studied

Reference( s)

Ig
Egc,lg
Egc,Ig

Area effect
Al vs. alloys
Ni alloying

356
1102
1192

Humid gas
O.lNNaCI
O.OIMNaCI
5%NaC!
Seawater
0.IMNa2S04,O.lM
KCI, O.IM KN03

E distribution
E, I distribution
E distribution
Egc,lg
E distribution
E distribution, Ig

Kelvin probe
Effect of Resistance
Modeling
Equipment
Cathodic protection
Corrosion rate

1245
556
1233
1100
516
38

Fe

Seawater
3% NaCI

E distribution
I, impedance

Modeling
Zn-rich coating

415
5,317

Steel

CO~-,SO~-,N03 in

Egc,lg

Potential reversal

410,194

Egc,lg
Egc' C
Ec,lg
Ea, Ec,Ig
Ig
Ig
Igo pH
Egc,Ig

Potential reversal
Transient E-t
Cathodic protection
Galvanic E-I curve
Potential reversal
Corrosion products
Variation of pH
Ni alloying
Galvanic protection
Paint adhesion
Surface properties
Solution effect
Effect of paint
Flow rate
RH,ctime
Galvanic protection
SCC of steel d
Thin electrolyte
Protective power

682,709,925
905
1232
1231
457
286
364,1081
1123
1242
1084
905
1099
827
1237
1261
357
1234
522
1240

Al
Al alloys

3.5% NaCI
3.5%NaC!
0.6NNaCI

Cu

Cu-Ni alloy
Cu, brass

hot water
Hot tap water
Seawater
Concrete
pH 3.8-9.5
0.05MNaC03
3.5% NaC!
O.IMNaCI
0.6MNaCI
IN NaCI
Painted
Synthetic seawater
NaC!, MgS04, etc.
5%NaC!
Seawater
Cyclic wet/dry
Soils
3% NaCI
O.OIM Na2S04
O.OOIM Na2S04

Weight loss, E-I


Morphology
Transient E-t
Weight loss

Ig
Ig
Morphology
Weight loss

Ec
E, I distribution
E, I distribution

Stainless steel

O.OINNaCI
Soils

Ig
Ig

Area effect
Effect of soi I
resistivity

518
517

Passive zinc
Pb, Fe
Cd, Cu, Ni
Sn, stainless steel
Ti-6AI-4V
AI, Cu, Pd, Fe
Cu, Pd. Ni. Mg, anodized AI, Sn, Cr,
steel, stainless steel,
carbon-filled polyethylene

0.IMKCr04

Ig

Pitting

205

Soils
3.5% NaCI
Atmosphere
O.lNNaCI

fR drop

Ec-Ea

1239
1103
293, 1093
518

Ege , Ig' weight loss


Weight loss

Ig

Corrosion rate
Corrosi on rate
Area effect

aRef. 1294.

bIg. Galvanic current; Egc. potential of couple; Ec' potential of cathode; Ea. potential of anode; C, capacitance.
cRH. relative humidity.
dSCC, Stress corrosion cracking.

CORROSION FORMS

185

In this section, the theoretical and practical information on galvanic corrosion of zinc
and its alloys coupled to other metals, particularly steel, is organized, and a conceptual
and elemental analysis of galvanic coupling between zinc and steel is presented. Various
factors that may play roles in galvanic action between zinc and coupled metals are
systematically discussed. The principles and practical applications of galvanic protection
of steel by zinc coatings, zinc anodes, zinc-rich paints, and other means are also reviewed.
A conceptual and elemental analysis is also made for the galvanic action between zinc
and steel for geometries of particular importance to applications.
7.2.2. Theoretical Aspects
7.2.2.1. Factors in Galvanic Corrosion. When two dissimilar metals in electrical
contact with each other are exposed to an electrolyte, a current, which is called a galvanic
current, flows from one to the other. Galvanic corrosion is that part of the corrosion which
occurs on the anodic member of such a couple and is directly related to the galvanic
current by Faraday's law [5241.
Under a galvanic corrosion condition, the simultaneous additional corrosion taking
place on the anode of the couple is called the local corrosion or the self-corrosion. The
local corrosion mayor may not equal the corrosion taking place when the two metals are
not electrically connected, called the normal corrosion. The difference between the local
corrosion and the normal corrosion is called the difference effect and may be positive, if
the local corrosion decreases when galvanic current flows, or negative. A galvanic current
generally causes a reduction in the total rate of corrosion of the cathodic member of the
couple. In this case the cathodic member is cathodically protected.
The polarity and direction of galvanic current flow between two connccted bare
metals is determined by the thennodynamic reversible potentials of the metals. The metal
that has a higher reversible potential in the electromotive force (emf) series is the cathode
in the gal vanic couple. Table 7.2 shows the standard emf series of common metals [I 124].
In real situations, owing to the formation of a surface oxide or salt film on the surface
or owing to the difference in the local electrolytes around the two coupled metals, the
polarity may be different from that predicted by the electromotive series. Thus, the relative
position of each metal or alloy in a galvanic series depends on the environment. Table
7.3, for example, gives the galvanic series of some commercial metals and alloys in
seawater [524, 1246J. As can be noted, some metals that are low in the electromotive
series, such as titanium, are actually high in the galvanic series in seawater.
Compared to normal corrosion, galvanic corrosion is generally more complex owing
to the fact that, in addition to material and environmental factors, it involves also
geometric factors. The fundamental relationship in galvanic corrosion is described by
Kirchhoff's second law:
(7.1 )
where R" is the resistance of the electrolytic portion of the galvanic circuit, R", is the
resistance of the metallic portion, Ec is the effective (polarized) potential of the cathodic
member of the couple, and E" is the effective (polarized) potential of the anodic member.
Generally, R", is very small and can be neglected. Ea and Ec are functions of the galvanic

186

CHAPTER 7

TABLE 7.2. Standard emf Series of Common Metals a

i
Noble or cathodic

Metal-metal-ion
equilibrium
(unit activity)

Electrode potential
VS. NHE at 25C

Au-Au 3+
Pt_Pt 2+
Pd_Pd 2+
Ag-Ag+
Hg-Hg~+
Cu-Cu 2+

+1.498
+1.118
+0.951
+0.799
+0.797
+0.342

HrH+

0.000

Pb_Pb 2+

Active or anodic

J.

Sn-Sn2+
Ni-Ni 2+
Co-Co2+
Cd-Cd2+
Fe-Fe 2+
Cr-Cr3+
Zn-Zn 2+
AI_AI 3+
Mg_Mg 2+
Na-Na+
K-K+

(V)

-0.126
-0.138
-0.257
-0.277
-0.403
-0.447
-0.744
-0.762
-1.662
-2.372
-2.714
-2.931

"Data from Ref. 1224.

current I; hence, the potential difference between the two metals when there is a current
flow through the electrolyte does not equal the open-circuit cell potential.
In addition to the potential difference between the two coupled metals, many factors
play roles in determining galvanic corrosion. Depending on the circumstances some or
all of the factors illustrated in Fig. 7.1 may be involved in galvanic corrosion. Generally,
for a given couple, the factors in categories a, b, and c vary less from one situation to
another than the factors in categories d, e, and f. The effect of the geometric factors on
the galvanic actions could, in many cases, be mathematically analyzed. On the other hand,
the effect of the factors related to electrode surface condition and its effect on the reaction
kinetics in real situations can be very difficult to determine.
7.2.2.2. Analysis. The mathematical description of galvanic corrosion can be very
complex because of the many factors involved. It can, however, be simplified for the
galvanic corrosion of zinc. In real applications, galvanic corrosion of zinc occurs mainly
in two situations: when zinc is used as a coating and when it is used as a sacrificial anode.
The specific geometries involved in these applications may be generalized by the scheme
illustrated in Fig. 7.2a. The case in which the distance between zinc and steel, d, equals
zero, represents galvanized steel on which the zinc coating is partially removed, as shown
in Fig. 7.2b. On the other hand, the case in which d (xae - d) and d Xce (the lengths
of the zinc and steel electrodes) can be considered as representing the situation when the
zinc is used as a sacrificial anode, as shown in Fig. 7.2c.

CORROSION FORMS

187

TABLE 7.3. Galvanic Series of Some Commercial Metals and


Alloys in Seawater"

i
Noble or cathodic

Platinum
Gold
Graphite
Titanium
Silver
[

Chlorimet 3 (62Ni, ISCr, ISMo)


Hastelloy C (62Ni, l7Cr, 15Mo)

IS-S Mo stainless steel (passive)


[ IS-S stainless steel (passive)
Chromium stainless steel, 11-300/0 Cr(passive)
[

Inconel (passive) (SONi, 13Cr, 7Fe)


Nickel (passive)

Silver solder
Monel (70Ni, 30Cu)
Cupronickels (6D-9OCu, 40-1 ONi)
Bronzes (Cu-Sn)
Copper
Brasses (Cu-Zn)
[

Chlorimet 2 (66Ni, 32Mo. lFe)


Hastelloy B (60Ni, 30Mo, 6Fe, IMn)
Inconel (active)
Nickel (active)

Tin
Lead
Lead-tin solders
[

18-8 Mo stainless steel (active)


18-8 stainless steel(active)

Ni-resist (high-Ni cast iron)


Chromium stainless steel. 130/0 Cr (active)
[ Cast iron
Steel or iron
2024 aluminum (4.5Cu, 1.5Mg, 0.6Mn)
Active or anodic

J,

Cadmium
Commercially pure aluminum (1100)
Zinc
Magnesium and magnesium alloys

"Reprinted from Fontana and Greene [524], with permission from The McGrawHill Companies.

188

CHAPTER 7

(f) Geometric factors


- area of zinc and steel
- distance between zinc and steel
- location
- shape and orientation

(a) Reversible electrode


potentials

(b) Reactions
- zinc dissolution
- O 2 reduction
---- hydrogen evolution

/
zinc~
(e) Electrolyte properties
- ionic species
- pH
- conductivity
- temperature
- volume
- flow rate

/
(c) Metallurgical factors
- alloying
- heat treatment
- mechanical working

(d) Surface conditions


- surface treatment
- passive film
- corrosion products

FIGURE 7.1. Factors involved in galvanic corrosion of a zinc/steel couple. After Zhang [12941.

The basic relationships for the geometrical arrangement shown in Fig. 7.2a can be
expressed as follows:
(7.2)
and
x a S O,xc;::: 0

(7.3)

where Ea and Ec are the corrosion potentials of zinc and steel, respectivel y, under separate
open-circuit conditions; I] a andl] care the anodic and cathodic overpotentials under the
coupled condition; and ~ VR is the ohmic potential drop across the electrolyte between
x" on the zinc surface and XC on the steel surface. (" the total anodic current, and (., the
total cathodic current, are given by
Ia

(7.4)

=f iaCxa)l dx a
d

Ic =

(7.5)

f i,. (xC)l dx'


o

in which l is the width of the electrodes, and (,(x") and (eX are the respective current
densities on the anode and cathode. Assuming that both the anodic and cathodic reactions
are activation-controlled, they can be expressed by the Butler-Volmer equation [1100]:
C

(a)

x.c
d

o
X

XC

insulating
material

(c)

~~

(b)

zinc coating

electrolyte

~~

zinc anode coupled to a steel cathode. After Zhang [12941.

FIGURE 7.2. (a) General geometry of a zinc/steel galvanic couple. (b) Geometry of zinc-coated steel. (c) Geometry of a

'1..&

electrolyte

QC
\C

[/)

3:::

o
;>:l

.."

(5

[/)

;>:l

o;>:l

190

CHAPTER 7

(7.6)
(7.7)
in which i o" and ioc are the exchange currents for the anodic and cathodic reactions,
respectively, Paa, Pac' Pea' and Pee are the kinetic constants, and B" and Bc are the area factors.
The area factors vary between 0 and I, being equal to I when the whole surface is active
and being close to zero if the surface is fully passivated. In the cases in which the cathodic
reaction is limited by oxygen diffusion in the electrolyte, Eq. (7.7) is replaced by

(7.8)
with F the Faraday constant, Do the diffusion coefficient of oxygen in the electrolyte,
Co, the oxygen concentration in the bulk, and J the thickness of the diffusion layer.
- The total ohmic potential drop in the electrolyte between any two points on the
surface of the anode and the cathode for the situation in Fig. 7.2a consists of three parts:
(7.9)
where Ll Va' Ll Vc ' and Ll Vd represent the ohmic potential drop in the electrolyte in the x
direction across the anode, across the cathode, and across the distance between the anode
and cathode, respectively. They can be further expressed by:
x"

(7.10)

Ll V,,(x") = i,,(x") dR(x")


d

Ll

Vc(x c')

(7.11)

= ic(x

dR(x

(7.12)
where Rd = pd/tl, with p the resistivity of the electrolyte, t the electrolyte thickness, d the
distance between the anode and cathode, and l the width of the electrodes, andi" andic,
given by the following equations, are the sums of the current from x" to Xae on the anode
and from XC to Xee on the cathode, respectively.

ia =

f i" (x")l dx"

(7.13)

:/'

ic =

ea

f ic

(xC)l dx c

(7.14)

CORROSION FORMS

191

It can be seen that the factors listed under categories (a)-( e) in Fig. 7.1, contribute
to galvanic action through their effects on the electrochemical reaction kinetics given by
Eqs. (7.6) and (7.7). For example, changing the pH of the solution may cause a change
in the kinetic parameters, io", ioc ' Pall' Pac' Pca' P"o, or it may cause a change in the effective
area, 0" or ()" through passivation. On the other hand, the geometric factors under
category (0 affect the galvanic corrosion through the parameters in all the equations from
(7.4) to (7.14).
Equations (7.4)-(7.14) apply to a rather general geometry. For a specific application,
they can be further simplified. In the case of Fig. 7.2b, representing the galvanic action on
zinc-coated steel where the bare steel surface is next to the zinc-coated steel surface, that is,
d = 0, the term i1Vd in Eq. (7.9) becomes zero. For the geometry in Fig. 7.2c, representing
the situation of galvanic protection of steel by a zinc anode when d (xae - d) and d > > x,"
!"and!, in Eqs. (7.4) and (7.5) simply become i,,Aaandi,A, with A,,=l(xae-d) and
A, = {xce , the areas for the anode and the cathode, respectively. In addition, i1 V" and t, Vc
in Eq. 0.9) can be taken as zero because they are very small compared to i1V". In such a
case, the geometry of the galvanic cell (i.e. shape and orientation of electrodes, size of
the electrode etc.) become insignificant in determining the galvanic action of the couple,
and the galvanic corrosion of the anode, as well as the galvanic protection of the cathode
surface, becomes uniform. Thus, the galvanic action can be fully described by the
polarization characteristics of the anode and the electrolyte resistance. In this case, the
relation between the effective potentials, galvanic current, and resistance can be graphically represented by the anodic and cathodic polarization curves as shown in Fig. 7.3.
When the solution resistance R is infinite, no current flows, and E( - E" is the
open-circuit value of the cell potential. As R is made smaller, ! increases and E( - E"
becomes smaller because of polarization. When R is zero, Ec - E" becomes zero, and the
galvanic current reaches the maximum, known as the "limiting galvanic current," and is
at the intersection of the polarization curves of the anode and cathode. The exact shapes
of the anodic and cathodic polarization curves depend on the electrochemical reaction
kinetics of each metal in the electrolyte and are thus functions of pH, temperature, solution

IR
1________________
Corro$ion potenlial _

,
I

Limidng
currenl

:
'-l

,
I

Galvanic current
FIGURE 7.3. Graphic estimation of galvanic current.

192

CHAPTER 7

concentration, diffusion, formation of passive films, etc. Normally, the anodic dissolution
of zinc is activation-controlled, with a relatively small Tafel slope (around 40 m V)
[43-45]. The cathodic reactions on the steel surface, on the other hand, can either be
activation- or diffusion-controlled depending on the conditions, particularly solution pH
and aeration conditions. The typical shapes of the anodic polarization curve for zinc (Ea)
and the cathodic curve (Ec) for steel are illustrated in Fig. 7.3.
A galvanic-corrosion system may operate under different control mechanisms. If the
anode does not polarize and the cathode does, then, in solutions of low resistivity, the
current flow will be controlled entirely by the cathodic electrode. Such a situation is
considered to be under cathodic control. If the anode polarizes and the cathode does not,
the status is reversed, and the system is said to be under anodic control. If neither electrode
polarizes and the current flow is controlled by the resistivity of the path, mostly in the
electrolyte, then the system is said to be under resistance control.
7.2.2.3. Potential and Current Distribution. The galvanic corrosion of the anode and
the galvanic protection of the cathode are essentially governed by the potential distribution across the surface of the electrodes. The galvanic current distribution can be
determined from the potential distribution when the potential-current relationships for
the electrodes are known. The exact description of the potential and current distributions
on the surfaces of a galvanic couple can be obtained by solving Laplace's equation:
(7.15)
There are a number of mathematical models using Laplace's equation for galvanic
systems with different cell geometries [321, 1233, 1235, 1236, 1244, 1248]. In these
models the polarization parameter, L i , is often used:
(7.16)
where p is the specific resistivity of the electrolyte; Ii is the current density and '7 j is the
overpotential of the anode or the cathode. The polarization parameter, defined originally
by Wagner [751], has the dimension of length and provides an electrochemical yardstick
for classifying electrochemical systems. Waber [321, 1248] and other authors [415, 1235]
used the polarization parameter to describe the behaviors of galvanic corrosion cells.
According to Waber [321, 1248], whether the anode and cathode behave "microscopically" or "macroscopically" is determined by the ratio of the dimension of either
electrode, C i, divided by the polarization parameter L i The mathematical modeling
indicated that, when the ratio C/L j is small, the variation of current density across an
electrode is small; i.e., the electrode behaves microscopically. On the other hand, when
the characterizing ratio is large, i.e., when the electrode dimension is much larger than
L i , the electrode process can be regarded as macroscopic, and the variation of current
density across the electrode surface is large.
MaCafferty [1235] modeled the potential distribution of a concentric circular galvanic corrosion cell by assuming a linear polarization for both the anodic and the cathodic
reactions. Figures 7.4 and 7.5 show the calculated potential distribution and current
distribution as a function of electrolyte thickness for the case in which the polarization
parameter of the anode, La> is I cm and that of the cathode, Le, is 10 cm. It can be seen

193

CORROSION FORMS

E~

-1.0

ELECTROLYTE
THICKNESS

'3w
...J
<{

z 0.5

f-

a0w

ac::
f-

FIGURE 7.4. Distrihution of electrode potential for L" = I em and Lc


= 10 em for different electrolyte
thicknesses. h. Anode radius a = 0.5
em. cathode radius c = 1.0 em; E~;
= 0 V. E~) = I V. After McCafferty

112351

W
...J
W

E~

-0.0 e==I::1==::.o~--c--.-<.J

1.0
RADIUS, r (em)

that, in the bulk electrolyte, the potential variation across the electrodes is small, but both
the anode and the cathode are strongly polarized; the actual electrode potentials are far
away from E~ and E? . Under a thin electrolyte, the potential variation is large from the
anode to the cathode, but both the anode and cathode are only slightly polarized except
for the areas near the boundary between the anode and the cathode. The galvanic current
increases with increasing electrolyte thickness. Also, the current is distributed on the
electrode surface more uniformly in bulk solution than in thin-layer solutions, where the
current is more concentrated near the contact line in the thin electrolyte. According to the
calculation of Doig and Flewitt [1233], the potential distribution is uniform in the
thickness direction under a thin layer of electrolyte, e.g., I mm. It is nonuniform when
the cell is under a thick electrolyte. Similar results were reported by Morris and Smyrl
[1236] for a galvanic cell with coplanar electrodes. The potential distribution of galvanic
corrosion with more general geometrical conditions has been calculated by Munn and
Devereux using a finite-element method [415,1244].
One problem in mathematical modeling is the assumption that both the anode and
the cathode have a linear or Tafel polarization behavior over the entire potential range.
However, the polarization characteristics of a metal electrode are generally different for
the anode and for the cathode, and they vary in different potential ranges. Sometimes they
also vary with the physical elements in the galvanic cell, such as electrolyte thickness. In
addition, the electrode properties of the coupled metals usually change with time due to
changes on the surfaces and in the solution. These elements need to be taken into
consideration in using a mathematical model for predicting long-term behavior in a real
galvanic system.

CHAPTER 7

194

0.3 r - - - - -- -----,,----- -- - - - ,
CATHODE

ANODE

BULK
ELECTROLYTE
~

~
.- 0.2
~-

iii

zw

IZ

a:
a:
::J
u
...J

0.1

b=0010/

u
0

...J

o
o

~~/

0.5

1.0

RADIUS , r (em)

-0.2

FIGURE 7.5. Current distribution


for different electrolyte thicknesses
under the same conditions as in
Fig. 7.4. After McCafferty [1235] .

r-----------------..."...--~

165/lm

-0.4

T 70/lm

+ Bu lk electrolyte

ill

>

~ -0.6

Q)

a.
-0.8

Zn

10

20

30

40

50

60

Specimen length, mm

FIGURE 7.6. Distribution of potentials on the electrode surface of a galvanic CulZn couple in a O.IN NaCI
solution for different electrolyte thicknesses. Data are taken from Ref. 556.

CORROSION FORMS

FIGURE 7.7. Schematic representation of the electrochemical


cell used for obtaining data on potential and current distributions. D
is the distance between the zinc
and the steel electrode. W is the
width of the steel electrode, and X
is the position on the steel electrode. Reprinted from Zhang and
Valeriote [5221, with kind permission from Elsevier Science Ltd.
The Boulevard, Langford Lane.
Kidlington OXS 1GB. United
Kingdom.

195

Re

/
/

The potential distribution on the electrode surface of a galvanic couple can be


experimentally determined. Rozenfeld [556] showed that the potential variation of the
surface of a coplanar zinc/copper couple greatly increases with decreasing electrolyte
thickness on top of the surface, as shown in Fig. 7.6. The sharpest potential changes take
place on the copper cathode, while the anode does not polarize at all. Zhang and Valeriote
[522] measured the potential and current distributions of a coplanar zinc/steel couple
under thin-layer electrolytes of various thicknesses and salt concentrations using the cell
design shown in Fig. 7.7. The potential distributions on the zinc and steel are similar to
that measured on the zinc/copper couple shown in Fig. 7.6. Figure 7.8 shows that the
galvanic current is higher for thinner electrolytes, which is opposite to the prediction of
the mathematical models [1235, 1236]. In these models, the rate of cathodic reaction on
the cathode is assumed to be independent of the electrolyte thickness. However, under
thin-layer electrolytes, the oxygen diffusion rate is increased since oxygen reduction is
the main reaction on the steel cathode.
Figure 7.9 shows that the galvanic current is larger for a thinner electrolyte when the
anode and the cathode shown in Fig. 7.7 are close together, but the opposite is observed
when the two electrodes are far apart [522]. This change of the relative galvanic current
values for small and large distances is due to the change of the rate-limiting process from
500 -0.09

mm + 0.14 mm *0.27 mm

r--0.54mm *1.08mm +2.16mm

('oJ

E 400
FIGURE 7.8. Effect of changes in
electrolyte thickness on galvanic
current density vs. time curves
measured in O.OIM Na2S04 on
two electrically connected zinc
and steel strips which were 1 mm
apart. Reprinted from Zhang and
Valeriote [5221, with kind permission from Elsevier Science Ltd,
The Boulevard, Langford Lane,
Kidlington OX5 1GB, United
Kingdom.

~
~
-300
~

'(ij

c
~200

~r'-+~+1---+--~--+---~-+

~ 100 -~~i=~~~==~l====I====~~~I====I
u

OL---~--~----~--~----~--~----~~

50

100

150

200

Time (sl

250

300

350

196

CHAPTER 7

J.36mm

o ~'--~--~--~--~--~--~--~--~----

0.5

1.5

2.5

3.5

4.5

Distance (em)

FIGURE 7.9. Galvanic current as a function of the distance between zinc and steel in O.OOIM Na2S04 solutions
of different electrolyte thicknesses, t, for a steel width of 1 mm. Reprinted from Zhang and Valeriote r522 J, with
kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington OX5 1GB, United
Kingdom.

oxygen diffusion at a small distance to ohmic conduction in the electrolyte at a large


distance [522].
The galvanic corrosion of zinc under thin-layer electrolytes measured experimentally
for the couple illustrated in Fig. 7.7 is summarized in Fig. 7. 10 [522]. The galvanic current
(J) increases with the area of the steel (WI) up to a certain size and then decreases slightly
with further increases in the area. It decreases sharply as the distance between zinc and
steel (D) increases because the system becomes ohmic-resistance-controlled. It is relatively less sensitive to the variation of electrolyte layer thickness (t). The width of the zinc
has little effect on the galvanic current because most anodic reactions take place at a very
narrow area at the edge closest to the steel.

7.2.3. Practical Factors


7.2.3.1. Effect of Coupled Metals. Different metal alloys have different electrode
potentials. However, the extent of galvanic corrosion of a metal does not always parallel
the potential difference between the coupled metal alloys. Table 7.4 shows that, although
the potential difference between steel and zinc is much less than that between stainless
steel and zinc and that between Ti-6AI-4V and zinc, the amount of galvanic corrosion
is much larger in the zinc/steel couple than in the other two couples [1103]. This indicates
that the difference between the corrosion potentials of the uncoupled metals is not a
reliable indicator of the extent of galvanic corrosion. Similar results have been reported
on the galvanic corrosion of zinc when coupled to various metal alloys in different
atmospheres [293]. As shown in Table 7.4, the amount of corrosion is greater when zinc
is coupled to mild steel than to copper, although the potential difference between zinc and
steel is smaller than that between zinc and copper. In these situations, other factors, such
as reaction kinetics and formation of corrosion products, rather than just the potential
difference between the two metals, are the rate-determining factors in the galvanic
corrosion. The different galvanic corrosion rates of the anodes coupled with different
cathode materials can be explained in some cases in which the cathodic reaction is oxygendiffusion-limited by the different diffusion rates of oxygen through the oxide films. On

197

CORROSION FORMS

200 ~A

D = O.OSmm

,......
,, , ,

\
I

I"

"

'

'
\
\

200 ~A

50

D =Smm

mm

(a)

200 JJA

w
D =40mm

(b)

w
(c)
FIGURE 7.10. Three-dimensional plots of the galvanic corrosion current of zinc in 0.00 IM Na2S04 solution
for different distances (D) between the zinc and the steel. Reprinted from Zhang and Valeriote 1522). with kind
permission from Elsevier Science Ltd, The Boulevard, Langford Lane , Kidlington OX5 I GB , United Kingdom.

the other hand, when diffusion is not the limiting process, the variation in galvanic
corrosion rate can only be due to the cathodic efficiency for oxygen reduction in the oxide
scale on the cathode surface 11103].
As a result of the galvanic corrosion of the anodic metal, the corrosion of the coupled
metal or alloy is generally reduced, that is, cathodically protected. The extent of protection
for different metal alloys galvanically coupled to zinc has been investigated in atmospheric environments [293,545], in seawater [1246], and in soils [357, 517]. The galvanic
corrosion of zinc is, however, not always beneficial to the coupled metal. It has been
reported that, although zinc is anodic to aluminum, the amount of aluminum corrosion
in 3.5% NaCI solution is increased when aluminum is coupled to zinc, compared to that
in the uncoupled condition [356]. Similar results were reported by Mansfeld et al. [1102]
for a zinc/aluminum alloy couple in 3.5% NaCI solution. The higher dissolution rate of
the coupled Al alloy compared to the uncoupled one is attributed to increased alkalinity
on the surface of the Al alloy due to the cathodic reaction.

CHAPTER 7

198

TABLE 7.4. Galvanic Corrosion Rates of Zinc (20 cru 2) Coupled to Various Alloys of Equal
Size Tested in 3.5% NaCI Solution for 24 Hours"
Galvanic corrosion rate
r~

Coupled alloy
None
Stainless steel 304
Ni
Cu
Ti-6AI-4V
Sn
4130 steel
Cd

"

t.v d

!.j1mJyr)

!.j1mJyr)

(mV)

0
244
990
1065
315
320
1060
600

101
705
1390
1450
815
810
1550
660

905
817
811
729
435
483
258

"Data from Ref. 1103.


hMeasllred as galvanic curren(.
cMeasured as weight loss.
dPotential difference between zinc and the coupled metal before testing.

As can be noted in Table 7.4, the weight loss of zinc when galvanically coupled to
other metal alloys can be much larger than the sum of the galvanic corrosion calculated
from the faradaic current plus the normal corrosion measured in a noncoupled condition.
This implies that self-corrosion (or the local corrosion) of zinc is enhanced by galvanic
alloy coupling to another alloy.
7.2.3.2. Effect of Alloying. Addition of alloying elements in zinc changes its electrochemical properties, such as electrode potential, dissolution kinetics, oxygen and
hydrogen reduction overpotentials, and formation of solid surface films. Since zinc is
widely used in applications in which the galvanic protection of steel is an essential
requirement, alloying is usually engineered to improve the normal corrosion resistance
but not to reduce by much the electrode potential difference between zinc and steel. In
general, additions of small amounts of alloying elements change the corrosion potential
of zinc little. With additions of alloying element to about 10%, the potential of the zinc
alloy may change by 50-100 illV, usually to a more noble value than the corrosion
potential of zinc, as shown in Table 5.4 in Chapter 5. For alloys with more noble elements
such as Cu, Ni, and Fe, the potential can be much more positive when the concentration
is high. For example, Baldwin et ai. [1222, 1223] found that Zn-Ni alloys galvanically
corrode in O.6NNaCI solution when coupled to aluminum alloys or steel up to about 14%
Ni concentration, above which the polarity reverses and the corrosion of the coupled
aluminum alloys or steel is accelerated.
The corrosion potential of an alloy in an electrolyte is a function of time. It tends to
change to more positive values as the time of immersion increases because, in most cases,
the preferential dissolution of zinc causes an enrichment of the more noble elements on
the surface. The polarization behavior of zinc can also be significantly affected by
alloying. High polarization resistance is often not desirable for zinc when used for
galvanic protection of steel because a large polarization of the zinc anode reduces the
potential available for the polarization of steel.

CORROSION FORMS

199

600

Eu
OJ

- - Weight loss
- - - Number of coulombs

500 '"

.c
E

ZnFe

E
vi

400 ~
o
u

.cC1>

300 '0

'"
.2

Q;

'iii
it
u

.c

200 ~

'6
0
c:
<:

100
o L-------------~----~------~----~0

20

40

60

80

100

Area of steel cathode , cm 2


FIGURE 7.11. Effect of area of steel cathode on the weight loss of Zn anode (100 cm" in area) and on the
number of coulombs flowing through the Zn/steel couple over a 96hr period in 1M NaCI solution at 25 C.
After Pryor and Keir r1242\.

7.2.3.3. Effect of Area. The effect of anode and cathode areas on the galvanic
corrosion depends on the type of control over the system. If the galvanic system is under
cathodic control, variation in the area of the anode will not change the total amount of
corrosion much but variation in the cathodic area will. The converse obtains if the system
is under anodic control. The total amount of corrosion will be affected by a change in the
area of either electrode if the system is under mixed control. When the system is primarily
under resistance control, the corrosion will only change with changes in electrode area if
the resistance of the electrolyte also changes with changes in the areas of the electrodes.
Pryor and Keir [1242] studied the effect of the areas of zinc and steel electrodes on
the galvanic corrosion of zinc in 1M NaCI aerated solution. Figure 7.11 shows that the
20 r-~-----------------------, 400

mg , cm 2
+ mg

15

300

OJ

<Ii

vi

.2

.cOJ

10

'"
.2
200 ~

'0;

3:
'iii

'0;

it
u
c:

0>

100

L-____- L_ _ _ _

20

______

40

60

_ _ _ _ _ L_ _ _ _ _ _

80

~~o

100

Area of zinc anode, cm'

FIGURE 7.12. Effect of area ofZn anode on the total corrosion and on the intensity of corrosion of Zn coupled
to mild steel (100 cm 2 in area) over a 96hr period in 1M NaCI at 25C. After Pryor and Keir 11242\.

200

CHAPTER 7
120r---------------------------------~--,

.s 100
Ol

c
0

80

(ij

.2

.s
Q)

40

:l

"C
<n
<n

.2

20
0

20

60
40
Area of anode, em'

80

100

FIGURE 7.13. Effect of area ofZn anode on the weight loss due to local action in Znlsteel couple in 1M solution
at 25C. The area of the steel cathode was 100 cm2. After Pryor and Keir [1242].

galvanic corrosion of zinc increases with increasing steel cathode area. On the other hand,
the galvanic corrosion of zinc changes only very slightly with increasing zinc anode area,
as shown in Fig. 7.12. It is noted in Fig. 7.13 that much of the increase in the weight loss
of zinc with increase in the area of the zinc anode shown in Fig. 7.12 is due to the increase
in the local corrosion. The local corrosion is obtained by the total weight loss minus that
calculated from the charge passed through the couple. These results indicate that the
galvanic corrosion of zinc is mainly cathodically controlled. This is confirmed by the
polarization curves of the zinc/steel couple shown in Fig. 7.14. The shape of the curves
suggests that variation of the steel area will significantly change the galvanic current but
variation of the zinc area will change the current only slightly.

\'l

>,-0.6

ca
E
-S -0.7
0..

10 em'

-0.8
-0.9 L -_ _ _--'-_ _ _ _' - -_ _ _--'-_ _ _ _' - - - - - '

400

800

1,200

1,600

Current, JlA

FIGURE 7.14. Effect of area of steel cathode on the polarization curves for the Znlsteel couple in 1M NaCl.
The area of the Zn anode was 100 cm2. After Pryor and Keir [1242].

CORROSION FORMS

201

TABLE 7.5. Corrosion Rates of Equal-Area Zinc/Steel Couples in Various Solutions"


Corrosion rate (jlm/yr)"
Uncoupled
------

Solution
O.05MMgS04
O.05M Na1S04
O.05MNaCI
O.OOSMNaCI
Carbonic acid
Calcium carbonate
Tap water

Coupled

-----_.

Zinc

+'
285
254
112
10.2

+
+

Steel
66
254
254
178
73.7
150
71.1

----~

-----~------

Zinc
86.4
838
762
218
38.1

+
+

~----

---

Steel

----------

+
+
+
+
+
+
+

"Ref 1099.
!'Specimens of equal area partially immersed for 39 day~.

(Pills signs indicate that specimens 1!ained weight.

The effect of area varies from situation to situation. It has been found that the polarity
of the zinc/steel couple in hot water reversed faster for a larger steel-lo-zinc area ratio
[925]. Schick 1518] found that corrosion of galvanized steel coupled to 301 stainless steel
in a solution containing 266 mg cnl and 70 mg SO~-Il is controlled by both anode and
cathode areas. Mansfeld and Kenkel 1356] found that, for the Zn/ Al couple in 3.5% NaCI
solution, the galvanic current density changed little with the variation of both the zinc
anode area and the aluminum cathode area, largely owing to the inactive surface of the
aluminum.
7.2.3.4. Effect of Solution Factors. In aqueous solution, zinc is normally anodic to
most other common metal alloys and corrodes galvanically. However, in some solutions
in which passivation occurs, zinc can be cathodic to other metal alloys owing to the higher
corrosion potential of the passive surface. Table 7.5 shows the corrosion loss of zinc and
steel in coupled and uncoupled conditions in various solutions [1099]. In all the sol utions,
the galvanic action results in a protection of the steel, but the amount of zinc cOiTosion
varies with the composition of the solution. The difference in the corrosion rates in
magnesium sulfate and sodium sulfate indicates the significant effect of cations on the
reaction kinetics. The presence of metal ionic species more noble than zinc, such as Cu 2+
in solution, is known to enhance the corrosion of zinc due to the mini-galvanic cells
between zinc and the copper islands deposited on the zinc surface [737 J.
In the cases in which there is only a limited amount of electrolyte, the composition
of the electrolyte may significantly change as a result of the galvanic action. Massinon
and co-workers 1364, 1081] found an increase of pH at a confined electrolyte after a
certain time of galvanic action for a zinc/steel couple. Pryor and Keir [12421 pointed out
that when the distance between the anode and cathode is small compared to the dimensions of the electrodes, the galvanic corrosion is small because of the limitation in the
mass transport of the reactants and reaction products.
The position of a galvanic couple in the solution can also affect the galvanic actions
between the coupled metals. Shams El Din et al. [205] found that there is a larger potential
variation near the solution surface between a zinc anode and a copper cathode, which are

202

CHAPTER 7

half-immersed in the solution, due to the higher oxygen concentration near the surface
than in the bulk solution.
7.2.3.5. Effect of Surface Condition. Formation of a surface film, whether a salt film
or an oxide film, may significantly change the properties of the surface. It may not only
affect the rate of galvanic corrosion but may also affect the polarity ofthe galvanic couple.
Usually, in low-pH solution, zinc corrodes without the formation of solid corrosion
products on the surface. The corrosion products formed in neutral and slightly basic
solutions are oxide and hydroxides, usually only loosely attached to the surface [404,
462]. The corrosion products formed on the zinc surface in the pH range between 9 and
13 have varying degrees of compactness and can result in passivation of the zinc surface
[497, 128]. The presence of certain ionic species, such as carbonate, phosphate, and
chromate, can enhance the formation of a passive film in a broader pH range [710]. As a
result of passivation, the potential of zinc can shift to more positive values, thus changing
the galvanic behavior of zinc when coupled to another metal. Typically, for example, if a
stable and compact zinc oxide is formed, the zinc electrode may show a potential more
noble than -0.5 VSCE ' This potential is considered to be related to the semiconducting
properties of zinc oxide. Because zinc oxide is an n-type semiconductor and has a flatband
potential of between -0.4 and -0.6 VSCE [30, 514, 526], at equilibrium a positive
overpotential is required to balance the charge accumulation at the solid/electrolyte
interface.
In certain cases, when the formation of a surface film is not complete, a part of the
zinc surface is passivated and acts as the cathode to form a local galvanic cell, causing an
enhanced corrosion of the rest of the nonpassivated zinc surface [1240]. Shams El Din et
al. [205] found that galvanic current was developed between a passivated zinc sample and
a partially passivated sample positioned in a two-compartment cell, containing O.IM
K2Cr0 4 in one compartment and O.IM K2Cr04 and some NaCl in the other. Pits were
found to be generated as a result of such a corrosion situation.
The galvanic action can vary also with the surface condition of the metals coupled
to zinc. Different kinds of surface films can form on the metals to change the surface
condition. For example, aluminum has a low reversible electrode potential but is usually
cathodic to zinc in neutral or acidic solutions, owing to the formation of a passive
aluminum oxide film. Formation of iron oxide of the form Fe203 may not change the iron
corrosion potential much but may change the electrode behavior of iron because Fe 20 3,
like ZnO, is an n-type semiconductor, which facilitates the cathodic reaction but hinders
the anodic reaction [1124,476].
Jordan [286, 740] studied the effect of corrosion products of zinc and steel on the
galvanic corrosion rate of zinc. He found that the galvanic corrosion rate is dependent on
the behavior of the corrosion products. In general, fresh, hot-dip galvanized zinc coating
is slightly anodic to zinc that is covered with white rust. The rust of steel, as compared to
bare steel, may accelerate the galvanic corrosion of zinc. According to Stratmann and
Muller [1124], oxygen reduction on an iron electrode is greatly increased by the formation
of the rust because oxygen can be reduced in the iron oxide scale, which has a much larger
effective surface area.

CORROSION FORMS

203

TABLE 7.6. Polarity of Zinc/Steel Galvanic Couples in Various Waters


~-------

Couple

-----

Water

T(0C)

pH

CO 2(1

02a

8
8

A
P

B
P

7.7
7.7
7.7

B
A
B

B
A
P

7.7
7.7
7.7
7.2

B
B
B
P
P

P
A
B
P
P

Polarity

---~----

Zn/stcel
Zn/steel

Distilled
Distilled

85
85

Zn/steel
Zn/steel
Znlstecl

Hard
Hard
Hard

Room

Zn-Fc/steel
Zn-Fe/steel
Zn/Zn-Fe
Zn/stcel
Zn/stccl

Hard
Hard
Hard
Tap
Tap

85

85
8S

85
8S

74
70

Always anodic
Sometimes
cathodic
Always anodic
Always anodic
Sometimes
cathodic
Strongly cathodic
Always anodic
Always anodic
Anodic
Cathodic

Refcrence

-".-,_.-

709
709
709
709
709
709
709
709
4S7

682

._----_.II

A, ahscnt: P. present; 8, absent or present.

7.2.4. Polarity Reversal


The polarity of a zinc/steel galvanic couple may reverse under certain conditions;
that is, the steel becomes the anode, and the zinc becomes the cathode. The phenomenon
was first reported by Schikorr [682] in 1939. Many studies have since been undertaken
to determine the different conditions for polarity reversal, with particular attention to that
occurring in hot water and diluted solutions [194, 709, 457, 657]. Table 7.6 lists the
polarity of zinc in various waters as reported in the literature.
The change in the zinc electrode potential is chiefly responsible for the reversal of
polarity since the potential of the steel remains relatively unchanged [709]. It is generally
found that polarity reversal does not occur in distilled water up to 6S 0 e, and, without the
presence of oxygen, it does not occur in hot distilled water [709, 409, 457]. Many other
factors, besides temperature, such as dissolved ions, pH, and time of immersion, are found
to affect the polarity of a zinc/steel couple.
In normal circumstances, such as in cold water, the zinc is anodic to the steel and
thus provides sacrificial protection for the steel. When polarity reversal occurs for a
galvanized steel, the steel is not protected cathodically by the zinc coating, and the
corrosion of the steel underneath the zinc coating occurs when the coating is completely
penetrated by localized corrosion such as pitting. On the other hand, the corrosion of zinc
is reduced because the zinc coating is in a passive state. The zinc, instead of being
galvanically corroded, is actually cathodically protected by the steel.
7.2.4.1. Temperature. Temperature is a critical factor in the reversal of polarity. The
critical temperature, i.e., that at which polarity reversal occurs, is determined by the
solution composition [409, 4lO]. At a given composition, the tendency for potential
ennoblement of the zinc electrode over time increases with increasing solution temperature as shown in Fig. 7.15. Gilbert [709] found that polarity reversal in hard hot water
occurs when the solution temperature is over 60C and disappears if the previously heated
sample is cooled to below 60C. In some solutions, polarity reversal can also occur at
room temperature. For example, polarity reversal was found to occur in a 600 ppm
bicarbonate solution at 28C after 10 days [471].

CHAPTER 7

40C

50 C

60C
70C'------------2

3
Time,

hours

FIGURE 7.15. Effect of temperature on short-circuit current of zinc1steel couple. Solution composition: 70-80
ppm HCO;, 10 ppm SO~-, and 10 ppm NO;. From Hoxeng and Prutton [4IOJ. Copyright by NACE
International. All Rights Reserved by NACE; reprinted with permission.

7.2.4.2. Solution Composition. Solution composition is one of the most important


factors determining the polarity of a zinc/steel couple. In distilled water, polarity reversal
of the zinc/steel couple does not always occur at high temperatures; this is due to the less
protective nature of the corrosion products [709]. The presence of certain ionic species is
required for the reversal to occur, as shown in Table 7.6.

-0.2,----------,-------------,
potential increasing agents

potential decreasing agents

-0.4

~
]~ -06
.

potential
range of
steel

E
~

&:

-0.8

1
I

-1

~
I
I

130 HCO,-

I
10 so:5 NO,-

130 HCO,-

Solution Composition (in ppm)

40 HCO,5S0 4 2 10 NO,100 HCO,10 SO,'-

FIGURE 7.16. Effect of various chemical agents on the potential of zinc in different solutions.

205

CORROSION FORMS

As previously noted, the change in polarity occurs mainly through change of the zinc
electrode potential, because the steel potential is relatively independent of factors such as
solution composition. The zinc potentials in solutions containing various amounts of
different species are summarized in Fig. 7.16, based on data reported by Hoxeng and
Prutton [409,410]. A very small amount of chemical agents, at levels of a few parts per
million, can cause the potential of zinc to change significantly. For a given solution
composition, increasing the concentration of the species on the left-hand side of the figure,
i.e., bicarbonate and nitrate, causes an increase in the potential of zinc, while increasing
the concentration of the species on the right-hand side, i.e., chloride, sulfate, calcium, and
silicate, decreases the potential. The effect of a particular agent on the potential of zinc
depends strongly also on the presence of other agents. This may explain the very different
responses of zinc/steel couples in different waters (Table 7.6), because natural waters can
contain many different chemical agents at various concentrations.
Hoxeng and Prutton [409,410] also investigated the effect of some chemical species
in hot water in the presence of oxygen and found that sulfates and chlorides decrease the
probability of reversal whereas bicarbonates and nitrates increase it. The addition of even
small amounts (up to 20 ppm) of calcium salts or silicates can also decrease the probability
of reversal in zinc/steel couples. In the absence of oxygen, the zinc is always anodic to
the steel.
There appears to be an interaction between ions with opposite tendencies to either
promote or inhibit the polarity reversal in hot solutions. Kurr [471] noted that when
gypsum was added to a solution at room temperature in which polarity reversal had
occurred, the normal zinc potential was restored, as shown in Fig. 7.17. It was also
observed in another study that the addition of 5 g of NaCI per liter immediately restores
the original polarity for a zinc/steel couple in 0.05M NaHCO, solutions [457]. In a study
by von Fraunhofer and Lubinski [457], it was found that polarity reversal occurs at an
NaHCO,:NaCI concentration ratio of20: 1. Table 7.7 lists some of the solutions in which
the tendency for polarity reversal has been investigated.

_OA

>

'/
I

Saturated CaS04
added as gypsum

~ -0.6

'"

:g
(j)

J2

-0.8

-t

L __ _ _ _ _ _ _ _

20

_ _ _ _ _ _ _ _- L_ _ _ _ _ _ _ _ _ _L __ _ _ _ _ _ _ _

40

60

80

Time, days

FIGURE 7.17. Restoring effect of added gypsum on zinc potential adversely affected by bicarbonate-rich
solution (600 ppm HCO'j. 73 ppm NO.1, 20 ppm CO.1 ) at room temperature. From Kurr [4711. Copyright by
NACE InternationaL All Rights Reserved by NACE; reprinted with permission.

206

CHAPTER 7

TABLE 7.7. Polarity of Zinc in Zinc/Steel Couple in Various Solutions

ot

Polarity

Reference

Cathodic

Anodic

Cathodic

Cathodic

Cathodic

Cathodic

457
457
194
410
471
449

Solution

T(C)

pH

CO 2"

0.05M NaHC03

50-75
50-75
65
50-70

8.5
8.5
8.1
8.5

0.05M NaHC0 3

O.OIM NaHC03

40 ppm HCO; + 5 ppm NO;


600 ppm HCO; + 73 ppm NO;
140 ppm HCO; + 10 ppm NO;- +
10 ppm SO~-

Room

50-75

8.2

"P, present: A. absent.

7.2.4.3. pH. The pH of waters or solutions in which reversal occurs is generally


slightly basic, between 7 and 9, as can be seen in Tables 7.6 and 7.7. Hoxeng [409]
evaluated the effect of pH on the zinc potential in solutions containing bicarbonate ions
and found that at about 60C the most noble potential is reached in the pH range between
7 and 8. This pH range is partially determined by the presence of carbonate and
bicarbonate, as they have a buffering effect. Also, as Glass and Ashworth [194] pointed
out, raising the solution temperature of a solution that contains HCO;- will result in an
increase in pH due to the loss of CO 2 to the atmosphere according to the reaction

7.2.4.4. Time of Immersion. The length of time required for polarity reversal varies
greatly with the conditions. The time for polarity reversal generally decreases with
increasing temperature. In a solution containing 600 ppm bicarbonate, the zinc potential
reaches -0.7 VSCE within 1 h at 80C, whereas at 28C it take 280 h to reach that value
[471]. In distilled water, it was found that with daily heating to 85C for 8 h, polarity
reversal occurs at about 3-4 days [709]. Von Fraunhofer and Lubinski [457] observed a
relationship between the temperature (T) and time (t) for polarity reversal in 0.05M
NaHC0 3 solution, as expressed in the following equation with a and b as constants:
tT=bt+aT

Once the polarity is reversed, it may require a certain time to return to the normal
polarity after the solution is cooled. According to Gilbert [709], it takes a few minutes for
the zinc to become the anode again after the previously ennobled zinc sample, which has
been immersed in hot water for a few days, is cooled to below 60C. This process can
take a few hours if the sample has been immersed in hot water for a period of weeks.
When the zinc specimen is immersed for a long time, e.g., 6 months, the zinc can remain
cathodic to the steel for as long as 44 days after the solution is cooled.
Shulerner and Lehrman [925] found that for solutions of the same composition, the
reversal potential is achieved at a faster rate when the ratio of the area of iron to that of
zinc is increased.

CORROSION FORMS

207

7.2.4.5. Mechanism. The mechanisms of polarity reversal of a zinc/steel couple in


hot water and very dilute solutions have been the subject of several investigations [194,
559,458, 449J. It is generally concluded that:
I. Ennoblement of zinc only occurs in certain waters and solutions; it occurs readily
in the presence of bicarbonate and less readily, or not at all, in the presence of
chloride or sulfate.
2. The presence of oxygen is necessary for ennoblement.
3. For a given solution, the tendency for ennoblement increases with increasing
temperature.
The ennoblement of the zinc potential with increasing temperature can be simplistically explained according to Fig. 5.4 in Chapter 5. The anodic polarization curve changes
from I" to 3" as a result of the surface passivation. Assuming that the kinetics of the
cathodic reaction is little affected, then the corrosion potential will increase from E~orr to
E~orr'
Most investigators have attributed the ennoblement of zinc in hot waters to surface
passivation through the formation of a solid corrosion product film. Potential ennoblement is usually a characteristic of surface passivation. The increasing potential with
temperature indicates a transition from an active state to a passive state, due to the
formation or transformation of hydroxide, oxide, or salt films on the surface.
Very different types of corrosion products can form in supply waters depending on
the type and amount of impurities. Gilbert [458] found that different amounts of zinc
oxide and hydroxides can be formed depending on whether the zinc is in distilled water
or supply water, hot or cold, or whether the water contains CO 2 In cold water, zinc
hydroxide is usually the corrosion product, sometimes accompanied by zinc oxide. The
corrosion product at 85C is usually zinc oxide, sometimes accompanied by some form
of hydroxide.
The fact that potential reversal is faster when the area ratio of iron to zinc is increased
[12381 indicates that the formation of the passive film needed for this change is related
to the cathodic reactions on the steel surface. Before potential reversal occurs, the passive
film formation on the zinc surface requires an anodic current that equals the cathodic
current on the steel. The larger the steel surface is, the larger the cathodic current and the
shorter the time for the formation of such a passive film.
Cathodic depolarization has been proposed by some authors as a cause of the
ennoblement of zinc in hot water [709, 458, 449]. Gilbert [458] argued that the change
of the surface state from active to passive is not the main cause of the ennoblement, and,
instead, cathodic depolarization must be mainly responsible for the ennoblement. The
cathodic depolarization at higher temperatures is attributed to an increase in the electronic
conductivity of the semiconducting oxide film. Changes in solution composition lead to
a change in the composition and/or the physical form of the oxide and, therefore,
determine the extent of ennoblement. Trabanelli et al. [4491 characterized the semiconductor property of the corrosion products by measuring the photo-potential response of
the zinc electrode in bicarbonate solutions. They found that the surface film formed at
room temperature had p-type characteristics. It was only at temperatures above 50C, as
polarity reversal occurred, that the film on zinc showed n-type characteristics, which
facilitate the cathodic reactions. Also, examination of the corrosion products by X-ray

208

CHAPTER 7

diffraction indicated that the film was ZnO at high temperature while it was
Zn5(OHMC03)2 at room temperature.
On the other hand, polarity reversal is not necessarily associated with a cathodic
depolarization because the corrosion product films may have poor conductivity. According to Glass and Ashworth [194], the formation of zinc corrosion products (a very thin,
transparent/white film and predominantly basic zinc carbonate) in O.OIM NaHC0 3
actually increases polarization. Also, it is very likely that not all of the passive corrosion
products formed in different solutions are semiconducting. It seems that if the formation
or transformation of a passive film contributes to ennoblement, the effect of electronic
conductivity is only secondary to the effect of passivation, because without passivation
any significant departure of the potential of the zinc from its reversible electrode potential
must be accompanied by a severe dissolution. On the other hand, only a minimal level of
conductivity, whether it is semiconducting, ionic, or via defects, is required for a potential
ennoblement to occur once the surface is passivated. Thus, although semiconducting
behavior of some corrosion products seems to be associated with the ennoblement of zinc
in hot solutions, it is most likely the result, rather than the cause, of the ennoblement.
The main role of oxygen in the ennoblement of zinc in hot waters is deemed to be
related to cathodic depolarization. In the absence of oxygen, there are not enough
oxidizing agents in water for the passivated zinc surface to achieve more noble potentials.
A similar role is believed to be played by nitrate and some other oxidizing agents.
According to Evans and Davies [401,402], one important effect of the oxygen reduction
is that relatively high pH is developed at the local reduction areas, which then regulates
the formation of corrosion products, different physical forms of hydroxides and oxides
being formed at different pH values [404].
7.2.4.6. Effect of Reversal on the Steel Corrosion Rate. When polarity reversal
occurs, the zinc surface is passivated, and the zinc ceases to provide sacrificial protection
to the coupled steel. The effect of this reversal on the corrosion of the steel depends on
the type of environment and the arrangement of the zinc/steel couple. If a bulk piece of
zinc is used as a sacrificial anode and coupled to a steel article, the fact that the zinc surface
became the cathode may not alter the general corrosion rate of the steel very much
compared to uncoupled steel. This is because the zinc surface area in such a couple is
usually small compared to that of the steel. However, if the steel surface is also in a passive
state and the passive film has weak points, the corrosion at these points can be accelerated
as a result of the polarity reversal.
When zinc is used as a coating, polarity reversal will result in a large zinc cathode:
steel anode area ratio at any coating discontinuity. If passivation of the active steel does
not occur, much faster penetration rates at the areas of discontinuities can be expected;
this is the main cause of premature failure of some hot water tanks.

7.2.5. Galvanic Corrosion in Natural Environments


7.2.5.1. Atmospheric Environments. Field exposure data are valuable for a realistic
evaluation of the relative severity of galvanic corrosion. Compared to other types of
corrosion, galvanic corrosion in the field has not been well investigated. This is probably
due to the more complicated situation; in addition to all the factors that may affect the
normal corrosion of a metal, other factors, such as the kind of cathodic materials, the size

CORROSION FORMS

209

of the electrodes, anode and cathode arrangement, etc., are also involved in a galvanic
corrosion system. In addition, this complexity makes application of the field corrosion
data limited because in a real situation it is seldom, or very rare, that the whole
arrangement of material, dimensional, and geometric factors, plus the environmental
factors, is closely similar to that of an earlier field test.
A test program on galvanic corrosion under atmospheric conditions was started as
early as 1931 by the American Society for Testing and Materials (ASTM) [293]. Since
then, a number of extensive exposure programs, most of which took zinc as one of the
metals for the galvanic corrosion couples, have been carried out all over the world [336.
515,545,620]. In general, galvanic corrosion under atmospheric conditions is evaluated
by weight loss measurements. In assessing galvanic corrosion in various other environments, such as in water or soil, the potentials and/or the galvanic current of the two coupled
metals can be measured, but it is very difficult to measure the potentials of the metals
under atmospheric conditions.
For weight loss measurements, two types of assemblies have been mostly used: plate
type and wire-on-bolt type [293]. In the plate type of assembly, a strip of one metal IS
attached by bolts to a panel of another metal. The bolts are insulated from the strip and
panel. The galvanic corrosion is evaluated by visual examination or by weight loss
measurement for the strip or panel. In the wire-on-bolt type of assembly, a wire of the
metal to be tested is tightly wound in the threads of a bolt of the other metal in the couple.
The galvanic corrosion is quantitatively estimated by comparing the weight loss of the
coupled wire to that of the same wire wound on the threads of a plastic bolt.
Galvanic corrosion of galvanized steel occurs at areas where the coating is damaged
and the steel underneath is exposed, such as at cuts or at scratches. At these areas, the
exposed steel is cathodically protected while the surrounding zinc coating is galvanically
corroded. However, in most cases, for galvanized steel the amount of coating loss due to
galvanic corrosion, compared to the loss due to normal corrosion, is small because the
exposed areas of bare steel are usually too small to cause significant corrosion of the
relatively much larger zinc surface area. As a result, the atmospheric corrosion rate,
including the contributions of both galvanic and normal corrosion, of galvanized zinc
coatings is usually very similar to that of uncoupled zinc.
Galvanic corrosion can, however, be a significant contributor to the total atmospheric
corrosion of zinc when it is connected to other metals of similar size. Data in Table 7.8,
reported by Kucera and Mattsson [293], show the galvanic corrosion rate of zinc wires
when coupled to bolts of various metals in different atmospheric environments. Depending on the connected metal and the type of atmosphere. the galvanic corrosion can be as
much as five times the normal corrosion of zinc in a rural atmosphere and three times that
in a marine atmosphere. It can be seen from Table 7.8 that the amount of corrosion is not
directly related to the difference between the reversible potentials of zinc and the coupled
metal. Among the metals, mild steel acts as the most efficient cathodic material, largely
owing to the voluminous rust, which can absorb pollutants and retain moisture and thus
give rise to an aggressive electrolyte of good conductivity.
Table 7.9 shows the galvanic corrosion of zinc and iron in four different atmospheric
environments, assessed using metal disks clamped together with insulating washers
[1229]. In this galvanic cell, the corrosion rate of zinc disk samples is increased by a factor
of l.7-3.7. For zinc/steel couples, the galvanic corrosion of zinc is generally insignificant

210

CHAPTER 7

TABLE 7.8. Galvanic Corrosion Rates of Zinc Coupled to Other Common Commercial Metals
in Various Atmospheric EnvironmentsO (/1mJyr)
Galvanic corrosion rate (Ilm/yr)
Coupled alloy

Rural

Urban

Zinc freely exposed


Mild steel
Stainless steel
Copper
Lead
Nickel
Aluminum
Anodized aluminum
Tin
Chromium
Magnesium

0.5
3.0

2.4
3.3
1.8
2.0
2.4
1.9

l.l

2.2
1.6
1.5
0.4
0.9
1.0
0.7
0.02

Marine

1.3
3.9
2.0
3.2
3.4
2.8
1.1
1.0
2.4
1.9

l.l

1.9
2.6
1.4
0.04

l.l

"Data from Ref. 293.

compared to the decrease in the corrosion of steel resulting from the galvanic action. Also,
galvanic protection of the steel is more effective in industrial and marine atmospheres than in
rural ones. This suggests that the pollutants in the atmosphere are beneficial to the galvanic
protection of steel, although they are very harmful to the normal corrosion of the uncoupled
steel.
Compton and Mendizza [545] showed that the extent of galvanic corrosion of zinc
coupled to different metal alloys does not vary much, even though there are wide differences
in the reversible potentials among the alloys. They suggested that, under atmospheric
conditions, other factors, such as corrosion products on zinc and the other metals, are more
important in controlling the galvanic corrosion of zinc than the differences in the metal
potentials.

TABLE

7.9. Corrosion of Galvanic Couples in Various Atmospheric Environments after Seven


Years' Exposureo.h

Couple
ZnlZn
ZnlPb
ZnlCu
ZnlAI
Zn/Fe
Fe/Fe
Fe/Zn

Rd

W'
187
313
292
362
332
1825
43

Industrial, marine

Rural

Industrial

1.7
1.6
1.9
1.8
1
40

27
47
48
100
81
470
147

1.7

1.8
3.7
3.0

1.

195
328
338
440
349
1534
5

Industrial, humid

1.7
1.7

2.3
1.8
300

43
83
100
141
127
1406
6

1.9
2.3
3.3
3.0
230

"Ref. 1229.
"Samples consisted of two 1.5-in.-diameter disks, in. in thickness, clamped together with l-in.-diameter Bakelite washers,
giving an exposed area of in. all round the edge of the disk and an annular area, ~ in. deep, of 1.275 sq. in.
'Weight loss (in milligrams) of the first metal in a couple, e.g., Zn in Zn/AI.
"Ratio of the weight loss of metal M in the couple MIM' (Metal alMetal b) to that measured for M/M.

Tt;

Tt;

CORROSION FORMS

211

Zinc is usually anodic to other metal alloys in atmospheric environments; however,


aluminum in urban and marine atmospheres and magnesium in all atmospheres are
usually anodic to zinc, and hence their connection to zinc will reduce the corrosion of
zinc [293, 544, 551]. It is shown in Table 7.8 that the amount of zinc corrosion is smaller
for zinc connected to aluminum than that for a free-standing zinc sample, indicating a
galvanic protection of zinc by aluminum. Owing to the formation of a passive film,
however, aluminum is cathodic to zinc in many environments. Doyle and Wright [515]
have reported that aluminum, when tested with wire on a zinc bolt, is cathodic to the zinc
in most industrial atmospheres and some of the marine atmospheres, but the resulting
galvanic corrosion of zinc is usually very small. Galvanic action is most significant in
marine atmospheres because of the high conductivity of seawater, as shown in Table 7.8.
In a marine atmosphere, the galvanic corrosion rate of zinc is found to increase at the
beginning of the exposure and then remains at a relatively constant value afterward [620].
Rain, compared to other types of moisture, is particularly effective in causing galvanic
corrosion. Table 7.10 shows that the galvanic corrosion rate is several times that of normal
corrosion rates in an open exposure while the rates under a rain shelter are similar. This
can be explained by the fact that the electrolyte layer formed by rain is thicker and has a
smaller lateral electric resistance than the moisture formed by condensation.
7.2.5.2. Soil Environments. Like the atmosphere, soil is a complex medium. There
are many sources of variability in soils that can affect the electrochemical behavior of
metal alloys and, therefore, the galvanic actions between the alloys. The zinc electrode
potential can vary by hundreds of millivolts depending on the type of soil [357, 1239].
Thus, the galvanic series measured in soils often do not follow the emf series [11251.
Table 7.11, contained in a study by the National Bureau of Standards (NBS), shows
the annually averaged galvanic corrosion rates of zinc coupled to steel with different
anode!cathode surface area ratios in different soils [357, 1239]. Figure 7.18 illustrates the
potentials of zinc/steel couples in three different soils, recorded over a period of five years.
The cathode material was a 10-in. steel ring made of a 0.5-in.-diameter rod. Zinc anodes
of different surface areas were located I in. below the steel. It was found that the amount
of galvanic corrosion of zinc generally increased with decreasing soil resistivity. However,
the degree of galvanic protection for the steel was lower in a soil of higher resistivity.
Table 7.1 J shows that, although the total corrosion increases slightly with increasing zinc
surface area, the corrosion density decreases fairly significantly, along with a significant

TABLE 7. 10. Galvanic Corrosion Rates of a Zinc Wire on a Steel Bolt after One Year of
Testing with and without Rain Sheltera
Rural
R"

( Jlm/yr) ( Ilffi/yr)

Zinc/iron
Zinc freely exposed

0.5
0.4

Urban

OC
3.0
0.5

"Ref. 293.
"Galvanic corrosion rate with rain shelter.
'Galvanic corrosion rate without rain shelter.

RIO
0.17
0.8

(Ilmlyr) ( Jlmlyr)

1.3
1.3

3.3
2.4

Marine
RIO
0.39
0.54

()

(Ilm/yr) (Jlm/yr)

0.4
0.5

3.9
1.3

RIO
0.1
0.38

CHAPTER 7

212

TABLE 7.11. Galvanic Corrosion of Zinc1Steel Couple in Soilsa


Galvanic corrosion (glyr)

Soil characteristics
Location

pH

R
(,Qcm)

Zinc:steel
area ratiob

Lousville, Miss.

4.3

9390

West Austintown, Calif.

7.1

2582

0/20
1120
2120
3/20
0120

821

2120
3/20
0120

1120

Latex, Tex.

4.5

1120

2120
3/20

Cathode

Anode

10.1
8.3
5.1
4.94
11.2
2.58
1.48
1.45
21.4
0.57
0.19
0.45

0.12
1.55
3.36
6.1
0.1
4.9
7.38
7.79
0.25
13.7
20.7
20.3

"Data from Ref. 357.


hSteel area = 2100 cm2

reduction in the corrosion of the steel. Escalante [517] also found that a linear relationship
existed between the galvanic current and the resistivity of soils for a zinc/stainless steel
couple separated 30 cm from each other at a depth of 0.8 m,
There is a tendency for the galvanic current to decrease with the time of exposure.
This is attributed to the formation of anodic and/or cathodic reaction products that have
the effect of hampering the electrochemical reactions. In some soils, a protective cathodic
film, which inhibits the cathodic reactions, can be formed on steel that is galvanically
coupled to zinc.

-0.2 r - - - - - - - - - - - - - - - - - - - : - - - 1
-Louisville

>'

oj -0.4

West Austintown

+ Latex

.~ -0.6
'0
Q)

c..
:>
8 -O.B

-1.2 '---_ _-'-_ _ _.l...-_ _- ' -_ _ _.!..-_ _--'-_ _- - - '

Time, years

FIGURE 7.18. Variation electrode potentials of zinc/steel couples in three soils over five years. Data are taken
from Ref. 357. See Table 7.11 for soil characteristics.

CORROSION FORMS

213

Galvanic corrosion of zinc in soil is also found to occur when zinc is connected to a
nonmetallic conductive material. Schick [518] reported that in underground telephone
cable plants, rust formed on the galvanized steel used to support nonmetallic conductive
material hardware that was electrically connected to rebar in concrete. Also, increased
corrosion of galvanized steel posts was observed when they were connected to a
carbon-black-filled, polyethylene-jacketed power cable.

7.2.6. Galvanic Protection of Steel by Zinc


The galvanic corrosion of zinc generally results in galvanic protection of the coupled
alloy. This property of zinc has been used in many applications, especially for the
protection of steel. Coating steel with zinc is one of the most common ways to prevent
the steel from corroding in natural environments. The steel is protected by the zinc coating
through a barrier effect and a galvanic effect, in which zinc acts as the sacrificial anode
while steel acts as the cathode. Besides galvanizing, zinc is also widely used cathodically
as a bulk sacrificial anode material for cathodic protection of steel structures. The
principles of protection of steel structure through the use of sacrificial zinc anodes are in
essence the same as those for protection through impressed current by a rectifier. When
a cathodic current is passed through steel, the potential of the steel will be changed to
more negative potentials. When the potential is in the region in which iron is thermodynamically stable, the steel becomes inert. The amount of current required for cathodic
protection depends on many conditions including all the factors illustrated in Fig. 7.1.
The relations among polarization, electrolyte resistance, and cathodic protection of iron
have been systematically studied by Holler [1125].
In most natural environments, zinc corrodes much less than steel, by a factor of
10-100 times [357,539]. The protection of steel by a zinc coating is, thus, mainly through
the barrier effect. However, at the places where the zinc coating is removed or defective,
leaving the steel exposed, the galvanic action between steel and zinc can protect the
exposed steel from corrosion. The galvanic corrosion of galvanized steel is quite unique
because in galvanized steel, unlike other galvanic couplings, the combination of materials
and the geometry do not change much with different applications. The galvanic corrosion
rate of zinc and, at the same time, the extent of galvanic protection for the steel can be
determined based on dimensional and environmental factors.
In atmospheric environments, galvanic action on galvanized steel depends on factors
such as the concentration of the electrolyte, the thickness of the electrolyte, the dimension
of the bare steel surface, and the distance between the zinc and steel. These factors have
been systematically studied by Zhang and Valeriote [522, 1103] using a coplanarly
coupled zinc/steel cell under thin-layer electrolytes (Fig. 7.7). Figure 7.19 shows that the
protection area, on which the potential is below -700 mVSCE, proportionally increases
with increasing width of the steel up to a certain value and then decreases with further
increase in the steel width. It increases with increasing electrolyte thickness, although
less sensitively than with changing of steel width. For a smaller steel width, the whole
steel surface could be effectively under galvanic protection even when the zinc is at some
distance. The protection distance (the distance between the zinc and steel electrodes when
the potential of the steel is below -700 m VSCE) increases with increasing conductivity
and thickness of the electrolyte and with decreasing area of the steel.

214

CHAPTER 7

X (mm)
20

0.001 M Na,sO,

X (mm)
20

12

t (mm)

FIGURE 7.19. Three-dimensional plots of protection width X (galvanic protection area =X x L. see Fig. 7.7)
as a function of the electrolyte thickness (t), steel width (W), and the distance between the zinc and steel (D):
(a) D = 0; (b) D = 5 mm. After Zhang and Veleriote [1104].

Atmospheric exposure testing of mild steel wire on zinc bolts indicated that galvanic
action reduced the corrosion of the steel wire by a factor of 10-40, depending on the type
of atmosphere [293]. Similar results can be seen in Table 7.9, which presents the findings
of a seven-year exposure test with disks of the metals clamped together with insulating
washers, exposing an annular area of each metal-ls in. wide. The galvanic action reduced
the corrosion of steel by 40 times in industrial, 230 times in humid-industrial, and 300
times in seacoast industrial atmospheres. The reduction was only about threefold in rural
atmospheres. The much lower galvanic effect on the corrosion of steel in rural atmospheres is largely due to the relatively nonconductive nature of the moisture. Table 7.9 also
shows that the accelerating effect of galvanic action on the corrosion of zinc, a factor of
1.6-3, is generally insignificant compared with the reduction of steel corrosion.
The galvanic protection of steel by zinc anodes in soils can be seen in Table 7.11
[357]. The extent of protection depends on the resistivity of the soils and the zinc:steel
area ratio. In the soil with a resistivity of 821 Qcm, the corrosion of steel is virtually
stopped at a zinc/steel area ratio of 1120, but in the soil with a resistance of 9390 Qcm,
at an area ratio of 3/20 the corrosion of the steel is reduced only to half of that of the
uncoupled steel. It can be noted in Table 7.8 that the reduction in the amount of steel
corroded is accompanied by consumption of a similar amount of zinc owing to the
galvanic corrosion. This is quite different from the galvanic effect in atmospheric
environments, where the amount of galvanic corrosion of zinc is insignificant compared
to the reduction in the corrosion of steel (Table 7.9).
Zinc is a common material for making sacrificial anodes. Historically, zinc anodes
have been mostly used in seawater-oriented applications [1193]. They are also used for
cathodic protection in hot water tanks [447], fuel storage tanks [281], steel-reinforced
concrete [392, 1232, 1284], and underground steel structures [471,473]. Anode composition, shape, size, and position can be tailored to specific applications [469, 1192, 1193].
More information on zinc anodes can be found in Chapter 15 (Section 15.4).
When zinc is used as an anode material in an electrolyte of low resistivity, it has the
advantages, compared to other anode materials such as aluminum and magnesium, of
high efficiency and little hydrogen evolution [1247]. Self-corrosion due to hydrogen

CORROSION FORMS

215

evolution is significant for magnesium in solutions with pH below 12 and for aluminum
in solutions with pH above 10 [I, 1247, 12431. Owing to the small self-corrosion rate in
most natural environments, the zinc anode has a high galvanic efficiency, 95% to almost
100%, because zinc suspended in seawater or buried in the ground does not corrode
rapidly by self-corrosion [1105]. Another advantage of zinc as an anode material is its
generally low overpotential for anodic dissolution. Because the overpotential on the anode
is low, most of the potential difference between zinc and steel is available to polarize the
steel. In some situations, the smaller potential difference between zinc and steel compared
to that between steel and aluminum or magnesium has the advantage of not causing
overprotection, which could result in an excessive hydrogen evolution on the steel and a
high anode consumption rate.
When a zinc anode is employed in the cathodic protection of a steel ship hull in
seawater, an empirical rule is to employ 1 unit of zinc anode area for 100 units of surface
area of a painted steel hull or for 5 units of bare surface area [1105]. According to such
a rule, the zinc anode consumption rate is about 0.1-0.2 Ib/yr per square foot of painted
steel surface. The presence of Fe in a zinc anode, even in a very small amount, e.g.,
0.00 I %, is harmful to the performance of the anode [230, 1194]. The presence of Fe
causes a reduction of current output and ennoblement of the anode potential owing to the
formation of an insulating dissolution product film on the surface of the anode. Addition
of Al can reduce the effect of Fe in the zinc anode [470].
Steels are often protected by zinc-rich paints, in which the zinc dust serves as a
pigment material and provides some galvanic protection to the painted steel. For zinc-rich
coatings, three conditions must be satisfied in order for the galvanic process to occur [5,
1054J:
I. The zinc particles in the coating must be in electrical contact with each other.
2. The zinc particles must be in electrical contact with the steel.
3. A continuous electrolyte must exist between the zinc particles and the steel.
These conditions imply that the galvanic protection of steel by zinc-rich coatings
improves with increasing amounts of zinc. Thus, high zinc contents, higher than 70%,
are needed for good galvanic protection of steel [5, 700, 1054]. Galvanic action of the
zinc dust is usually effective in the early stage; with time, the oxidized zinc particles
causes a gradual loss of the electrical continuity between the dust particles in the interior
of the paint and the steel. However, the transformation of the zinc particles from the
metallic form to the oxide form exerts a sealing effect on the paint and adds more
resistance to the permeation of aggressive agents from the environment [569, 700].
The recent search for ways of producing more corrosion-resistant automobiles has
led to the development of many new zinc alloy coatings [351, 985, 1007, 1202]. By
comparison with pure zinc coatings, these alloy coatings are in general more resistant to
normal corrosion yet are still effective in providing enough galvanic protection to the
coated steel. Table 7.12 lists some studies on the galvanic action of different zinc alloy
coatings coupled to steel.
Hayashi et ai. [869] measured the time-dependent galvanic current and potential of
cold-rolled steel, coupled to Zn-Fe at various concentrations in differentially aerated
solutions. They found that the initial galvanic current decreases with increasing iron
content in the coating. The extent to which Zn-AI coatings provide galvanic protection

CHAPTER 7

216

TABLE 7.12. Studies on Zinc Alloys Coupled to Steel in Various Electrolytes


Zinc alloy

Electrolyte

Zn-Co.Zn-Mn
Zn-Fe
10% Mg, Ti, Cr, Fe, Ni, Cu
Zn-rich paint
13%Ni
Zn-Ni,Zn-Fe
Zn-Al
5%AI,55%AI
Painted alloy
5%AI,55%Al

Zn-Fe
Zn-Ni

5% NaCl
Supply water
5% NaCl
O.IMNaCI
IN NaCI
0.03MNaCl
IN NaCl
5% NaCl
Atmosphere,
humid
chamber
5% NaCl
0.6MNaCl

Measurement( s)o
Egc.lg
Egc,lg

19

E gc ' E-I curve


Egc
Egc,lg
Egc

19
Area of rust
. Egc,lg
Egc,lg

Effect studied

Reference

Alloying
Polarity reversal
Edge protection
Galvanic protection
Coating stability

390
709
1241
965
43

Time of protection
Galvanic protection
Galvanic protection
Galvanic protection

167
14
827
1237

Current transient
Alloying

869
1223

"Egc ' couple potential; I" galvanic current.

depends on the Al concentration. Coatings with more than 60% Al behave like aluminum
and provide little galvanic protection to the steel in atmospheric environments [248].
Zn-Ni coatings with Ni concentration below 14% provide galvanic protection to steel in
0.6N NaCl solution, and the polarity is reversed when the Ni concentration is higher than
18% [1223].
Suzuki et al. [1241] investigated the galvanic effect of different electroplated zinc
alloy coatings on the edge protection of painted panels of different zinc-alloy-coated
steels. The extent of galvanic corrosion at the cut edge of the painted steel subjected to a
wet-dry cyclic test varied with the alloying elements. Figure 7.20 shows the effect of

12

Cu

10

'#.

'In

8
Ti

Ni

-0
Q)

a:

Fe

Zn

Mg.
0
-1.05

Cr

-1

-0.95

-0.9

Immersion potential,

-0.85

-0.8

VSCE

FIGURE 7.20. Correlation between immersion potential and red rust ratio on the cut edge of various
zinc-alloy-coated steels. From Suzuki et al. [124IJ. Copyright by NACE International. All Rights Reserved
by NACE; reprinted with permission.

CORROSION FORMS

217

alloying elements on the red rust ratio on the edge. The throwing power of galvanic
protection is generally larger for coatings with more negative electrode potentials.
In summary many applications of zinc and its alloys, whether as a coating, an anode,
or a zinc-rich paint, involve galvanic corrosion, which is desirable for zinc, unlike many
other metals, because it is required for protecting another metal, usually steel. Galvanic
corrosion is complex because it is a function of many factors, including values of electrode
potentials, number of reactions and their kinetics, metallurgical conditions, surface
conditions, electrolyte properties, and geometric factors. Depending on the circumstances, some or all of the factors may playa role in the galvanic corrosion. Generally,
the effect of the geometric factors on the galvanic action can, in many cases, be
mathematically analyzed. On the other hand, the effect of the factors related to electrode
surface condition and reaction kinetics in real situations can be very difficult to determine.
For galvanized steel, unlike other galvanic couplings, the combination of materials and
the geometry do not usually change from one situation to another, simplifying the analysis
of the galvanic action. In real situations, each of the various factors needs to be considered
in order to maximize the beneficial effect resulting from the galvanic corrosion of zinc.
7.3. PITTING CORROSION

7.3.1. Introduction
Pitting corrosion is a form of localized attack that results in holes in a corroding
metal. It is a serious type of corrosion: though the extent of the reaction may be small,
the damage may be severe, particularly when the metal concerned is used as a container
for fluid (e.g., a heat exchanger tube or underground service pipe). Generally, a pit may
be described as a cavity or hole whose surface diameter is about the same as or less than
the depth [524]. One characteristic feature is that pitting generally occurs on metals and
alloys whose surface is in a passive state, with severe dissolution of the metal at localized
points and relatively little dissolution of the rest of the exposed surface. Another feature
is that it usually occurs in a medium that contains aggressive anions, such as chloride
ions, which cause local breakdown of the passive surface. A third characteristic feature
of pitting is the existence of a threshold potential below which pitting does not occur but
above which pitting occurs [197].
The susceptibility of a metal or alloy to pitting can be evaluated by several methods:
(I) determination of the characteristic pitting potential; (2) determination of a critical
temperature of pitting; (3) measurement of pit density; (4) evaluation of the size and depth
of the pits; and/or (5) determination of the critical concentration of aggressive ions.

7.3.2. Occurrence of Pitting


7.3.2.1. In Distilled Water. Pitting is a common form of corrosion of zinc in distilled
water at room temperature, as was reported as early as 1919 by Bengough and Hudson
[654]. According to their observations, if zinc panels are placed vertically in distilled
water, corrosion pits form, often arranged in straight rows of unconnected pits as shown
in Fig. 7.21. According to Evans and Davies [402], the pits are the result of the
gravitational sinking of solid corrosion products, which lodge at points where they shield
the underlying zinc from oxygen, facilitating the anodic attack at those points. Pitting

218

CHAPTER 7

o
,
I

,,

"

,.

,
"

,
J

FIGURE 7.21. Vertical arrangement of pits in zinc


exposed to distilled water. After Bengough and Hudson
[654].

does not occur without oxygen or under high oxygen pressure. The local depletion of
oxygen was found to be necessary for pitting corrosion. Zinc samples moving in distilled
water saturated with oxygen show no pitting.
The effect of oxygen depletion on pitting formation was demonstrated by Evans and
Davies [402] by covering the zinc surface with a polyethylene fiber to produce two parallel
lines of pits as shown in Fig. 7.22. At the two crevices on each side of the contact line,
the metal, being locally shielded from oxygen, become anodic to the main surface and
corroded preferentially [402].
When distilled water contains very small amounts of dissolved salts, general corrosion, rather than pitting corrosion, occurs [401]. The corrosion is no longer confined to

-~~PJl
Zinc

or trench

Pits~

Trench~
I

ij

Surface afler removal of fiber

FIGURE 7.22. Two lines of pits produced in


zinc in distilled water by contact with polyethylene thread. After Evans and Davies
[402]. Copyright by NACE International.
All Rights Reserved by NACE; reprinted with
permission.

CORROSION FORMS

219

the initiation points and tends to spread outward over an arch-shaped area owing to the
separation of the anode and cathode when salt is present in distilled water. This is also
confirmed by another study [559] in which the pitting corrosion rate for zinc samples
immersed in distilled water for 30 days was 22 f.1m/yr, while it was only 6.S f.1m/yr for
samples immersed in water containing 0.002% NaCI.
Kenworthy and Smith [4001 reported the effect of dissolved carbon dioxide on the
formation of pitson zinc and galvanized steel in pure water at room temperature. The
depth of pits was found to decrease, but the number was found to increase, with increasing
amounts of dissolved carbon dioxide in the water. After 57 days of immersion, the pit
depth on zinc samples was about SO f.1m at 0.6 ppm CO 2 and was only I f.1m at 36 ppm
CO 2 ; in contrast, the total amount of corrosion was four times more at 36 ppm CO 2 than
at 0.6 ppm. Compared to the average pit depth zinc samples, that on galvanized steel
samples was much less, being about 10 f.1m at 0.6 ppm CO 2 compared to SO f.1m for zinc;
however, the total corrosion rate was similar to that for pure zinc. For galvanized steel,
as pits penetrate to the zinc-iron alloy layers, the rate of penetration decreases owing to
the more noble nature of the alloy materials; this causes the pits to spread instead of
penetrating further.
7.3.2.2. In Hard Water. Kenworthy and Smith [400] found that the pitting penetration in zinc in hard water is only a fraction of that in distilled water. This agrees with the
general observation that the corrosion attack is more uniform when the water contains
salts [401].
7.3.2.3. In Hot Water. Pitting is a common form of corrosion for zinc in hot water.
This can be a serious problem for galvanized steel hot water tanks [400,655,656, 709].
Gilbert [709] investigated the corrosion of zinc in hot water and found that the corrosion
attack is highly localized and very deep. After five months of immersion in hard supply
water, saturated with 95% air and l.5% (about 14 ppm) carbon dioxide at S5C, the
maximum pit depth was about 0.3 mm on zinc samples and about 0.15 mm on galvanized
coatings.
In hot soft water, pitting corrosion is likely to lead to rapid penetration of galvanized
coatings. Because of the reversal of polarity for zinc/steel galvanic couples in hot water,
the pitting can continue through the substrate steel. In hard water, the corrosion is likely
to be stined by the deposition of a protective scale, depending on the heating method.
The presence of copper in the water was found to enhance the pitting corrosion of
galvanized coatings in hot water [257].
The extent of pitting in hot water also varies with the content of dissolved carbon
dioxide. Kenworthy and Smith [4001 reported that the depth of pitting is greatest with 5.6
ppm carbon dioxide and is least with 10 ppm carbon dioxide. The pits formed on solid
zinc are, in general, much deeper than those formed on galvanized coatings. As a result
of pitting, the corrosion failure of galvanized steel in hot water tanks may occur even
when SO% of the coating still remains.
7.3.2.4. In Solutions. The occurrence of pitting corrosion on zinc in solutions
depends on the pH and composition. In general, pitting corrosion is not likely to occur in
acidic solutions and is most often found in slightly basic or basic solutions. Lorking and
Mayne [334] reported that solutions prepared from distilled water with additions of
NaOH, at pH values from 5.8 to II, can cause pitting on zinc. Figure 7.23 shows that the
pitting rate increases with NaCI concentration from about 20 j.1m/yr at NaCi concentra-

220

CHAPTER 7

200,---------------------------------------,
- Average rate

+ Maximum pit depth/ year


...: 150

<::
E

Ol.

ai

1 100
c

'Vi

50

O~

____
0.001

0.0001

_L~

____

0.01

____

_ _ _ _ _ _L __ _ _ __ L_ _

0.1

10

Concentration of NaCI, %

FIGURE 7.23. Pitting corrosion rate of zinc in NaCI solutions at room temperature as a function of solution
concentration. Data are taken from Ref. 559.

tions of a few parts per million to about 160 f.1mJyr at 3.5% [559]. The pitting rates in the
solutions are much faster at higher temperatures. Pit density generally increases with
increasing halide ion concentration [45, 355,402]. The number of pits and pit size greatly
depend on the initial surface condition [355].
7.3.2.5. In Atmospheric Environments. Pitting corrosion in atmospheric environments has been seldom reported as the main cause offailure of zinc products. In an ASTM
atmospheric exposure program, pits were observed on high-grade zinc and on a 1%
Cu-zinc alloy after two years of exposure in four different environments [294,549]. The
pits were shallow, their depths generally being smaller than their diameters. Most of them
were more like dimples than pits by definition. The depth of the pits increased with time,
but the ratio of pit depth to the surface-averaged corrosion penetration decreased with
time.
Williams et al. [235] reported that pitting corrosion occurs on Zn-22Al alloy at 100%
relative humidity and 50e. Pitting was found to be more severe when sulfur dioxide was
also present. Pit depth increased from an average of 30 f.1m without sulfur dioxide to about
100 f.1m with 100 ppm sulfur dioxide after 1000 hours of exposure. External stress did
not noticeably change the pit depth.
7.3.2.6. In Soils. Corrosion of zinc in soils occurs unevenly over the surface, as soil
is a very inhomogeneous environment. This unevenness may lead to the development of
pitting, the extent of which varies significantly depending on the soil chemical composition and texture. In a study by the U.S. National Bureau of Standards, it was found that
zinc pits in most soils, and the maximum pit depth is, in general, five or more times the
average corrosion penetration [357].
7.3.2.7. In Zinc Batteries. In zinc cans used in zinc batteries, pitting often occurs
along lines of particular stress introduced during can formation [1145]. In inadequately
sealed cells, deep pits tends to form at the zinc/electrolyte/air boundary. The occurrence
of pitting in Leclanche types of electrolytes is a function of many factors, particularly the
presence of impurities such as Pb 2+ and Cd 2+ in the electrolytes [884, 885]. An increase

221

CORROSION FORMS

in ZnCI 2 concentration favors a more isolated pit formation. A trend from general
metallographic etching at low concentrations of ZnCI 2 0.0 1M) to severe pitting at high
concentrations (> I.OM) is found. The formation of pits on zinc cans is very detrimental
since perforation of the zinc may develop very rapidly, leading to electrolyte extrusion
and/or drying out of the battery [1145].

7.3.3. Pitting Potential


As described in Chapter 3, the potential at which the passivation begins to break
down at localized areas, leading to pitting, is defined as the pitting potential. Pitting
potential can be determined from an anodic polarization curve as the potential at which
the current begins to sharply rise with increasing potential. Figure 7.24, as an example,
shows the potentiodynamic polarization curves for zinc in KOH solution as a function of
chloride concentration in the solution [46]. The pitting potential, at which current sharply
rises, decreases with increasing KCl concentrations. Table 7.13 shows the pitting potentials measured in various solutions.
The value of pitting potential can be used as an indication of pitting tendency of a
metal in an electrolyte. Generally, the more positive the pitting potential, the more difficult
it is for pitting to occur at the rest potential. The values of the pitting potential can be
affected by many factors such as pH, solution composition, and temperature. Solution
composition (i.e., the nature and concentration of dissolved chemical species) has been
found to be a particularly critical factor in determining the value of the pitting potential.
Several investigations have shown that pitting potential becomes more negative with
increasing chloride concentration, as shown in Figs. 7.24 and 7.25 [46,355]. According
to Keitelman et al. [652], a linear relation exists between the pitting potential and the
5

NE
<)

=<:t

.-

0()

...J

FIGURE 7.24. Potentiodynamic anodic polarization curves for Zn electrodes in 10- 2M KOH solutions containing various concentrations of KCI.
After Ahd El Haleem [46].

0.8
E, Vsee

1.6

222

CHAPTER 7

TABLE 7.13. Pitting Potentials Measured from Potentiodynamic Polarization Curves in


Various Solutions
Solution

pH

0.02M NaCI04 + 0.2M H 2B0 3

-0.85
-0.73
-0.76
-0.9-0.95
+1.1
-0.7
0.6
-0.66
-0.7
-0.8
-0.4
+1
0.1

8
9
9
9
9.2
9.2
10
II
11
II
11.5
12
12

1M NaN0 3a
O.IMNaCl a
105 ppm NaCI + borate-NaOH buffer
210 ppm NaCI + borate-NaOH buffer
1M Na2S04 + 0.2M Na2HPO/
0.02M NaCI04 + 0.2M H 2B0 3
O.OOIMKOH"
O.OOIM KOH + 5 x 1O- 4M KCI"
3M NaCI"
O.OIMNaOH + 0.00IMNaC1"
O.OIM KOH + O.OOIM KCI"

Reference

Epit (V SCE )

16
16
45
45
355
355
701
16
46
46
3
37
46

aDeaerated.

logarithmic concentration of Cl- in borate-buffered sodium chloride solutions. Abd EI


Haleem [46] also showed that the pitting potential of zinc in D.DIM KOH decreases
linearly with logarithmic concentration of CI-; similar relationships were observed for
Br- and r- as shown in Fig. 7.26. Many other ionic species have been identified as
pitting-enhancing agents. Augustynski et al. [16] found that, in addition to Cl-, Br- and
r-, CIO:;, SO~-, NO), CIO), F", and CH}CO;: added to NaOH-H3BOr buffered solutions
reduce passivation breakdown potential.
The presence of certain anions in the solution, however, can inhibit the pitting corrosion
of zinc. Figure 7.27, reported by Abd EI Haleem [37], shows the effect of three anions that

2
pH 9.2
.., pH 11
w

:;l

>

cij
~
Q)

0
a. 0

_1L-------L-------L-------L-------L-----~

200

100

300

400

500

Chloride concentration, ppm

FIGURE 7.25. Pitting potential of zinc


and Lotlikar [355].

VS.

chloride concentration in borate-buffered solutions. After Davies

CORROSION FORMS

223

1.2

CIT Br-

0.8

- I-

0.4

"'
M

>

-0.4
-0.8
-1.2

-4

-3 .5

-3

-2.5

-2

log C

FIGURE 7.26. Variation of the critical pitting potential, 1" of zinc with the concentration of cr. Br-, and lions in 1O- 2M KOH solutions. After Abd EI Haleem [46).

enhance pitting resistance. In this work, the pitting potential was found to increase with
the concentration of the anions in the solution in the order PO;- < crO~- < C05-. In another
report, the inhibiting effect was reported to be in the order WO~- < PO~- < crO~- [1091. All
these anions were found to promote the formation of a solid oxide or salt film on the zinc
surface, which results in passivation [3, 57, 65, 93, 98, 127]. The amount of passivating
agents required for pitting inhibition depends on the concentration of the aggressive
species. It has been reported that in concentrated NaCI solutions. although the total
corrosion is reduced with addition of chromate up to 4 gil, the pitting rate is little affected
compared to that without chromate addition [559].

2~--------------------------------~

CO,"

1.6
1 .2

>~ 0 .8
~

_ _- -. CrO/"

0.4

o
-0.4

-0~2L.5------'
_2------1-L.-5------.1..1-------!0.5

FIGURE 7.27. Variation of Epitting for Zn-Ti alloy with the concentration ofC05-, CrO~-, and HPO~- in O.OIM
NaOH + O.OOSM NaCl solutions. After Abd El Haleem [37).

CHAPTER 7

224

It is important to note that the pitting occurring at the pitting potential is very different
from that occurring at the corrosion potential (the rest potential). The pitting potential can
be an indication of the pitting tendency in a particular environment but, by itself, does not
provide a basis for predicting whether pitting actually does occur at the corrosion
potential. This is because pitting potential is measured with an enforced external anodic
current. In real applications, zinc products, except for zinc anodes, are normally used
under a condition of no enforced current, i.e., at the corrosion potential. However a
comparison between corrosion potential and pitting potential can provide a practical
indication about the tendency for pitting corrosion at the corrosion potential. Pitting is
less likely to occur when the corrosion potential of the metal is well below the pitting
potential than when the corrosion potential is close to the pitting potential.

7.3.4. Morphology
The morphology of pits can be characterized by the number, shape, size, and
distribution of pits and corrosion products. Evans and Davies [402] investigated the pit
arrangement and characteristics of corrosion products on zinc sheets in distilled water.
The distribution and arrangement were found to be different for horizontally and vertically
placed samples. For the vertically placed sample, the pits were found to arrange in straight
rows and disconnect (Fig. 7.21) owing to the oxygen-shielding effect of the corrosion
product deposited from an upper pit. The pits were filled and were generally surrounded
by rings of the white corrosion product. Outside the white rings, the zinc was unattacked
but showed interference colors by specular reflection. The arrangement of the colors
showed the film to be thickest close to the pits. When the rings of white matter were
brushed off, the film was found to continue below them as shown in Fig. 7.28. The white
product around the pits was zinc oxide with some p-zinc hydroxide. The white matter
obtained from within the pits gave no lines in X-ray diffraction patterns, suggesting that
it was amorphous. The film giving the interference colors was identified to be zinc oxide.
Pits on galvanized hot water tanks were found to be usually formed under gas bubbles
that adhered to the surface during immersion of the zinc articles so that bubble cups (at
the base of the gas bubbles) of corrosion products were formed around the pits [400,709].
Negatively charged
colloid particles
Ring of
white matter

Ring of
white matter

Color film

Color film
Bare ring

------------~--~/

r~~----------------

Zinc
Pit

. Loose white matter

FIGURE 7.28. Schematic illustration of the morphology around a pit. After Evans and Davies [402].
Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.

CORROSION FORMS

225

These bubble cups consisted of white zinc corrosion products, and when failure of the
coating occurred, it was always beneath these white products. Very often, the bubbles
were removed when the water was changed, but fresh bubbles tended to appear in the
same places as before. The center of the cups on the zinc sample had the same color as
the surrounding metal, while the center of the bubble cups on galvanized steel had a darker
red coloration than the surrounding metal, indicating the dissolution of the Zn-Fe alloy
layers in the galvanized coating.
In Leclanche battery electrolytes, pitting on zinc has the following characteristics,
according to Baugh et al. [884, 885]. (a) Pits appear to be distributed indiscriminately
across the surface of the zinc electrode with no preferred tendency to form on or near the
grain boundaries. (b) The sites of initiation seem to be at the points of emergent
dislocations. (c) Similar pitting characteristics were observed on cleaved and polished
single-crystal electrodes, indicating that the effect of surface preparation on the pitting is
of secondary importance. (d) The number of pits formed under the open-circuit condition
is smaller than that formed under anodically polarized conditions. (e) Almost all the pits
observed exhibited radii greater than their depths.
The pits formed on zinc often have certain crystallographic features. Both rectangular and hexagonal pits can develop [29,45]. Alvarez and Galvele [45] observed that, after
passivity breakdown, zinc dissolution does not propagate in a flat front, but in certain
crystallographic planes. Pitting in O.IM NaCl solution, at a pH of9, was found to develop
along {OOOI} planes as well as along {IOID} planes. Baugh et al. [885] found that the
pits formed on single-crystal zinc electrodes with the basal plane exposed, at slightly
anodic potentials in Leclanche types of electrolytes, were hexagonal, and the regularity
of the hexagonal shape depended strongly on the composition of the electrolytes.
7.3.5. Mechanisms
The detailed processes in pitting corrosion are complicated. Whether pitting corrosion occurs depends on many factors. For a metal in solution, it can generally be said that
(1) physical imperfections on a metal surface determine the initial points of attack and
(2) chemical factors in the solution determine whether the attack will occur, be healed
through repassivation, remain localized and develop into a pit, or spread out and lead to
a general attack.
The mechanism involved in the pitting on zinc in cold distilled water is in essence
different from that in hot water containing trace amounts of salts. In distilled water the
zinc surface is initially not passivated. Because of the high resistivity of the water, the
electrochemical reactions are highly localized, causing the formation of pits. The areas
surrounding these pits then become passivated due to the rise of the local pH as a result
of the cathodic reactions [402]. The localized passivation is a result of pitting. In hot water,
on the other hand, the whole surface is generally passivated with or without the formation
of pits. Pitting occurs at the places where the passive film has undergone breakdown.
According to Evans and Davies [333,401,402], when corrosion starts at imperfections, the dissolution of zinc results in an acidification inside the pits due to the formation
of zinc hydroxides. When ionic transport is efficient, such as in impure water, the anodes
and cathodes can be physically separated, and ionic transport will allow the acid at the
anodes to spread out and dissolve the preexisting oxide film, hence leading to a general

226

CHAPTER 7

attack. In distilled water, the ionic conductivity is very poor, and hence the anodes and
cathodes stay very close together. In this case, the acid at the anodes is neutralized by the
base generated at the cathodes before it can spread out. While the corroding sites keep
active, the surrounding areas remain unattacked until pits are developed. The purer the
water, the fewer and smaller the pits are. Motion of the sample in water can promote ionic
movement along the surface and, therefore, reduce the chance of pitting, as was confirmed
by Evans [401].
In solutions, pitting of zinc only occurs on passivated surfaces [45,46,355]. To cause
pitting, breakdown of the passive film must occur. The electric field (anodic potential)
required for the breakdown depends on the presence of certain ionic species, particularly
halide ions, in the solution. The role of halide ions has been postulated to involve specific
adsorption on certain sites of the passive film, followed by penetration of the film under
the influence of an electric field. As a result. the field required for passivation breakdown
at these sites is lowered, and when an anodic potential is applied on the electrode and the
corresponding field exceeds the critical value. film breakdown occurs, leading to pitting
corrosion. When the concentration of chloride ions is high, the whole film can be
weakened and breakdown can occur over the whole surface, leading to a more uniform
corrosion.
The theory oflocal acidification seems to be widely accepted for the pitting corrosion
of zinc. To initiate and maintain an active pit. local acidification is considered to be
necessary [45, 128. 402, 652]. This local acidification is a result of metal dissolution:
(7.17)

(7.18)
The degree of acidification required depends on the bulk solution pH, the amount of
chloride ions. the electrode potential, and the presence or absence of inhibiting agents.
According to this model. the concentration of chloride ions determines the extent of
passivity breakdown; the pH regulates passivation and repassivation; and the electrode
potential controls not only the extent of breakdown but also the rate of dissolution at the
places where breakdown occurs. If the dissolution in a pit is not fast enough to maintain
sufficient acidity to prevent repassivation, a new passive film will form and the pit can
become stifled. The fewer aggressive ions and the more passivating agents in solution,
the higher is the potential needed to give a sufficient dissolution rate to maintain the acidic
condition. The pitting potential is considered to be the minimum potential at which such
acidification can be maintained [45].
According to Galvele and co-workers [45, 652], the pitting potential of zinc in
aqueous solutions is determined by the local acidification resulting from reactions on the
electrode surface and consists of three components: .
(7.19)

E;

where
is the corrosion potential of the acidified pitlike solution, '7 is the anodic polarization
necessary to draw enough current through the pit to maintain the local solution chemistry for
an active pit, and t1J is the potential drop resulting from the resistance in the electrolyte. The

CORROSION FORMS

227

relative contribution of each component to the pIttmg potential, Ep;1' vanes with pH,
concentration of buffering agents, and solution conductivity.
The formation of pits on a zinc surface in the Leclanche type of solutions was
concluded by Baugh et al. [884, 885] not to be due to the breakdown of passivity by
aggressive ions since neither the CI- concentration nor pH does significantly changes the
pitting characteristics. Instead, pitting was considered to be due to the presence in the
electrolyte of small amounts of impurities, such as Pb 2+ and Cd"+ ions, emanating from
ZnCI 2 The deposition of Pb and Cd on the zinc surface forms an metallic film. This film
allows dissolution of the zinc at a few unprotected sites, which leads to pitting. The
different effects of various anions on the pitting characteristics are attributed to their
specific adsorption on the zinc surface, which affects the number of nucleation sites and
the rate of dissolution inside the active pits.
7.4. INTERGRANULAR CORROSION
7.4.1. Introduction

Intergranular corrosion is defined as the localized corrosion at or adjacent to grain


boundaries, with relatively little corrosion of the grains [524]. It results, at best, in a loss
of ductility and strength and, at worst, in a very rapid, complete destruction of the metal.
In most cases, intergranular corrosion of metals is associated with the presence at the
grain boundaries of a phase different from the base metal in its electrochemical behavior.
The extent of intergranular corrosion is usually evaluated through metallographic examinations of the surface and a cross section of the corroded sample and quantified by the
depth of penetration along the grain boundaries. It can also be assessed by measurements
of strength and ductility.
7.4.2. Occurrence

Intergranular corrosion of zinc alloys was first reported in the first quarter of this
century as a serious problem for die-cast alloys used in hot water and warm humid
atmospheres [535]. It was later realized that the problem was associated with the presence
of aluminum along with certain impurities such as lead and cadmium in the alloys.
Although most reported cases have been related to die-cast zinc-aluminum alloys,
intergranular corrosion has been found to occur also in some other zinc products such as
zinc coatings under warm and humid conditions [723, 1289].
Intergranular corrosion of zinc-aluminum alloys has been observed to occur in
different environments: in the atmosphere, in water, in solution, and in concrete. Defrancq
[203] observed intergranular corrosion of lead-containing zinc alloys in warm domestic
waters. Ahmed et al. [154] reported that severe intergranular attack occurred in a
Zn-0.3AI-0.03Cd anode after two years of service in seawater. The same material was
attacked intergranularly after 6 hours of anodic polarization in 70 a C seawater or after one
month of immersion in 70 a C distilled water. Roberts [535] found that intergranular attack
on a Zn-O.l Al alloy occurred in 150 a C dry water vapor. Mercille [220] observed
intergranular corrosion of Zn-4.2AI and Zn-12AI alloys that were embedded inside
concrete and exposed in a rural atmosphere. Intergranular corrosion was also found to
occur under a cathodic current as well as under an anodic current [154,224].

CHAPTER 7

228

TABLE 7.14. Intergranular Penetration Rates of Some Zinc-Aluminum Alloys in Various


Environments
Percent Al

Environment

Zn purity

0.1
4
20
21.1

99.999%
99.999%
99.999%

0. 1
0.075
4
0.04
4.2

99.99%
99.999%
99.99%
99.99%
99.99%

95C water vapor


95C water vapor
95C water vapor
100% relative humidity
at 50C
95C tap water
95C water vapor
95C water vapor
90C, 0.05M KCI
Concrete

Duration of
exposure

Rate (mmJday)

10 days
10 days
10 days
42 days

10 days
10 days
10 years

Reference

0.18
0.033
0.028
0.002

722
722
722
235

0.1
0.02
0.066
0.07
5 j1mJyr

49
535
535
48
220

Table 7.14 lists the intergranular corrosion rates of some zinc-aluminum alloys in
various environments that have been reported in the literature. The penetration rate of
intergranular corrosion depends on many factors, among which the composition and
structure of the alloy are the most significant. The presence of very small amounts of lead,
tin, and some other elements may greatly accelerate the intergranular corrosion rate of
AI-containing zinc alloys. Figure 7.29 shows the depth of intergranular corrosion for three
zinc-aluminum alloys in water vapor at 95C as a funtion of time [722]. It is noted that
the intergranular penetration rate can be several orders of magnitude larger than the rates
of general corrosion or pitting corrosion in similar environments.

10
- 0 .1 % AI

E
E
C

--r

4 % AI

"* 20 % AI

0
.;:;

Q)

Q)

0.

'0
.r:

a.

0 .1

Q)

0.01

10

Exposure time, days


FIGURE 7.29. Intergranularcorrosion of zinc-aluminum alloys (prepared with 99.999% purity Zn and 99.98%
purity AI) in water vapor at 95C. Data are taken from Ref. 722.

229

CORROSION FORMS

7.4.3. Metallurgical Effects


7.4.3.1. Alloying Elements. Zinc of high purity is not susceptible to intergranular
corrosion [48, 722]. The presence of other elements, particularly aluminum, as alloying
elements or impurities is necessary to cause the intergranular corrosion. Intergranular
corrosion has also been found to occur in zinc alloys containing only lead or magnesium
[203,535].
For Zn-AI alloys, intergranular corrosion is observed to occur in a concentration
range between 0.03% and 50% Al [48,49, 225, 535]. Below 0.03%AI, intergranular
corrosion does not occur. According to Devillers and Niessen [48,49], the presence of
impurities is not required for the occurrence of intergranular corrosion on zinc-aluminum
alloys. They found that intergranular corrosion occurred on a Zn-O.I % AI alloy prepared
with 99.9999% purity Zn and 99.999% purity Al in the hot-rolled and as-cast condition
in 95C water. In addition, the penetration rate for this very high purity alloy was
essentially the same as that for a Zn-O.l % Al alloy prepared with 99.99% purity Zn and
99.9% purity AI, indicating that the residual impurities in the lower purity alloy are not
the primary cause of the intergranular corrosion of Zn-AI alloys.
The intergranular corrosion of zinc-aluminum alloys is attributed to the preferential
attack on the aluminum-rich phase at the grain boundaries. Since 0.03 wt. % aluminum
is close to the solubility limit of aluminum in zinc at room temperature, in zinc alloys
containing an aluminum concentration higher than 0.03%, the aluminum precipitates at
the grain boundaries and is thus responsible for the increased corrosion rate at the grain
boundaries [49]. Moreover, as reported by Devillers [48], in two-phase Zn-AI alloys,
e.g., Zn-5% AI, preferential corrosion occurs not only at grain boundaries of the zinc
200r-----------------------------------------~

>:

{l150

.3,
c

e
1ii

100

a.

"0

a:

50

OL----------L~

0.001

0.01

________L __ _ _ _ _ _ _ _

_ _ _ _ _ _~

0.1
AI content (wt%)

FIGURE 7.30. Intergranular corrosion rate of zinc-aluminum alloys (prepared with 99.99% purity Zn and
99.9% purity AI) in tap water containing 0.05M KCl at 90C as a function of aluminum content. Data are taken
from Ref. 48.

230

CHAPTER 7

matrix but also in the AI-rich phase. However, for two-phase alloys the penetration rate
is much lower than that for single-phase alloys of low Al content [535].
Devillers [48] found that in 0.05M KCI solution at 90C the penetration rate of
intergranular corrosion for Zn-AI alloys increases with increasing concentration of
aluminum until it reaches a maximum at 0.2% AI, and then it decreases with further
increase in Al content as shown in Fig. 7.30. Similar results obtained in another study
[535] are shown in Fig. 7.31, where the corrosion penetration increases, peaking at 0.27%
AI, and then decreases to much lower values with further increases in the Al content.
According to Devillers [48], aluminum's effect on reducing the grain size is at least
partially responsible for the reduction of intergranular corrosion rates for alloys with Al
content greater than 0.2%. A decrease in the size of grains results in an increase in the
number of grain boundaries; hence, to reach the same depth of corrosion penetration, the
total grain boundary path is likely to be longer in a fine-grain alloy than in a coarse one.
Melton and Edington [225] found that the intergranular corrosion of alloys containing
40% and 50% Al is much less severe than that of alloys containing 22% AI.
The intergranular corrosion rate of zinc-aluminum alloys can be greatly influenced
by the presence of small amounts of other elements. More specifically, experimental
results have indicated that Mg, Cu, Au, Ni, Pt, and Co are beneficial, whereas Sn, Tl, In,
Pb, Bi, Hg, Cd, and Na are harmful [48,49,535,J. Mo, Zr, Ti, Ba, Si, Be, Te, Li, Sb, and
Ag are found to have little effect. Tables 7.15 and 7.16 contain data reported by Devillers
and Niessen [49] on the effects of some beneficial and harmful elements on the intergranular corrosion of Zn-O.I % Al alloy exposed to 95C water vapor. According to these
results, Mg is the most effective element in suppressing intergranular corrosion, whereas
Sn has the most harmful effect. The beneficial elements and the harmful elements are
found to counter each other's effects when both exist in the same alloy.
It has been noted that all the harmful elements have very low solubilities in zinc, and
when their solid solubility limit in zinc is exceeded, they will form precipitates in their
elemental form at the grain boundaries of the zinc matrix. Another common characteristic
of all these elements is that they have higher hydrogen overvoltages.
1.2.------------------,
E 1

.,
g

.50.8

~ 0.6
Q)

...00.4
a.

-5

a.
Q)
00.2
OL----~---~---~--~-~

0.001

0.01

0.1

AI concentration, wt%

10

FIGURE 7.31. Average depth of intergranular corrosion penetration of zinc-aluminum alloys (prepared with 99.999%
purity Zn and 99.98% purity AI) in water
vapor at 95C for 10 days as a function of
aluminum content. Data are taken from Ref.
535.

231

CORROSION FORMS

TABLE 7.15. Effect of Some Beneficial


Additions on the Grain Boundary Corrosion
Rate of a Zn-O.IAI Alloy at 90C"
Addition (WI. %)
0.05% Mg
0.3% Cu
5%Au
0.5 Ni
O.I%Pt

CorrosIOn rate (%l

15
15

5
65
75

"Reprinted from Devillers and Niessen [491. with kind permission from Elsevier Science Ltd, The Boulevard, Langford
Lane, Kidlington OX5 1GB, United Kingdom.

hEx pressed as a percentage of the corrosion rate of the pure


Zn-AI alloy.

The presence of small amounts of a single harmful element, other than aluminum,
in pure zinc does not cause significant intergranular corrosion [48, 535], However, the
intergranular corrosion can be severe when there are also other elements in the alloy. As
observed by Roberts [535], Mg is the most beneficial element in inhibiting intergranular
corrosion of Zn-AI alloys in warm water. If a small amount of magnesium is present
alone in pure zinc, it causes only mild intergranular corrosion. However, when a small
amount of lead, tin, or cadmium is also present along with the magnesium, severe
intergranular attack occurs in the zinc-magnesium alloy, The intergranular corrosion of
zinc-magnesium alloys has not been systematically studied.
7.4,3.2. Effect of Mechanical and Heat Treatments, Williams et al. [235] tested the
effect of stress on the grain boundary corrosion of Zn-22AI in 100% relative humidity
at 49C and found that the depth of intergranular penetration is greater under tensile stress
than under compressive stress. Devillers [48] found that stress has little effect on the grain
boundary corrosion of Zn-O.I Al in 80C water.
Heat treatment generally atlects the number of grain boundaries and the extent of
grain boundary segregation, Devillers [48] found that the corrosion penetration rate of
annealed specimens is similar to that of quenched and aged specimens, whereas it is higher
TABLE 7.16. Increase of Boundary Corrosion Rate in a
Zn-O.I %AI Alloy with Addition of Harmful Impurity
Elements at a Level of 0,01 at.%"
Element
Sn
Pb
Bi
Cd

Hg
TI

Percent increase in
corrosion rate

400
300
260
250
50
50

"Reprinted from Devillers and Niessen [49], with kind permission from
Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington OX5 1GB,
United Kingdom.

232

CHAPTER 7

than that of quenched samples without aging, indicating that the segregation of aluminum
at the grain boundaries can be significantly reduced by annealing and quenching.
Annealing and quenching have little effect on the intergranular corrosion rate in water at
high temperatures, because the diffusion of Al to the grain boundaries is sufficiently fast
to cause a segregation.
Melton and Edington [22S] found that the intensity of intergranular corrosion,
measured as the dimensional growth of the test specimens, of extruded 22% AI-zinc alloy
is lower in the extrusion direction than in the perpendicular direction owing to the
directional microstructure of the material.
7.4.4. Effect of Environmental Factors
As noted in Table 7.14, intergranular corrosion of zinc-aluminum alloys can occur
under different environmental conditions. Among the environmental factors, temperature
seems to be the most significant. Roberts [722] studied the effect of water vapor
temperature from 80 to IS0C in an autoclave and found that the corrosion penetration
rate increased exponentially with increasing temperature as shown in Fig. 7.32. The
relationship may be expressed as
log d= a - KIT

(7.20)

where d is the rate of intergranular penetration, T is the absolute temperature, and a and
K are constants. Due to this exponential dependence on temperature, the penetration rate
can be very high at elevated temperatures. For example, as shown in Fig. 7.32, it can be
as much as 4 mrn/day at ISOC for Zn-O.lAI. A similar exponential relationship is found
to exist between the corrosion rate of Zn-AI alloys and temperature in the range
80-120C in tap water [48]; at higher temperatures, up to 200C, the relationship becomes
linear.
Alkaline environments appear to be the most aggressive in intergranular corrosion
of zinc-aluminum alloys. Figure 7.33 shows the effect of pH on the grain boundary
corrosion rate in 9SoC tap water [49]. Between pH Sand 10, the corrosion penetration
10

.,>

- 0.1 %AI

E
E
C

.,

.0

+ 4%AI
* 20%A1

CI
~

Q)

c. 0.1

<0

a:

FIGURE 7.32. Intergranular corrosion rate


of zinc-aluminum alloys (prepared with
0.01
70

90

11 0

130

Temperature, C

150

QQ QQQ% nnrltv 7n !lnti QQ QRo;,., nJlr;tv A n!1~

233

CORROSION FORMS
250~----------------------------------------'

:;:;

.g

200

2-

~
150
LD
0)

co
Q)

~ 100
c
0

'en

2
0

50

oL-__ ____ ____ ____


~

- L_ _ _ _- L____

10

12

__

14

pH

FIGURE 7.33. Effect of pH on intergranular corrosion rates of a rolled Zn-O. I wI. % Al alloy in 95C tap
water. Reprinted from Devillers and Niessen [491, with kind permission from Elsevier Science Ltd, The
Boulevard, Langford Lane. Kidlington OX5 1GB, United Kingdom.

rate is almost constant. Below pH 5, it decreases with decreasing pH. On the other hand,
for pH values above 10, it increases drastically with increasing pH.
Grain boundary corrosion can also occur in dry water vapor where condensation does
not occur, although the general corrosion rate is only about one-tenth of that in wet water
vapor [722]. It is, therefore, believed that the intergranular attack of zinc alloys in water
vapor can be both chemical and electrochemical in nature, depending on condensation
conditions. When there is no condensation, the intergranular attack is slow and is chemical
in nature. When the surface is wet due to condensation, on the other hand, the grain
boundary corrosion is mainly of an electrochemical nature, and the corrosion rate is
greatly increased.
The presence of air, and thus oxygen, in water vapor was found by Roberts [722] to
have no effect on the corrosion penetration rates of zinc alloys of three different Al
contents, 0.1 %, 4%, and 20%. Williams et al. [235] studied the effect of sulfur dioxide
on the grain boundary corrosion rate ofZn-22AI alloy in air with 100% relative humidity
at 49C. They found that intergranular corrosion occurs in the humid air without S02 but
does not occur with 100 ppm S02; in the latter case, the general corrosion is more severe.
It was also noted that the corrosion product formed in the humid air containing S02 was
very soluble, whereas that formed in air without S02 was not soluble.
External current, both anodic and cathodic, can affect the occurrence and the
penetration rate of intergranular corrosion of zinc-aluminum alloys [154, 224]. Ahmed
et al. [154) observed intergranular corrosion in Zn-0.3AI-0.03Cd in 70C seawater under
an anodic current of 0.5 mA/cm2 . Devillers and Niessen [224] reported that under
relatively high applied cathodic current density (>0.5 mA/cm2) hot-rolled Zn-AI alloys
in hot water suffer very rapid grain boundary attack and blistering, as shown in Fig. 7.34.
This effect does not occur in pure zinc, but the addition of 0.03% Al is sufficient for this
attack to occur. The intergranular corrosion rate under cathodic conditions depends less
on alloy composition than is the case for corrosion without applied current. However, the

234

CHAPTER 7

~200

-0

..;

~
c
o

.~

(5
u

1a

-0

C
:J

co
3
Appl ied cathodic current density, rnA I em'

FIGURE 7.34. Intergranular corrosion rate of a hot-rolled Zn-AI alloy (Zn-O.l wt. % AI) in hot water as a
function of cathodic current density. From Devillers and Niessen [224]. Copyright by NACE International.
All Rights Reserved by NACE; reprinted with permission.

rates were found to be strongly dependent on water composition and are particularly
sensitive to the presence of carbonates.

7.4.5. Effect on Mechanical Properties


As a result of intergranular corrosion, the strength of zinc alloys can be drastically
reduced. Table 7.17 contains data reported by Kehrer [1290] showing the reduction in the
strength of zinc-aluminum-lead alloys as a function of aluminum content after 10 days
of exposure in 95C steam. There is practically no mechanical strength left after the test
for the alloys containing more than 0.05% Ai. Similarly, Melton and Edington [225]
observed a 77% reduction of the original strength of a 22% AI-zinc alloy after 7 days in
hot steam.

TABLE 7.17. Influence of Aluminum Concentration on the


Strength of a Zn-I.l % Pb Alloy after 10 Days' Exposure at 95 C
in Air or Steama
Relative strength
Al concentration
0
0.01%
0.02%
0.03 %
0.05%
0.08%
0.12%
Ref. 1290.

In air
14
13
12.5
II
12
12
II

In steam
II
7.5

6
3
0
0
0

CORROSION FORMS

235

7.4.6. Mechanisms

lntergranular corrosion of metals is, in most cases, associated with the presence at
the grain boundaries of a phase whose electrochemical behavior is different from that of
the matrix. This can happen in two-phase alloys where the second phase forms a
continuous path along the grain boundaries or in unstable solid solutions where the solute
segregates toward grain boundaries.
The intergranular corrosion of zinc-aluminum alloys in hot steam or water has been
attributed to the increased electrochemical reactivity of the grain boundaries caused by
the segregation of impurities or precipitation of phases. In earlier days, due to the
unavailability of high-purity zinc, the role of different impurities in intergranular corrosion was not clearly identified, and their segregation at the grain boundaries was generally
regarded as the formation of the cathode with the adjoining grains as the anodes. It was
not until later, as high-purity zinc became available, that the role of aluminum in causing
intergranular corrosion was clearly identified.
Several theories have been proposed to explain the processes involved in the
intergranular corrosion in zinc-aluminum alloys. Roberts [535, 722] proposed that the
intergranular corrosion is due to the preferential oxidation of the AI-rich phase at the grain
boundaries. The oxidation can be brought about by a direct chemical reaction between
water vapor and the interface material, as manifested by the fact that intergranular
corrosion can occur in dry steam, where there is no condensation, to form a continuous
electrolyte. The much faster corrosion penetration rate in wet steam is attributable to
electrochemical reactions in which the aluminum-rich phase is the anode and the zinc
matrix acts as the cathode. The effect of the corrosion-accelerating elements such as lead
and tin is explained by either a buildup of a cathodic phase adjacent to the grain boundary
zone or the increase in width of the distorted boundary zone resulting from the segregation
of these elements at the grain boundaries.
With regard to the reason for the effect of the impurity or alloying elements on
intergranular corrosion, Roberts noted that there is no coherent explanation. The observation has been made that all inhibitive elements tend to form intermetallic compounds
with aluminum, while none of the corrosion-accelerating elements do. By the formation
of intermetallic compounds, the amount of aluminum available for diffusion to grain
boundaries is reduced. According to Roberts, the inhibiting effect of magnesium on the
intergranular corrosion of Zn-Al alloys is likely due to its role in reducing grain size
because the segregation of impurities is distributed over a greater area with a corresponding reduction in the effective concentration of the impurities in the grain boundary. The
severe intergranular corrosion that occurs with aluminum-free zinc containing small
amounts of Mg and Pb or Sn is attributed to the precipitation of magnesium intermetallic
compounds, probably Mg 2Pb and Mg 2Sn. These intermetallic compounds are found to
be unstable in moist air and, hence, are rapidly attacked.
In another theory, proposed by Devillers and Niessen [48,49], the preferential anodic
dissolution of the aluminum particles precipitated at the alkaline grain boundaries is
considered to be the primary cause of the intergranular corrosion of zinc-aluminum
alloys. The anodic dissolution is coupled with a cathodic reduction of hydrogen at the tip
of the corroding grain boundary. The alkalinity of the grain boundary electrolyte is
supposed to be caused by the formation of OH- ions from a cathodic reduction of

236

CHAPTER 7

1.6

1.2
4:

E
,;.

.i;i 0.8
c

C!>

1)

C
C!>

~ 0.4

()

AI

-1.5

-1

Potential, V

-0.5

FIGURE 7.35. Potentiostatic anodic polarization of Al and galvanostatic cathodic polarization of Zn, Pb, Sn, In, and Cu in a pH 10.5
solution. Reprinted from Devillers and Niessen
[49], with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane,
Kidlington OX5 1GB, United Kingdom.

hydrogen. This theory is supported by the fact that the intergranular corrosion rate
increases with increasing pH (Fig. 7.33).
The effect of the different elements on the corrosion rates, in Devillers and Niessen's
theory, is explained according to their effects on the corrosion potential of aluminum in
the active-passive transition range in alkaline solutions as shown in Fig. 7.35. In the
active-passive transition potential range, Al shows an anodic dissolution peak followed
by a passivation. The corrosion rate of the AI-rich phase at the grain boundaries is,
therefore, strongly dependent on the cathodic reaction rates, which depend on the
presence of impurities. According to Fig. 7.35, the presence of impurities with higher
hydrogen overpotential than zinc will increase the corrosion rate and thus prevent the
passivation of AI. Elements with more noble potentials, such as Cu, if present in large
enough quantity at the grain boundaries, will reduce the corrosion rate by shifting the
corrosion potential at the grain boundaries to more noble values, thus facilitating the
passivation of AI. The special effect of Mg, i.e., its inhibition of intergranular corrosion
in Zn-AI alloys, is believed to be the result of its formation of a solid solution with AI,
hence lowering the critical potential for passivation, and also its anodic polarization
characteristics, allowing easier passivation. Mg has no solid solubility in Zn but has
appreciable solid solubility in AI.
7.5. WET STORAGE STAIN
"Wet storage stain" is a term used in the galvanizing industry to describe the zinc
corrosion products formed on a galvanized steel surface during the period of storage. It
is also referred to as "white rust," which is a term generally applicable to all zinc corrosion
products. Wet storage stain is voluminous, white, powdery, and bulky and is formed when
closely packed galvanized articles are stored under damp and poorly ventilated conditions
[51-53]. The crevices formed between the articles can attract and absorb moisture and
retain the wetness more readily than the surface area exposed to the open air. This noneven
coverage of moisture on the galvanized steel articles results in localized patches of

237

CORROSION FORMS

corrosion. Although wet storage stain can seriously affect the appearance of the galvanized steel articles in some situations, it is generally not harmful in terms of the long-term
corrosion performance [53].
The moisture necessary for the formation of wet storage stain may originate in
various ways. It may be present on the galvanized parts at the time of stacking or packing,
as a result of incomplete drying after quenching. It may result from direct exposure to
rain or seawater or from condensation caused by atmospheric temperature changes. Close
packing can result in moisture being retained by capillary action between the surfaces in
contact, where drying is delayed by the lack of circulating air. In practice, wet storage
stain is most often formed by condensed moisture in a confined environment where the
relative humidity is high. The corrosion process experienced during storage is generally
different from the corrosion process in an open atmosphere, where the zinc surface is
periodically wetted and dried.
Gilbert and Hadden [437] did an extensive study on white rust formation. They found
that the form of moisture is important for extensive white rust formation. Trapped water,
either from rain or from condensation, between the surfaces of zinc articles is not readily
dried and is usually responsible for considerable amounts of white rusting. The amount
of corrosion under the stains may significantly vary with circumstances. Gilbert noted
that zinc sheets, exposed in a glass tank in which the relative humidity was 100%, for I
week were only tarnished. Considerable corrosion occurs within a few hours, however,
when drops of water are deposited on the surface of zinc. The conditions that facilitate
condensation promote corrosion. In one example, specimens placed over dishes containing ice corroded at a rate of 0.15 pm/day. In another example, the external surface of a
zinc-coated steel pipe, in which cold supply water was passed, corroded at a rate of 0.4
pm/day.
White rust formed in uncontaminated air usually consists mainly of a basic zinc
carbonate, often mixed with zinc oxide; in the absence of carbon dioxide, it consists only
of zinc oxide [52, 437J. Under confined conditions, due to the lack of air circulation and
drying, white rust is less compact and less likely to be converted into the more protective
zinc carbonates. Table 6.8 in Chapter 6 shows the chemical composition of the white rust
formed in various types of tests.
The primary electrochemical reaction leading to the formation of white rust in the
presence of air and moisture is, according to Gilbert and Hadden [437],
Zn 2+ + 20W ~ Zn(OH),

(7.21 )

The zinc hydroxide is precipitated by interaction of the products from adjacent anodic
and cathodic areas. Secondary reactions which then occur are:
(7.22)

or
(7.23)

or

238

CHAPTER 7
tf>

120

>-

'"

Test

"0

-0
<1l

'"ca

-r

'Y Salt spray

<:

16
Vi

-t-

Condensationcabinet

80

Q)

Waterlilm
Watertog

Q)

'-'
~

:;
II>

?f-

40

'"
.9
Q)

i=

0
0.5

1.5

2.5

Total chromium on sheet surface, 119/cm'

FIGURE 7.36. Effect of chromating on performance of galvanized sheet in various accelerated corrosion tests.
Data are taken from Ref. 52.

(7.24)
Gilbert reasoned that the formation of zinc carbonate is a secondary reaction because the
amount of corrosion in a water spray test is the same whether ordinary air or COr free air
is used to saturate the water, but the corrosion product is basic carbonate in the one case
and zinc oxide in the other. Zinc hydroxides do not tend to form in damp air, where the
conditions favor the formation of zinc oxide or, in the presence of carbon dioxide, basic
zinc carbonate. Severe white rust can be produced in moist air containing contaminants
such as sulfate and chloride [52, 437]. The presence of flux residues, from galvanizing,
also enhances the formation of white rust [437].
Chromating has been used in the galvanizing industry as an effective surface
treatment to prevent wet storage stain from forming during storage or transportation
periods [51-57,437]. Figure 7.36 illustrates the effect of chromating on the formation of
white rust in four different tests [52]. Different types of tests, with various degrees of
simulation and acceleration, can be used to study surface staining by corrosion, as also
shown in Fig. 7.36. The difference between the zinc losses in the water film and
condensation tests indicates that oxygen concentration may play an important role in the
formation of wet storage stain because the electrolyte is open to air in the condensation
test while in the film test, the electrolyte is isolated from the air.
More information on white rust formation in humid air and the characteristics of zinc
corrosion products can be found in Chapters 8 and 6.
7.6. HYDROGEN EMBRITTLEMENT AND CORROSION CRACKING
Hydrogen embrittlement is a loss of mechanical strength and ductility in a metal
caused by the interaction between the metal and hydrogen, which may be a by-product
of the corrosion process. Corrosion cracking refers to the cracking caused by the
simultaneous presence of tensile stress, either residual or applied, and a corrosive medium.

CORROSION FORMS

239

Hydrogen embrittlement and corrosion cracking are rarely encountered in practical


applications of zinc and its alloys, largely because zinc is not usually used as a structural
material to bear stresses. In one of the few studies on the subject, Foster et al. [426 J showed
that a galvanizing alloy (eta-phase alloy) was susceptible to stress corrosion failure when
exposed to 60C circulating potable water and that the failure always occurred in a
direction normal to the applied stress. The time to failure decreased exponentially with
increasing applied tensile stress. Since the cracking was observed to be reduced or
completely eliminated for the samples tested under a cathodic polarization, it was
suggested that the stress corrosion process is controlled by anodic dissolution.
Both intergranular and trans granular cracking of zinc have been observed for certain
alloys in corrosion tests in laboratory environments [154,203]. The intergranular cracking
of the zinc alloys in most investigations was found to be caused by intergranular corrosion
[225, 203, 235, 426]. It was also attributed to hydrogen embrittlement in some studies
[154,224J.

8
Atmospheric Corrosion
8.1. INTRODUCTION
Atmospheric corrosion is the most prevalent type of corrosion for zinc, owing to
extensive outdoor applications of galvanized steels. Numerous research programs have
been carried out in the past to investigate the corrosion behavior of zinc in various types
of atmosphere. The testing methods used in these investigations can be generally divided
into two groups: field exposure testing and simulated testing. Data from field exposure
represent a real corrosion rate in an atmospheric environment, while those from simulated
testing provide specific information on the effects of the atmospheric variables. In view
of this intrinsic difference, information from real field exposure testing and that from
simulated testing are presented separately in this chapter. A brief description is provided
first on the atmospheric factors involved in a corrosion process. The mechanisms for
atmospheric corrosion of zinc are discussed at the end of the chapter.
8.2. ATMOSPHERIC FACTORS
8.2.1. Tvpe of Wetting
Atmospheres in different geographic locations vary greatly with respect to solar
radiation, temperature, moisture, wind, air constituents, and air pollutants. Atmospheric
corrosion of a metal alloy is affected by many factors [540-542], particularly those
affecting surface wetness and precipitation of pollutants [312, 315, 509]. The duration
and form of wetness are also important in determining the corrosion behavior of a metal.
Experimental results have indicated, for example, that, for the same time of wetness, rain
could cause more corrosion of zinc than dew [614, 1094].
Wetness in an outdoor atmosphere is usually generated by condensation, fog, or rain.
In the case of condensation, the level of wetness depends on the water content of the air
and the temperature. Figure 8.1 shows that the amount of water in air increases with
increasing temperature and relative humidity. When the water content reaches the saturation level, i.e., at a relative humidity of 100%, water will condense in the form of dew.
The rate of condensation on a surface is a function of substrate temperature. Figure 8.2
shows that the rate of water condensation decreases with increasing sample temperature
[556]. Condensation depends also on the hygroscopicity of the surface contaminants and
241

242

CHAPTER 8

50

100%

40

90%

'OJ

30

65%

20

....

.5
....

B
~

Ph

45%

~30%

10
-10

10

20

30

Temperature (0C)

FIGURE 8.1. Absolute water vapor content in the air


at different temperatures and relative humidities. From
Rozenfeld [556]. Copyright by NACE International.
All Rights Reserved by NACE; reprinted with permission.

corrosion products. All water-soluble salts are hygroscopic to some extent. A rough or
porous surface structure can result in capillary. condensation, where water condenses
inside a pore at a vapor pressure below the saturation level. It has been calculated that the
relative humidity for condensation decreases from 98% to 50% as pore radii decrease
from 360 to 15 A [542]. Wind is found to increase the relative humidity at which
condensation occurs [615].
Rain has an effect of not only wetting the surface of an exposed metal but also
washing away pollutants and corrosion products. In some areas, the rain can be polluted
(e.g., acid rain), which increases its corrosiveness. Compared to rain, snow has a
negligible effect on the corrosion of metals because snow is not an electrolyte and a
corrosion process cannot proceed without an electrolyte [13].
Metal surfaces can also receive water through adsorption. It has been reported that
a zinc surface at OC adsorbed 15 molecular layers of water at 55% relative humidity, 17
layers at 93%, and 92 layers at 100% [542]. The amount of adsorption changes drastically
with changes in sample temperature, as shown in Fig. 8.3 [1265].
Moisture, in the form of rain or dew, on a metal surface will evaporate under the
drying effect of temperature, radiation, or wind, thus causing an exposed metal surface
to continuously undergo cycles of wetting and drying. One effect of drying is that solutes
(pollutants and corrosion products) are solidified. The duration of wetting and drying as

-.:..c;
Ne
~
~

0.6

~....

0.4

c::

'i
'"c::

'8"0

0.2

10

15

Temperature (0 C)

20

FIGURE 8.2. Dependence of condensation on metal temperature. Air temperature is 25C, and relative humidity
is 100%. From Rozenfeld [556]. Copyright by NACE
International. All Rights Reserved by NACE; reprinted
with permission.

243

ATMOSPHERIC CORROSION

20~------------------------------,

(f)

Ql

>-

15

cu

o
c

E 10

-0
Ql

.D

o(f)

~
FIGURE 8.3. The amount of physically
adsorbed moisture on the surface of zinc
at 93% relative humidity and various
temperatures. After Strekalov el aL.
112651.

oL-----~--~--~--~~--~----~

10

20

30

40

50

60

70

80

90

Temperature (oe)

well as the frequency of the cycles has a significant effect on the morphology and
compactness of corrosion products [614, 1194].

8.2.2. Air Pollutants


The normal composition of air is summarized in Table 8.1 [540], In many places,
such as in cities, the air is polluted. The major air pollutants are sulfur dioxide, hydrogen
sulfide, oxides of nitrogen. and aerosols. Near the seacoast, the air is laden with sea salts,
particularly NaCI. In industrial areas, appreciable amounts of S02 and lesser amounts of
H 2 S, NH 3 , N02 , and other suspended salts are encountered [542]. The concentration of
pollutants varies greatly from location to location. Table 8.2 shows typical levels of
gaseous pollution.
The type and concentration of pollutants may vary with the form of water precipitation. It is reported that fog contains the highest levels of H+ ions [492]. Rain receives a
higher degree of mineralization than mist precipitation. As the raindrops fall, they wash
a large volume of air and catch salt particles suspended in the air. The amount of salt

TABLE 8.1. Approximate Constitution of the Atmosphere at 10C and 760 mm Hg Pressure,
Excluding Impurities"
------_._----

Amount present
Constituent
Air
Nitrogen
Oxygen
Argon
Water vapor
Carbon dioxide
"Ref. 540.

glm

1172
879
269

IS
8
0.5

Weight %
100
75
23
1.26
0.70
0.04

Amount present
Constituent
Neon
Krypton
Helium
Xenon
Hydrogen

mglm

14
4
0.8
0.5
0.05

ppm
12
3
0.7
0.4
0.04

244

CHAPTER 8

TABLE 8.2. Typical Concentrations (mg/m3) of Impurities in the Atmospherea


Hazy winter day

Summer

Impurity

Town

Country

Sulfur dioxide
Solid particles
Carbon monoxide
Sulfur trioxide
Ammonia

1.2
1.2
10.0
0.01

0.15
0.24

Town

Country

0.2
0.2
2.0
0.0001

0.03
0.05

0.01

"Ref. 540.

dissolved in raindrops varies between 20 and 200 ppm, depending on the geographic
location [556]. Nitrate appears to be predominantly a result of gaseous deposition;
deposition of organic substances on a metal surface is higher indoors than outdoors; and
sulfate ions may come more from dew than from other forms of precipitation [492].
The most important corrosive constituent of industrial atmospheres is sulfur dioxide,
which originates predominantly from the burning of coal, oil, and gasoline [13]. Table
8.3 shows that the concentration of sulfur dioxide in the air is high inside cities and falls
off with distance from a city because of the higher fuel consumption inside cities [1266].
Sulfur dioxide slowly oxidizes homogeneously in the presence of oxygen and moisture
to form sulfuric acid. The heterogeneous reaction of S02 in the presence of moisture and
oxygen is found to be significant after adsorption onto a substrate, such as an airborne
particle or an exposed metal surface [542]. The adsorption of S02 on a metal surface
increases with increasing relative humidity. Comparative data shown in Fig. 8.4 indicate
that surface adsorption on zinc is somewhat lower than that on Fe, but higher than that
on Cu and AI.
Other air contaminants are hydrogen sulfide and nitrogen compounds. Hydrogen
sulfide is produced by putrefaction of organic sulfur compounds or by the action of
sulfate-reducing bacteria in anaerobic conditions. Since hydrogen sulfide is easily oxidized, it generally occurs at very low concentrations [540]. Nitrogen compounds are
produced during electric storms and by decay of organic matter. Their most significant
contribution to corrosion is probably through the formation of ammonium sulfate.

TABLE 8.3. Variation of S02 Content of Air with Distance from Center of Citya.b
S02 content (ppm) at a distance (in kilometers) of:
City
Detroit
Philadelphia-Camden
Pittsburgh
St. Louis
Washington, D.C.

0-8

8-16

16-24

24-32

32-40

40-48

0.023
0.030
0.060
0.111
0.003

0.012
0.018
0.030
0.048
0.001

0.006
0.016
0.015
0.029
0.001

0.004
0.021
0.018
0.020
0.001

0.004
0.012
0.009
0.018
0.001

0.005
0.012
0.010
0.014
0.002

"Ref. 1266.

urhe level of S02 has decreased over the years due to environmental regulations.

ATMOSPHERIC CORROSION

245

Fe

1.5

FIGURE 8.4. Rate of adsorption of S02 on Fe. Zn. Cu,


and Al surfaces after 200-min exposure to S02 at different
relative humidities. Reprinted from Sydberger and Vannerbcrg [1267], with kind permission from Elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington OX5
1GB. United Kingdom.

0.5
60

70

80

Relative Humidity (%)

Dust is an important solid contaminant of many atmospheres. Industrial atmospheres


carry suspended particles of carbon, carbon compounds, metal oxides, H2S04 , (NH4)2S04'
NaC!, and other salts [13]. Marine atmospheres contain salt particles that may be carried
for miles inland, depending on the speed and direction of the wind. These substances,
combined with moisture, initiate corrosion by forming galvanic or differential aeration
cells, either by their hygroscopic nature, hence forming an electrolyte on the metal
surface, or by the capillary effect of inducing condensation at lower relative humidities.
The amount of pollutants in air has dropped over the years due to environmental
regulations [623],
8.3. CORROSION IN OUTDOOR ENVIRONMENTS
8.3.1. Typical Corrosion Rates

The corrosion rate of a metal varies significantly from one geographic location to
another and from time to time. The corrosion rate of zinc is lowest in dry, clean
atmospheres and highest in wet, industrial atmospheres. Seacoast atmospheres not in
direct contact with salt spray are mildly corrosive to zinc [546, 549]. Locations near sea
level are subject to salt spray, and hence the corrosion rate can be much higher at such
locations. The typical atmospheric corrosion rates of zinc and its alloys are [4 J:
Rural

0.2-3 JlmJyr

Marine (outside the splash zone)

0.5-8

Urban and industrial

2-16

Table 8.4 lists the rankings of 45 different atmospheres around the world with respect
to their corrosivity for zinc and steel in the period between 1960 and 1962 [539].

246

CHAPTER 8

TABLE 8.4. Rankings of 45 Locations with Respect to Corrosivity for Steel and Zinc from
Two Years' Exposure a ."
Weight loss (g)

Ranking
Location
Norman Wells, N.W.T., Canada
Phoenix, Ariz.
Saskatoon, Sask., Canada
Esquimalt, Vancouver Island. Canada
Detroit, Mich.
Fort Amidor Pier, Panama Canal Zone
Morenci. Mich.
Ottawa, Ont., Canada
Potter County. Pa.
Waterbury. Conn.
State College. Pa.
Montreal, P.Q., Canada
Melbourne, Australia
Halifax (York Redoubt). N.S., Canada
Durham. N.H.
Middletown. Ohio
Pittsburgh. Pa.
Columbus. Ohio
South Bend. Pa.
Trail. B.c.. Canada
Bethlehem. Pa.
Cleveland. Ohio
Miraflores, Panama Canal Zone
London (Battersea). England
Monroeville, Pa.
Newark, N.J.
Manila, Philippine Islands
Limon Bay, Panama Canal Zone
Bayonne, N.J.
East Chicago, Ind.
Cape Kennedy, Fla., ~ mile from ocean
Brazos River, Tex.
Pilsey Island, England
London (Stratford), England
Halifax (Federal Building), N.S., Canada
Cape Kennedy, Fla., 60 yards from ocean,
60-ft. elevation
Kure Beach, N.C., 800-ft. lot
Cape Kennedy, Fla., 60 yards from ocean,
30-ft. elevation
Daytona Beach, Fla.
Widness, England
Cape Kennedy. Fla., 60 yards from ocean,
ground level
Dungeness, England
Point Reyes, Calif.
Kure Beach, N.C., 80-ft. lot
Galeta Point Beach, Panama Canal Zone
"Ref. 539.
"Specimen size. 4 x 6 in.

Loss ratio
Zinc Steel/Zinc

Steel

Zinc

I
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

2
3
4
15
5
II
7
\3
31
10
28
6
20
19
12
30
27
21
18
14
33
8
29
24
35
16
32
39
22
9
23
40
42
43
38

0.73
2.23
2.77
6.50
7.03
7.10
9.53
9.60
\0.00
11.00
I I. 17
11.44
12.70
12.97
13.30
14.00
14.90
16.00
16.20
16.90
18.3
19.0
20.9
23.0
23.8
24.7
26.2
30.3
37.7
41.1
42.0
45.4
50.0
54.3
55.3
64.0

0.07
0.13
0.13
0.21
0.58
0.28
0.53
0.49
0.55
1.12
0.51
1.05
0.34
0.70
0.70
0.54
1.14
0.95
0.7S
0.70
0.57
1.21
0.50
1.07
0.84
1.63
0.66
1.17
2.11
0.79
0.50
0.81
2.50
3.06
3.27
1.94

10.3
17.0
21.0
31.0
12.2
25.2
18.0
19.5
18.3
9.8
22.0
10.9
37.4
18.5
19.0
26.0
13.1
16.8
20.8
24.2
32.4
15.7
41.8
21.6
28.4
15.1
39.8
25.9
17.9
52.1
84.0
56.0
20.0
17.8

37
38

26
36

71.0
80.2

0.89
1.77

80.0
45.5

39
40
41

25
44
37

144.0
174.0
215.0

0.88
4.48
1.83

164.0
39.0
117.0

42
43
44
45

34
17
41
45

238.0
244.0
260.0
336.0

1.60
0.67
2.80
6.80

148.0
364.0
93.0
49.4

Steel

17.0

33.0

247

ATMOSPHERIC CORROSION

Compared to the corrosion rate of steel, that of zinc exhibits less variation with geographic
location. Table 8.4 shows that the corrosiveness of the atmosphere from one location to
another varies by as much as a factor of 100 for zinc and 500 for steel. It also shows that
the atmospheric corrosion rate of zinc in most locations is at least 10 times lower than
that of steel, which is why zinc is commonly used, through the galvanizing process, to
effectively protect steel from corrosion. The difference in corrosion rates between zinc
and steel tends to be larger in marine environments and less in rural environments.
Furthermore. it can be seen from Table 8.4 that the relative corrosiveness of different
atmospheres toward steel may be very different from that toward zinc. For example.
among the locations compared. Waterbury. Connecticut has a ranking of only 10 for steel
but 31 for zinc; Cape Kennedy has a ranking of 31 for steel but only 9 for zinc. These
variations and differences among atmospheres and materials demonstrate the very complex nature of atmospheric corrosion.
Even in atmospheres with little contamination and within one geographic region.
corrosion rates can vary significantly. For example. the corrosion rate of zinc measured
at 26 sites in one rural area in Spain varied from 0.7 to 2.4 jlm/yr 13061. Also. atmospheric
conditions. such as the amoum of rain or the pollution level. change over time. and thiS
can cause the corrosion rate to change from year to year. The yearly averaged corrosion
rate of zinc does not vary much for a given atmosphere 1190.217.550 I. Figure 8.5 shows
that the corrosion loss of zinc measured at a single site in British Columbia was almost
linear as a function of time [1901. It shows also the seasonal effects due to variations in
time of wetness and the amount of air pollutants.
Due to environmental regulations and awareness. the level of pollution in many
developed countries has been considerably reduced over the years; similar trends have
generally been found for corrosion rates of many metals. The corrosion rate of zinc was
found to be higher between 1960 and 1970 than it was in the 1980s 1241. 248. 6161.
Table 8.5 presents a comparison of the corrosion rates of zinc and other common
commercial metals in various atmospheric environments. Accordingly. thc corrosion
resistance of zinc is higher than that of iron and cadmium in all environments, highcr than
1,000
May t959

Q;

+ Mar. 1960

800

(1)

..e0>
E

Dec, 1958

Sep. 1960

X No . 1960

600 r

III
III

.2
C

0
~

400 L

'iii

g
0

tJ

FIGURE ~.5. Corrosion loss versus time curve for zinc specimens
exposed at different times of the
year. Dala are taken from Ref. 190.

200 ~

0'
0

.
20

"
+
40

60

80 100 120 140 160 180 200

Exposure time (weeks)

248

CHAPTER 8

TABLE 8.5. Comparison of Typical Corrosion Rates of Zinc and Other Common Commercial
Metals in Various Atmospheric Environments
Relative corrosion rate
Metal

Industrial

Zn

I
2
0.23
0.13
2.4
0.07

Cd
Sn
Al
Cu
Pb
Ni
Sb
Mg
Fe

0.06
0.31
30

Marine

Rural

2
1.6
0.3
0.72
0.3
0.6

2.4
1.9
0.09
0.38
0.28
l.l

1.8
50

1.9
15

Reference(s)
297
292,543
543
543
543
543
543
544
217,539

that of copper in industrial environments, and higher than that of tin and magnesium in
marine and rural environments. There is no single set of rules for a reliable estimation of
the corrosion rate of a metal in the atmosphere for every geographic location. Field
exposure data are the best basis for reliable prediction of the corrosion resistance of zinc
and its alloys in the atmosphere. The corrosion rates of zinc and its alloys in many
locations can be found in an Intemation Lead Zinc Research Organization (ILZRO)
publication, Zinc: Its Corrosion Resistance [217], and review articles [294, 331]. Other
corrosion data have been reported from many countries, including Australia and New
Zealand [300], Brazil [301], China [302], Finland [303], Norway [304], South Africa
[305], Spain [306], Czechoslovakia [308], Poland [297], the Soviet Union [556], and
Japan [623].
8.3.2. Effect of Time of Wetness

The corrosion of zinc is negligible when the relative humidity is low but is significant
when the surface is wet at a high relative humidity. A zinc surface can become wet at a
relative humidity below 100%. Because of the presence of hygroscopic impurities in the
air or in the metal itself, the critical relative humidity for condensation to occur is usually
much lower than 100%.
The time of wetness can therefore be defined as the total time during which the
relative humidity is higher than the critical value. Guttman and Sereda [190, 191] have
experimentally determined that the time of wetness for zinc is the time during which the
humidity exceeds 86%. They also derived an empirical equation for calculating the
corrosion loss of zinc as a function of the time of wetness. Time of wetness can be
measured with various types of miniature moisture sensors made of bimetal couples,
which often include zinc as the anode material [201,210,537,538,615,312]. It can also
be estimated from meteorological data [548].
Time of wetness depends on the position and orientation of the exposed surface. For
a particular building, time of wetness was found to be longer on the roof than on a side
wall by about a factor of 2 [541]. Time of wetness was about 50% higher near the top of

ATMOSPHERIC CORROSION

249

TABLE 8.6. Average Corrosion Losses of Zinc Coatings on Buildings in Various Locations and
Atmospheric Environments after 10 Years"
Corrosion loss (mill
55%AI-Zn

Zinc
Location

Environment

Rural
Urban
Industrial
Severe industrial
Inland, shore of Rural
lake or marsh Urban
Industrial
Severe industrial
Coast
Rural
Urban industrial
Inland

Seashore

Severe industrial
Severe industrial

Roof

Wall

Roof

Wall

0.42
1.48
1.40
1.59
0.59
1.97
1.40
2.12
0.74
2.47
1.75
2.65
2.06

0.17
0.59
0.56
0.64
0.24
0.79
0.56
0.85
0.29
0.99
0.70
1.06
0.82

0.15

0.06

0.15

0.06

0.20

0.08

0.20

0.08

0.25

0.10

0.25

0.10

0.46

0.18

"Reprinted from Suzuki [312], by courtesy of Marcel Dekker.lnc.


hI mil = 25.4 ,urn.

the building than at the bottom [541]. As seen in Table 8.6, the corrosion was at least twice
as severe on the roof as on the side wall [312].

8.3.3. Effect of Pollutants


Sulfur dioxide, which is one of the most serious atmospheric pollutants, induces an
abnormally high corrosion rate in zinc [4, 217]. Other air pollutants, such as NOx , have
a relatively insignificant effect on the corrosion of zinc, largely because of the much lower
content of these species in the air [331, 616]. Sulfur dioxide has a high solubility in water
and, when dissolved, forms sulfurous acid, H 2S0 3 For example, when the concentration
of S02 in air is O.l %, the concentration of H2S03 in water is 1.6 x 10-3 molll, and the
resulting pH at equilibrium is low (2.8). It has been found that there is a direct correlation
between the corrosion weight loss and the amount of S02 adsorbed by zinc, as shown in
Fig. 8.6 [217]. Figure 8.7 shows that the corrosion rate followed the winter rise and
summer fall of humidity and of the S02 content in air, due to the higher fuel consumption
in the winter than in the summer [555].
Haynie and Upham [509] proposed a simple linear function to relate S02 concentration with zinc corrosion rates measured over four years in eight major cities in the United
States, as shown in Fig. 8.8:

Y =0.00103(RH - 48.8) x [S02]


The variable Y is in microns per year, and [SOz] is in micrograms per cubic meter. Below
a relative humidity (RH) of 48.8%, the zinc surface was considered not wet, and no

250

CHAPTER 8

.-...
"'E
.....
~

..

"'0
a.>

S02 ab orption
weightlo

.c
0

'"

"'0

'"
0
CIl
"'0

;a
'"

'"
.9

.c!>I)
v
~

Nov.

Oct.

April

Month of the year


FIGURE 8.6. S02 adsorption and corrosion loss of zinc at Stuttgart at different times of the year. Data are taken
from Ref. 217.

corrosion should occur [217, 509, 556]. Similar empirical equations have been derived
in other investigations [304, 625].
Experimental results indicate that there is a stoichiometric relation between Zn 2+ and
SO~- in the corrosion products [426,626]. Spence and co-workers [626-628] analyzed
the ionic species in the runoff water collected from zinc samples under field exposure,
obtaining the data shown in Table 8.7, and found that zinc loss is directly related to the
amount of pollutants precipitated on the zinc surface. The pollutants react with the zinc
~

"t:!'"

ME

.....

c::

.E-0
0

Relative humidity
S02 absorption
Weight loss

.-...

0'"

0.4

90

0.3

70

:a

c::

0.2

50

..c::

'"
.9

0.1

30

.c
'"

CIl
"'0

'"'"

.c

.~
a.>

Oct.

Oct.

April

1940

April

Oct.
1941

Oct.

April

10

s
;::I

a.>

;>
.~

Q)
~

1942

Time (month of the year)


FIGURE 8.7. Monthly corrosion rate and S02 adsorption rate of zinc at Berlin (Dahlem). Data are taken from
Ref. 217.

ATMOSPHERIC CORROSION

251

8
>;

..:!.
OJ

c
0
'v;

::;

slope = 1.03

4
0

'!.

00

FIGURE 8.8. Effect of atmospheric S02 concentration and relative humidity (RH) on the
corrosion of zinc. From Haynie and Upham
[509]. Copyright by NACE International.
All Rights Reserved by NACE; reprinted with
permission.

Atmospheric factor (RH - 48.8)~S02l (mg/ml )

corrosion products (oxide, hydroxides, and carbonates) and cause their dissolution when
the surface is wet. The dissolved species are then washed away in the event of rain. Spence
and co-workers derived a linear damage function for galvanized steel structures based on
the chemical analysis of the runoffs and on the assumption that a stoichiometric relation
exists between the amount of zinc corrosion and the amount of corrosive species (H+,
SO;, CO~-) precipitated on the surface. The reactions responsible for the dissolution of
zinc corrosion products are: (I) zinc replacing hydrogen ions in rain, [H+lrain: (2) solution
of basic zinc carbonate by dissolved carbon dioxide in rain, [HCOllain ; (3) formation of
solution of soluble zinc sulfate from rain, [S02Ler; and (4) dry deposition of sulfur dioxide
gas, [S02]dry' The overall corrosion rate, R, is

in which ao, at, a2, and a3 are constants. These constants are determined by a number of
factors such as wind direction and shape and size of samples r123-125, 522, 556].

TABLE 8.7. Ionic Species Detected in Galvanized Steel Runoff Water at Sites in Ohio (OH)
and North Carolina (NCt
Rate of detection [nmoll(cm 2 day)]
Site

HCOO-

Cl-

NO;-

SO~-

Na+

K+

H+

Ca2+

Zn 2+

NC
OH

1.17
0.36

1.04
2.90

0.60
0.62

2.63
13.99

0.21
0.41

0.23
0.47

0.12
0.09

0.38
7.29

5.34
10.05

"Ref. 626.

252

CHAPTERS
~

6
><

~
~
~ ~
~

1'i

20

...
<:

10

weight loss
salinity

30

..9 S

.!:!'
Q)

40

10

1000

100

Distance from the sea (m)

FIGURE 8.9. Corrosion loss of zinc and air


salinity as a function of distance from the
seashore. From Brown and Masters [542]. Reprinted by permission of John Wiley & Sons,
Inc.

8.3.4. Effect of Elevation and Distance from Seawater


Near the seacoast the major pollutants are chloride salts. In relative terms, sea salts
are less corrosive than S02' The typical corrosion rate of zinc in a marine atmosphere is
about 2.5 flm/yr, which is about 25 times less than that of iron. The corrosion rate increases
when zinc is exposed closerto seawater [191,436,542]. Figure 8.9 shows thatthe amount
of corrosion decreases with distance from the seashore because the salt content in air
drops significantly with distance from the shore [542]. Since the time of wetness also
drops with distance, the difference between the corrosion rates inland and at the seashore
cannot be entirely attributed to the difference in salt concentration. For example, it has
been found that, due to the mist near the sea, the time of wetness at 800 ft from the shore
is only two-thirds that at 80 ft [191]. The longer time of wetness near the sea is partially
due to the mist and partially to the higher salt concentration, because salt promotes
condensation at lower humidities. Table 8.8 shows that the corrosion rate of zinc varies
little with changes in elevation, in contrast to the corrosion rate of steel, which is very
sensitive to the elevation above sea level [539].
8.3.5. Effect of Initial Weather Conditions
The conditions at the time of initial exposure exert marked effects on the corrosion
of zinc, as illustrated in Fig. 8.10 [618, 191]. This figure shows that samples that suffered
the more severe attack initially continued to corrode at a higher rate than those which

TABLE 8.8. Weight Losses (in Grams) for Steel and Zinc at Different Elevations and Distances
from the Oceana b
60 yards from ocean
Material

~ mile from ocean

Steel
Zinc
Steellzinc
aRef. 539.

bAt Cape Kennedy, Florida.


'Elevation.

42
0.51
82.3

Ground

30 f{

60 ftc

215
1.83
117

80.2
1.77
45.3

64
1.94
32.9

253

ATMOSPHERIC CORROSION

0.4

'"

'"
!

August

0.3

.c<>0

v
i3:

0.2

0.1

40

FIGURE 8.10. Effect of conditions at the


time of initial exposure on the atmospheric
corrosion of zinc. After Elli s 16181.

80

160

120

200

Time (days)

were less severely corroded initially. Long-lasting rainfall or a relative humidity at or near
100% during the first days tends to cause a high corrosion rate. Figure 8.10 also shows
that the corrosion rate is, in general, higher at the beginning of the exposure and decreases
thereafter. This effect can also be seen in Fig. 8.11 [436]. The decrease of corrosion rate
with time is generally due to the gradual formation of protective corrosion products.

4 .---------------------------------------------~

marine atmosphere

<0

a:2
c

.2

'"~

0,
(;

OL---~---L--~----L---~---L--~----~

20

40

60

80

100

1 20

1 40

1 60

__~__~

1 80

200

Time of Exposure (months)


FIGURE 8.11. Variation of corrosion rate with time in a marine atmosphere. Data are taken from Ref. 436.

254

CHAPTER 8

8.3.6. Effect of Climate


One of the climatic effects on an exposed metal surface is the pattern of periodic
wetting and drying. Depending on weather conditions, wetting can take the form of
condensation or rain. In the case of condensation, the moisture has little effect on the
accumulation of corrosion products. Rainwater, on the other hand, can wash away the
corrosion products. The duration of wetting and drying can also have a significant effect
on the compactness of the corrosion products. Depending on the type of wetting and
drying pattern, the amount and the form of corrosion products can differ greatly from one
atmosphere to another. For example, the corrosion rate of sheltered (without effect of rain)
zinc samples may be several times less than that of unsheltered samples [307].
Other climatic factors, such as wind and radiation, may also affect condensation and
rate of drying as well as the amount of contaminants and corrosion products retained on
the surface. Recent studies have shown that wind can be important in the corrosion of
zinc coatings, because wind velocity determines the amount of sulfur dioxide deposition
on the zinc surface [626].
The corrosion rates also vary depending on the time of day when the samples are
exposed [1268]. Table 8.9 shows that the highest corrosion rate was exhibited by
specimens exposed in the early morning period when the probability of wetting was high
owing to dew formation. In another study, it was reported that nighttime corrosion is about
three times more than daytime corrosion [620]. The increased rate was attributed to high
humidity during the night.

8.3.7. Effect of Sample Configuration


The size, shape, and orientation of test samples may considerably affect the corrosion
rate of zinc [617]. The corrosion rate is higher on the skyward surface than on the
groundward surface even though the wetting time is longer on the groundward surface
[187, 190,315]. This may be attributed to the effect of rain and the larger amount of
precipitation of pollutants on the skyward surface. While the corrosion rate of a skyward
sample is generally linear with respect to time, the groundward surfaces do not show
linearity between the corrosion rate and time [295].
Thin wire is normally found to corrode faster than thick wire. The corrosion rate in
terms of weight loss per unit area of a 0.5-mm-diameter zinc wire was found to be about
four times that of a 12.5-mm-diameter wire, which had about the same rate as flat surfaces

TABLE 8.9. Weight Losses of Rolled Zinc Alloys Exposed during Different Periods of the Day
for Six Years a
Weight loss (g)/panel
Exposed period

a. 8:15 A.M.-4:15 P.M.


b. 4:15 P.M.-12:15 A.M.
c. 12: 15 A.M.-8: 15 A.M.
Sum of a. b, and c
24 hr, continuous
QAt

Zn-0.01Pb-0.OO3Cd

Zn-0.5Pb-0.2Cd

0.7582
0.8248
1.1691
2.7521
2.7224

0.8493
0.9378
1.2578
3.0449
2.8731

Palmerton. Pennsylvania, a mildly industrial environment; data are from Ref. 539.

255

ATMOSPHERIC CORROSION

6 ~-----------------------------'

Open aIr

FIGURE 8.12. Comparison of the corrosion


loss of zinc under sheltered and unsheltered
conditions in Madrid, Spain. Data are taken
from Ref. 306.

Louvered box

~L-----~------~------3
L-----~4--~

Time, years

1217]. The effect of diameter appears to be significant only on wires with a diameter less
than a few millimeters [1098]. Flat sheet specimens were reported to corrode more slowly
than bent sheet specimens and spherical specimens [217]. Spence and Haynie 1627, 6281
related the effect of sample orientation, shape, and size on the corrosion rates with the
changing deposition rates of pollutants. The deposition rate was the highest for fencing
wire, followed by small corrosion panels, and was the smallest for large sheets.
8.3.8. Effect of Sheltering

Sheltering prohibits rain from falling directly on the sheltered sample surface and
thus can significantly reduce the corrosion rate. In one study, Feliu and Morcillo [306]
found that the amount of corrosion of zinc samples exposed in Madrid, Spain, was
considerably lower under a sheltered condition than under an unsheltered condition, and
the differences seemed to increase with time, as shown in Fig. 8.12. The effect of

3.5 .-----------------------,
3

:l..2.5
oS

Sheltered

Unsheltered

~
c

.2

~ 1.5

u 1
FIGURE 8.13. Corrosion loss for zinc
plates exposed for five years in different
types of atmosphere under sheltered and
unsheltered conditions. Data arc taken
from Ref. 1259.

O : L--...lI........J~r_~~
Rural

Urban

Indus.

Type of Atmosphere

Marine

CHAPTER 8

256

sheltering depends on the type of atmosphere as reported by Johansson and Gullman


[1259] and shown in Fig. 8.13. Sheltering (hanging on a rod under a roof) reduced the
amount of corrosion by 3.8 times in a rural, 2.4 times in an urban, and 2.2 times in an
industrial environment.
Interestingly, as shown in Fig. 8.13, in the marine environment the amount of
corrosion under a sheltered condition actually increased by 1.4 times. This was explained
as possibly due to the fact that zinc chloride is very hygroscopic, and the formation of
zinc chloride in the marine environment causes the zinc surface to remain moist for a
longer time in a rain-sheltered condition [1259]. However, in another study by Kucera
and Mattsson [293] it was found that the corrosion of zinc in a marine environment under
a sheltered condition (in a meteorological box with slatted walls) was 2.6 times less than
that under an unsheltered condition. On the other hand, the amount of corrosion reduction
with sheltering was only 1.25 and 1.85 times in rural and urban environments, respectively. Kucera and Mattsson attributed the significant effect of sheltering in the marine
environment to the virtual prevention of transport of NaCl particles and droplets into the
meteorological box. The results from these studies indicate that the method of sheltering
is also important in affecting the corrosion. It needs to be distinguished whether samples
are shielded from rain, wind, orland air for a given sheltering condition.

8.3.9. Galvanized Steel


The corrosion of the zinc coating on galvanized steel is generally similar to that of
zinc panels. Data from field exposure tests indicate that the time to red rust is largely a
linear function of the zinc coating thickness [1098]. Typically, the number of years to
reach 50% surface red rust is suggested as the average life of a zinc coating [217].
Figure 8.14 shows that, for a given thickness, the life of the coatings produced by
different processes does not vary significantly [1269]. The results of a test program by
ASTM [1098] in 11 different geographic locations in the United States over 21 years
show that zinc-coated steel wire of different diameters has a coating life directly proportional to the coating thickness, irrespective of the method used to produce the coating.
Zinc coating produced by thermal spray has been found to corrode slightly more than that
produced by hot-dipping [623, 624]. It was reported that the zinc coating deposited from
a cyanide bath had better corrosion resistance than that deposited from a sulfate bath [297].

'"i:
~

15

on

"'"

10

'"
..::.>
:J

..

Coating process:
sherardizing
hot dipping
zinc spray
a plating

'"

"
50

100

150

Coating thickness (j.tm)

200

FIGURE 8.14. Effect of coating thickness


and coating methods on coating life. After
Hudson [1269].

ATMOSPHERIC CORROSION

257

TABLE 8.10. Effect of Various Annealing Treatments on the Time for Appearance of Yellow
Rust and the Black Layer and the Life of Galvanized Steel Wires"
Temperature and
time of annealing
As received
450C/30 min
500CI20 min
550C/30 min
600C/20 min
650Cl30 min

Weeks to appearance

Average iron
content(%)

Yellow rust

4.2

113

10.9

44

12.6
15.4
18.0
21.6

46
2

Black layer

147"
75
75
53
75
53

Life (weeks)

159
166
168
221
253
>503

"Ref. 433.
"Slight blackening only.

From the time of initial rusting, it takes a longer time for galvanized steel with thicker
coatings to become fully covered with red rust [510, 525. 617]. Thicker zinc coating is
produced by longer dipping time and has a thicker alloy layer; therefore, it takes a longer
time from initial rusting to complete rusting of the galvanized steel.
The zinc-iron alloy layers in a galvanized coating are often found to be more
corrosion-resistant than the pure zinc layer. Table 8.10 shows that the life of hot-dipped
galvanized steel can be substantially increased by a suitable annealing treatment [433],
although dicoloration of the coating is accelerated. The life of a coating containing 21.6%
iron is more than three times the life of a nonannealed coating. In another study [434], it
was reported that coatings containing less than 10% iron are inferior to those containing
20% iron.
The corrosion products formed on zinc-iron alloy coatings, e.g., annealed galvanized steel, are yellowish brown in color, reflecting the presence of the corrosion products
of iron [433, 394]. Brown stains also appear during atmospheric corrosion of hot-dip
coatings when the free zinc layer is consumed or in coatings on steels of high silicon
content [219, 394]. Annealed galvanized coating develops yellow rust, which then
gradually changes to a uniform dark gray to black color [433]. The black film formed
after six years' exposure in an urban environment was found to be about 80 f.1m in average
thickness, about three times that on nonannealed samples. The black layer, which is
insoluble and adherent, does not completely cover the coating unless the coating itself is
rich in iron.
The steel substrate can have a significant effect on the galvanizing process. Siliconcontaining steel (or silicon-killed steel), known to galvanizers as "reactive steel," gives
rise to the formation of a thick coating containing a high proportion of zinc-iron alloy
phases. A I5-year urban atmospheric exposure of six different kinds of silicon-containing
steels (0.21-0.45% Si) revealed that corrosion rates are comparable to that of nonreactive
steel [219]. However, the iron content in the coatings causes the galvanized silicon-containing steels to show various degrees of brownness (60-100% surface coverage with
brown corrosion products), while regular galvanized steel appears mainly gray-white after
15 years of exposure.

258

CHAPTER 8

8.3.10. Effect of Alloying


The effect of alloying elements on the atmospheric corrosion performance of zinc is
complex. Some elements may have beneficial effects in one situation while having
harmful effects in another. Some elements have little effect on atmospheric corrosion but
may enhance corrosion when another element is present. For example, trace amounts of
lead in zinc, which are otherwise harmless, induces intergranular corrosion when aluminum is present. For galvanized steel, the iron-zinc alloy layers are more resistant to
atmospheric corrosion than pure zinc. In general, trace amounts of impurities have very
little effect on the corrosion rate of zinc in atmospheric environments [294].
8.3.10.1. Lead. As a common impurity or alloying element in zinc products, lead
has little effect on the corrosion of zinc when synergistic effects with other elements are
absent. No significant differences in corrosion rates were found among different grades
of zinc with lead concentrations of 0.0055, 0.049, and 0.84% after 20 years of atmospheric
exposure [546]. Similar findings were reported in another study for galvanized coatings
containing 0.57 and 0.68% lead [1093].
8.3.10.2. Copper. Copper has been found to have a beneficial effect on the atmospheric corrosion resistance of a galvanized coating. It was reported that addition of up to
0.82% copper increased the corrosion resistance by as much as 20% in a two-year
industrial exposure test, as shown in Fig. 8.15 [489]. Adding 1% copper to rolled zinc
sheet was found to have little effect on the average rate of corrosion except in a severe
marine environment, where the copper-containing sheet showed a higher corrosion rate
than unalloyed zinc [549]. This enhancement of the corrosion rate was attributed to the
galvanic cell effect between zinc and copper, the latter acting as an effective cathode in
the low-resistivity electrolyte formed by the marine environment. Copper-bearing zinc is
more likely to develop distinct pits during corrosion than unalloyed zinc [294].
8.3.10.3. Aluminum. Aluminum is a widely used alloying element for zinc. When
present in small quantities, aluminum reduces the atmospheric corrosion resistance with
a maximum effect at a concentration of 0.32%, as shown in Fig. 8.15 [489]. Addition of
a small amount of copper was found to offset the effect of aluminum at a similar
concentration. As shown in Fig. 8.16, with the addition of more than 1% AI, the
200 r-------------------------------~

- 150

E:

.c

Ol

50

pure

ZinC

coating,

O L--L__~~__~__~_~___ L_ _~~_ _

0.1

0.2 0.3 0 .4

0 .5 0.6

0.7

Concentration (wt % )

0.8

0.9

FIGURE 8.15. Coating loss in 2~


years for galvanized steel produced in
baths containing different amounts of
various alloying elements. After
Radeker et al. [489J.

ATMOSPHERIC CORROSION

259
25 ~--------------------------,

-- Sevele marine
...,- Indusui al
*

M arine

... Ind u5trial

en
en

15

Rural

FIGURE 8.16. Effect of Al content on corrosion of Zn-AI alloy coatings after 5-yr
exposure in various atmospheric environments. After Zoccola et at. [1891.

::::
.3
40
50
60

O ----~~~~L-~--~--~--~--~

10

20

30

70

80

Concentration of Al (wI %)

atmospheric corrosion resistance of zinc coatings increases with aluminum content up to


4-7% Al (eutectic composition is 5%), beyond which corrosion resistance decreases with
aluminum content to about 21 % (the eutectoid composition being 22% AI). Between 21 %
and 70% AI, corrosion resistance increases almost linearly with Al content [189].
Two major commercial zinc-aluminum alloy coatings, Galfan and Galvalume, have
been developed for the production of more corrosion-resistant steel sheets. Table 8.11
compares the corrosion resistance of galvanized, Galfan, and Galvalume coatings in
various atmospheric environments [242]. In general, Galfan is about two times more
corrosion-resistant than a galvanized coating, and Galvalume is two to four times more
corrosion-resistant than a galvanized coating. The corrosion of zinc/aluminum alloys
proceeds through several stages. First, the zinc preferentially dissolves, leaving an
aluminum-rich porous structure. During this stage, the steel is cathodically protected.
After the zinc is depleted in the coating, depending on the compactness of the remaining
structure and the type of atmosphere, red rust may start to form since the aluminum may
be passivated and hence is cathodic to steel. In a chloride environment, the aluminum left
in the zinc/aluminum coatings after depletion of zinc can still galvanically protect the

TABLE S.II. Corrosion Losses for Galvanized Steel, Galfan, and Galvalume after One-Year
Exposure in Four Different Atmospheric Environments"
CorrosIOn loss (um)
Galfan
Environment

Galvanized

Max.

Min.

Galvalume

Rural
Urban
Marine
Severe marine

1.0
3.0
2.4
5.4

1.0
1.4
2.8
3.8

0.3
0.7
1.4
2.8

0.3
0.6
1.1
2.6

"Ref. 242.

CHAPTER 8

260

140 --

II>
II>

pure 2 me COJtlng

80 ~

.2

1:
60 Ol

-Ag

+Sb

'iii

5: 40

*Mg
+ Bi

20 -

O ~~--~~--L--L--~~--~-L--

0.1

0.2 0.3 0.4 0.5

0.6 0.7 0.8 0.9

Concentration (w t %)

1 .1

FIGURE 8.17. Effect of various


alloying elements on corrosion of
galvanized steel in an industrial
environment. After Radeker r489].

steel until all the coating is consumed, since the passive film of aluminum is not stable
and aluminum is anodic to steel [189].
8.3.10.4. Tin. Small additions of tin (from 0.27 to 0.96%) in a galvanizing bath have
been reported to have very little effect on the atmospheric corrosion rate of the gal vanized
coating [217]. However, Fig. 8.15 indicates that with an addition of 0.3-0.9% Sn, the
corrosion rate of zinc in an industrial environment increases by about 20%. Laminar
coatings of zinc and tin or zinc alone exhibit equivalent or better resistance to industrial
atmospheric corrosion than similar tin-zinc alloy coatings [621].
8.3.10.5. Other Elements. Figure 8.17 shows the effects of a number of other
elements on the corrosion rate of galvanized steel [489, 1093]. A IS-year urban atmospheric exposure showed that addition of 0.08% vanadium slightly increases the corrosion
resistance of galvanized coatings [219]. Addition of 0.04 wt. % Mg was found to have no
significant effect on the life of galvanized coatings in various atmospheric environments
[332].

8.3.11. Effect of Surface Treatment


The surface of zinc can be treated to increase the corrosion resistance. The process
most commonly used is chromating. However, due to environmental concerns, the use of
chromating is becoming increasingly limited. The chromating process involves dipping
in a solution containing chromate, resulting in the formation of a conversion coating
consisting of chromium oxide and chromate. Chromate treatment is very effective in
delaying the onset of corrosion on a zinc surface and in preventing the formation of wet
storage stain. However, chromating does not significantly change the long-term atmospheric corrosion performance of zinc, as can be seen in Table 8.12 [593]. Phosphating is
another surface treatment extensively used for surface treatment of zinc and its alloys.
Because phosphating significantly alters the surface appearance, it is used less as a surface
finishing process and more as a pretreatment for painting.
Plating a more corrosion-resistant metal or alloy layer on the zinc surface can also
be used to increase the service life of zinc products. Copper, nickel, and chromium
electroplated coatings have been found to improve the corrosion resistance of zinc

ATMOSPHERIC CORROSION

261

TABLE 8.12. Outdoor Exposure Data on Cronak-Treated Zinc"


Months to rusting"
Coati ng type

----_.

__.

Zinc thickness (11m)

Untreated

Cronak treated

30
30

30

-----------"----------

Electrodepositcd

Hot-dip

8.6

12.9
17.2

42

2S.R

48

22.R
32.7

54

60

30
.+2
54

60
60
------

"Ref. 593.
"In New York City.

die-casting alloys [188]. Pitting at defects of the coatings is the predominant corrosion
feature of these coated products.
8.3.12. Effect of Corrosion Products
For most metals, the corrosion resistance is largely determined by the stability,
adherence, and compactness of the corrosion products (e.g., oxides, hydroxides, and salt
films). In generaL the corrosion rate of zinc after atmospheric exposure decreases with
time because of the formation of protective corrosion products on the zinc surface [4,
618]. During the exposure, as the silver-gray surface turns dull gray, a thin adherent
surface film (identified most often as zinc carbonate) gradually forms and inhibits further
corrosion of zinc.
The formation of solid zinc carbonate is a slow multistep process. Zinc is first
converted to zinc hydroxide, fonning a gel, which may then be converted by a small
amount of carbon dioxide to a tough, thin film of basic zinc carbonate. Under conditions
of limited access of air, and hence slow drying and insufficient carbon dioxide, the zinc
hydroxide is converted to a fluffy zinc oxide, known as "white rust." Because of the loose
nature of the white rust, it has little barrier effect on the access of solution to the zinc
metal. Furthermore, the buildup of white rust will prolong the time of wetness by reducing
the critical relative humidity for condensation, retaining more moisture, and retarding the
drying process. The formation and characteristics of various corrosion products are
discussed in Chapter 6.
8.3. J3. Forms of Corrosion
The corrosion of zinc in most atmospheric environments is usually general corrosion;
that is, the corrosion occurs uniformly across the zinc surface. The corroded surface after
years of exposure may be covered with dimples, for which the ratio of depth to diameter
is small [2941. The dimple size can be a few millimeters in a marine environment and
much smaller (about one-tenth) in a rural environment. Pitting is not a common form of
corrosion of zinc in atmospheric environments.
Another common corrosion form of zinc is galvanic corrosion. On galvanized steel
at places where the coating is damaged, such as at cut edges, the exposed steel is
cathodically protected while the surrounding zinc coating is galvanically corroded.

262

CHAPTER 8

Although a common form of corrosion, galvanic corrosion is not a major contributor to


the corrosion of zinc coatings because the exposed areas of bare steel are usually too small
to cause significant corrosion. Usually, therefore, the atmospheric corrosion rate of
galvanized zinc coatings is essentially the same as that of zinc.
Galvanic corrosion can be a significant contributor to corrosion when a small piece
of zinc is connected to a similar or larger piece of another metal. Table 7.8 in Chapter 7
shows that the corrosion of zinc wire is increased by electrically connecting to bolts of
other common metals, with the exception of aluminum and magnesium. The corrosion
rate of zinc decreases when it is connected to magnesium in all types of atmospheric
environments and to aluminum in urban and marine environments [293, 544, 551]. In
other atmospheric environments, zinc is anodic to aluminum, owing to the passive film
on aluminum. A zinc rod wired by aluminum is anodic to the aluminum wire in most of
the industrial and some of the marine atmospheric exposures [515]. Coatings with more
than 60% AI behave like aluminum and provide no galvanic protection to the steel [248].
The galvanic effect is most significant in marine environments because of the high
conductivity of surface moisture. The thinner the moisture film on the metal surface, the
more localized is the attack at the galvanic couple contact [293]. A detailed discussion of
galvanic corrosion is presented in Chapter 7.
Premature darkening is a phenomenon sometimes encountered in galvanized roofing
sheets, when the darkening, which would otherwise take a few months, occurs very
rapidly after a few days of exposure [119, 120]. The characteristics of premature
darkening have been reported as the following:
1. The darkening occurs only in rural environments.
2. The darkening occurs within one week of the initial exposure to the atmosphere.
(It normally takes a few months to darken the surface.)
3. Only the skyward surface is darkened.
4. Sheets adjacent to the darkened sheets remain bright.
The exact cause of premature darkening is not fully understood. The initial surface
conditioning during the storage period may be responsible for premature darkening. The
darkening may also be due to the particular kind of galvanized coating [119].
Wet storage stain is the voluminous, white, and powdery corrosion product formed
when closely packed galvanized articles are stored under damp and poorly ventilated
atmospheric conditions [51]. It is most often found on stacked and bundled items such as
galvanized sheets, plates, angles, bars, and pipe. Weathered zinc surfaces are seldom
attacked. The formation of wet storage stain generally has little effect on the corrosion
life of galvanized steels other than changing the surface appearance. It has been observed
that when galvanized steel sheets are exposed for six months, the sharp color contrast
between the stained and nonstained areas is less noticeable. After two years, there is little
difference between ordinary and wet storage stained sheets [53]. More information on
wet storage stain is presented in Chapter 7.

8.3.14. Highway Environment


The highway environment, experienced by automobiles and highway structures, is
particularly aggressive owing to the high pollution level from gas exhaust and, in the

263

ATMOSPHERIC CORROSION

wintertime, from the deicing salts [571]. The salt solution not only increases the corrosivity but also the time of wetness.
In the highway environment, zinc is used primarily as a coating material for steel
structures such as guardrails, automotive body panels, and rebar. The zinc coating on
guardrails is directly exposed in the highway environment. On the other hand, the zinc
coating on automotive body panels or rebar is covered with paint or concrete, respectively,
and is normally indirectly exposed to the environment. The discussion presented here is
limited to the corrosion caused by direct exposure to the highway environment. Information on the corrosion behavior of zinc under paint or inside concrete is presented in
Chapters II and 13, respectively.
The Society of Automotive Engineers (SAE) conducted an extensive under-vehicle
test for various zinc-coated steels [5721. Table 8.13 shows that the corrosIOn performance
of zinc-coated steels is superior to that of bare steel or steel coated with other types of
coatings. The reduction in weight loss for zinc-coated steel compared to the bare steel is
about sixfold. Also, it can be noted in Table 8.13 that a thicker zinc coating is beneficial
in reducing the percentage of base metal attack and the pitting depth. As a rough estimate,
the corrosion rate of the zinc coatings in an under-vehicle environment, being about 8.5
Jim/yr, is comparable to the corrosion rate in a relatively severe marine atmospheric
environment.
The corrosion performance of zinc- or alloy-coated steel varies with coating type.
In one study, it was reported that, after two-year under-vehicle testing, galvanized and
galvannealed steels have less than 5% surface area showing base metal corrosion and pit
depth less than 30 Jim. On the other hand, in the case of the electroplated Zn and Zn-Ni
coatings, the base metal corrosion is between 50 and 80% and the pit depth is about 80
Jim [3371.
Table 8.14 contains data reported by Allegra and Townsend [485] on the corrosion
rates of galvanized and 55% AI-Zn-coated steels obtained from under-vehicle tests. The
corrosion rate of the zinc coating, varying from 1.1 to 3.3 wn/yr and being higher than
that of the zinc-aluminum alloy coating, is comparable to the corrosion rate in a rural
environment, which is typically in the range of 0.2-3 Jimlyr [4].

TABLE 8.13. Corrosion of Various Coated and Uncoated Steel Panels after 2.3-Y r
Under-vehicle Exposure Test"
- - - - - - - - - - - - - - - - - - - - - - - - - - -

Weight loss
Sample

G90
G60
Anodic electrodeposited primer
MOPB
Zinc-rich primer
Bare steel

Img/cm 2 (jim)]

13.9(19.6)
13.5 (19)

82.7 (ll7)

Average % area of base


metal attacked
5.1
16.6
22.8
40.9
58.6
100.0

- - - - - - - - - _... _ -

Average pit depth


8.5
23.1
58.0
93.1
52.0
132.1

-.------------------------------

"Data from Ref. 572.


"G90 and GOO. Galvanized zinc coating specified in ASTM Standard AbS3: MOPE. metallo-organic petroleum base coated
steel.

264

CHAPTER 8

TABLE 8.14. Corrosion Rates ofZn and 55%


Al-Zn-Coated Steel Panels in Under-vehicle
Exposure Tests"
Corrosion rate (j1m/yrl
Sample
A
B

C
D
E
F

Average

Galvanized

55%AI-Zn

3.33
1.56
1.54
1.4
l.l
1.9
l.81

0.47
0.42
0.21
0.68
0.65
0.24
0.45

"Data from Ref. 485.

The severity of a highway environment varies with location. German [186] reported
the results of a 7-yr field exposure test for galvanized steel placed along various highway
locations in Ontario and Quebec. It was estimated that the zinc coating life is about 5
yr/mil (corresponding to a corrosion rate of about 5 flm/yr) in an urban highway
environment and that about 10-20% longer life is likely in a rural environment.
The corrosion of automotive materials is also affected by motion. Talati and Patel
[299] found that the corrosion rate of zinc on a moving coach in Bombay and Ahmedabad
was several times higher than that under a static condition. The higher corrosion rate was
attributed to the falling off of the corrosion product, which has the effect of inhibiting
corrosion, from the specimen during movement at high velocity.
8.4. CORROSION IN INDOOR ENVIRONMENTS
In normal indoor environments, such as inside residential houses, zinc corrodes very
little. Generally, a visible tarnish film forms slowly, starting at spots where dust particles
have fallen on the surface [404]. Over a period of time, such films grow gradually until
the surface has lost much of its original luster. The appearance of the surface and the
degree of corrosive attack are related to the relative humidity. Relative humidity up to
about 70% has little influence on the corrosion. Above this point, corrosion activity may
occur as it becomes possible for moisture to precipitate on the surface, especially when
the surface is covered with zinc corrosion products and contaminants.
The corrosion rate of zinc in a clean indoor atmosphere is typically below 0.1 flm/yr.
Figure 8.18 presents the testing results obtained by British Steel in 15 residential houses
in England over a period of three years, showing that the average corrosion loss of
galvanized steel samples over three years is 0.12 flm and the rate decreases with time
[1292]. A linear regression analysis of the data in Fig. 8.18 yields an expression for
corrosion loss with respect to time of the form:
Weight loss = 0.059 t 064

265

ATMOSPHERIC CORROSION

0.4 . - - - - - - - - - - - - - - - - - - - . . . ,
Carr. loss ~ 0.059 to'"

E 0.3

:J..

iii
II)

E
c 0.2
o

'iii

g
o

u 0.1

0.5

1.5

2.5

3.5

Time, years

FIGURE 8.18. Corrosion loss of galvanized steel. exposed in the loft area of 15 residential houses located in
three different geographic locations in England, as a function of time. The data are the mean values of 6 samples
for each house. The equation in the figure is the best fit from linear regression analysis. Data are taken from
Ref. 1292.

Similar low corrosion rates have been found in other indoor situations. In one case It was
reported that the average corrosion of galvanized steel on telephone equipment exposed
for up to 40 years to the New York City environment was much less than 0.1 j.HnJyr [406J.
In another case the weight increase of zinc exposed in a basement room with a window
open to the street at an average relative humidity lower than 70% was found to be linear
with respect to time, with a rate of about 0.1 f.lrnJyr [746].
As with outdoor corrosion, many factors may affect indoor corrosion. These include
the type of climate, thermal insulation, heating, air conditioning, and amount of air
contaminants. Corrosion rates higher than the normal values may be found in situations
in which moisture precipitates regularly or the air is polluted. Table 8.15 shows that the
corrosion rates in industrial indoor environments are in the range of 1 j.HnJyr, much higher
than in clean indoor environments [559]. In another case, it was found that in an indoor

TABLE 8.15. Corrosion of Different Alloys in Industrial Indoor Environments for 10 Years"
---------------------------;c------Corrosion rate (/lm/yr)"

Test location
Pyrometer shed-ZnO furnace building
ZnO furnace building over furnaces
Cement crusher building
Cement kiln furnace building
Coal breaker building

A
0.773
2.06
0.258
0.515
0.515

0.773
2.32
0.258
0.515
0.515

C
0.773
2.58
0.258
0.515
0.773

D
0.773
2.58
0.515
0.258
0.773

"Data from Ref. 559.


"A, Hot-rolled high-grade zinc; B, Hot-rolled brass special zinc; C, Hot-rolled selected zinc; 0, Hot-rolled zinc plus 1% Cu
and 0.0 I % Mg.

266

CHAPTER 8

environment the average formation rate of corrosion products on the exterior surface of
a galvanized pipe for cold water was about 0.4 J1.rn/yr over a period of 32 years, while it
was only 0.1 J1.rn/yr on a hot water pipe [1293]. It is easier for moisture to condense on a
cold surface than a warm surface at a given relative humidity. It has also been reported in
another study [613] that no significant corrosion was observed for zinc samples placed
above a lead-acid battery electrolyte (concentrated sulfuric acid) reservoir under a
nonsealed condition (with a small hole in the cover) in a room for electronic equipment
because the normal air flow in the room prevented the generation of concentrated acid
vapor.
8.5. CORROSION IN SIMULATED ENVIRONMENTS
Atmospheric corrosion is a slow process, usually taking several months or years to
show its effects. Thus, laboratory testing methods are used to accelerate the process under
simulated conditions. However, because atmospheric corrosion is a complex phenomenon
involving numerous variables that are noncontrollable and change irregularly with time,
it is very difficult to accelerate and, at the same time, to closely simulate atmospheric
corrosion.
In many cases, the corrosion rates obtained from a laboratory simulated test bear
little resemblance to those resulting from normal exposure. For example, as shown in Fig.
8.19 [547], the corrosion rate of zinc subjected to a spray of natural seawater in the
laboratory can be much higher than the rates obtained from real exposure in a marine
atmospheric environment. Comparative results shown in Fig. 8.20 indicate that the
corrosion rate of zinc greatly varies depending on the kind of test used, with the salt spray
(SS) test being the most severe [52]. The accelerating factor of each test is very different
for different metals. Thus, caution must be exercised in evaluating and comparing the
corrosion rates obtained for different materials from these corrosion tests. For example,
the corrosion rate of zinc is only about 2 times lower than that of steel in a salt spray test
but is 10 to 100 times slower than that of steel in real atmospheric exposure tests.

13

15

00

,s
'"'"
..9

10

.E

.!1!l

~m
500

1000

1500

Exposure time (hours)


FIGURE 8.19. Comparison of corrosion of zinc exposed in a marine atmosphere (25 and 250 m from the sea)
and in sprays of natural seawater and 3 and 20% NaCl. After Baker and Lee [547].

267

ATMOSPHERIC CORROSION

til
til

0>

' ijj

FIGURE 8.20. Comparison of corrosion of galvanized steel in different accelerated tests: Sp, steam pressure; W
film, water film; CC, condensation cabinet; W fog, water fog; SS, salt spray.
Data are taken from Ref. 52.

0.1

o.ol L0.5

50

Time (hours)

However, laboratory experiments can be designed to study the effect of the variables
on specific aspects of atmospheric corrosion. The most common tests used to evaluate
atmospheric corrosion of a metal are salt spray test, humidity chamber exposure, and
wet/dry cyclic test. These test methods have one element in common: they form a thin
layer of electrolyte on the metal surface, which is what occurs under real atmospheric
exposure. Also, all these tests use enclosed chambers so that the required humidity, spray,
and level of pollutants can be generated and controlled.
Humidity chamber exposure is a simple test to evaluate the effect of relative humidity
and temperature, but it lacks the dynamic effect of raining and drying. On the other hand ,
salt spray tests simulate the effect of continuous raining but do not represent the chemical
composition of rain nor the effect of condensation and drying. The standard salt spray
test, ASTM B 117 [259] , although widely used, is a very severe corrosion test and bears
little similarity to atmospheric corrosion. The wet/dry cyclic test, incorporating the effect
of condensation, spray, and drying, is closer to real atmospheric exposure. Because a large
number of variables are involved, the conditions in different cyclic tests can vary greatly.
Thin-layer electrolytes can also be used to study the electrochemical reactions and
changes in solution chemistry during the corrosion process.

8.5.1. Humidity Chamber Exposure


A humidity chamber provides the conditions to create surface wetness by water
condensation . Thus, the corrosion inside a humidity chamber is similar to that caused by
natural dew. A zinc surface becomes stained when tested in a humidity chamber. The
percentage of stained area increases with increasing relative humidity and temperature as
shown in Fig. 8.21 [120]. Figure 8.22 shows that the corrosion rate of zinc in a humidity
chamber is linear as a function of time [1270]. At room temperature in a closed chamber
containing distilled water (relative humidity 100%), zinc is only tarnished after many
days of exposure [437]. However, when the zinc article is cooler than the surrounding
humid air, large droplets of water may form and cause more corrosion.
The presence of pollutants in the air can significantly increase the corrosion rate .
Figure 8.23 illustrates that, in a chamber at 21 C and at a relative humidity between 95

268

CHAPTER 8

60
o

25C

40

60

FIGURE 8.21. Effect of relative humidity on surface staining of wetted galvanized steel. Reprinted from HelwIg
r120], with kind permissIOn from Elsevier Science lnc., 655 Avenue of the
Americas, New York.

80

Relative humidity (% )

and 100%, the corrosion rate of zinc increases significantly with increasing sulfur dioxide
concentration [1271]. Gilbert and Hadden [437] reported that in a closed tank containing
various amounts of saturated S02 solution, the corrosion of a zinc sample, placed
horizontally above the solution, increased with increasing S02 concentration in the tank
as shown in Table 8.16. The sulfur dioxide appears to facilitate the wetting of the surface
and to damage the initial oxide film, which would otherwise maintain its protective
properties in pure air [556]. The formation of sulfate compounds as the corrosion products
is equally important in promoting corrosion because these compounds are more soluble
and more hygroscopic.
In moist H2S gas, zinc is very stable in comparison to many other metals, as shown
in Fig. 8.24. Thus, a zinc coating can offer effective protection to steel and iron against
corrosion in an environment containing hydrogen sulfide [556].

---u
E

1.2

Sen
en

0.8

'il
~

0.4

.3
1:
OJ)

97% relative humidity


85% relative humidity

.'
3

12

Time (months)

FIGURE 8.22. Weight gain versus duration of exposure for zinc in a humidity chamber at two humidities. After
Rajagopa1an and Ramaseshan [l270].

269

ATMOSPHERIC CORROSION

SO, volume %

20

.
0

00
2;

15

'"
:cOIl

10

0.00
0.01
0.05
0.1
0.5

'"
.S!
.;:;
~

FIGURE 8.23. Effect of S02 content in


air on corrosion of zinc at 100% relative
humidity. After Barton and Beranek
[12711.

10

30

20

Time (days)

Askey et ai. [992] studied the effect of fly-ash particulates on the corrosion of zinc
in a humidity chamber and found that the fly ash caused a slight increase in the corrosion
rate. Coal fly-ash particulates, containing less than 0.3% leachable ionic species. arc
generally very much less corrosive than oil fly ashes, with corresponding leachable
species contents greater than 1.5%. Liquid-phase catalytic oxidation of SOz by the ionic
species leached from fly-ash-bascd particulates is not significant.
In addition to relative humidity and pollutants, other factors can also affect the
corrosion of zinc in a humidity chamber. It was reported that the corrosion rate of zinc in
moving water vapor (40 cm/min) containing 1.8% S02 was high (635 !lm/yr) at the
beginning of the test and reached a fairly constant value of 135 {lm/yr after 500 days
[499]. Photo-enhanced corrosion was observed when a zinc surface was illuminated at
100% relative humidity, but the enhanced corrosion does not appear to be a dominant
factor in the overall corrosion process [331]. The corrosion rate of zinc in moist air can
be increased by the presence of a very thin film of noble metals on the surface [835].
TABLE 8. I 6. Corrosion of Galvanized Steel Placed for One Week in Sealed Tanks Containing
Various Amounts of SOrSaturated Solution"
SO? solution
added (mll b

Weioht loss
Odor in tank initially
Strong
Strong
Easily detectable

20
8
2

0.25
0.1

Just detectable
Not detectable
Not detectable

Condition of specimens

(m~/cmz)

Large pools of liquid quickly appeared on surface


Large pools of liquid quickly appeared on surface
Smaller pools of liquid quickly appeared on
surface
Very fine droplets appeared on surface
Only a very little liquid detectable
No liquid observed on the surface
No liquid observed on the surface

7.72
4.7
1.95

"Ref. 437.
b Average

for both sides. Attack was most severe on the upper surface.

0.92
0.15
0.12
O.IS

270

CHAPTER 8

20 .-------------------------,

1
:"5
~ r
o

Zn

AI

C,

Brass

Ni

Cu

FIGURE 8.24. Corrosion of metals in a mixture of


air and 5% H 2S saturated with moisture (100%
relative humidity) for 35 days. Data are taken from
Ref. 556.

Fe

8.5.2. Water and Salt Spray


Water spray provides a flow of electrolyte to the metal surface and has the effect of
washing away corrosion products and surface contaminants. In a detailed study, Gilbert
and Hadden [437] found that in a 100% relative humidity chamber the corrosion rate of
zinc sheets, sprayed for 3 s twice a day (5 days/week), was high for the first 5 days and
then decreased to a low value, as shown in Fig. 8.25. The water spray, although short in
duration, significantly increased the corrosion rate compared to that without spray, as
shown in Table 8.17.
Table 8.17 also shows that spray in dry air induced much less corrosion than spray
in humid air owing to the shorter time of wetness. Table 8.18 shows that CO 2 in ordinary
air does not change the corrosion rate in the water spray test. This indicates that the effect
of carbon dioxide on the corrosion of zinc is not direct but rather involves a secondary
reaction with the zinc hydroxides to form zinc carbonate. When a large amount of CO 2

1.5

Number of limes sprayed

0.5
I

10

14

18

10

22

15

38

28

20

Time of exposure (days)


FIGURE 8.25. Weighlloss versus time of exposure to distilled water sprays for galvanized steel. After Gilbert
and Hadden [437].

ATMOSPHERIC CORROSION

271

TABLE 8.17. Corrosion Rate of Zinc under Water Spray"


Test modeh

Corrosion (pml5 days)

Two WS./day in 100% RH air


Two WS./day in lab air (RH about 60%)
Two WS./day in 100% RH air + 3 h/day in lab
Two 3% NaCI spray in 100% RH air
Two pH 4 WS./day in 100% RH air
No spray in 100% RH air

1.8
0.09
0.9
3.0
2.0
0.2

"Data from Ref. 437.


"Abbreviations:

w.s .. Water spray for three seconds; RH. relative humidity.

is present in air, the corrosion rate in the water spray test is considerably reduced [437].
The corrosion of galvanized steel under continuous water spray was found to be similar
to that of galvanized steel dipped in distilled water and then kept in a 100% relative
humidity tank.
In the same study, it was found that freshly prepared galvanized steel sheet was
slightly less resistant to corrosion in the water spray test than that stored for 18 months.
Also, electroplated coatings corroded 35% more than hot-dipped coatings [437].
Johansson and Linder [937] used dropping water to simulate rain and found that in
the pH range of 4-7 the corrosion rate varied little, but that it was considerably higher at
pH 3. The amount of corrosion under the synthetic rain was found to be a linear function
of time.
Salt spray causes more corrosion than water spray as shown, for example, in Fig.
8.20 and Table 8.17. When zinc is tested in the ASTM salt spray test, using a 5% NaCI
solution spray, its surface becomes completely covered with a white corrosion product in
less than one hour. Figure 8.26 shows that the zinc coating loss in a salt spray test is almost
linear with exposure time [630]. The corrosion rate is about 1000 pm/yr, about 200-1000
times higher than the rates experienced in real atmospheric environments, indicating the
severity of the salt spray test. As also indicated in Fig. 8.26, more than half of the corrosion
product is washed away by the salt spray, and the amount of corrosion product remaining
on the surface become almost constant after a few days of exposure.

TABLE 8.18. Effect of CO 2 in 100% Relative Humidity


Air on Corrosion of Galvanized Steel in Air with Two
Distilled Water Sprays Daily for a Week"
Loss of coating
Atmosphere

(llm)

Ordinary air
CO 2-free air
Air containing 1.5% CO 2

1.85
1.87
0.72

----"--------------_._-----

"Data from Ref. 437.

CHAPTER 8

272

Weight loss
Corrosion product

;:;

"0

c:

rn
~

5 t-

:o~:
o

00 00

l001~1~loo1oo200

Exposure Time (Hours)

FIGURE 8.26. Weight loss and amount of


corrosion product remaining on the surface as
a function of exposure time inside a salt spray
chamber. After Zhang [630 J.

8.5.3. Cyclic Test

The conditions in a wet/dry cyclic test can vary greatly according to the wetting and
drying methods (condensation, dipping, or spray for wetting, heating or dry air blowing
for drying), speed and duration of wetting and drying, and chemical composition of the
wetting solution. Therefore, great caution has to be exercised in evaluating and comparing
the corrosion rate data obtained from different cyclic tests.
Lyon et al. [213] reported that the corrosion rate is about 200 j.lfnJyr when galvanized
steel is exposed to a cyclic test with one-hour salt spray (0.35% sulfate and 0.35% chloride
solutions) followed by one-hour drying, a factor of20 to 100 times the typical atmospheric
corrosion rates [213]. Haynie et al. [185] used a condensation/light -drying environmental
chamber to study the effect of different pollutants. Sulfur dioxide and relative humidity
were found to be the most important factors among 15 possible direct or synergistic effects
on the corrosion of galvanized steel. With a more sophisticated chamber in which solar
radiation, dew, rain, and photochemical smog could be simulated, Spence et al. [626]
found that the zinc content collected in the condensate is a linear function of S02 content,
in agreement with the data from field exposure (Fig. 8.8). It was postulated that for every
S02 molecule that is deposited on galvanized steel during periods of surface wetness, one
zinc atom reacts to form zinc sulfate (ZnS04 ).
Zhang and Tran [1094] studied the effect of cyclic wetting and drying on the
corrosion of zinc and steel. They found that on both the zinc and steel surfaces the ratio
of weight loss to corrosion product (WIP) decreases with exposure time, indicating that
the corrosion products formed initially have the effect of enhancing the ability of the
surface to retain corrosion products. Spraying has a strong effect of washing away the
corrosion products on the samples, especially for zinc (see Fig. 8.27); the W/P ratio
increased with increased spraying time for a given time of wetness. Also the variation of
the relative proportion of the time for water condensation and the spraying time for a
given time of wetness has a significant effect on the morphology of the corrosion products
[1094].
Figure 8.28 shows that hydrogen ions in spray solutions of pH greater than 3
contribute more to the cathodic reaction and less to the dissolution of the corrosion

ATMOSPHERIC CORROSION

273

11

we ight loss

.A - - - - } (

W/P

0.9

~--L-~--~--~~--~--~--~~---J O.4

10

12 14

16

18

20

Spraying lime (m in. )

FIGURE 8.27. Weight loss and ratio of weight loss to amount of corrosion product remaining on the sample
surface for zinc and steel with variation of spraying time. Drying time was 10 min, and total wetting time was
20 min. After Zhang and Tran r10941.

lO r--------------------------------------, 10
N

E
u
0,
E

__ weight loss

- - - - corr. product

:>

"0

a:o

6
"

4~
~J

steel

.c

.. -

0>

~ o

zinc

1.5

2.5

3.5

4.5

5 .5

6 6.5

pH

FIGURE 8.28. Effect of pH of the spraying solution on weight loss and weight of corrosion product after I-day
exposure in a wet/dry cyclic test. Cyclic pattern: 5 min of spraying, IS min at 100% relative humidity, and 10
min of drying. After Zhang and Tran r1094].

CHAPTER 8

274

products on the steel surface while the opposite is true on the zinc surface. The results in
Figs. 8.27 and 8.28 demonstrate that the cyclic wetting and drying pattern is one of the
important factors determining the corrosion rates of zinc alloys and steels. This may
explain why the atmospheric corrosion rate of a metal alloy is sometimes quite different
at different geographic locations with similar times of wetness and similar levels of air
pollutants [539]. The fact that a particular wetting and drying pattern has different effects
on zinc and steel might be partially responsible for the different rankings of the corrosiveness of the atmosphere toward the two materials at various locations around the world
as shown in Table 8.4.
8.5.4. Thin-Layer Electrolytes
A thin-layer electrolyte is an electrolyte having a thickness less than a few millimeters. The corrosion of metals under a thin layer of electrolyte is very different from that
in the bulk solution because the diffusion of oxygen is greatly enhanced under a thin
electrolyte layer in open air [336, 556]. Also, experimental results have indicated that the
solvation capacity of a thin electrolyte layer for the dissolved species is very limited,
which can affect the process of formation of corrosion products [183].
In an early study, Gilbert and Hadden [437] reported on the corrosion of zinc under
water drops. For a given volume of water, corrosion increased with the number of drops
as shown in Fig. 8.29. For the same volume, more drops cause a larger surface area to
corrode. In addition, a smaller water drop has a smaller thickness, which facilitates oxygen
diffusion and enhances the corrosion. The corrosion generates a band of localized pitting
around the perimeter of the drops, which is attributed to the difference in oxygen diffusion
at the perimeter compared to the center of a water drop.
Valencia et al. [945] found that the formation of various zinc corrosion products can
be simulated using different synthetic solutions in an immersion-drying method, originally developed by Pourbaix et al. [258]. Rozenfeld [556] examined corrosion under
frequent wetting by periodically immersing the sample in a solution and then exposing it
in a chamber of controlled relative humidity and oxygen pressure. The thin electrolyte
layer formed by such a method is about 30 J.1m thick. Figure 8.30 shows that the corrosion

20

rr
Cl

~1+
I
5 L _ - - - ' - _ - - - 1_ _"---_-'-_-'-_--'--_---'_~
o 10 20 30 40 50 60 70 80
Number of Water Drops

FIGURE 8.29. Corrosion caused by


equal quantity of distilled water distributed on galvanized steel surface as various numbers of drops for one week.
Total volume of water was 70 ml. After
Gilbert and Hadden [437].

275

ATMOSPHERIC CORROSION

4,------------------------------,

FIGURE 8.30. Effect of wetting frequency


(number of wettings per hour) on zinc corrosion (oxygen consumption) and comparison to zinc corrosion resulting from
immersion in bulk solution. Duration of test
was 6 hr. Data are taken from Ref. 556.

... 0.5 N NaCI


x 0.5 N Na,SO.
/'

Bulk immersion

9 10 11 12 13 14

Number of Wettings

rate of zinc under frequent wetting is many times higher than that for the fully immersed
condition and that it increases with the frequency of wetting. In this figure, the corrosion
rate under frequent wetting is also seen to be higher in sulfate solution than in chloride
solution.
The relative humidity of the surrounding air determines the retention time of the thin
electrolyte on the sample surface and, hence, the corrosion rate. It also affects the
corrosion features of zinc. At lower relative humidity, corrosion is more uniform. The
effect of temperature on the corrosion rate under a thin-layer electrolyte was attributed
by Rozenfeld [556] to, among other things, changes in the diffusion-layer thickness
resulting from changes in convection and solubility of oxygen.
Mansfeld and Tsai [210] measured the weight loss of zinc plates covered by a
O.5-mm-thick electrolyte of various chemical compositions after drying them out at
various relative humidity values. The corrosion rate was much higher under a thin
electrolyte compared to that for bulk immersion in O.OIM NaCl or O.OIM Na2S04
solution. However, the converse was found for O.OIM Hel or O.OIM H2S04 solutions. In
8r-------------------------------~

6
Cl

E
U)
U)

24

FIGURE 8.31. Weight loss for zinc under


0.5-mm thin layer of distilled water in air
of different humidities. Data are taken
from Ref. 210.

o
o

__ Bulk immersion
I

10

20

30

40

50

Relative Humidity (%1

60

70

80

276

CHAPTER 8

16,------------------------------,
14

~ 12

'"
.210

4L-__
-4

-3.5

__

_ L _ _ ~_ _ _ _ ~_ _ _ L _ _ _ i _ _ ~

-3

-2.5

-1.5

-2

-1

Log c

-0.5

FIGURE 8.32. Weight loss for zinc under


thin layers of Na2S04 solutions of different concentrations. Reprinted trom Mansfeld and Tsai r210]. with kind permission
from Elsevier Science Ltd. The Boulevard, Langford Lane, Kidlington OX5
1GB, United Kingdom.

the salt solutions the main oxidizing agent is oxygen, the concentration of which near the
surface is a function of diffusion-layer thickness (electrolyte thickness), whereas in the
acidic solutions it is hydrogen ions, the amount of which depends on the volume of the
solution. Figure 8.31 shows that the corrosion of zinc increases with increasing relative
humidity since the time of wetness is longer at a high relative humidity. At a given relative
humidity, the corrosion under a thin-layer electrolyte increases with increasing concentration of the electrolyte, as shown in Fig. 8.32.
Stiles and Edney [183] reported that corrosion of Zn under a 0.3-mm thin electrolyte
increases with the initial concentration of H+ (Fig. 8.33), independent of the kind of acid.

0.4

HNO" pH = 3

><

0.3

+'

N(:i

0.2

(:i

.~
o.l

g
8

g
N

0.1

HN0 3 , pH = 4

rg

El

~~/~o~o--------------~------
j

HN03 pH = 5

500

1000

Residence time (8)

1500

FIGURE 8.33. Zinc concentration as a function of time in thin-layer HN0 3 solutions of


different H+ concentrations. From Stiles and
Edney [183]. Copyright by NACE International. All Rights Reserved by NACE; reprinted with pennission.

ATMOSPHERIC CORROSION

277

250

g
OJ

150

100
50
25
5

FIGURE 8.34. Effect of electrolyte film thickness on


cathodic polarization of zinc in a 0.1 N NaCl solution. Data
are taken from Ref. 556.

__-ULL____~____L-~

-0.8

-1

-1.2

-1.4

Electrode potential (V",)

The pH of the thin-film electrolytes increased in less than 100 seconds to about 6.7. The
ratio of the final [Zn 2+] to the initial [Wj was about 0.48, which is consistent with the 1:2
stoichiometric ratio ofZn to H required for the corrosion process in an acidic environment.
At the end of the test, the thin-layer electrolyte was found to be saturated with zinc ions.
Data from electrochemical measurements reveal that the cathodic current of zinc
under a thin-layer electrolyte is greatly increased with decreasing thickness of the
electrolyte owing to the enhanced diffusion rate of oxygen, as shown in Fig. 8.34 [336,
522,556]. The effective diffusion-layer thickness under a thin electrolyte layer is about
0.3 mm, which is smaller than that in a bulk solution because under a thin electrolyte layer
the diffusion is also enhanced by convection caused by evaporation [556]. The anodic
polarization behavior of zinc under a thin-layer electrolyte is similar to that in the bulk
electrolyte. The reduction of dissolved S02 in solution was also found by Rozenfeld [556]
to increase under a thin-layer electrolyte. Instead of being a reducing agent as is
commonly assumed, S02 shows oxidizing properties and acts as a powerful cathodic
depolarizer.
The galvanic action of a metal couple under a thin-layer electrolyte differs also very
significantly from that in bulk electrolyte. Because of the difference in polarization
resistances between the anodic and the cathodic reactions, the potential and galvanic
current distributions over a coplanar zinc/steel couple under thin-layer electrolytes are
very different for zinc and steel, as shown in Figs. 7.4 and 7.5 in Chapter 7. Because of
this kind of potential distribution, when a piece of galvanized steel is connected with a
piece of bare steel under a thin-layer electrolyte, the corrosion will only occur at the edge
of contact and will proceed in the direction away from the contact line. Walter II 079]
found a correlation between the corrosion current, electrochemical impedance, and
weight loss of zinc for a zinc/steel galvanic couple under a thin water layer (0.5 mm thick).

278

CHAPTER 8

8.6. CORROSION MECHANISMS


Zinc corrodes very slowly in clean dry air at room temperature with the formation
of an oxide film. On a cleaved plane of a single crystal, an amorphous film is slowly
produced [404]; after a few weeks, it reaches a thickness of approximately 100 A. It is
amorphous on the surface but crystalline in the interior. If the zinc surface is anodically
polished, the film thickens faster and reaches a thickness of a few hundred angstroms in
several days. The oxide films prevent further oxidation of the metal.
Unlike oxidation in dry air, the oxidation under moisture is of an electrochemical
nature. When in contact with condensed clean moisture in the form of rain, mist, or dew,
zinc corrodes with the formation of zinc hydroxide according to the following reactions:
Zn + 20W

Zn(OHh + 2e-

(8.1 )

(8.2)
and/or
(8.3)

Over a certain period of time, ranging from a few days to a few weeks, this hydroxide
layer then dehydrates to form oxide or reacts with the carbon dioxide dissolved in the
water to form the relatively less soluble zinc carbonate [173, 331, 437]:
(8.4)

or
(8.5)

or
5Z00 + 2C0 2 + 3HzO ~ 2ZnCO y 3Zo(OHh

(8.6)

The formation of a carbonate film slows down the corrosion rate. In the early months of
exposure, when the carbonate film is forming, the corrosion rate is relatively high, but it
gradually reaches a lower constant value as the formation and dissolution rates of the
carbonate film become equal. Thereafter, the corrosion is determined by the rate of
chemical dissolution of the carbonate film, resulting in a linear relationship between the
atmospheric corrosion and exposure time.
The atmospheric corrosion of hot-dip galvanized sheet occurs in three distinct stages:
(1) a short initial period during which a protective surface layer is formed; (2) a long
period of corrosion of the zinc coating; and (3) corrosion of steel on the exposed area
where the zinc coating has been consumed. As corrosion proceeds, the surface changes
in appearance from silver-gray (zinc) to gray-white (corrosion product of zinc) to
brown-gray (corrosion product of zinc-iron alloy phases) to red rust (corrosion product

ATMOSPHERIC CORROSION

279

of steel) [550]. Regular galvanized steel is gray-white in appearance for most of the life
of the coating because unalloyed zinc constitutes the main pan of the coating.
Regular galvanized coatings consist of several layers of different ZnlFe compositions
and structures. These layers, from the free surface to the steel surface, are free zinc
followed by (, J, and Tintermetallic phases [3941. Initially, the free zinc layer corrodes.
This is followed by the corrosion of the ( layer, which is more corrosion-resistant. At
some local sites. the ( crystals may break away, leaving the J layer, and. subsequently,
the Tlayer, exposed to attack. In the meantime, the surface becomes more and more brown
because of the formation of iron oxides. When the T layer is consumed initially at a
localized area, the steel base is attacked, leading to the formation of brown voluminous
corrosion products, which protrude out of the zinc layer as brown pimples. However.
because the zinc surrounding the pimples provides cathodic protection, extensive corrosion of the steel at these pimples does not occur until most of the zinc coating around the
pimples is consumed.
The whole atmospheric corrosion process follows two irregular cycles: on the one
hand, the cycle of wetting (dew or rain) and drying (radiation, temperature, or wind) and,
on the other hand, the cycle of zinc dissolution, hydroxide formation, carbonate formation, and dissolution of the carbonate film. The interaction between these two cycles
determines the panicular corrosion rate of zinc in a given atmospheric environment. The
various parameters involved in atmospheric corrosion affect the corrosion of zinc by
changing the patterns of these two cycles. The amount of water per unit area in dew is
small, and this water can be quickly saturated with dissolution products to form precipitates. Salt saturation does not easily occur in the surface water formed by rain owing to
the flowing nature of rainwater. Therefore, while the length of the wetting time will
determine the amount of corrosion within one cycle, the form of wetting determines the
amount of corrosion products retained on the surface. The speed and the extent of drying
will determine the composition as well as the compactness of the corrosion products.
since the formation of zinc oxide and zinc carbonate involves dehydration. Long periods
of wetting with infrequent drying result in relatively more corrosion and formation of less
compact corrosion products. For the same wetting time, wetting with much rain will result
in more dissolution of the carbonate film and less accumulation of corrosion products.
while wetting with much dew and little rain will be more likely to result in a pileup of
corrosion products. The corrosion process and the formation of corrosion products can
cause condensation at a lower relative humidity and increase the absorption of moisture
as well as pollutants.
Under a thin-layer electrolyte, dissolution and precipitation are two fundamental
kinetic processes in the corrosion of a metal. They. in turn, are functions of two chemical
variables, acidity and ionic strength, which are determined by the atmospheric conditions.
In a nonpolluted atmosphere, hydrogen reduction is not the major cathodic process owing
to its high overpotential on pure zinc, and the corrosion rate of zinc is normally controlled
by the reduction of oxygen. Under thin-layer electrolytes, such as those formed by rain
and dew, the rate of oxygen reduction is greatly increased compared to that in bulk
electrolyte because of the thinner diffusion layer. As a result of the cathodic reactions, the
pH of the electrolyte increases, reducing the solubility of the corrosion products and thus
facilitating solid precipitation. Carbon dioxide is another air constituent that affects the
corrosion of zinc. With a concentration within the range of 300-500 ppm in the

280

CHAPTER 8

atmosphere, it equilibrates with the rainwater to form a buffered solution of pH approximately 5.6 [631]. Although carbon dioxide is beneficial in forming the protective zinc
carbonate film, the dissolution of much CO 2 in moisture may lower the pH to a range in
which the carbonate is not stable and can be dissolved [331, 627].
Abnormal corrosion rates of zinc occur when there are considerable amounts of
pollutants present in the air. The major effects of pollutants are increasing the time of
wetness, increasing the solubility of zinc oxide and carbonate, decreasing the pH of the
wetting solution, and contributing directly to the cathodic process. Low-pH moisture,
such as acid rain, is very corrosive to zinc because the protective zinc carbonate film
cannot form. Laboratory testing has shown the existence of a stoichiometric relation
between the amount of zinc dissolution and the concentration of hydrogen ions in a
solution [183, 626].
Salt near the sea and sulfur dioxide in urban and industrial areas are two pollutants
that can cause a significant increase in the corrosion rate. The presence of both salt and
sulfur dioxide increases the time of wetness and the solubility of zinc in the wetting
solution. However, unlike sulfur dioxide, salt does not enhance the cathodic reaction
process, which may explain why much more salt, as compared to S02' is needed to cause
a comparable amount of corrosion and why the atmospheric corrosion in unpolluted
marine environments is much lower than that in heavily polluted industrial areas.
There are two different theories on the role of sulfur dioxide in the corrosion process
of metals. The commonly accepted one involves the acidification of moisture through
hydration and oxidation of sulfur dioxide to sulfuric acid and the reduction of hydrogen
ions in the moisture to promote the corrosion of zinc [331, 626, 629]:
(8.7)

(8.8)
(8.9)

(8.10)

and
(8.1l)

The entire process can be considered as corrosion under an acid rain, in which hydrogen
acts as the cathodic depolarizer and the sulfate ions promote the solvation of the dissolved
zinc ions. The deposition of S02 on zinc surfaces has been found to be enhanced by the
presence of 0 3 and N0 2, which also oxidize surface sulfite formed during the corrosion
[940].
The theory based on the reactions in Eqs. (8.7)-(8.11) agrees with a number of
experimental observations. Data from both laboratory testing and field exposure indicate
a one-to-one stoichiometric relation between Zn 2+ and SO~- in the runoff solutions [626].
Furthermore, zinc sulfates are usually found in the corrosion products of zinc in almost
all types of atmospheric environments [173, 297, 331, 437].

ATMOSPHERIC CORROSION

281

Another theory, originally proposed by Rozenfeld [556], suggests that S02 acts
directly as a cathodic depolarizer:
(S.12)
(S.13)
(S.14 )
This theory is supported by the fact that the whole process of reduction of S02 is enhanced
under thin-layer electrolyte conditions. McLeod and Rogers [499] observed that the
formation of sulfur is the first reaction to occur on a zinc surface that is exposed to moist
air containing S02, while the formation of 2ZnSO,5H 1 0 and then ZnS0 4 H 20 occurs
more than one month later, indicating that the reduction of S02 is an important contributor
to the corrosion. They also observed the formation of sulfide and thiosulfate in deaerated
sulfurous acid as a result of the corrosion of ZInC. However, Fiaud [291] pointed out that
the S02 concentration (0.1 %) in Rozenfeld's experiments is much higher than that in real
atmospheres (around 10-4 %) and that the reduction reaction at very low levels of S02 may
be very different. It was reasoned that under a thin-layer electrolyte the reduction of S02
at atmospheric pollution levels is greatly affected by the presence of oxygen.
While laboratory electrochemical and analytical experiments indicate the depolarizing effect of S02, which should lead to the formation of sulfur species of lower valence,
only sulfates have heen reported to exist in the corrosion products of zinc from field
exposure [173, 331]. This indicates either that the reduction of SOl does not make a
significant contribution to corrosion in field exposure or that the products of the reduction
reaction are not stable and are oxidized eventually to form sulfates.

9
Corrosion in Waters and
Aqueous Solutions
9.1. INTRODUCTION
Waters are commonly classified as pure water (e.g., distilled water or deionized
water), natural fresh water, and seawater. Waters containing artificially introduced salts
are usually called aqueous solutions.
Zinc-coated steel articles and structures, such as galvanized tubing and water tanks,
are commonly used in waters. They are not usually used in aqueous solutions in practical
applications. However, a large number of studies on the corrosion of zinc have been
carried out in solutions, primarily (a) to simulate a corrosion phenomenon, (b) to
accelerate a corrosion process, or (c) to conduct electrochemical measurements.
The corrosion data presented in this chapter are organized into three main sections:
(i) pure water, (ii) natural water, and (iii) aqueous solutions. The corrosion rates in aqueous
solutions are largely limited to the data obtained with gravimetric methods. The data
obtained by electrochemical methods and the general electrochemical behavior of zinc
electrodes in solutions are discussed in Chapter 2, 3, and 5. In addition, the information on corrosion forms, which are also mainly studied in solutions, is presented in
Chapter 7.
9.2. CHARACTERISTICS OF WATERS
9.2.1. Fresh Waters

Waters contain solids, gases, and sometimes colloidal or suspended solid matter. The
concentrations of dissolved substances are relatively low but vary considerably from one
source to another. Even distilled water, depending on its aeration condition, can contain
varying amounts of oxygen and carbon dioxide. The important constituents in water can
be classified as follows: (1) dissolved gases (e.g., oxygen); (2) mineral constituents,
including calcium and sodium salts, salts of other metals, and silica; (3) organic matter,
including that of animal and vegetable origin; and (4) microbiological life forms [558].
Among the dissolved gases, oxygen is probably the most significant constituent in
relation to the corrosion of metals, owing to its cathodic depolarizing effect. In surface
283

284

CHAPTER 9

water, the oxygen concentration approaches saturation. The solubility is slightly less in
the presence of dissolved solids than in pure water, but this effect is not very significant
in natural waters containing less than 1000 ppm dissolved minerals. Dissolved carbon
dioxide is also very important; however, its effect must be considered in relation to other
constituents, especially calcium hardness. The amounts of dissolved air and oxygen in
water as a function of temperature are listed in Chapter 2 in Table 2.10 [496].
The principal ions found in waters are calcium, magnesium, sodium, bicarbonate,
sulfate, chloride, and nitrate. The hardness of water is usually referred to the calcium
carbonate content. Water containing less than 50 ppm of CaC0 3 is considered soft, and
water containing more than 150 ppm is considered hard [558 J. CaC0 3 tends to precipitate
on the surface of the water container to form scale. The amount of scale is less with waters
of high carbon dioxide concentration because of the higher solubility of calcium carbonate. Table 9.1 shows the constituents of typical waters. Typical resistivities of different
waters are shown in Table 9.2 [659].

9.2.2. Seawater
The most characteristic feature of seawater is its high salt content. The salt content
of open-sea water, away from inshore influences such as melting ice, freshwater rivers,
and areas of high evaporation, is quite constant and is roughly equivalent to that of a 3.5%
solution of sodium chloride. As a result of its high salt content, seawater has a very low
resistivity compared to other waters, as shown in Table 9.2. Table 9.3 shows the major
constituents of seawater. The average temperature of the surface water of the oceans tends
to vary directly with the latitude, ranging from about -2C at the poles to 35C right on
the equator.

TABLE 9.1. Typical Analyses (ppm) for Natural Fresh Waters"

Water
Soft lake water
Moderately soft surface
water
Slightly hard river water
Moderately hard river
water
Hard borehole water
(chalk formation)
Slightly hard borehole
water containing
sodium bicarbonate
Very hard underground
water
aRef.558.
b Also 51 ppm nitrate (NO,).

Alkalinity
to
methyl
Total
Calcium
orange hardness hardness
pH value (CaC0 3 ) (CaC0 3 ) (CaC0 3)

Sulfate
(S04)

Chloride
(CI)

Silica
(Si0 2)

Dissolved
solids

6.3
6.8

2
38

53

5
36

6
20

II

0.3

33
88

7.4
7.5

90
180

120
230

85
210

39
50

24
21

3
4

185
332

7.1

250

340

298

17

400"

8.3

278

70

40

109

94

12

620

7.1

704

559

451

463

149

1670

10

Trace

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

285

TABLE 9.2. Resistivities of Various Types of Water"


Resistivity (Q'cm)

Type of water

20.000,000
500,000
20,000
1-5000
200
30
20-25

Pure water
Disti lied water
Rainwater
Tap water
River water (brackish)
Seawater (coastal)
Seawater (open sea)

"Approximate values; data are trom Ref. 1105.

The amount of dissolved oxygen in seawater is a function of temperature. The


amounts dissolved at equilibrium are tabulated below [563]:
T(C)
Dissolved O2 (mill)

-2

8.52

8.08

10
6.44

5
7.16

IS
5.86

20
5.38

30
5.42

Figure 9.1 shows the temperature, salinity, pH, and oxygen concentration as a
function of depth in the sea. The concentration of oxygen decreases with increasing depth
up to about 2000 ft but then increases slightiy with further increases in depth. The
concentration of dissolved oxygen is also affected by the degree of water movement and
by the amount of biological activity. Photosynthesis increases the oxygen concentration.
while some bacterial activities can reduce it to zero [659J. The metabolism of some
bacteria produces hydrogen sulfide (H 2S), ammonia. and other nitrogenous compounds;
H2S concentrations of 30-35 ppm are quite common in seawater.
The pH of surface seawater, in equilibrium with carbon dioxide in the atmosphere.
normally lies between 8.1 and 8.3, because of the existence of excess amounts of basic
radicals, mainly carbonates, but may fall to 7 in stagnant basins [659]. The pH decreases

TABLE 9.3. Major Constituents of Seawater,,h


Species

Concentration (ppt)

Chloride (Cn
Sulfate (SOh
Bicarbonate (HCUjl
Bromide (Br-)
Fluoride (F-)
Boric acid (H oB0 3 )
Sodium (Na +)

Magnesium (Mg 2+)


Calcium (Ca2+)
Potassium (K+)
Strontium (Sr 2+)
"Ref. 563.
"Chlorinity, 190/r; density at 20C. 1.0243.

18.98
2.65
0.14
0.065
0.0013
0.026
10.56
1.27
0.40
0.38
0.013

286

CHAPTER 9

Oxygen, mil I
2

1,000

~I

~ :/~~

'I i~

~...,., ~

2,000

-..

.c.

a;

~I

3,000

~I

a.

~ r-- PH

4,000

5POO

~
6pooO

I~
I

~~ ~

~~ I

12

14

16

18

342

34.4

34.6

34.8

7.6

7.8

8.0

82

Temperature, C
33.0

332

33.4

33.6

33.8

34.0

Salinity, ppt
6.4

6.6

6.8

7.0

7.2

7.4

pH
FIGURE 9. I. Oceanographic data taken in the Pacific Ocean at a site west of Port Hueneme. California. After
Fink and Boyd [660].

with depth as shown in Fig. 9.1. The presence of carbon dioxide also affects the formation
of scales. However, precipitation of calcium carbonate in seawater does not occur as
readily as in fresh water since the solubility of calcium carbonate in seawater is about 530
times that in fresh water [659].
9.3. CORROSION IN PURE WATER
The corrosion rate of zinc in distilled water varies widely, ranging between 15 and
150 f1rn1yr [217, 559, 654]. The degree of distillation or deionization has negligible effect

on the corrosion rate. In one study, about the same corrosion rate of zinc was found in
deionized water with resistivities of 1 MQ'cm and 18 MQ'cm [1263].
The corrosion rate depends strongly on the amount of dissolved oxygen and carbon
dioxide as shown in Table 9.4 and Fig. 9.2 [400,559]. According to Kenworthy and Smith

287

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

TABLE 9.4. Effect of Oxygen on the Corrosion of Zinc in


Distilled Water".!'
- - - - - - - - _.... _ - - - - - - - - - - - - - -

Test condition
Boiled distilled water: specimens
immersed in sealed flasks
Oxygen bubbled slowly through
the water

Temperature
(OC)

Corrosion rate
(pm/yr)

Room
40
65
Room
40
65

25.4
48.3
83.8
218.4
348.0
315.0

"Ref. 559.
hHigh-grade zinc specimens, in duplicate, immersed for 7 days. The corrosion rate
was calculated after removal of corrosion products.

[400], the form of corrosion in distilled water changes from pitting to unifonn attack with
increasing carbon dioxide concentration.
Figure 9.3 shows that the corrosion rate of zinc in distilled water increases only
slightly with temperature up to about 50C, then increases quickly with temperature,
reaching a maximum at about 65C, before decreasing [412]. According to Cox [4121,
the sharp increase in corrosion rate from 50 to 60C may be attributed mainly to an abrupt
change in the nature of the corrosion products from being protective to being nonprotective, leading to a sharp increase in the corrosion rate. At room temperature the corrosion
products precipitate in the form of a gel with an indefinite amount of absorbed water.
When the precipitate is heated, it gradually loses water and undergoes changes in its
physical characteristics. As shown in Fig. 9.4, Grubitsch and Illi [707] similarly found
that a peak in the corrosion rate occurs at around 60C and reported that the presence of
a sufficient amount of oxygen is required for the occurrence of the peak.
0.2 .---------------------------------------,
>-

'"

5 0.1 6

0,
E

Hard Supply Wator

-0- Distillad Water

-g 0.12
".

"0
(/)

III

Ci

u
.::

0.08

"0

:E 0.04
Ol

12

18

24

30

36

42

Free Carbon Dioxide Content of Water, ppm

FIGURE 9.2. The effect of free carbon dioxide in distilled water and hard supply water on the dissolution of
zinc (immersed for 56 days at 18C). After Kenworthy and Smith [400].

288

CHAPTER 9

3 ,-------------------------------,
>

E 2.5
E

Qi

a::
c
0

en

1.5

I1J
Cl

ro

~ 0.5

<{

oL---~======~~~----~--~~

20

40

80

60

100

Temperature, C

FIGURE 9.3. Corrosion of zinc in airsaturated water as a function of temperature. The test samples were
rotated at a speed of 56 rpm . After Cox
[412J.

In distilled water at room temperature and open to air, zinc corrodes with the
formation of pits [400-402, 654]. The formation of pits depends on the oxygen content;
when the water is depleted of oxygen, there is little corrosion, and when oxygen pressure
is high, the corrosion is of a uniform type. The corroding area spreads with time as the
surface gradually becomes covered with a thick layer of hydroxide and carbonate. In a
few weeks, the specimen is completely covered with the white material. On removal of
this layer, pits are found on the surface of the metal.
9.4. CORROSION IN NATURAL WATERS
9.4.1. Cold Fresh Water

In general, the corrosion rate of zinc is lower in hard water than in soft water or
distilled water [400, 217]. An example is shown in Fig. 9.2. This lower rate is largely

0.4 . ------------------------------------ ,

+ Ul"1de-r PUt'
0.32
NE

Under aif, CO 2 ,,

+ Under Nl.+ O.6vol% 02


. . U nd~r pvre 0 2

~0 . 24

<Ii
en

.2

.E 0 .1 6
Cl

w
~

0.08

oL-____ ____ ____


o
20
40
~

______

60

Temperature, C

L __ _ _ _

80

~~

100

FIGURE 9.4. Effect of several gases


on the temperature dependence of zinc
corrosion in hot distilled water. After
Grubitsch and IIIi [707].

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

289

TABLE 9.5. Corrosion Rates of Zinc and Zinc Coatings Immersed in


Various Industrial and Domestic Waters a
Type of water
Mine water, pH 8.3, 110 ppm hardness, aerated
Mine water, 160 ppm hardness, aerated
Mine water, 110 ppm hardness, aerated
Demineralized water
River water, moderate soft
River water, moderate soft
River water. treated by chlorination and copper sulfate
River water, treated by chlorination and copper sulfate
Tap water, pH 5.6, 170 ppm hardness. aerated
Spray cooling water, chromate treated. aerated
Hard water
Soft water

Corrosion rate
(jlm/yr)
31
30

46
137

97
61"
81

64b
142
IS
16
15

"Refs. 203 and 217.


"Galvanized steel.

attributed to the formation of a protective scale in hard waters. As shown in Table 9.5, the
corrosion rate can vary significantly, from as low as 8 j1m/yr to as high as 140 j1m/yr, in
different waters.
Dissolved carbon dioxide in hard water generally increases the corrosion rate but
has less effect than in distilled water (Fig. 9.2). Also, the corrosion of zinc in hard supply
water is much more uniform than that in distilled water, in which pitting usually occurs.
This was attributed by Evans [401] to a more effective ionic exchange in supply waters,
which prevents the localization of corrosion activities. According to Kenworthy and
Smith [400], in waters oflow carbon dioxide content, the calcium bicarbonate in the water
can precipitate as a protective carbonate scale, but at high carbon dioxide contents, this
precipitate will not easily form, owing to the lower pH. and the dissolution of zinc
proceeds unhindered.
Flowing water causes more corrosion than still water. Fujii [561] found a linear
relation between the corrosion of a zinc coating and the cubic root of the flow velocity in
tap waters, as shown in Fig. 9.5. This figure also shows that, at a given flow rate, the
corrosion rates in the tap water of Berlin and Dortmund were many times higher than in
that of Tokyo.
9.4.2. Hot Fresh Water
Galvanized tanks have been widely used to store hot water and can usually last as
long as 30 or 40 years. However, these tanks have also been found to have a very short
life in some supply waters [655,656]. Among the many factors affecting the performance
of galvanized steel water tanks, temperature and composition of the water appear to be
the most important [707, 400, 412, 458, 709].
The area-averaged corrosion rate in hot hard water is usually low. Kenworthy and
Smith [400] reported that the corrosion rate of zinc in 75C hard supply water, free of

290

CHAPTER 9

Tokyo

;;;-

--E

OJ

ai

"

c
0
-iii

e
0

0 .2

0.4

0_8

0.6

1.2

Flow velocity, (m/s)'/l


FIGURE 9.5. Effect of flow velocity on the corrosion rate of zinc-coated steel tubes in tap waters. After Fujii
[5611.

carbon dioxide, is lower than I jJ.m/yr. Dissolved carbon dioxide has a significant effect
on the corrosion rate, as shown in Fig. 9.6 [400]. Gilbert [709] found that the corrosion
rate of zinc over a five-month period in hard supply water saturated with 1.5% (about 14
ppm) CO 2 at 85C was between 25 and 43 jJ.m/yr.
9.4.2.1. Pitting Corrosion. Pitting is a common form of corrosion for zinc in hot
supply water. It is, in general, much worse on pure zinc than on galvanized specimens;
the zinc-iron alloy layers in galvanized coatings have greater resistance to the pitting

-+ C0 2 free

-- 5 . 6 ppm CO2

E
'-'
Oi

E
-0
Q)

>

-10.2 ppm

20.2 ppm

.... 35 ppm
-<>-62 ppm

'0
(/)
(/)

'6

c'-'

' _~L- ~ ------------------------------__~__~__L -_ _L-~~

O~~--~---L---L

20

40

60

80

100

120

140

160

180

Duration of test, days


FIGURE 9.6. Effect offree carbon dioxide on the dissolution of zinc in hot hard supply water. After Kenworthy
and Smith [400].

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

291

attack [400, 709 J. During a corrosion test, as described by Gilbert [709], bubbles were
found to adhere to the surfaces at the beginning of the test, and "bubble cups" of corrosion
product formed at some of the points of attachment. These bubble cups consist of white
zinc corrosion products, and in some cases several bubbles adhere in close proximity such
that fairly large areas covered with white zinc compounds are produced. When failure
occurs, it is always beneath this white corrosion product.
Bonilla [686] reported that although the total amount of corrosion in hot water is
much less on galvanized steel than on black steel, pitting may be more severe on
galvanized steel. In hot water, the zinc surface may be passivated, leading to a polarity
reversal of galvanized steel. When pits are initiated, they may penetrate through the
coating and into the steel substrate because of the galvanic action between zinc and steel.
The pitting rate in hot water can thus be quite high, especially when carbon dioxide is
present. The presence of copper in hot water also enhances pitting. Campbell [257]
reported that, in tap water with 0.1 ppm copper and 28 ppm free carbon dioxide at 75C,
a galvanized coating showed numerous pits, extending into the steel to a depth of 0.3-0.4
mm after 33 weeks. The general theories on pitting and polarity reversal are discussed in
Chapter 7.
Intergranular corrosion is sometimes found to occur on zinc alloys in hot water,
especially on zinc-aluminum alloys. It has been established that intergranular corrosion
of zinc alloys in hot water occurs in alloys with more than 0.03% Al [48,491. The intensity
of intergranular corrosion is increased by the presence of small amounts of impurities
such as Ph, Sn, In, Cd, and Hg. The tensile strength of zinc-aluminum alloys can be
significantly reduced as a result of intergranular corrosion [225 J. Detailed information on
the intergranular corrosion of zinc alloys is presented in Chapter 7.
9:4.2.2. Effect of Dissolved Copper. The presence of trace amounts of copper in
water can substantially increase the corrosion of zinc. Kenworthy [737] found that the
amount of copper dissolved in hot water increases with increasing carbon dioxide
concentration. As little as 0.1 ppm copper causes a definite increase in corrosion rate.
With amounts of copper up to about 0.3 ppm, the amount of corrosion is proportional to
the amount of copper. The copper appears to deposit as small metallic particles on the
surface of the zinc. Enhanced corrosion occurs because of the larger cathodic activity
generated by the copper particles.
9.4.2.3. Other Factors. The flow of water over a zinc surface has a significant effect
on its corrosion. Nielsen and Y ding [720] found that galvanized pipes in which hot water
circulated continuously showed more coating corrosion than pipes without circulation.
Also, pitting corrosion was observed to be more severe on the bottom of the pipes. In
another work, Weast and Shulman [226] reported that in pressurized waters the corrosion
rate of zinc as a function of temperature does not pass through a maximum as it does in
nonpressurized water.

9.4.3. Seawater
9.4.3.1. Corrosion Rate. The corrosion rate of zinc in seawater is typically between
20 and 110 f.1rn1yr, varying with location, length of exposure, type of zinc, etc., as shown
in Table 9.6. It is generally higher at the beginning of exposure and decreases with time.

292

CHAPTER 9

TABLE 9.6. Corrosion Rates of Zinc in Seawater


Material

Duration of
expos ure (yr)

Location

99%Zn
99%Zn
Galvanized

Pacific Ocean
Pacific Ocean
Kure Beach. Hawaii

0.5

99.I%Zn

Eastport, Maine

I
3

Cast zinc
Cast bar
Galvanized
Zinc
Galvanized
Zinc

Panama
Bristol Channel
Bristol Channel
Digha, India
Digha, India
Tropical

Corrosion rate
(p.mJyr)

Reference

110
70
53

0.5

575
575
436
436
436
436
436
436
436
662
662
663
663
217

28
28
25

28
92

4
4
0.1
0.1
4
16

64
34
41
21

18
IS

8
10

Monel
Zinc
Si bronze
Lead
Low brass
Cu-Ni
Aluminum

.!:!

'E
c

Gi
c
CD
n.

.,

24
CD

>

Rc:::: 002 mpy

16

Exposure Time, years


FIGURE 9.7. Average penetration of wrought nonferrous metals after 16 years' continuous immersion in
seawater. After Fink and Boyd [660].

293

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

140,----------------------------------------- Lake fresh water

120

Qi
cQ)

0..

Sea water (immersed)

+ Sea water (tidal)

"': 100
c
o
~ 80
60

Q)

Ol

'>Q;"

40

<t:

20
0

Years exposed

FIGURE 9.8. Comparison of zinc corrosion rates in lake water and seawater in the Panama Canal Zone. After
Alexander et al. [690].

The data in Table 9.6 indicate that in the Pacific Ocean and at Kure Beach, the average
corrosion rates for a l-yr exposure were only half of those for a 0.5-yr exposure.
Figure 9.7 shows that zinc corrodes relatively fast compared to several other common
metals and alloys. It also shows that the corrosion rate of zinc in seawater decreases with
immersion time [6601. Alexander et al. [690] investigated the relative corrosion rates of
zinc in fresh water and seawater. The corrosion rate in seawater was initially higher than
that in fresh water but after about two years of exposure it became similar to that in fresh
water, as shown in Fig. 9.8.
The data in Table 9.7, reported by Anderson [436], show that a small amount of iron
in rolled zinc had little effect on the corrosion rate. Campbell et al. [434], on the other
hand, found that heat treatment of a galvanized coating, which introduced 10-20% iron,
resulted in rapid failure of the coating in seawater.
9.4.3.2. Corrosion Form. Both uniform and localized forms of corrosion can occur
on zinc in seawater, depending on the material [217, 661, 690]. Pitting was found to occur

TABLE 9.7. Corrosion Rates of Iron-Containing Zinc in Seawater"


Corrosion rate l.}1m1yd
Iron (%)
0.0003
0.0008
0.0014
0.0021
0.006
0.011

Total immersion

Tidal zone

Flowing water (2 ft/s)

35.6
22.4
22.4
22.4
22.4
35.6

20.3
20.3
22.9
22.9
20.3
22.9

76.5
50.8
50.8
50.8
50.8
50.8

"From Anderson [436]. Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.
b Average

for two I x 4 x

~-inch specimens exposed for

I yr.

CHAPTER 9

294

on 99.7% pure zinc in tropical seawater, the ratio of the deepest pit to the average corrosion
rate being 14 [690]. Kweon and Coddet [691] observed blistering of a flame-sprayed 85%
Zn-15% Al coating on an aluminum alloy after immersion in seawater for 10 months.
9.4.3.3. Effect of Flow. Tidal Zone, and Depth. In one study, similar corrosion rates
were found for iron-containing zinc in a totally immersed condition and a tidal zone
condition, as shown in Table 9.7 [436]. Water flow significantly increased the corrosion
rate; at a flow velocity of 2 ftls the corrosion rate was more than double that in stagnant
water.
Khan et al. [662] reported that the corrosion rate at half-tide level is five times more
than that in a fully immersed condition. Anderson [436] also reported a higher corrosion
rate at mean tide level than in a fully immersed condition. The splash and tidal zones are
usually much more corrosive than the submerged zone since they are continuously wetted
with well-aerated seawater [660]. The differences between the data of Suzuki and those
of others may be due to the fact that Suzuki used sprayed zinc, which is very porous.
However, according to Suzuki [562]. the corrosion rate of a sprayed zinc coating, after a
two-year exposure, was highest in the submerged zone, being about three times more than
that in the splash zone, as shown in Fig. 9.9.
The depth of immersion in the sea also affects the corrosion. Reinhart [575] reported
that the corrosion rate of zinc in seawater decreased with depth from 113 f1rn1yr near the
surface to 58 f1rn1yr at about 2000 ft but increased to 168 f1rn1yr at about 5500 ft. It is
noted that the change of the corrosion rate with depth appears to correspond to the change
of oxygen concentration with depth as shown in Fig. 9.1.
9.4.3.4. Effect of Temperature. Mor and Beccaria [179] reported that the corrosion
rate in synthetic seawater decreased with increasing temperature from 25 to 60C, as
shown in Fig. 9.10. The decrease of corrosion rate with time was found to be associated
with the accumulation of corrosion products on the zinc surface. With increasing temperature, the amount of zinc oxide and hydroxide decreased while calcium carbonate
50,----------------------------------------,

40
Zinc spray-coated
steel

~ 30
enen

.Q

c
020

'iii
~

(;

10

OL---------~--------~------

Submerged zone

Tidal zone

___ L_ _ _ _ _ _ _ _~

Splash zone

FIGURE 9.9. Corrosion rates of zinc coating under various seawater conditions (2 years' exposure at Akashi,
Japan). Data are taken from Ref. 562.

295

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

3r----------------------------------------,
. 25C
... 40C

400
Time, hours

FIGURE 9.10. Weight loss of zinc in synthetic seawater at pH 8.2 at different temperatures. From Mor and
Beccaria 1179]. Copyright by NACE International. All Rights Reserved by NACE: reprinted with permission.

increased in quantity. Other corrosion products were formed whose presence was not
detected at room temperature, such as basic sulfates, ZnS043Zn(OH)2AH20. By precipitating together with CaC0 3 and MgS0 4 , this basic sulfate possibly makes the barrier layer
more compact and thus inhibits the subsequent anodic dissolution of the zinc.
9.4.3.5. Effect of Bacterial Activity. The effect of sulfide, a common by-product of
bacterial activities, on the corrosion of zinc was investigated by Mor et al. [209]. Figure
9.11 illustrates that the corrosion of zinc in synthetic seawater is accelerated in the
presence of sulfide at pH values greater than 7.2, whereas at lower pH values the corrosion
process is partially inhibited. Mor et al. attributed this finding to the different concentra-

Aerated , no S2

0.8

... Aerated , with S 2-

De-aerated, with S 2 XDe-aerated, no S 2

~0 . 6

...

<Ii

'"o

E 0.4
OJ

'w
~

0 .2

O~------------~------------~--------------~

pH
FIGURE 9.11. Weight loss of zinc specimens after immersion for 48 hours in aerated and deaerated artificial
seawater at different pH values, with and without addition of sulfide. After Mor et al. [2091.

296

CHAPTER 9

tions of S2- ions in the two pH regions. At pH values above 7.2, the S2- ions (at a
concentration of 1O- '2M) cause the corrosion products to be predominately ZnS, which
reduces the adhesion of corrosion products and, therefore, enhances the corrosion.
9.5. CORROSION IN AQUEOUS SOLUTIONS

9.5.1. Effect of Dissolved Species


Solutions differ from natural waters in that a solution contains solutes, usually
artificially introduced. A solution is formed when chemical compounds are dissolved in
water. As a rough classification, solutions containing more than 0.1 mol per liter of solute
can be considered to be concentrated while those containing less than 0.01 mol per liter
can be regarded as dilute. As seen in the previous sections, the presence, even in trace
amounts, of chemical compounds dissolved in water can greatly change the corrosion
behavior of zinc. The corrosion behavior of zinc can vary drastically, depending on the
solution composition and test conditions, from virtually noncorroding to rapidly dissolving.
The major factors affecting the corrosion form and the rate of zinc dissolution are
type of dissolved species, concentration, pH, and temperature. The presence of various
chemical species can change the solubility of zinc dissolution products by forming
complexes with zinc ions, increasing the electrolyte conductivity, modifying the composition, structure, and compactness of the corrosion products, forming an insoluble salt
film on the surface, providing reactants for the anodic and cathodic reactions, and
changing the reaction kinetics through catalytic or inhibitive adsorption.
The corrosion processes of zinc in a solution are greatly influenced by the nature of
the anions present. Depending on the specific effect, anions may be classified into three
groups: (a) anions that increase the solubility of zinc, such as chloride or sulfate; (b) anions
that reduce the solubility of zinc and thus promote the precipitation of zinc salts which
may be protective, such as carbonate or phosphate; and (c) anions that react with the zinc
surface and, depending on the reaction products, may form a passive film, such as
chromate.
Table 9 .8 lists the corrosion rates of zinc in some common solutions. The particularly
low values in phosphate and chromate solutions are due to the formation of passive films
on the zinc surface. It appears that in neutral solutions, with chemical agents that are
neither electrochemically reactive nor capable of forming insoluble salts or complex ions
with zinc, the corrosion rate of zinc is not very different from that in distilled water.
Figure 9.12 shows the effect of concentration of several salts on the corrosion rate
of zinc [559]. The corrosion rate of zinc in chloride and sulfate solutions increases with
concentration up to about 5 gil and then decreases with further increase in concentration.
At salt concentrations higher than 150 gil, the corrosion rate is actually less than half of
that in pure water. In nitrate solutions, the corrosion rate decreases with increasing
concentration. Nitrate is known to promote the passivation of zinc in neutral solutions
[409,410]. However, as reported by Goodrich and Schmid [223], the presence of nitrate
in ammonium-containing solutions enhances the corrosion of zinc because the formation
of a zinc-ammonia complex in the solution reduces the pH near the surface and prevents
the formation of a passive film. In LiBr solution, the zinc corrosion rate increases with

297

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

TABLE 9.8. Corrosion Rates in Different Solutions in the Neutral pH Range


Solution

Duration of exposure

Distilled water
Distilled water
Distilled water
0.INNa2S04
O. I N benzoate
O.INNaCI
O.IN Na3P04
O.IN chromate
NaCI.5 gil
KCI, 5 gil
NaN0 3, 5gIl
Na2S04' 5 gil
K2S04,5 gIl
3.5% NaCl
8% Na2S04,IOH20
5% NaCI
3% NaCI
3% Na2S04
--.------

------

Corrosion rate (um/yr)

4 weeks
I month
3 weeks
4 weeks
4 weeks
4 weeks
4 weeks
4 weeks
2 months
2 months
6 months
2 months
2 months
I month
I month
3 weeks
I day
I day

Reference

- -_. , - - - -

46
55
48
70
59
62
1.8
0.4
90
92
18
65
52
88
83
89
175
144

.... - ~---.

7\0
559
722
710
710
7\0
710
7\0
559
559
559
559
559
208
208
722
97
97

solution concentration, peaking at about 3M, and then decreases with further increasing
concentration [1262].
Lorking [710] found that, although the formation of corrosion products is related to
the solubility of zinc hydroxides and zinc salts, there is not necessarily a definite
relationship between corrosion rate and solubility. He found that the corrosion rate in
sulfate solutions was related to the amount of dissolved zinc in the solution. The corrosion
rate of zinc specimens in zinc oxide-saturated sulfate solution was 2 pm/yr. which is less

100
...... NaCI

>:

60

+KCI

-6: KN0 3

,3.
Ql

<ii

. . Na,SO. 10H.0

60

.I<,SO.

II:

c
0

'iii

e(;

()

40
20
0

50

100

150

200

250

300

Concentration (g/I)
FIGURE 9.12. Effect of salt concentration on corrosion rate of zinc (T = 8-13C, 57-189 days, nonaerated).
Data are taken from Ref. 559,

298

CHAPTER 9

than one-tenth of that without zinc oxide in the solution. On the other hand, there was not
much difference between the corrosion rates in zinc oxide-saturated and zinc oxide-free
benzoate solutions.
Leidheiser and Suzuki [97] reported that the presence of small amounts of Co 2+, up
to 1O-4M, in 3.5% NaCI solution slightly inhibited the corrosion of galvanized steel, but
a significant increase in the corrosion rate was found at higher concentrations of C0 2+.
The effect of cobalt was postulated to be due to its incorporation into the zinc oxide, which
reduces the flux of electrons for the cathodic reactions. At high cobalt concentrations, in
or on the surface of the oxide, elemental cobalt aggregates to form metallic cobalt, which
serves as a catalyst for the cathodic reaction.
Some chemical agents can inhibit the formation of corrosion products on zinc
surfaces. Boto and Williams [128] reported that the addition of a complexing agent
(EDTA) prevented the formation of corrosion product in neutral or basic solutions and
that the corrosion rate was limited only by the diffusion of oxygen. On the other hand,
the presence of some inorganic or organic species may promote the formation of a surface
film, which then acts as a barrier to corrosion. Besides chromates and phosphates, certain
other inorganic agents (such as carbonates [194,331], molybdates [367, 102], tungstates
[199,420], silicates [574,599], and cerium salts [605]) and certain organic species (such
as gluconate [100], esters [601], phosphines [64], and benzene thiols [164]) are found to
inhibit the corrosion of zinc.

9.5.2. Effect a/pH

In the absence of reducing or passivating agents, the corrosion of zinc in aqueous


solutions is primarily determined by the pH of the solutions. The results in Fig. 9.13,
reported by Roetheli et at. [497], show that the corrosion rate of zinc in water of pH 6-12
is relatively low. At pH values lower than 6 or higher than 12, the corrosion rate increases
substantially. The low corrosion rate at pH values between 6 and 12 is primarily due to
the formation of passive corrosion products on the surface of the zinc. According to
Roetheli et at. [497], the decrease in the corrosion rate at pH values near 14 shown in the
figure is due to a decrease in the solubility of oxygen in strongly alkaline solutions.

(ij
Q)

0.5

>-

c::
0

0.4

a; 0.3
c:

OJ
Cl.

c;;

Qj

0 .2

is
Q)

0>

'"

0.1

Qj

pH

12

16

FIGURE 9.13. Corrosion rate in distilled water as a function of pH (addition of N aOH or HCI for pH
adjustment). After Roetheli et al.
[497].

299

CORROSION IN WATERS AND AQUEOUS SOLUTIONS


1 ,000
-Na 3 P04

+ Na 2SO.
100

'6' Benzoate

"E
<.J

c;

10

(II
(II

.2
1:

OJ

'OJ
~

0 .1

FIGURE9.14. Effect of pH on weight


loss of zinc in O. IN solutions. Data are
taken from Ref. 710.

0.01

10

12

14

pH

The form of corrosion is related directly to the range and s-tability of passivity, which
is a function of pH. Below pH 6 or above pH 12, the corrosion is normally of a general
type, while in the pH range between 6 and 12 the corrosion tends to be more localized.
Figure 9.14 shows the corrosion rate measured by Lorking [710] for zinc in various
solutions as a function of pH. The pH dependence of the corrosion rates of zinc in sulfate,
chloride, and benzoate solutions is similar to that in water, i.e., relatively low in nearneutral or slightly alkaline solutions and high in acidic or strongly alkaline solutions. In
phosphate solutions, the corrosion is inhibited between pH 4 and 12, owing to the
formation of a less soluble and more protective zinc phosphate film. In chromate
solutions, the corrosion of zinc is inhibited almost in the entire pH range of 1-13 owing
to the formation of a passive chromate-incorporated chromium oxide film.
As shown in Fig. 2.3 in Chapter 2, pH determines the solubility of zinc in water, the
solubility being lowest at pH 9. Over pH range 0-4, zinc enters solution as a divalent
cation. From pH 4 to 12, zinc corrodes with precipitation of zinc hydroxides, due to their
relatively low solubility. At pH values greater than 12, zinc enters solution as the zincate
ion [ 1, 710]. It may be noted, by comparing Figs. 9.13 and 2.3 that the pH value at which
the corrosion rate of zinc is the lowest does not coincide with the pH value for the lowest
solubility. The lowest solubility for zinc hydroxide in water is near pH 9 [1], while the
lowest corrosion rate is around pH 12. This indicates that solubility of corrosion products
is not the only factor to determine the corrosion rate. If corrosion rates were determined
primarily by the solubility of corrosion products, the lowest corrosion rate should be
observed at a pH value around 9. pH not only dictates the solubility and stability of the
corrosion products but also determines the physical structure.
Besides the effect on the formation of corrosion products, the pH of a solution affects
the corrosion of zinc by affecting the cathodic reaction. It has been established that in
aerated solutions of near-neutral or higher pH values, hydrogen reduction is not the main
cathodic reaction in the corrosion process, owing to the relatively high overpotential for
hydrogen reduction on a zinc surface. The main reducing agent for the cathodic reaction
of the corrosion process in aerated solution is oxygen [116, 128]. As the hydrogen ion

FIGURE 9.15. Distribution of anodic and cathodic areas on an electrode surface at the waterline. Data are taken from Ref. 403.

'0

:>:l

tTl

"1:)
....,

n
::c

'"oo

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

301

concentration increases, hydrogen evolution becomes the more important cathodic reaction.

9.5.3. Ejject of Immersion Conditions


It has been known that when a zinc sheet is partially immersed vertically in a chloride
solution, attack is soon observed at numerous points, distributed sporadically, and spreads
out from some of these points, producing corroding areas a short way below the waterline.
A corrosion-immune zone is formed along the waterline with the attacked zone below it.
As reported by Thornhill and Evans [403], the cause of waterline immunity is generally
attributed to the fact that alkali, a cathodic product, is preferentially found in the zone
where oxygen, the cathodic stimulator, can be renewed most readily. The zinc salts formed
by anodic action in the lower parts must travel a considerable distance before precipitating
as hydroxides, whereas any zinc salts formed momentarily in the upper parts are
precipitated by the excess alkali in physical contact with the metal, thus stifling any further
anodic attack. As a result, interference colors due to the hydroxide film are visible in this
regIOn.
The electrochemical nature of this reaction process of zinc in 1O-3M NaCl solution
can be seen in Fig. 9.15 [403]. The area close to the head of the meniscus is the cathode,
while the area around the foot of the meniscus is the anode. This kind of distribution is
determined mainly by differential aeration since oxygen can easily reach the head of the
meniscus without passing through a thick layer of liquid [403, 708]. From Fig. 9.15, it is
seen that at the outset, most anodic attack is at the foot of the meniscus; later, it will be
directed mainly on any physically loose areas in the lower part of the specimen; finally,
when the supply of unstable matter on these areas has been exhausted, it returns to the
foot of the meniscus. The distance between the attack and the cathode is determined by
the pH and conductivity of the solution. High conductivities favor attack at a distance
from the cathode.
Bianchi [708] pointed out that formation of macro cells of differential aeration is
connected with passivation of zinc in the more aerated zones. Differential-aeration
corrosion cannot exist without passivation of the more aerated zones of the zinc surface.
If buffered solutions are used, so that the pH is outside the range for passivation of zinc,
the formation of differential-aeration macrocells may not occur.
Conditions of aeration and sample rotation have an important influence on the
corrosion rate of zinc. Table 9.9 shows the effect of rotation and aeration on the corrosion

TABLE 9.9. Effect of Aeration and Rotation on the Corrosion Rate (j1m/yr) of Zinc in Distilled
Water and 3.5% NaCI Solution"
Distilled water

---------------------------- - - - - 3.5% NaCI

------

Type of test

15 days

30 days

15 days

30 days

128
249
193
194

124
165
128
122

104
362
127
194

133
242
118
226

--------~-

No aeration; no rotation
No aeration; rotation
Aeration; no rotation
Aeration; rotation

"From Anderson and Reinhard [559J. Reprinted by permission of John Wiley & Sons, Inc.

302

CHAPTER 9

rate. As found by many authors [112. 113. 116, 128,445], the corrosion of a bare zinc
surface in neutral nondeaerated solutions is oxygen-diffusion-controlled. while that of a
corrosion-product-covered surface may be independent of oxygen concentration in the
solution.
9.5.4. Effect of Surface Treatments

Surface treatments can have a significant effect on the corrosion of zinc, especially
in the early stages of a corrosion test. Among the many surface treatment processes,
chromating is the oldest and the most effective one in increasing the corrosion resistance
of zinc in solutions [57, 65]. Chromate conversion coating on galvanized steel that is
immersed in distilled water can delay the appearance of white rust for many weeks [591].
Dipping in phosphate, molybdate, tungstate, and carbonate solutions has also been found
to be beneficial in increasing the corrosion resistance of zinc [93, 94, 101. 102, 178,404,
420]. Anodization can also produce surface films that inhibit the corrosion of zinc in water
[493,596].
9.5.5. Effect of Metallurgical Factors
9.5.5.1. Crystal Orientation. Using single-crystal zinc, Ashton and Hepworth [222]
measured the corrosion currents of different crystal planes in 0.5M NaOH. They found
that the corrosion rate decreased in the order (1120) > (1020) > (0001), which is the same
as the order of decreasing planar packing density. Abayarathna et al. [446,721] found the
order to be the same in terms of the corrosion rate of zinc single crystals in three different
orientations. The oxide or hydrated oxide film which formed on the (0001) surface
appeared to be the most protective. They also found that particular planes of a zinc single
crystal in near-neutral 1M (NH4)S04 solution were corroded preferentially, so that the
corrosion on the basal plane resulted in hexagonal pits with facets parallels to the (10TO)
planes. On the (10TO) surfaces, elongated and striated structures were observed, with the

3 r--------------------------------------.
As roll ed
- 3% per pass

en

oi
~

2 .8 . 5%

2.6

.
2 .4
en

e
o

u 2 .2

~------_L

20

______

~~

40

______

_ L _ _ _ _ _ _~ _ _ _ _~

60

80

Total reduction , %

FIGURE 9.16. Corrosion rate in 7.5% HCI solution as a function of total reduction of zinc sheet immediately
after cold rolling. After Kobayashi et al. [411].

CORROSION IN WATERS AND AQUEOUS SOLUTIONS

303

3 .---------------------------------~

5% per pass

(/) 2.8

.E
b

oi

+ As rolled
~ ,

Month

2.6

li!

.~ 2.4
(/)

O 2 .2

25L-------1JO--------1L5--------2~
O -------2~5~----~
30
Residual stress, MPa

FIGURE 9. 17. Relation between corrosion rate and the original residual stress for cold-rolled zinc sheets. After
Kobayashi et al. [411].

elongation aligned along the (1010) direction. Corrosion on the (1120) surfaces resulted
in a crested ridge structure. Also, regardless of the initial surface orientation, the morphology of the corroded surface was consistent with the (1120) plane being the most
active surface.
Leidheiser and Suzuki [97] found that for different galvanized steels with the amount
of (1000) grain orientation varying from 50% to 98%, the corrosion rates in 3% NaCI
solution were approximately the same. On the other hand, the corrosion resistance of
electrolytic zinc coatings, reported by Girin and Panasenko [459], significantly increased
with a reduction in the average scattering angle for the axial texture with the (1120) axis
and with an increase in the percentage of crystals with a disordered orientation.
9.5.5.2. Deformation. Fedrizzi and Bonora [715] measured the polarization resistances of zinc coatings subjected to different extents of deformation. The polarization
resistance of deformed samples in 3.5% NaCI solution was slightly lower than that of the
undeformed ones. This was attributed to the increased surface area caused by the coating
cracks created by the deformation. Kobayashi et al. [411] studied the effect of cold rolling
on the corrosion rate of zinc in 7% HCI solution. The corrosion rate exhibited a maximum
at about 30% reduction for the samples immediately after rolling (Fig. 9.16) but was
independent of the amount of rolling when the samples were held at room temperature
for one month after rolling. The difference III the corrosion rates was found to be related
to residual stress, which decreases with time after rolling, as shown in Fig. 9.17.

10
Corrosion in Soil
10.1. INTRODUCTION
As a corrosion environment, soil differs from other natural environments, such as
atmospheric environments or waters, in two major aspects. Firstly, soil has a much wider
range of chemical and physical properties. For example, the pH of soil may vary from as
low as 2.6 to as high as 10.2, and the resistivity from several tens of ohms to near 100 k.Q
[357]. As a result, the corrosion rate of a metal may drastically vary from soil to soil.
Secondly, soil is a highly inhomogeneous environment, both microscopically (e.g., on the
scale of a clay particle) and macroscopically (e.g., on the scale of a rock). Thus, the
corrosion in soil is seldom uniform across the metal surface.
Galvanized steel is commonly used for structures in soil such as lamp posts, culverts,
and land enforcements. Zinc anodes are also frequently used in soil for cathodic protection
of underground structures such as pipelines and storage tanks.
There have been few systematic studies of the corrosion of zinc and its alloys in soils.
The most extensive investigations were carried out by the U.S. National Bureau of
Standards (NBS) between 1910 and 1955 [357,565]. In these investigations, the corrosion
of rolled zinc, cast zinc, and galvanized steels, with different coating thicknesses, was
evaluated in more than 50 different soils at various locations across the United States. The
data presented in this chapter are abstracted mainly from the reports of these investigations.
10.2. CHARACTERISTICS OF SOIL
Soil makes up the first few feet of finely divided, weathered (by climatic and
biological processes) rock material covering the level and moderately inclined portions
of the earth [357]. It is usually named and classified according to the size range of its
particulate matter: the principal categories are sand, silt, and clay, which derive their
names from the predominant size range of the inorganic constituents. Particles between
0.07 and about 2 mm in mean diameter are classified as sands. Silt particles range from
0.005 to 0.07 mm, and clay particles range from 0.005 mm down to colloidal matter. The
relative proportions of the three types determine many properties of the soil. A common
classification scheme is shown in Fig. 10.1 for various proportions of sand, silt, and clay.
305

306

CHAPTER 10

40
PERCENT SAND

20

FIGURE 10.1. Proportions of sand, silt, and clay making up the various groups of soils classified on the basis
of particle size. After Romanoff [357].

More detailed information on the classification and the chemical and physical properties
of soils related to corrosion can be found in several references [314, 357, 565, 987].
Besides the differences in texture, soils differ in their chemical composition and their
interaction with other environmental factors. A large number of chemical compounds
exist in soils. The inert compounds are chiefly those of silicon. aluminum, and iron. Iron,
in its various oxidation states, is responsible for the color of many soils. Furthermore. this
color is an indicator of the degree of aeration of the soil; red, yellow. and brown colors
indicate the oxidized forms of iron. Poorly aerated soils are predominantly gray. indicating the presence of reduced forms of iron.
The soluble base-forming elements are sodium, potassium, calcium. and magnesium.
The acid-forming species are carbonate. bicarbonate, chloride, nitrate, and sulfate. The
nature and amount of soluble salts, together with the moisture content of the soil, generally
determine the ability of the soil to conduct an electric current, which is very important in
relation to corrosion processes. Another important factor influencing the corrosion of a
metal is soil acidity. The development of acidity in soils is a result of the natural processes
of weathering under humid conditions. In regions of plentiful rainfall, not only are soluble
salts removed from soil, but the base-forming constituents normally present in the
colloidal materials are partially removed. resulting in increased acidity [565].
Bacteria also affect the chemical properties of soils through oxidation and reduction
reactions. Bacterial activities tend to decrease the oxygen content and replace oxygen
with carbon dioxide. Most bacterial activities occur in the upper six inches of soil [1280].

307

CORROSION IN SOIL

50,-----------------------------------------~

40
)(

30

ai
u
c

~ 20
<fl

'r;;

ill

0:

10

O~----~------~======~====~
-20
-10
o
10
20
Temperature, C

FIGURE 10.2. Effect of temperature on soil resistance. Data are taken from Ref. 1105.

The physical properties of soils that are important in relation to corrosion are mainly
those that determine its permeability for air and water. The particle-size distribution in a
soil is an important factor with respect to air and moisture content. Soils of fine texture,
due to a high clay content, contain more closely packed particles and have less pore
capacity for gaseous diffusion. It is generally assumed that the composition of the gases
in the upper layers of a soil are similar to that in the atmosphere above the soil. Soil
moisture may derive from free groundwater, which is constantly present at certain depths
below the surface, gravitational water, which enters the soil from the surface through
rainfall, and capillary water, which is the usual form of water retained in a soil owing to
capillary action.

1,000

a
)(

E
u

100

'iIi

10

ill

a:

1L---------~--

25

________

________

45

____

65

Moisture content of soil, %

FIGURE 10.3. Effect of moisture content on resistivity of a clay soil. Data are taken from Ref. 1105.

308

CHAPTER 10

Various types of parameters have been used to evaluate the corrosion of metals in
soils [987]. It is generally accepted that aeration, moisture content, pH, and resistivity are
the main factors affecting corrosion, although there is still a lack of good correlation
between the corrosion of a metal and these factors. Among them, resistivity, which is
directly related to the amount of soluble salts, and moisture content seem to have the best
correlation with the corrosion rate. The resistivity of a soil is typically a function of
temperature and moisture, as shown in Figs. 10.2 and 10.3 [1105].
10.3. CORROSION RATES
Tables 1O.l and 10.2 contain the results obtained in the NBS investigation for the
corrosion rates of zinc and galvanized steel in various soils. The average annual corrosion
rates of rolled zinc and zinc coatings in most soils are below 10 11m/yr. The maximum
penetration rates, shown in Table 10.2, are 3-30 times the surface-averaged corrosion
rates.
Table 10.2 also shows the beneficial effect of zinc coatings on pitting corrosion of
the steel. The area-averaged corrosion rates of galvanized steel are 3-6 times lower than
those of bare steel whereas the pitting corrosion rates of galvanized steel are 4-20 times
lower than those of bare steel.
As also shown in Table 10.2, the average corrosion rates of galvanized steel are
similar to those of zinc, but the pitting penetration rates of galvanized steel are noticeably
lower than those of zinc. This indicates either the effect of galvanic protection by the zinc
coating or the higher corrosion resistance of the Zn-Fe alloy layers at the coating/steel
interface.
It was found in the NBS study [357] that when most of the outer coating corrodes
away, the penetration rate slows down, indicating a higher corrosion resistance of the
intermetallic layers.

10.3.1. Effect of Soil Factors


The factors which may affect the corrosion of zinc and galvanized steel in soils are
numerous, but the correlation between the corrosion behavior and the various factors is
in general rather poor. The NBS results appear to suggest that the corrosion rates tend to
be lower in soils with high resistivities, as shown in Fig. 10.4. Insofar as pH is concerned,
there is little correlation between corrosion rate and soil pH, as shown in Fig. 10.5. (It
should be noted that the corrosion rate of zinc in aqueous solution has a clear dependence
on pH as shown in Fig. 9.13 in Chapter 9.) This lack of correlation between corrosion
rate and soil pH and the very weak correlation between corrosion rate and soil resistance
reflects the very complex nature of corrosion in soil.
The various corrosion rates in different soils are essentially determined by the
synergistic effect of soil resistivity, pH, moisture content, aeration, soluble salt concentrations, and other factors [987, 1085, 1086, 1088]. In general, as stated by Romanoff
[357], poorly and very poorly aerated soils are more corrosive to zinc, and a high corrosion
rate is not always associated with deep pitting. Soils of fair to good aeration, but containing
high concentrations of chlorides and sulfates, tend to induce deep pitting. The extreme
corrosion rates and deep pitting seem to be associated with soils having very low pH

309

CORROSION IN SOIL

TABLE 10.1. Average Corrosion Rates, R, of Galvanized Steel Pipe Specimens" Embedded in
Various Soils for 10 Yearsh
No.
I
2
3
4
5
6
7
8
9
10
II
12
13

14
15
16
17
19
20
22
23
24
25
26
27
28
29
30
31
32
33
35
36
37
38
40
41
42
43
44
45
46
47

Soil type
Allis silt loam, Cleveland, Ohio
Bell clay, Dallas, Tex.
Cecil clay loam, Atlanta, Ga.
Chester loam, Jenkintown, Pa.
Duhlin clay adobe, Oakland, Calif.
Everett gravelly sandy loam, Seattle, Wash.
Maddox silt loam, Cincinnati. Ohio
Fargo clay loam, Fargo, N. Dak.
Genesee silt loam, Sidney, Ohio
Gloucester sandy loam, Middleboro, Mass.
Hagerstown loam, Loch Raven, Md.
Hanford fine sandy loam, Los Angeles, Calif.
Hanford very fine sandy loam, Bakersfield,
Calif.
Hempstead silt loam, St. Paul, Minn.
Houston black clay, San Antonio, Tex.
Kalmia fine sandy loam, Mobile, Ala.
Keyport loam, Alexandria, Va.
Lindley silt loam, Des Moines, Iowa
Mahoning silt loam, Cleveland, Ohio
Memphis silt loam, Memphis, Tenn.
Merced silt loam, Buttonwillow, Calif.
Merrimac gravelly sandy loam, Norwood,
Mass.
Miami clay loam, Milwaukee, Wis.
Miami silt loam, Springfield, Ohio
Miller clay, Bunkie, La.
Montezuma clay adobe, San Diego, Calif.
Muck, New Orleans, La.
Muscatine silt loam, Davenport, Iowa
Norfolk fine sand, Jacksonville, Fla.
Ontario loam, Rochester, N.Y.
Peat, Milwaukee, Wis.
Ramona loam, Los Angeles, Calif.
Ruston sandy loam, Meridian, Miss.
St. John's fine sand, Jacksonville, Fla.
Sassafras gravelly sandy loam, Camden, N.Y.
Sharkey clay, New Orleans, La.
Summit silt loam, Kansas City, Mo.
Susquehanna clay, Meridian, Miss.
Tidal marsh, Elizabeth, N.J.
Wabash silt loam, Omaha, Nebr.
Unidentified alkali soil, Casper, Wyo.
Unidentified sandy loam, Denver, Colo.
Unidentified silt loam, Salt Lake City, Utah

p (Q'cm)

--------

pH

R (jlm/yr)

--------

1,215
684
30,000
6,670
1,345
45,100
2,120
350
2,820
7,460
11,000
3,190
290

7.0
7.3
5.2
5.6
7.0
5.9
4.4
7.6
6.8
6.6
5.3
7.1
9.5

lUI
1.5
1.7
7.9
7.7
0.5
10.R
3.2
5.0
5.2
3.7

3,520
489
8.290
5,980
1,970
2,870
5,150
278
11,400

6.2
7.5
4.4
4.5
4.6
7.5
4.9
9.4
4.5

1.1
1.5
4.2
14.8"
2.9
4.9
5.2
40.6"
1.1

1,780
2,980
570
408
1,270
1,300
20,500
5,700
800
2,060
11,200
11,200
38,600
970
1,320
13,700
60
1,000
263
1,500
1,770

7.2
7.3
6.6
6.8
4.2
7.0
4.7
7.3
6.8
7.3
4.5
3.8
4.5
6.0
5.5
4.7
3.1
5.8
7.4
7.0
7.6

1.45
2.9
3.9
8.8
25.5"
1.9
0.67
2.4
7.4
1.3
1.0
8.7
0.85
4.0
2.2
3.0
5.5
1.9"
7.5
0.7
4.3

2.2 d

3.7

------

"Average coating thickness, 121 JIm.


"Ref. 357.
'Original identification.
dSheet specimens.
"Coating corroded completely: data included the corrosion of steel.

CHAPTER 10

310

TABLE 10.2. Average Annual Surface-Averaged Corrosion Rates (CR) and Maximum
Penetration Rates (MPR) of Zinc and Galvanized Steel Embedded in Soils for 9 Years"
Rolled zinc
No.

51
53
55
56
58
59
60
61
62
63
64
65
66
67
70

Soil type
Acadia clay
Cecil clay loam
Hagerstown loam
Lake Charles clay
Muck
Carlisle muck
Rifle peat
Sharkey clay
Susquehanna clay
Tidal marsh
Docas clay
Chino silt loam
Mohave fine
gravelly loam
Cinders
Merced silt loam

CR

(Qcm)

MPR

Galvanized stee(
CR

MPR

Steel
CR

MPR

(pmlyr) {flm/yr) (pmlyr) (pm/yr) (pmlyr) (pm/yr)

pH

190
17,790
5,210
406
712
1,660
218
943
6,920
84
62
148
232

6.2
4.8
5.8
7.1
4.8
5.6
2.6
6.8
4.5
6.9
7.5
8.0
8.0

22.9
5.3
3.3
21.5
35.3
21.9
112d

22.9
3.8
3.9
26.3
43
14.1
77.4"
7.2 d
4.3
9.6
7.6
7.6
11.9

22.5
16.9
16.9
36.7
180.6
22.2
234+
34
16.9
22.6
28.2
16.9
16.9

83
16.2
19.6
132
82.7
35.8 d

5.2
6.2
9.6
6.7
6.7
28 d

79
36.6
22.5
81.8
163
61.1
416+
39.5
33.9
69.4
223
158
124

455
278

7.6
9.4

131d
17.2

749+
237

58
9.9'

286
18.5

151
64

85.5
20.1
25.3
51.1
43"
33.4
58.8 d

361+
209
260
409+
277
56.4
164
135
192
226
226
183
409+
409+
344

"Ref 357.
"Identification in the original report.
'Nominal coating thickness. 130 pm.
dData at 4 years.
'Data at 11.2 years.

30,-----------------------------------~

~
~

c
o
'iii
~10
(;

",

O~--------L---------~~------~-----~--~

10

100

1,000

10,000

100,000

Resistivity, ohm-em
FIGURE 10.4. Average annual corrosion rates of galvanized steel embedded in 53 different soils across the
United States for 10 years, as a function of soil resistivity. Data are taken from Ref. 357.

CORROSION IN SOIL

311
30~------------------------------'

2i

E
c
o

.iii

::10

FIGURE 10.5. Average annual corrosion rates of galvanized steel embedded in 53 different soils across
the United States for 10 years, as a
function of soil pH. Data are taken
from Ref. 357.

"

;' ,: .'

OL---~--~'~'~----~----~----~----~--~.

10

pH

values. Muddy clay and peat (as compared to sand) are, in general, more corrosive to zinc
[ 1085].
Camitz and Vinka [1085] investigated the corrosion rate of galvanized steel samples
in seven different soils and found that the position of galvanized steel samples with respect
to the groundwater level had no distinct effect on the corrosion rate, as shown in Fig. 10.6.
The level of the water table was found, however, to affect the corrosion rate of the steel
samples. Camitz and Vinka also found that the corrosion rate was lower on panels
embedded in a homogeneous sandfill than on panels placed directly in the original soil,
which was attributed to the formation of a more protective surface film, due to better
aeration in the sand.
60

o above G W table
o above G W table, in sand fill

'-

Q)

'">E 40

below GW table
below G W table, in sand fill

:J

<Ii

'
c

g20
o

clay 1

ctay 2

S clay

Mclay 1 Mclay 2

B ~

peal

sand

Soils

FIGURE 10.6. Corrosion rates on zinc-coated panels after about three years of exposure above and below the
groundwater (GW) table in soil (Data for the Kramfors test site are for only one year of exposure.) AfterCamitz
and Vinka [1085] .

CHAPTER 10

312

Different corrosion rates may be found on different parts of a single structure owing
to a highly inhomogeneous soil environment. Edgar r1088] investigated the corrosion of
culverts that were put into service some 30 to 40 years ago and reported that the corrosion
rate for most pipes was below 5 Jim/yr. The corrosion was most severe in the center of
the culvert, which was usually the joint between two sections. He reported that the worst
corrosion typically occurred at the invert, or the bottom, of the pipe. Edgar also pointed
out that the use of poor backfill material was the most obvious defective practice causing
poor corrosion performance of galvanized culverts.

10.3.2. Galvanic Corrosion


Galvanic corrosion of zinc in soil occurs when the zinc coating on galvanized steel
is partially removed, exposing the steel underneath, or when it is used as an anode
connected to a steel structure for cathodic protection of the steel. In contrast to selfcorrosion, which seems to lack a clear correlation with the various factors in soil, galvanic
corrosion has a definite dependence on the resistivity of the soil. This is because, for a
given potential difference between the anode and cathode, the amount of galvanic current
flowing between the coupled metals is directly proportional to the resistivity of the soil.
The galvanic corrosion of zinc coupled to various metals in different mediums is discussed
in detail in Chapter 7 (Section 7.2).
10.4. ELECTROCHEMICAL MEASUREMENTS
The electrode potential of zinc in a soil depends on the chemical composition of the
soil. In Table 10.3, the electrode potentials reported by Romanoff r357] for zinc, steel,
and a zinc-iron alloy layer of galvanized steel in 12 different soils are tabulated. The
specimen of the zinc-iron alloy with surface condition A was prepared by electrolytic
removal of the outer zinc coating, and the specimen with surface condition B was achieved

TABLE 10.3. Potentials (V seE) of Zinc-Iron Alloy, Zinc, and Steel in "Air-Free" Soils
Zinc-iron alloy, surface A.
outer zinc coating removed
electrolytically
Soil

Initial

Steady

51
55
56
58
60
61
62
63
64
65
66
70

-0,96
-0,86
-0,91
-0,87
-0,96
-0.92
-0.90
-0.92
-0.97
-0.95
-0.80
-0.95

-0,62
-0,60
-0,61
-0,63
-0,64
-0.66
-0.54
-0.65
-0.61
-0.64
-0.61
-0.60

Zinc-iron alloy, surface B,


outer zinc coating removed
by corrosion
Initial

Steady

-0.96

-0.103

-0.90
-0.99
-0.85
-0.98

-0.75
-0.96
-0.84
-0.88

Zinc

Steel

-1.02
-1.02
-1.04
-1.04
-1.02
-1.02
-0.92
-0.94
-1.08
-1.01
-0.94
-0.99

-0,71
-0,75
-0,73
-0,74
-0.68
-0.72
-0.72
-0.64
-0.73
-0.71
-0.72
-0.76

313

CORROSION IN SOIL

-0.4.-------------------.-------------------,
II

Coltosion in 1he lield

... Electrolytic stri pping

-0.6
w

>'"
~ -0.8

c:<l>
0

a.
-1

_1.2L---~----~----L----L----~----L---------~

-20

-10

- 15

-5

10

15

20

Current densi ty, JiA I em'

FIGURE 10.7. Anodic and cathodic polarization curves of zinc-iron alloy exposed by corrosion in the field
(surface B) and electrolytic stripping in the laboratory in an "air-free" environment (surface A). After Romanoff
[357)

by using the section of the galvanized steel that had its outer zinc coating corroded by 13
years of exposure to soiL The measurements were made in water-saturated soils from the
test sites. Air was excluded, to prevent the soil from drying in the laboratory, by sealing
the test celL The potential of zinc in the 12 soils was found to be between -0.92 and -1.08
VseE' The potentials of the zinc-iron alloy are generally more positive than those of zinc,
reflecting the effect of the presence of Fe in the alloy. The difference between the
steady-state potentials for the two surface conditions of the alloy was attributed by
Romanoff to the presence of a cathodic film formed on surface B during field exposure.
Figure 10.7 shows that, because of the presence of this surface cathodic film, the
polarization resistance is much larger for surface B than surface A. This figure also
explains the higher steady potential values in Table 10.3 for surface A; because of the
relatively higher reactivity, the zinc-iron alloy layer is soon corroded away, resulting in
the exposure of the substrate iron.

TABLE 10.4. Weight Loss (in Grams) of The Samples with Different Surface Conditions of
Galvanized Steel and Bare Steel Samples in Aerated Soil for 60 days"
Soil

Surface Ai>

Surface Be

Surface Cd

Galvanized
steel

Steel

.-----~---.-----------------------------------

Lake Charles clay. pH


7.1. p = 400 Qcm
Merced silt loam. pH
9.4. p = 280 Qcm

2.231
1.444

0.166

3.234

2.255

0.Q45

2.431

2.03

"Ref. 357.
/lGalvanized steel with ourer zinc layer removed electrolytically.
(,Galvanized steel with outer zinc layer removed by soil in the field.
"Galvanized steel with zinc layer removed by soil in the laboratory.

CHAPTER 10

314
0.8r-----------------------------------------~

>- 0.6

a.

ai

1ii

~ 0.4

'iii

u 0.2

OL-____

_______ L_ _ _ _ _ _

10

20

_ _ _ _ _ _L __ _ _ __ L_ _

30

40

50

Time, months

FIGURE 10.8. Corrosion rate (in mils per year) determined by the linear polarization technique versus time.
After Rabeler; data are taken from Ref. 1086.

To further evaluate the properties of the cathodic film. samples with surface condition
C, which is similar to that from field exposure, were prepared by exposing a galvanized
steel sample to a very corrosive soil under a laboratory condition. Table 10.4 shows the
beneficial effect of the surface conditions Band C. The higher corrosion resistance of
surface conditions B and C was attributed to the formation of a protective film in the soil
by the galvanic action between the zinc coating and the bare steel. It was further found
that a similar film was formed on a steel surface galvanically coupled to a piece of zinc
immersed for one year in tap water containing 3% NaCl. This film was white and was
identified by X-ray analysis to be primarily zinc silicate. The formation of a protective
cathodic film on the surface of the zinc-iron alloy may perhaps explain the fact that in
many soils the corrosion rate of galvanized steel can be high initially but remain relatively
low in value thereafter for a lengthy period of time.
Corrosion rates in soils are sometimes determined as corrosion currents by electrochemical methods. Serra and Mannheimer [313] measured the corrosion currents of
galvanized steel in four different water-saturated soils. which were held in a sealed cell.
They found that, except for the relative higher values initially, the corrosion current was
quite constant during the one-month measurement. Rabeler [1086] conducted corrosion
current measurements in the field on galvanized steel anchors supporting electrical power
lines. The results, shown in Fig. 10.8, indicate that the corrosion rate decreases with time
to a relatively low value and were found to be in good agreement with weight loss
measurements on anchors embedded for about four years in the field.

11
Under-Paint Corrosion
11.1. INTRODUCTION
Knowledge about painted zinc-coated steels has rapidly accumulated during the past
two decades owing to the extensive research and development of more corrosion-resistant
car bodies. There are numerous published studies on the corrosion perfonnance and
corrosion mechanisms of various painted steel products. This chapter aims to compile the
results reported in these studies. The basic characteristics of paint and its function as a
physical barrier are briefly described in the first section. In subsequent sections, the tests
that are used to evaluate under-paint corrosion are described, and infonnation on the
effects of various components in a painted system on the corrosion perfonnance IS
presented. In the last section, the mechanisms of under-paint corrosion are discussed.
11.2. BASIC CHARACTERISTICS OF PAINT
Paint is a coating material consisting essentially of pigments dispersed in a solution
of a binding medium. The main components are (a) a binder or vehicle, (b) pigments and
extenders, (c) volatile solvents or thinners, and (d) additives such as driers, hardening
agents, stabilizing agents, surface-activating compounds, and dispersion agents. The
characteristics of various types of paint and the specific functions of each paint component
have recently been reviewed by Van Eijnsbergen [1202].
A paint coating is used either to protect the substrate material from deterioration or
to give the surface a certain aesthetic appearance. It protects the metal substrate from
corrosion primarily by two mechanisms: (1) by serving as a barrier to water, oxygen, and
other corrosive chemical species and (2) by serving as a reservoir for corrosion inhibitors
[I 109]. The barrier effect of the paint can be improved by increasing its thickness, by
utilizing pigments and fillers that make the paint less penneable to water and oxygen, and
by increasing its resistance to degradation.
A metal surface corrodes very little when it is properly painted. Corrosion occurs
only at places where the paint is damaged or has deteriorated. Starting at the damaged
places, such as at cuts, scribes, and cracks, the corrosion propagates laterally under the
paint, causing paint delamination and blistering. The nature and extent of corrosion are
detennined by the type of substrate, type of paint system, and environmental conditions.
315

316

CHAPTER II

11.2.1. Components in Paint


The vehicle, which includes solvents and resin or oil, etc., determines the basic
physical and chemical properties of the paint, but these can be modified by the nature and
proportion of pigments [312, 1107, 1202]. Vehicle materials are usually polymers of
relatively low molecular weight. The function of the volatile solvents in a vehicle is to
control the viscosity of the paint for ease of manufacture and for subsequent application.
Additives are added to the vehicle to improve the brittleness and to inhibit crazing.
The primary function of the pigment for a finish paint is to provide color, but in a
primer paint the pigment contributes to the corrosion resistance and the durability of the
whole system. The corrosion-inhibiting effects of pigments are realized in several ways.
The reactions of pigments with vehicles reduce the number of hydrophilic groups, thereby
increasing the resistance of the paint to the permeation of aggressive species. In addition,
pigments and their reaction products inhibit the corrosion of the substrate metals by their
buffering ability, which fixes the pH of the electrolyte in the paint. Furthermore, some
pigments, such as chromates, can leach out ionic species to inhibit the reactions at the
paint/metal interface or to passivate the metal surface.
In general, a single paint does not possess all the required properties, and in real
applications it is usually necessary to use multilayer paints comprising a primer, a finish,
and possibly one or two intermediate paints. A primer is the first coat of a system. Its
principal functions are to provide adequate adhesion and afford good protection to the
substrate. It can contain rust-inhibitive pigments such as zinc chromate and zinc dust. The
finish or the final paint must make up for the deficiencies of the primer and provide the
required color and degree of gloss.

11.2.2. Barrier Properties of Paint


For an undamaged paint, corrosive species such as water and oxygen have to diffuse
through the paint in order to reach and react with the metal surface. The rates of diffusion
have been investigated in a number of studies. It has been estimated [364, 1108] that the
amount of water diffusing through various O.I-mm-thick paint films at 85-100% relative
humidity is between 0.2 and 1.8 g/( cm2yr). The diffusion rate for oxygen in various paint
films is between 0.002 and 0.05 g/( cm2 yr). Tanabe et al. [738] reported that the diffusion
coefficient of oxygen in an epoxy paint film in 3 wt. % NaCI solution is between 1.1 x
10-7 and 1.6 x 10-7 cm2/s and varies little with paint thickness and time of immersion.
The diffusion rate of ions in paint is, in general, much smaller than that of water and
oxygen but varies significantly with the type of resin and pigment used [312].
Water penetration into paint depends on osmotic pressure. Miyoshi et al. [772]
evaluated water penetration in distilled water, in atmospheres with 100%,95%, and 85%
relative humidity, and in a salt solution. Distilled water and an atmosphere with 100%
relative humidity bring about the highest osmotic pressure and were found to give the
highest water penetration, as shown in Fig. 11.1. Miyoshi et al. also found that water
penetration is independent of the substrate since similar amounts of penetration were
found for cold-rolled, galvanized, and zinc/iron-coated steels. The resistance to water
penetration may also vary with time. Some paints become more and more penetrated by
the solution during immersion while others maintain their diffusion-restraining properties
[244].

UNDER-PAINT CORROSION

317

3.0,-------------------,

Water
~

2.0

I-

I-

FIGURE II. I. Water penetration into


a 90-Jin1 coat on galvanized sheets exposed to distilled water. 5% NaCI solution, and atmospheres with relative
humidities of 100%, 95%, and 85%.
Water content is calculated from the
weight increase. Reprinted from Miyoshi el al. [7721, with permission,
SAE Paper 850007. 1985 Society of
Automotive Engineers. Inc.

a:

1.0

I-

5%NaCI Sol

95%RH
85%RH

0
TIME

(d)

Many studies have revealed [364, 1108] that the diffusion of water or oxygen is not
necessarily the rate-limiting process for under-paint corrosion, since the intake rate of
water and oxygen by a paint film can be much larger than that needed for the corrosion
of an unpainted metal surface.
The durability of a paint is primarily determined by (1) the internal strength of the
paint films and (2) adhesion to the substrate [3121. The common degradation mechanisms
of a paint include abrasion and impact, cracking or crazing at low or high temperature,
bond breakage within the polymer matrix due to hydrolysis reactions, oxidation, or
exposure to ultraviolet light, and freeze-thaw cycling. The result of such degradation
allows access of reactants to the coating/substrate interface without the necessity of
diffusion through the matrix. With increasing age, the elasticity of the film usually
decreases. Thus, expansion and contraction of the metal base caused by severe temperature changes can result in the formation of discontinuities in a relatively inelastic paint
film. In practice, however, premature failures of paints are usually due to insufficient
preparation and treatment of the metal surface. The physical and chemical properties of
various paint products and their failure modes have been described in detail elsewhere
[312,11091
11.3. CORROSION TESTS
Different types of corrosion tests can be used for evaluating and understanding
under-paint corrosion. Depending on the practical purpose, a corrosion test is designed
to have one or all of the following properties: (1) to simulate corrosion in a real
environment; (2) to accelerate the corrosion process; (3) to be reproducible; and (4) to
provide an estimation of corrosion life in service [754]. However, it is usually very
difficult to design a corrosion test that possesses all these properties. Practically, corrosion

318

CHAPTER II

tests can be divided into two groups: those conducted in environments in which the
conditions, such as temperature, relative humidity, salt spray, spray solution concentration, etc., can be controlled; and those conducted in real environments in which the
conditions cannot be controlled. Usually, the first group of tests provides better acceleration and reproducibility whereas the second group of tests provides better simulation and
life estimation. Miyoshi [754], in a review article, classified corrosion test methods for
painted automotive materials as follows:
Field survey: Used car bodies are evaluated for the degree of cosmetic damage
and number of perforated spots.
Monitor car test: Cars are tested under designed driving conditions.
Proving ground test: Cars are driven on salty, dusty, and muddy roads in the
daytime and kept in a humidity chamber overnight.
Vehicle exposure test: Cars are exposed at the seacoast and are periodically
sprayed with a solution.
Under- and on-vehicle test: Test coupons are mounted under or on a vehicle to be
driven under given conditions.
Laboratory tests: Two most frequently used tests are the salt spray test (SST) and
the cyclic corrosion test (CCT). A CCT includes various factors such as salt spray,
drying, humidifying, freezing, etc. There are many different CCTs. involving
various combin.ations of temperature, humidity, solution composition, pressure,
duration, chipping with stones, etc. (The test conditions for a typical CCT can be
seen in Fig. 11.13.)
A CCT is intended to accelerate as well as to simulate the corrosion processes
occurring in a car body. Currently, almost all of the world's automakers use some form
of laboratory cyclic corrosion test to evaluate coated steel sheet. In order to provide a
reliable and universally accepted ranking of automotive sheet steel products for cosmetic
corrosion, there has been an effort by the automotive and steel industries to develop a
standard CCT. The effects of the various factors in such a standard accelerated corrosion
test have been summarized by Townsend [1112].
Test specimens of various configurations may be used. depending on the practical
purpose. Shaped structures and lapped panels are usually used for evaluation of perforation damage, and flat panels are mainly used for evaluation of cosmetic damage. Since
under-paint corrosion usually starts at paint defects or damaged spots, defects are often
artificially introduced prior to the test by scratching, chipping, cutting, etc.
The common parameters for evaluating under-paint corrosion are the area of the
rusted surface, amount of paint loss, penetration depth in steel, and creep length from the
scratch. A cross-hatch test is normally used for evaluating wet adhesion of paint. In this
test, the paint is scribed with perpendicular lines spaced a few millimeters apart. After a
corrosion test, the number of the paint squares removed by an adhesive tape is used as an
indication of the loss of adhesion [772,762]. Besides exposure-type tests, electrochemical
measurements are often used to study the mechanistic aspects of under-paint corrosion
by measuring paint porosity, diffusion, impedance, anodic and cathodic delamination,
etc. [167,244,364,827,985].

UNDER-PAINT CORROSION

319

11.4. CORROSION BEHAVIOR

11.4.1. Characterization of Corrosion


The corrosion of painted metals, particularly automotive bodies, is generally characterized as perforation corrosion or cosmetic corrosion. Corrosion of a painted steel sheet
that initiates at an interior surface of a car body panel, penetrates the sheet, and eventually
shows through as rust at the exterior exposed surface is known as perforation corrosion
[lOOT]. It often occurs at locations that are difficult to clean, phosphate, and coat, such
as lapped parts and hem flanges, or at crevices that collect dirt, salt, and moisture [1113,
754].
The term cosmetic corrosion is applied to an attack that is initiated at the exterior
surface, usually at regions where the paint is damaged. Although this form of corrosion
may eventually lead to perforation, the main concerns are with appearance. Cosmetic
corrosion is usually related to (l) red rust-rust stain and bleeding at scratches in the
paint; (2) paint creep-undercutting of the paint and loss of adhesion at scratches: and
(3) chipping-removal of paint due to the combined effects of corrosion and Impact
damage by stones and road debris. Simplistically, the directions of corrosion propagation
in cosmetic corrosion and perforation corrosion can be described as being parallel to the
surface and perpendicular to the surface, respectively, as shown in Fig. 11.2.
J1.4.1.1. Peiforation Corrosion. The resistance of a painted material to perforation
corrosion is usually evaluated by weight loss or depth of corrosion penetration. Figure
11.3 shows the depth of corrosion penetration for cold-rolled steel and several coated
steels after a cyclic test [762]. The depth of penetration at the scribed line is much smaller
for zinc- and zinc-alloy-coated steel than for noncoated steels.
The corrosion penetration at the places where the paint is damaged, such as at the
scribed lines, can be compared to the corrosion of an unpainted surface. Figure 11.4 shows
that the weight losses of unpainted zinc-coated steel in a cyclic test are lower than those
of cold-rolled steel and decrease with increasing coating weight [1072]. Similarly, Fig.
1l.5 shows that the corrosion penetration of the samples painted with a primer is
decreased with increasing zinc coating weight [282].
Unlike cold-rolled steel, coated steel exhibits substantial corrosion penetration only
after a period of incubation [339]. The incubation time for coated steel is attributed to the
corrosion resistance and the galvanic action provided by zinc coatings. Thus, an increase
in coating weight generally increases the incubation time and delays the perforation of
the coated steel. After an incubation period, the coating is consumed, and the corrosion
rate of coated steel becomes similar to that of cold-rolled steel. Pure zinc coatings are
more effective than alloy coatings in reducing the corrosion penetration of the steel, owing
to the stronger galvanic action between the steel and the coating.

steel
FIGURE 11.2. Corrosion direction in
perforation corrosion (P) and in cosmetic corrosion (C).

1I1I11I11I!jjillllJlLiJP'"

'"""11111111111"/

---'\~\~\T"""""'\
-\
\~\~\-r-\\r--<,\--'-\~\-"-\\r-.:\:--"-~.....--:\"<""""<"""~'-

coating
.

crevice

CHAPTER II

320

perforation
I

E
E

a.

~ 0 .5
c:
o

'iii

e
o

CR

ZM

5Z

GA

FIGURE 11.3. Corrosion depth at scribe line on cathodic electropaint film after 60 cycles of cyclic corrosion
test. CR, Cold-rolled steel; ZM, Zincrometal; SZ, Ni-Zn electroplated steel (20 glm2); GA, galvannealed steel
(45 g/m2); G90, hot-dip galvanized steel (135 glm 2). Reprinted from Wakano et al. [762], with kind permission
from The Iron and Steel Institute of Japan, Chiyoda-ku, Tokyo, Japan.

11.4.1,2. Cosmetic Corrosion. Cosmetic corrosion is most commonly evaluated by


measuring the length of under-paint creeping. It is sometimes evaluated also by the extent
of rust formation or paint loss [282, 762, 772, 1090]. Many factors, such as coating
composition, coating thickness, surface treatment, test condition, type of paint damage,
and type of paint, can affect the cosmetic corrosion of painted steels, as will be discussed
in the following sections. Data from a field survey (Fig. 11.6), have shown that, in general,
steel panels coated with zinc or zinc-alloy coatings have much slower under-paint
creeping than cold-rolled steel [1072].

120.------------------------------------------,
-CR

100

FIGURE 11.4. Effect of drying time ratio Rdry on corrosion weight loss of unpainted cold-rolled (CR),
electrogalvanized (EG, two coating weights), and hot-dip galvanized (GI) steel sheets after three-week wet/dry
cyclic test. Reprinted from Ito et al. [1072], with kind permission from The Iron and Steel Institute of Japan,
Chiyoda-ku, Tokyo, Japan.

UNDER-PAINT CORROSION

321

0.4 , . . , C " " ' o " " ' l d - : - - - - - - - - - - - - - - - - - ,


.
roll ed Elec tropla ted
1"10 NaCI
(;l

In- I
0 0.1"10
aC I
0.3 0

'"
C 0.01% aC I
Ga lvannea led
'"c:
0.2

.s
""

-><

c..>

FIGURE 11.5. Perforation test results for


electrodeposition primed samples (dip
phosphate, cathodic electrodeposition, 20
pm). After Iric [2821.

:
-g

0.1

c..>

:J

CI:

20

4D

Galvani zed

60

80

t20

Coating weight (g/m')

11.4.2. Effect of Coating Type


Many different zinc alloy and composite coatings have been developed for automotive applications. Table 11, I lists the major zinc- and zinc-alloy-coated steel products used
for automotive bodies [1198]. The coated steel products used in automotive bodies must
meet many engineering requirements [1007]. Some of the key properties of these
products, besides corrosion resistance. are weldability, formability, bondability, and
paintability.
There is now a large amount of data indicating that almost all coated steels show
improved resistance to cosmetic and perforation corrosion in comparison to cold-rolled
steel. However, it has been noted that the differences between the various types of coated
steels are not always significant, and the corrosion performance rankings for the various
coatings reported by different investigators sometimes disagree [985].
In general, as shown in Fig. 11.7 [351], zinc and zinc-alloy coatings substantially
reduce the blistering width compared to that for cold-rolled steel. Robbins et al. [3461
tested painted steel samples coated with alloys of various types and thicknesses and found
that a pure zinc coating with heavier coating weight exhibited the best cosmetic corrosion
resistance, The relative cosmetic corrosion resistance of the different zinc and alloy
6 r---------------------------------------~

.. Cold Rolled Steel


.. G.lvanne.led Stee l

+ Electrogalv. St!.

Qj3

.~

:0

02
.s::

'5 1

.~

:J

_______

____

~o~------~----r-==~======~~~
o
2
3
Time (years)

FIGURE 11.6. Growth of underfilm corrosion of cold-rolled, galvannealed, and galvanized steel sheets used
for monitor-car body outer panel. Reprinted from Ito e/ al. [1072], with kind permission from The Iron and
Steel Institute of Japan , Chiyoda-ku, Tokyo, Japan.

322

CHAPTER 11
TABLE 11.1. Sheet Steels Used in Car Bodieso
Cold-rolled steel
Mild, unalloyed
High strength, e.g., micro-alloyed, dual-phase, P-alloy
Steel, hot-dipped
Zn + 0.1-0.2% Al
Zn + about 10% Fe !Galvanneall
Steel, electroplated
Zn
Zn + about 10% Ni
Zn + about 16% Fe
Zn + about 16% Fe + FeZn topcoat (83% Fe)
Steel, hot-dipped and electroplated
Zn + about 16% Fe (Galvanneal) + FeZn topcoat (83% Fe)
Steel, coated with conductive primer
Zinc-rich primer (Zincrometal)
Zn or ZnNi (electrolytic) + Zinc-rich paint (Bonazinc 2000)
Zn or ZnNi (electrolytic) + organic coating (1 ,urn, e.g .. Durasteel)
"Ref. 1198.

coatings, however, varies in different investigations. Johansson and Rendahl [1006] found
that Zn and Zn-Fe coatings have better overall corrosion resistance in the hot -dipped state
than in the electrolytic state [1006]. In another study, electrogalvanized zinc-iron and
zinc coatings were reported to have better paint adhesion properties than hot-dip zinc and
Zn-Fe coatings [1073].
Figure 11.7 indicates that the under-paint creeping after an outdoor exposure test is
less for the zinc coating than for the Zn-Ni coating. However, in another study it was
found that the creeping resistance of Zn-Ni-coated steel in atmospheric environments is

Zn-Ni

Free Zn

e~;~:nnoaled
__ ZnCoCr
with AI~O~

1.5

2.5

3.5

Duration (year)

FIGURE 11.7. Growth of blistering on cold-rolled (CR), galvannealed, and zinc- and zinc-alloy-coated steels
in the outdoor exposure test. Coating weight was 30 g/m 2. Reprinted from Yasuda et al. [351], with permission
from ASM International.

UNDER-PAINT CORROSION

323

comparable to that of painted pure zinc coatings [283]. Van Ooij et ai. [984J compared
the blistering of several zinc alloy coatings and found that (1) Zn-Ni coatings have
excessive cracking sensitivity and a high rate of paint blistering; (2) no delamination and
less blistering occur for a Zn-Fe coating; and (3) for Zn-AI coatings the blistering is
extensive owing to poor phosphate adhesion.
Figure 11.8 shows that 4% AI-O.I % Mg-misch metal prepainted steel sheet has a
better resistance to creeping compared to zinc- and other Zn-AI-alloy-coated steels
[1082]. The better corrosion resistance was attributed to magnesium depositing on the
a-AI area of the eutectic phase and decreasing the crystallization rate of the Zn-AI
eutectic phase.
Various methods have been used to further increase the corrosion performance of
coated steels. Robbins reported that trace amounts of Co codeposited in Zn-Ni coatings
increase the corrosion resistance of the coatings [346]. Yasuda et at. [351 J reported that
an AI 20 r dispersed Zn-Co-Cr composite coating exhibited excellent resistance to creeping. They attributed this resistance to the formation of more mechanically and chemically
stable corrosion products and the anodic passivation brought about by Cr and AI 20 r
Shastry and Townsend J8l5J found that the presence of an electroplated Cr + CrO, overlay
significantly improves the scribe creep resistance of zinc coatings, owing to the improved
paint adhesion by mechanical interlocking of the primer to Cr + CrO, on zinc. in addition
to the chemical stability ofCrO,. in the paint environment. Memmi et al. [340J reported
that Cr-CrO,-topped zinc coatings performed significantly better in terms of the resistance to creeping, under a cyclic corrosion test. than simple Zn, Zn-Ni, and Zn-Fe
coatings. For a Cr-CrO,-topped zinc coating, a phosphate treatment is not necessary
because paint adhesion is ensured by the Cr-CrO, layer. In addition, the surface of this
layer is chemically inert to phosphate solutions.
Duplex coatings, such as Fe-ZnJZn-Ni and Zn-Ni/Fe-P, have also been developed
[312, IllS]. Duplex coatings consist of a lower layer to provide good corrosion resistance
and an upper layer to provide good paint adhesion. Kokubo [IllS] found that Zn-Fe/ZnNi two-layer electroplated steel sheet has better resistance than one-layer coated steel to
paint blistering from a scribe, owing to the better corrosion resistance of the Zn-Ni
coating and the better wet adhesion of the Zn-Fe coating. Other alloy and composite
coatings such as Zn-Co [44.229.246], Zn-Mn [330, 349], Zn-Co-Mo [312], Zn-CoCr [425], Zn-Ni-Si0 2 [263], Zn-Ni-Ti [487], Zn-Si0 2 [284], and Zn-Co-Cr-AI 201
[351] have also been investigated.
Coating Weight. The resistance to under-paint creeping for a given coating generally
increases with coating weight [767,1064,1072,1073, 1090,1114]. Taking data reported
by Ito et al. [I072J as an example, Fig. 1l.9 shows that the resistance of zinc and
zinc-alloy-coated steels to creeping increases with coating weight. Franks [11 14] found
that the length of under-paint creeping decreased approximately linearly with coating
weight from 25 to 125 g/m2 in an on-vehicle exposure test. Davidson et ai. [1073 J reported
that the scribe creepage was 37 mm for a 3.5-,um-thick coating and only 2.4 mm for a
SS-,um-thick coating after a two-year outdoor exposure test. Light coating weight generally results in poor cosmetic corrosion resistance regardless of the type of paint defects
[346,1073J.

os

,.-

1.5

+ Zn

5AJMM

-~AJ

Exposure time (years)

Ichikawa-1

Ichikawa-2

FIGURE 11.8. Developmental changes in the width of blisters at the cut edge of prepainted steel sheets subjected to atmospheric
exposure tests. MM, Misch metal. Reprinted from Masuhara and Kumon [1082], with kind permission from The Iron and Steel Institute
of Japan, Chiyoda-ku, Tokyo, Japan.

iii

.~

.~

-c

-5

E
5

Kiryu

'M

>-

::0

tTl

""-l

n
:c

325

UNDER-PAINT CORROSION

a:
w

~~

f-

a5

..,

~'-'
3
Qw

3:rn
~iI 2
::l~
~~

~g

20

..

Zinc-Alloy
Coated Steel

,~___
40

Electrogalvanlzed
--_ _--"S:...\.t:.:.,:ee'"-I_ _

60

80

100

COATING WEIGHT (g/cm 2 )

FIGURE 11.9. Etfect of coating weight of zinc-alloy-coated steel sheets on paint adhesion of monitor-car body
outer panel after three years' running in Okinawa, Japan. Reprinted from Ito et al. r1072], with kind permission
from The Iron and Steel Institute of Japan, Chiyoda-ku, Tokyo, Japan.

11.4.3. Effect of Test Conditions


Currently, almost all of the world's automakers and steelmakers use some form of
laboratory accelerated corrosion test for evaluating coated steel sheets [1112]. The various
tests differ in the conditions employed, and thus the results vary accordingly.
Under-paint creeping strongly depends on the test conditions such as temperature,
rate of drying, impurity level. sequence and cyclic frequency, and type and duration of
moisture, as reviewed by Miyoshi [754]. It has been established by many investigators
[767,772, 827] that under a field exposure or in a cyclic test the under-paint creeping rate
for zinc-coated steels is much lower than for cold-rolled steel, but under a salt spray test
condition the converse can be observed, as shown in Fig. 11.10 [827]. Continuous salt
spray testing accelerates the corrosion process, but it generally does not renect the
corrosion performance of painted steels in real environments.
An important factor in a cyclic test is periodic wetting and drying. Figure 11.11
shows that corrosion penetration of unpainted samples varies with the percentage of
wetting time in the total time of testing [827]. A similar effect can be seen in Fig. 11.4,
which reveals the important effect of drying on the amount of corrosion: a wet/dry ratio
of around 0.5 seems to be the most severe condition, leading to the greatest weight loss
of the unpainted steels in the cyclic corrosion test. Figure 11.12 illustrates the effect of
wetness fraction on the creeping of painted samples [827]. It has been noted that for a
given total time of wetness, the corrosion rates of unpainted zinc and steel samples are
much higher when the samples are continuously wet than when they are periodically dried
[1129].
The type of wetting is also significant in determining the corrosion behavior [980].
It was pointed out by Standish et al. [767] that corrosion product buildup, which occurs
in atmospheric service, is prevented in a test in which the solution runs continuously, such
as the salt spray test. Zhang and Tran [1117] found that variations in the proportion of
spraying time and drying time result in significant changes in the amount of corrosion
product formed on the surface. The frequency of cycling was reported by Kurokawa et
al. [827] to exert a significant effect on corrosion penetration of cold-rolled and coated

~ 10

90

ok:::
o

180

270

450

T ime, days

360

540

630

Seashore exposure

~
720

10

20

Time, days

30

40

50

i~~-+--~--

SST

FIGURE 11. 10. Growth of blisters in Okinawa seashore exposure and salt spray test (SST). EG, Electrogalvanized steel; GA, galvannealed steel;
CR, cold-rolled steel. From Kurokawa et al. [827]. Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.

CD

]j 5

Q;

-~

i5

.J::.

-CR

GA

+EG

-------------------------------------r-------------------------------------,

15 r.

tTl
;:0

-l

n
>.."

:r:

N
0\

....

327

UNDER-PAINT CORROSION

~ 0.8

S
E

E_ 0.6

c
.2
ti

::l

'C
~

0.4

Wetness 100
fraction( %)

80

75

66

50

38

Corrosion SST
tests

CCTI

CCTI

CCT3

CCT4

CCT5

Modified
Volvo

'"

Okinawa

Seashore
exposure

FIGURE 11.11. Thickness reduction of the bare steel (specimen coated with IO-flm electropaint film and with
I O-mm-wide bare metal in the center) after one-month exposure in various corrosion tests. SST. Salt spray test;
CCT, cyclic corrosion test; modified Volvo: seawater spray twice a week; Okinawa. seashore exposure on
Japanese islands located in subtropical zone; EG. electrogalvanized steel; GA. galvannealed steel; CR,
cold-rolled steel. From Kurokawa et al. [827]. Copyright by NACE International. All Rights Reserved hy
NACE; reprinted with permission.

6.0 .----------------------------------------,
" CR

5 .0

Wetness 100
fraclion( %)
Corrosion SST
tests

80

75

66

50

38

ccn

ccn

CCTJ

CCT4

CCT5

Modified
Volvo

Okinawa

Seashore
exposure

FIGURE 11.12. Blister widths after one-month exposure in various corrosion tests. See caption to Fig. 11.11
for identification of tests and samples. After Kurokawa et al. [827].

328

CHAPTER II

Cyclic Test

40r-------------------------------,

Salt spray
at 35 "C . 10 min.

E
E

.p 30

Drying
at 60 "C , 155 min.

Cll

cQ)

0.

I
I

Q)

Wetting at 60 "C
and 95 % RH. 75 min.

~ 20

...Q)

I
I

~
~

o
0.

Drying at
60 "C. 160 min.

~10
"0'

5 cycles
OL-L-~~~

___ L_ _ _ _ _ _ _ __ L_ _ _ _ _ _ _ _
10

15

Wetting at 60 "C
and 95 % RH, 80 min.

Cyclic test creep length Imm)


FIGURE 11.13. Correlation between the results of a cyclic corrosion test (after 28 cycles, one cycle-24 h) and
one-year Okinawa exposure test. Reprinted from Sakauchi and Kunimi [1074), with kind permission from The
Iron and Steel Institute of Japan, Chiyoda-ku, Tokyo, Japan.

steels. The composition of the spraying solution also has a significant effect on the
corrosion rate, as shown in Fig. 11.5.
It is important for accelerated tests to produce results that show a definite correlation
to those in a real environment. Much research effort has been made to develop such tests
[754, 1112]. Figure 11.13, for example, shows that the results from a multistep cyclic
corrosion test correlate well with those from a field exposure test [1074]. In simulated
tests, wet/dry cycling is critical for creating corrosion products having a structure and
morphology similar to those formed under natural environments. To more closely
simulate the real corrosion performance of a car body, a chipping effect may also be
included in a cyclic test. According to Gray and Shaffer [589], a cyclic corrosion test with
a chipping effect is a more realistic test for assessing paint adhesion of coated steels. They
found that the results of the chipping corrosion test show no linear relationship to those
obtained from the scribed test and that the initial chipping damage is of critical importance
in the corrosion performance [169, 338].
11.4.4. Effect of Paint System
The possible variations in paint formulation are infinite, and it is difficult to
systematically classify the various paints according to their effect on the corrosion of the
substrate materials. In general, corrosion of painted products occurs at places where the
paint is damaged. The effect of a paint on corrosion is mainly determined by its internal
strength, stability, and adhesion to the substrate, which are not only a function of paint
formulation but also a function of the methods used in applying the paint.
A paint has to have a certain thickness in order to furnish adequate protection against
corrosion penetration, as shown in Fig. 11.14 [827]. A paint less than a few microns in

329

UNDER-PAINT CORROSION

perioration
E
E

.sc
a

il

:>
"0

0.5

<J)
<J)
Q)

C
.><

u
.s::.

I-

Paint thickness ,pm

FIGURE 11.14. Effect of paint thickness on the thickness reduction of the painted steels, measured after fi ve
months' exposure in a c yclic corrosion test (wet: 66%). EG, Electrogalvanized stee l; GA , galvanllealed steel;
CR. cold-rolled steel. From Kurokawa el (l1.1827J. Copyright by NACE International. All Rights Reserved
by NACE; reprinted with permission.

thickness is not very effective in suppressing corrosion since the barrier effect of the paint
against the permeation of oxygen, water, and other aggressive species increases with paint
thickness [287, 827,1199]. Paint thickness also affects wet adhesion [772].
Mabuchi et al. [287] found that hydrophilic resins generally promote the formation
of ZnCI 2 AZn(OHh, which shows good corrosion resistance, whereas hydrophobic resins
promote the formation of ZnO, which is less protective. However, hydrophilic resins
promote water permeation, which i sharmful to paint adhesion and corrosion performance
[312,754]. It has been reported that the addition of a certain amount of silica strongly
affects corrosion resistance by altering the types of corrosion products developed [287,
341]. Zinc-rich paint was found to reduce under-paint creeping of Zn-Ni coatings [768,
770] .
van Ooij and Sabata [978] found that paint adhesion changes with changes in the
curing conditions for the primer and with a rinse of the phosphated panel with silicates
and silanes. Chromating before painting has been found to have various effects on the
corrosion performance of coated steels [954, 120 I]. In one study, chromating of the zinc
coatings before application of organic paints was found to increase the time to occurrence
of red rust [287]. Chromate treatment of Zn-Ni alloy coatings prior to painting has been
reported to enhance corrosion resistance [341]. However, it was reported in another study
that chromate or nonchromate rinses have little effect on the scribe creep resistance of
zinc or zinc-alloy coatings in a cyclic test [815] .
11.4.4.1. Phosphating . Surface treatments of zinc- and zinc-alloy-coated steels are
particularly important for paint adhesion and corrosion performance [170, 762, 815].
Currently, zinc- and zinc-alloy-coated steels for car bodies are exclusively treated hy a
zinc phosphate process. In an early study, LaQue and Boylan [755] reported that
phosphating at least doubled the resistance against corrosion weight loss of various types
of painted bare steels after exposure in a marine atmosphere. It becomes more beneficial

330

CHAPTER 11

100

~
.2

+P

100

Q)

I + Wet adhesion

~
c
'iii
Q)
.c

.Q

I .... P-ratio

50

SO

"0

'"

Cl.

a;

Zn

20

40

60

80

Fe content of Zn-Fe layer (%)

0
100
Fe

FIGURE 11.15. Wet adhesion. A 90-,um three-layer coating on Zn-Fe alloy (5-100% Fe) electroplated steels
with a coating weight of 20 g/m 2 was evaluated with a cross-hatch test after IO days in distilled water at 40C.
Adhesion failure was measured by a video pattern analyzer. P ratio is defined as P/P+H, where P and H denote
phosphophyllite and hopeite, respectively, and was determined from X-ray diffraction peak intensities. Data
are taken from Ref. 772.

in cases in which the paint is damaged by scrubbing because phosphating improves the
wet adhesion of paints [762, 772].
Phosphating is particularly effective on zinc coatings containing high levels of iron.
Figure 11.15 shows that the wet adhesion of paint increases with iron content in Zn-Fe
alloy-coated steel [772]. This was attributed to the fact that the amount of the adhesionenhancing phosphate formed increases with increasing iron content in the coating. The
type of zinc phosphate tends to be mostly phosphophyllite for coatings with an iron
content above 60% and to be mostly hopeite for coatings with an iron content below 20%.
The adhesive property of a phosphated surface can be further improved by adding certain
amounts of ionic species to control the composition and crystal size of the phosphate layer
[1059]. The beneficial effect of phosphating is, however, not always observed because
other factors may become predominant in determining adhesion strength. As reported by
Hess and Soreide [1111], the phosphate coating weight, crystal structure. and size have
little correlation with paint adhesion performance. Also, Shastry and Townsend [815J
found that the scribe creep resistance of Zn-Ni and Zn-Fe alloy coatings is less sensitive
to the differences in phosphate treatments than is the case with zinc coatings.
Loss of paint adhesion can occur due to water penetration in the paint, with or without
the effect of corrosion [754]. As reported by Wakano et ai. [762] and shown in Table 11.2,
the loss of adhesion due to water penetration is generally more severe for zinc- and
zinc-alloy-coated steels than for cold-rolled steel because phosphophyllite, which enhances paint adhesion, easily precipitates on a bare steel surface but not on a zinc alloy
surface. Loss of adhesion was also found to be related to the phosphate coating weight
and to the uniformity of the galvanized coating surface crystal orientation [1065].
11.4.4.2. Effect of Paint Defects. Corrosion of painted structures usually starts at
damaged areas. Test results have shown that under-paint creepage depends on the kind of

331

UNDER-PAINT CORROSION

TABLE 11.2. Results of Paint Adhesion Tests".!'


Peeled area of paint film (%)
Substrate

Test I

Test 2

0
0
0
0
0

Cold-rolled steel
Electrogalvanized steel
Gal vannealed steel
Ni-Zn alloy electroplated steel
Zincrometal

0
76

100
44

"Reprinted from Wakano et al. [762], with kind permission from The Iron and Steel Institute of Japan,
Chiyoda-ku, Tokyo, Japan.
''Test I: As painted; test 2: after immersion in warm deionized water (40C or SOC) for 240 h.

mechanical damage [754]. There are several methods for generating paint damage:
scrubbing, gravel chipping, arrow chipping, and diamond chipping, In one study, it was
found that gravel chipping gave the largest creepage, followed by wide scrubbing, as
shown in Fig. 11.16 [754]. The extent of paint damage caused by chipping depends on
the type of paint, the paint thickness, and the substrate [169, 1075]. Standish et ai. [767]
found that there is, in general, more creeping at scribes which cut deeply into the metal
panels than at those which cut just through the paint.
Vrable [588] found that deformation of zinc- and Zn-Fe-coated steels prior to
painting, which causes cracks in the alloy coatings, has little effect on paint creeping
performance in a cyclic corrosion test. Deformation of the painted galvanized steel, on

O CRS

$10 I-

f2 Zn

:u

~ Zn-Fe

.~

'0

.....
.c

...

'0

'~

5 I-

Q)

01

III
Q,
Q)

~
d

Paint damaging method


FIGURE 11.16. Influence of paint-damaging method on the creepage width or diameter after one-year
weathering test ~f zinc-coated, zinc-alloy-coated, and cold-rolled (CR) steels with a salt solution spray. Coating
weight, 40 g/m-. (a) Wide scribing (both sides); (b) narrow scribing (both sides); (c) diamond shot; (d) arrow
chipping; (e) gravel chipping. Data are taken from Ref. 754.

332

CHAPTER II

the other hand, reduces the corrosion resistance, with complex strain conditions being
worse than linear strain conditions [428]. Cosmetic corrosion of coated steel at welds has
been studied by van Ooij et al. [986].
11.4.5. Galvanic Action

Galvanic action between the coating and the steel substrate is one of the critical
factors determining the corrosion behavior of the coated steels. As has been systematically
discussed in Chapter 7, the galvanic action of a zinc-coated steel involves a large number
of factors, and a specific test condition can be designed to evaluate each of these factors.
Generally, the extent of galvanic action can be estimated by measuring the coupled
potential and the galvanic current between the zinc coating and the steel, either painted
or unpainted.
Figure 11.17, as an example, shows the galvanic current between painted zinc-coated
specimens and a bare cold-rolled steel specimen in 5% NaCl solution [827]. The painted
zinc-coated specimen behaves as an anode while the painted bare steel behaves as a
cathode. Thus, the corrosion of bare steel, such as at a paint scratch, in contact with a
painted zinc-coated steel would be reduced by the galvanic action whereas in contact with
a painted cold steel the corrosion would be enhanced by the galvanic action. Wakano et
al. [762] reported that the galvanic current between a painted steel and a bare zinc-alloycoated steel increases with time. Jordan [286] found that the galvanic corrosion rate of a
zinc coating is higher when coupled to iron corrosion products than when coupled to bare
steel, indicating that the formation of steel corrosion products can accelerate the corrosion
of zinc coatings at areas where the paint is damaged. This is, however, a short-lived
transient effect that occurs during reduction of FeOOH [1299]. In a study by Sun and
Tsujikawa [1261], it was found that the width of the under-paint corrosion of a zinc
coating coupled to steel depended on the conditions of the cyclic test, such as the
concentration of NaCI, the relative humidity, and the length of drying. The drying
10,--------------------------------------,

EtJ 0.1

+GA

~0.01

~
~

.m

O~--------~~------------~~--------~

+~

Q)

i-o.o
(.)

-0.1
-1

11

Cycles (2 days for 1 cycle)

FIGURE 11.17. Galvanic currents between electropainted test specimen and cold-rolled sheet. EG, Electrogalvanized steel; GA, galvannealed steel; CR, cold-rolled steel. Data are taken from Ref. 827.

UNDER-PAINT CORROSION

333

condition in a cyclic test has a significant effect on the initiation of the under-paint
galvanic corrosion of the zinc coating.
Suzuki et at. [1241] determined the rest potential of various alloy coatings in 5%
NaCI solutions and found that the galvanic throwing power was generally higher for a
coating with a more negative potential. Massinon et al. [167] measured the galvanic
protection life of various zinc- and zinc-alloy-coated steels coupled to bare cold-rolled
steel in O.03M NaCI solution and found that the protection life generally increased with
increasing coating weight. They also found that, as a result of cathodic protection, the
steel becomes less efficient as the cathode owing to the precipitation of jJ-Zn(OHh or
ZnO on the steel surface. Hayashi et al. [869] studied the relationship between the
under-paint chloride penetration rate and the galvanic current for Zn-Fe alloy coatings.
They found that the galvanic current was limited by the kinetics of charge transfer within
the first 0.1 s and then was controlled by oxygen reduction at the cathode. The efficiency
of Zn-Fe alloy coatings as a cathode increases with increasing iron content. The
under-paint corrosion at the lap joints of galvanized steel over cold-rolled steel was found
by Vincent and Coon [1199] to decrease with paint thickness.
The galvanic action between the steel substrate and the zinc coating under paint
occurs essentially in a thin layer of high-resistance electrolyte (within the paint film).
According to Zhang and Valeriote [522], the area ofthe galvanically protected steel under
a thin-layer electrolyte depends on the resistivity and thickness of the electrolyte, the
distance between the steel and the zinc coating, and the area of the bare steel surface.

11.5. CORROSION MECHANISMS


The mechanisms of under-paint corrosion for zinc- and zinc-alloy-coated steels have
been the subject of many studies [167-169, 772, 869, 978, 983, 984]. The multicomponent interactions between paint, coating, steel, and environment make under-paint corrosion a very complicated phenomenon. In general, under-paint corrosion starts at places
where the paint is damaged. The corrosion proceeds with corrosion of the coating or with
paint delamination followed by corrosion of the substrate and, at longer times, leads to
perforation of the steel. Generally, the corrosion products built up at the corrosion front
may mechanically delaminate the paint. Delamination can occur at different interfaces in
a paint-coating-steel system, depending on the materials and the environmental conditions, as schematically shown in Fig. 11.18. The causes of the delamination at the
corrosion front, as reported by different investigators [167-169, 983, 984], can be
physical, anodic, cathodic, mechanical, or the combinations of them. The predominant
cause in a specific corrosion situation can be due to variations in paints r1081], coatings
1168,983], phosphates [772,978], and test conditions r168, 767, 827].
Shastry and Townsend [168] noted that significant differences exist in the mechanisms of paint creepage from the scribe on steel and coated steel substrates. In the case
of a steel substrate, interfacial failure results from the dissolution of phosphate, whereas
for zinc-coated steel, anodic dissolution of the coating seems to be the predominant factor
controlling scribe creep. According to van Ooij et al. [983, 984], the expansion of the
corrosion product causes delamination of paint. The rate of delamination is related to the
anodic corrosion rate of the zinc coating. For Zn-Ni alloy coatings, the first step in the
corrosion process is also delamination by anodic dissolution, but the dissolution rate is

334

CHAPTER II

FIGURE 11.18. Possible delamination


paint
phosphate _
coating

steel

~~~~~~~~~~

modes of a painted coated steel: (I) at


paint/phosphate interface due to loss of
adhesion; (2) within phosphate layer due
to mechanical fracture; (3) at phosphate/coating interface due to dissolution
of phosphate; (4) dissolution of coating:
(5) at coating/steel interface due to mechanical failure.

lower owing to dezincification of the coatings, leading to a better barrier effect of the
Ni -enriched layer and the fonnation of a more compact corrosion product. The differences
in perfonnance of various types of alloy coatings are, to a large extent, due to the
compactness and stability of the corrosion products formed in a corrosion environment
[983].
Cathodic delamination has been identified as a common cause for advancement of
the corrosion front [168, 772]. In a local corrosion cell at a scribe, the coating serves as
the anode, and the surrounding area of the painted surface serves as the cathode. Because
the cathodic reaction is primarily the reduction of oxygen diffusing through the paint, the
high alkalinity resulting from the cathodic reaction can cause the dissolution of phosphate
layers, resulting in paint delamination. Massinon and co-workers [167, 1081] found that
cathodic paint delamination can be produced by passing a cathodic current through the
phosphated galvanized steel. When the paint is thin or of a porous nature, oxygen and
water can easily penetrate through the paint and react with the coating, thus causing a
dissolution of the phosphate layer. Oxygen reduction at the delamination head is considered to be the rate-controlling reaction in the delamination process. Figure 11.19 schematically represents the corrosion process caused by a cathodic delamination. The
galvanic action between zinc and steel is also found to cause paint/substrate interface
deterioration due to a cathodic delamination [1084]. Thus, the ratio of the area of the steel
cathode to that of the coating anode at a micro defect may determine the extent of
blistering [1083].
The advancement of under-paint creeping can also be directly caused by anodic
dissolution of the zinc coating [168, 827, 1072]. Kurokawa et al. [827] found that, in a
cyclic corrosion test, the anodic dissolution front under the paint on the coated steel is
acidic while the corrosion product behind the dissolution front is basic. According to

Phosphate _
layer

FIGURE 11.19. Schematic representation of the degradation under


a paint film. After Massinon el ai.
[167) .

UNDER-PAINT CORROSION

335

van Ooij et al. [994], the mechanism of under-paint corrosion propagation is initially
anodic, i.e., dissolution of the coating, for all systems.
Mechanical paint delamination, due to corrosion product buildup, has been identified
as a cause of delamination between phosphate and coating surfaces in many circumstances
[983, 984]. Standish et al. [767] found that corrosion buildup, which results in the
delamination occurring within zinc phosphate crystals, is the cause of lateral spread of
corrosion in painted steels. The better performance of galvanized steel as compared to
that of cold-rolled steel is attributed to the lower volume and the barrier effect of the zinc
corrosion products under a cyclic or a natural exposure condition. Standish et al. also
found that corrosion product does not build up beneath the paint in a salt spray test, while
it does in cyclic or natural exposures.
van Ooij and co-workers [978, 994] noted that the adhesion of paint does not strongly
affect corrosion creeping at damaged areas such as at scribes of zinc-coated steels.
According to their corrosion model, as shown in Fig. 11.20, the creeping process (at a
scratch in a wet/dry test) starts with anodic dissolution of the exposed zinc layer, with the
bare steel as the cathode. The gradual formation of zinc corrosion product causes a
wedging effect to delaminate the phosphate crystals at the zinc and phosphate interphase.
The cathodic area in this initial stage is behind the corrosion front. As the corrosion front
only

~~Iili--~"""'''''U phosphate

'iimr~

Attacked phosphate
above chloride

Chloride only

Phosphate

I~~~~~~~~

. pro ducl / ' "


CorroSIOn

~~~$~~~~~~
Scribe
FIGURE 11.20. Schematic lOp view and cross section of cosmetic corrosion propagation mechanism in painted
precoated steels. From van Ooij and Sabata [311. Copyright by NACE International. All Rights Reserved by
NACE; reprinted with permission.

336

CHAPTER 11

propagates farther from the scratch, small corrosion cells form ahead of the corrosion
front, and, when the corrosive species start to diffuse through the paint in quantity, these
areas start to serve as the major cathodic areas on which an alkaline environment is
generated. At this stage, the cathodic area is ahead of the corrosion front. This results in
a partial, and eventually a complete, alkaline dissolution of the phosphate crystals. Once
the cathodic area is ahead of the corrosion front, and dissolution of phosphate crystals
occurs, the overall rate of under-film corrosion is increased.

12
Zinc-Rich Coatings
12.1. INTRODUCTION
Zinc-rich coatings, which contain zinc dust and a binding medium, have been widely
used since the 1930s for the protection of steel structures from corrosion [345, 1159).
They can be used as a primer or as a final coating and can be applied by spray, dip, or
brush. As coatings, they possess a number of advantages over galvanized coatings and
regular organic paints: they are easier to apply than galvanized coatings, particularly for
large or odd-shaped articles, and, unlike regular organic paints, they provide galvanic
protection at edges or the places where the paint coating is defective.
In contrast to other types of applications of zinc, where the resistance to corrosion
is of primary importance, the role of zinc dust in zinc-rich paints is not to resist corrosion
but to react with moisture and oxygen, especially in fresh coatings, and in the process to
strengthen the paint and to galvanically protect the coated steel. The corrosion of zinc
dust is thus an essential part of the corrosion protection mechanism of a zinc-rich paint.
The material presented in this chapter is organized into three sections. First, the basic
components and characteristics of zinc-rich paints are briefly described. The corrosion
protection mechanism is then discussed, and finally the effects of various factors on the
performance of zinc-rich paint are presented.
12.2. COATING CHARACTERISTICS
There are basically two broad categories of zinc-rich coatings-organic and inorganic. The classification of a coating as organic or inorganic depends upon the chemical
nature of the binder used to bond the zinc particles to each other and to the steel surface
(1159). The organic type uses binder materials such as phenoxies, catalyzed epoxies,
urethanes, chlorinated rubber, silicones, or vinyls. The inorganic type usually employs
sodium silicate or ethyl silicate as the binder. Potassium, lithium, and ammonium silicates
are also used in inorganic coatings.
Most modem inorganic zinc-rich coatings are of the self-curing, as opposed to the
postcuring, type. They are further divided into water-reducible and solvent-reducible.
Generally, those using alkali metal silicates are in the water-reducible group, which cure
during and after evaporation of water from the coating. The solvent-reducible group
consists primarily of the partially hydrolyzed alkyl silicates. A special case involves
337

338

CHAPTER 12

organic silicate vehicles, such as partially hydrolyzed tetraethyl silicate. When such a
vehicle cures, the organic portion is hydrolyzed or evaporated, leaving an inorganic
silicate film.
Because galvanic action is an important protection mechanism of zinc-rich coatings,
both organic and inorganic types of coatings must be electrically conductive to ensure
galvanic action between the zinc particles and the steel substrate. Binders of either type
do not have good electrical conductivity. Thus, the coatings must rely on a heavy load of
fine zinc particles in contact with each other to provide the electrical continuity required
for galvanic action to occur.
Additives and extenders are also added to zinc-rich coatings to obtain specific effects.
For example, addition of iron phosphate at levels of 10-20% improves the weldability of
paints such as epoxy ester paints [1159]. A small amount of carbon black is sometimes
added to enhance the conductivity. Red or yellow iron oxides may be added to give the
paint tint. Addition of a small quantity of calcium oxide can reduce gassing from the
reaction between zinc dust and water or acid. Zinc oxide, when added to zinc-rich paints
at levels of about 20%, has been found to be of special value when the paints are to be
applied to galvanized steel, which is more difficult to paint than steel because of adhesion
problems. Zinc oxide has also the effect of maintaining coating flexibility. The maintenance of mechanical properties by zinc oxide is attributed to the absorption of ultraviolet
light and to its reaction with organic decomposition products. which tend to catalyze
further degradation [569]. The chemical effect of zinc oxide in a coating can also be due
to pH buffering [839]. Other additives such as iron and lead have been used in epoxy-based
coatings to control operation conditions [507].
A sheet steel coated with zinc-rich coating, known as Zincrometal, has been widely
used in the automotive industry [5, 1159]. The Zincrometal coating consists of a basecoat
called Dacromet and a topcoat called Zincromet. Dacromet is a water-based pretreatment
consisting of a combination of zinc dust. chromic acid, and other chemicals. Zincromet
is a solvent-based zinc-rich organic coating consisting of approximately 50 vol %
(85 wt. %) zinc dust contained in a thermoplastic phenoxy resin. The typical thickness of
the coating is about 12 ,urn.
Zinc-rich coatings are most often used as primers. They are topcoated with other
paints, not containing zinc dust in many cases, to provide longer lasting protection.
Generally, organic zinc-rich paints require less rigorous surface preparation than inorganic zinc-rich paints. They are more compatible with a variety of topcoats. more flexible,
and less dependent on atmospheric moisture during curing but more difficult to apply
under high or low ambient temperatures, as compared to the inorganic ones [1159]. On
the other hand, inorganic coatings generally provide better corrosion protection, better
toughness and abrasion resistance, faster drying under optimum conditions, and better
temperature resistance. Different zinc-rich coatings, depending on the type of binder, offer
varying degrees of advantages in terms of economics, amount of protection afforded, and
ease of application and topcoating [506].
The formation process of an inorganic zinc-rich coating is very different from that
of organic coatings. According to Munger [345], several reactions are involved in the
formation process of inorganic zinc-rich coatings. The first reaction is concentration of
the silicates by evaporation of the solvent, providing the initial drying and primary
deposition of the coating. The second reaction is insolubilization of the silicate matrix by

ZINC-RICH COATINGS

339

TABLE 12.1. Oxygen Permeability of Zinc-Rich Paints (72% Zinc by Volume)"


-----------------------------------------------Oxygen reduction CUiTent (flA)
Fresh paint
Paint type

Dry

Aged paint"
Wet

Dry

Wet

---------------------------~------------------------~------------------

Chlorinated rubber, 125-175


flm thick
Epoxy polyamide, 100-150
11m thick

104

1.4

28

1.0

16

0.2

12

0.2

"Data from Ref. 726.


/, Aged by immersion in O.SM Nae] for six months.

reaction with zinc ions from the surface of zinc particles and iron ions from the steel
surface. At this stage, the vehicle hydrolyzes to a form of silicic acid with which zinc ions
react to form an insoluble silica-oxygen-zinc polymer. At the zinc silicate/steel interface,
an iron-zinc-silicate complex is formed. The zinc silicate and iron-zinc-silicate matrix,
surrounding the surfaces of zinc particles and the steel substrate, is very inert to water,
weather conditions, solvents, and many other chemicals. The third reaction that takes
place within the coating occurs over a long period of time--a number of months or years.
During this time, a reaction of zinc compounds with carbon dioxide from the air results
in formation of zinc carbonate, which fills the pores within the coating. In addition, a
reaction of the alkali in the coating with the moisture from the air releases silicic acid in
the silicate polymer. This reaction gradually proceeds through the coating to the steel
interface, increasing the adhesion of the coating to the steel.
The permeability of zinc-rich coatings is affected by many factors. The data in Table
12.1, reported by Ross and Wolstenholme [726], show that dry paints are much more
permeable to oxygen than wet paints and that the chlorinated rubber-based paint is more
permeable than the epoxy polyamide-based one. Aged paints have lower permeability
than fresh paints owing to the blockage of pores inside the paint by corrosion products.
Also, the permeability of the chlorinated rubber coatings to Cl- ions was found to be much
lower than that of the epoxy coatings.
12.3. PROTECTION MECHANISM
The protection of steel by a zinc-rich coating is mainly via two effects: the barrier
effect and galvanic action. While the coating serves as a physical barrier to the environment in which the coated steel is used, the zinc particles in the coating provide the galvanic
protection. In addition to the galvanic action and the barrier effect of zinc-rich coatings,
two other effects that are also beneficial in terms of corrosion protection have been
identified: (a) the absorption of oxygen by the zinc powder and (b) the inhibition of the
oxygen reduction by the zinc corrosion products on the steel surface [1191]. The
remaining zinc particles, after their insulation from the steel surface, later react with
oxygen and thus reduce the amount of oxygen, arriving at the steel surface. In the case of

340

CHAPTER 12

the oxygen that does arrive at the steel surface, the reduction reaction is hindered by the
zinc corrosion products.
In order for a galvanic process to occur on a steel with a painted zinc-rich coating,
three conditions must be satisfied:
1. The zinc particles in the coating must be in electrical contact with each other.
2. The zinc particles must be in electrical contact with the steel.
3. A continuous electrolyte must exist between the zinc particles and the steel.
The first two conditions are met by zinc-rich coatings containing a sufficiently high zinc
content. The third condition is fulfilled when a steel panel bearing a zinc-rich coating is
wetted by a film of electrolyte such as a salt solution.
There are two stages in the protective action of zinc-rich coatings [5,726,780]. The
first stage is a relatively short period in which galvanic protection of the steel by zinc
particles is in effect. After this period, the galvanic action between the steel and zinc
gradually disappears. The second stage is a long-term barrier protection that is attributed
to (1) greater resistance of the coating to the permeation of aggressive species such as
water, oxygen, and salts because the pores in the coating are blocked by the zinc corrosion
products or (2) inhibition of the steel surface by the zinc corrosion products.
The galvanic action generally decreases with time. The loss of galvanic protection
is due to (1) the loss of electrical contact between zinc and steel as a result of corrosion
of zinc particles and formation of nonconductive corrosion products at the interface; (2)
the loss of electrical contact between zinc particles as a result of the formation of corrosion
products on the surface of the zinc particles; or (3) blockage of the coating surface by
zinc corrosion products. However, the galvanic protective action of a zinc-rich paint at a
later stage is still latently present and may activate at any time after mechanical damage
of the paint or chemical removal of the barrier effect of the corrosion products. For
example, a commercial zinc-rich paint, after being weathered for 10 years, had a layer of
zinc corrosion products removed and afterward showed quite a normal galvanic action
[780].
The galvanic effect of a coating can be evaluated by measuring the potential of a
coated steel sample having a certain area of bare steel surface or by measuring the current
between a coated sample and an uncoated sample [780,965, 1172]. Alternatively, galvanic
protection can be evaluated by visually observing signs of red rust on a coated sample at
cuts or other defects on the coating.
The barrier protection of zinc-rich coatings is often evaluated by the use of electrochemical impedance techniques. Impedance spectra of zinc-rich coatings generally
display a behavior corresponding to a high-frequency charge-transfer process in series
with a low-frequency diffusion process [432, 805]. Owing to the formation of corrosion
products which block the pores inside the paints, the impedance of many zinc-rich
coatings gradually increases with exposure time in the testing environment [432, 805,
813, 1160]. Thus, in general, the barrier effect of zinc-rich coatings on the diffusion or
permeation of aggressive species increases with time. Armas et al. [805] reported that the
impedance of steel coated with zinc-rich paint decreased with immersion time in artificial
seawater. Pereira et at. [432] found that the high-frequency impedance of a polyamide
cured epoxy and an ethyl silicate zinc-rich paint in a marine environment was associated
with charge-transfer reactions and increa~ed with time. Feliu et at. [1160] measured the

ZINC-RICH COATINGS

341

impedance of several zinc-rich paints after three years of atmospheric exposure and found
that the barrier effect of the paints increased with exposure time. They also found that the
paints still maintained a large part of their capacity for providing cathodic protection,
especially in the case of an ethyl silicate vehicle.
12.4. PERFORMANCE

12.4.1. Effect oJZinc Content


Zinc dust content and particle size are two important factors determining the extent
of galvanic protection provided by a zinc-rich coating, through their effects on the specific
surface area and on the electrical continuity between the steel substrate and zinc particles
and also among zinc particles. In general, for each type of zinc-rich coating, there is a
certain zinc level below which the galvanic protection of the steel substrate by the coating
is not effective. Figure 12.1 shows that the resistance to red rust of steel samples coated
with two inorganic coatings under salt fog greatly increases with zinc content, particularly
in the range of 40-60% by volume [513]. Table 12.2 shows that the effectiveness of
inorganic zinc-rich coatings on a steel surface with mill scale increases with zinc content
[1162]. Ross and Wolstenholme [726] examined the effect of zinc content on the galvanic
protection of a bare steel area on steel panels coated with zinc-rich paints. They found
that the coatings with low zinc contents provided no cathodic protection to the bare steel
surface.
The effect of zinc dust content on the protection of steel strongly depends on the type
of binder and pigment materials [506,513,805]. Theiler [780] found that there exists an
optimum zinc content level for each binder material that yields the longest cathodic
protection for damaged coating spots, as shown in Fig. 12.2. According to Theiler, if too
much zinc is incorporated into the paint, then the anodic area is large in comparison with
the cathodic area, and, as a result, a small current density is drawn from the active zinc.

10

I'"

n.
o

OJ

'"

Rusting: 10 = no rust
completely rusted

o=

~-------------~~~------

-Pigment A
Pigment B

IT:

~~0~-----------5LO-------------6~0------------~70
% Zn in dry film

FIGURE 12.1. Results of 4000-hour salt fog tests on primers made from silicate vehicle with two types of
pigments and varying levels of zinc (vol. 'Yo) in the dry film. Data are taken from Ref. 513.

342

CHAPTER 12

TABLE 12.2. Percent Zinc in the Dry Film of Inorganic Coatings vs. Performance over Mill
Scale"
Zinc level (wt. %)

Dry film thickness (mil)

2.5

20
40
60

2.6

70
85

2.2
2.5

87.5

2.4

Time to failure h
2 months
2 months
2 months
4 ('ears
52 years
No failure, less than I % rust after ~
years
No failure, less than I % rust ~ years

2.1
2.3

"From Montie and Hasser [1162). Copyright by NACE International. All Rights Reserved by NACE; reprimed with
permission.

The pH of the electrolyte in the vicinity of the zinc particles does not change under this
low current density. The zinc surface is thus passivated by the formation of insoluble
corrosion products in the neutral or slightly basic electrolyte and loses its protective action
after a short time. On the other hand, if there are too few zinc particles present within the
paint, the anodic area is very small and the anodic current is very high. The active zinc is
consumed very rapidly, and the cathodic protection is quickly lost.
The available zinc content can also be evaluated by measuring the apparent polarization resistance. Figure 12.3 shows that the apparent polarization resistance of steel
samples coated with zinc-rich paints decreases with increasing zinc content. The total
zinc surface area is increased owing to the increase in the total number of zinc particles,
which results in more contacts among particles [780].
Thomas et at. [996] determined the available zinc content (the amount of zinc
available to afford cathodic protection) in zinc-rich coatings by measuring the galvanic
current between painted and bare steel samples. They defined the amount of zinc available
40 r-----------------------------------~

.,"'>-

'C

C 30
.9

POlystyrene

.. Chlorinated rubber

+ Vinyl copolymer

(3

o
Q)

~20

'6
a
.c:

iii

o
'010
QI

i=

O~----~------~-------L------~------~

50

55

60

65

Zinc content, vol%

70

75

FIGURE 12.2. Time of cathodic protection as a function of zinc content of


zinc-rich paints with different binders
under immersed conditions. Reprinted
from Theiler [780], with kind permission from Elsevier Science Ltd, The
Boulevard, Langford Lane, Kidlington
OX5 1GB, United Kingdom.

ZINC-RICH COATINGS

343

100,000.-------------------------------------- ,
~

Chlorinated rubber

s:}-

Polystyrene

+ Vinyl copolymer

10,000
NE

:-'

c:

1,000

0:0.

100

10L---------~----------~----------~~

50

60

70

80

Zinc content, vol%

FIGURE 12.3. Apparent polarization resistance in relation to the zinc dust content for zinc-rich paints with
different binders. Data are taken from Ref. 780.

for galvanic protection as that which is in direct contact with the steel substrate; hence.
the available zinc content is independent of coating thickness. Field exposure data show
that paints having less than I 0% available zinc fail to prevent rust formation in scribe
lines at 6 months whereas those with greater available zinc are all successful in preventing
red rust formation at 12 months. The variation of available zinc with vehicle type was
explained on the basis that some vehicles are more efficient wetting agents than others.
12.4.2. Effect oJZinc Particle Size

Zinc dust can be classified as regular. fine. or super fine. The typical properties of
each class are given in Table 12.3 [822. 1159J. There is a logarithmically linear relation
between the mean diameter and the specific area of zinc dust samples.
Zinc particle size affects the zinc content available for cathodic protection [996J.
Figure 12.4 indicates that the available zinc content is the highest for particle sizes
between 2.5 and 3 ,urn. It has been reported that a particle size between 5 and 7 ,urn
represents a good combination of particle-to-particle contact and specific area [506J. In
one study, the impedance responses of two epoxy zinc-rich paints were found to depend
strongly on the particle size distribution [432J. In another study, it was reported that a

TABLE 12.3. Typical Properties of the Three Classes of Zinc Dust"


Property
Median diameter (pm)
Percent smaller than 10 f1.m
Percent under 44 f1.m
Surface area (m 2/g)
Apparent density (glcm3)
"Ref. 822.

Regular
8.0
65.0
98.0
0.1
3.04

Fine
5.0
95.0
99.0
0.16
2.4

Super fine
4.0
99.0
100.0
0.2
2.2

344

CHAPTER 12

30
ill

:0
C1l

.(ij

20

u
c:

'0

*" 10
O~--------------~---------------L------~

Pigment fisher size, J.lm

FIGURE 12.4. Effect of particle size on percent available zinc for cathodic protection in silicate vehicle. Data
are taken from Ref. 996.

particle size of 2.5 .urn provides the best protection in a flowing electrolyte [728]. There
exists a balance between the magnitude and length of galvanic action. The optimum
particle size should be that which provides the right balance for the specific application.
12.4.3. Effect of Binders
Binder material is one of the most important factors determining the effectiveness
of the corrosion protection provided by a zinc-rich coating. Figure 12.2 shows that the
effectiveness of galvanic protection can differ greatly among coatings using different
binders.
The apparent polarization resistance for a given zinc content also varies with the type
of binder material, as shown in Fig. 12.3. The polarization resistance was considered by
Theiler [780] to be related to the wettability of the binder to the zinc particles: the better
the wetting, the less the active zinc surface area, and thus the larger the polarization
resistance. More particles are sealed by the binders with better wetting properties and thus
are not available for reactions with the electrolyte.
Feliu et al. [1160] measured the electrochemical behavior of zinc-rich paints exposed
in the field for up to three years. They found that the coatings retained a considerable
amount of their galvanic protection capacity, especially those of the ethyl silicate type,
and that the barrier effect increased with time, particularly for coatings of the epoxypolyamide type. Lindqvist et at. [965] reported that commercial ethyl silicate-based paints
provided much longer cathodic protective action than epoxy-based ones in a 60-day
immersion test in O.IM NaCI solution.
Some coatings still provide galvanic protection when a topcoat is applied, while
others do not. In one study [726], it was reported that a zinc-rich coating based on
chlorinated rubber with a top coating provides cathodic protection to a bare steel surface
at defective spots in the coating, while one based on epoxy polyamide does not; on the
other hand, when the coatings are not topcoated, both zinc-rich paints provide cathodic

ZINC-RICH COATINGS

345

-0.6r----------------------.,
a - Initial
immersion
b - After 1 day
c - After 3 days
d - After 14 days

-0.7r-_ _ __
w

-0.8

C
0

-0.9

Q)

c..

-1
-1.1
0

Current, f.lA

FIGURE 12.5. Polarization curves for the galvanic couple comprised of a steel specimen coated with 73% (by
volume) zinc-dust paint with chlorinated rubber binder coupled at the centerto a piece of bare steel (0.072-1.27
cm 2 in area) measured in 0.5M NaC!. Reprinted from Ross and Wolstenholme [726], with kind permission from
Elsevier Science Ltd, The Boulevard. Langford Lane, Kidlington OX5 1GB, United Kingdom.

protection to the bare steel area. Because of the lesser galvanic action between the bare
steel and the painted area for the epoxy polyamide paint, there is also less deposition of
the corrosion product on the steel surface than in the case of the chlorinated rubber paint.
In addition, it was found that the corrosion product deposited on the steel surface has an
effect of cathodic inhibition. Figure 12.5 shows that for a chlorinated rubber paint
immersed in 0.5M NaCl solution for 14 days, the open-circuit potential of the steel
significantly shifted in the cathodic direction. The fact that the open-circuit potential of
the steel after 3 days of immersion was more negative than -900 m V seE indicates that the
corrosion products deposited on the steel surface probably contained some metallic zinc.

12.4.4. Effect of Coating Thickness


In general, the life of a zinc-rich coating increases with coating thickness. Figure
12.6 shows an example for an organic coating in an accelerated test [996]. As another
example, Table 12.4 shows the effect of coating thickness on the corrosion performance
in salt fog for an ethyl silicate zinc-rich primer [1162]. According to Thomas et al. [996],
the amount of zinc particles available for galvanic protection is the amount that is in direct
contact with the substrate. Since the amount of zinc particles in direct contact with the
substrate does not vary substantially with coating thickness, the increase in the coating
life with increasing thickness must be associated with a stronger barrier effect for the
thicker coatings.

12.4.5. Effect of Additives


Fawcett et al. [10 10] investigated the effect of Fe2P as a pigment on the corrosion
performance of an epoxy ester zinc-rich coating. Figure 12.7 illustrates that a pigment
with a composition of 50 wt. % Fe2P and 50 wt. % zinc provides the best performance in
a salt spray test. The beneficial effect of Fe 2P is associated with Fe2P serving as cathodic
areas for O 2 reduction. In another study [1026], Fawcett et al. found that the presence of

346

CHAPTER 12

300,---------------------------------------,
(I)

:;

With FS 3.7/1 dust

250

'" With FS 2.4/1 dust

.s=:
ai 200

E
~

:=

150

"t:l
Q)

Cil

Q; 100

a;

()
()

<t:

50
o~----------~----------~------------~

0.05

0.15

0.1

Coating weight, g / em'

FIGURE 12.6. Effect of thickness of zinc-rich coatings on accelerated life for polystyrene vehicle. Data are
taken from Ref. 996.

ZnO in zinc dust at levels between 18 and 27% is beneficial to the performance of the
coating. The effect of zinc oxide was attributed to the reduction of the total anodic activity
of the zinc dust, thus balancing the cathodic and anodic activities. Chua and Ross [729]
also reported that the addition of zinc oxide reduces the zinc content in the coating without
reducing its functional properties.
The effect of the addition of pigments on the performance of zinc-rich coatings is
complicated. Fragata and Mussoi [813] found that the presence of extender pigments
affects only the physical and chemical characteristics of zinc-rich coatings, but not the
galvanic protection. However, in some cases, as shown in Fig. 12.1, the corrosion
resistance of zinc-rich coatings may vary significantly with changes in the type of pigment
used [513].
Szauer and Miszczyk [1161] modified zinc-rich paint by replacing a small percentage of zinc with carbon black or zinc phosphate. Carbon black was found to decrease the
duration of both the cathodic protection and the long-term barrier protection, whereas

TABLE 12.4. Ratings for Inorganic Zinc-Rich Coatings of Different Thicknesses after Salt Fog
Exposurea .b
Months in test

Dry film
thickness (urn)
5.33
7.11
10.67
12.45

9
9
10
10

15

6
7
10
10

2
3
7
7

4
5

1
4
3

"From Montie and Hasser [1162]. Copyright by NACE International. All Rights Reserved by NACE; reprinted with
permission.
"Rating scale ranges from 10 (best) to 0 (poor).

ZINC-RICH COATINGS

347

8,------ - - -- - - -- -- -- -- - - -- - - -------1

60

20

100

80

Wt % Zn in p igment

FIGURE 12.7. Degree of rust on steel substrate after 200-hour salt fog test at 50
(mixture of zinc and Fe2P)' After Fawcett el al. [10101.

e VS. zinc content of pigment

zinc phosphate was found to have no adverse effect on the period of cathodic protection
but increased the period of the long-term barrier effect.
12.4.6. Effect qf Surface Condition
Mayne [1054] studied the effect of surface condition on the galvanic protection of
steel provided by a zinc-rich coating. Figure 12.8 shows that the coating applied on a
rusty steel surface (exposed to the atmosphere for one year) provided less cathodic
protection to the steel than that applied on a clean surface (pickled). It was also found that
the coating on mill scale, which was removed to various extents through weathering,
-0.3 .----------------------------~P~ic~k7Ie-d~--------;

- - - Unpa inted specimens

>
Qj

-0.5

Rusty. wire-brushed

+ Rusty

Painted specimens

~,

... Electrolytio

zinc

T _

<?

... T ___ -..- _ _ _

(3

__ .... _____ ... ...

..

::.:: -0.7

"lii

!e.
~

~ -0.9

a
a.

-1 .1

--- . --- .- -- .---- -~--- .---.


L -____L-____

10

20

____

30

____

40

_____ L_ _ _ _

50

_ __ _

60

70

Time, days

FIGURE 12.8. Potential-time curves for painted and unpainted steel specimens partially immersed in natural
seawater. Painted specimens are steel samples with various surface conditions (indicated on the figure) on which
a zinc-rich coating was applied. Data are taken from Ref. 1054.

348

CHAPTER 12

provided cathodic protection to the steel. Two possible explanations were suggested:
either (a) the rust on the surface may not have been continuous so that zinc particles may
have been in contact with the steel surface or (b) the rust may have been reduced to a
certain extent by the zinc, thereby becoming conductive. Long-term exposure showed
that steel samples that had been weathered for six months before being coated remained
in good condition after four years of atinospheric exposure. According to Mayne. the best
performance was obtained when the exposed steel was coated in the summer. Hendry
[513] reported that zinc-rich coatings over mill scale perform well.

12.4.7. Other Factors


Figure 12.9 shows the potential of steel samples coated with four commercial ethyl
silicate zinc-rich coatings immersed in O.IM NaCl solution. For the samples with 91 %
and 87% zinc, the galvanic effect only lasted for about 10 days, while for the samples
with 83% and 93% zinc it lasted for more than 60 days. indicating that factors other than
the base binder material and zinc content can also be important in determining the length
of galvanic protection by a zinc-rich coating.
Tanabe et al. [743] reported that zinc-rich paints, when used as primers. greatly
increase the durability of epoxy top coatings, as shown in Fig. 12.10. Zinc-rich coatings
when used without a topcoat tend to perform better with higher zinc loadings in severe
conditions while coatings with lower zinc loadings tend to perform better when topcoated
in less severe conditions [506]. Generally, a topcoat reduces the galvanic activity between
the zinc particles and the steel substrate [726].
Ross and co-workers [727-729, 731] investigated the effect of solution flow rate on
the galvanic protection of a zinc-rich coating on steel, using a rotating coated sample with
a bare surface in the center. They noted that the anodic current of the zinc-rich coating
increased significantly with increasing flow rate of the 0.5% NaCl solution. At high flow
rates, an ennoblement of the zinc-rich coating was observed. The rise in the potential was
-0.5r------------------------,
91 wt%

-0.6

67 wt%

"""01-0.7

]! -0.8

"E
Ql

63 wt"A.

(5

c.. -0.9

93 wt%

solid zinc

-1

-1.1~--~--~--~--~--~--~-~

10

20

30

40

50

60

Time, days

FIGURE 12.9. Electrode potential as a function of time during exposure in aerated O.IM sodium chloride
solution at 25C and pH 7 for four commercial ethyl silicate zinc-rich paints on steel and for solid zinc metal.
The zinc contents of the paints are labeled on the curves. After Lindqvist et al. [965].

ZINC-RICH COATINGS

349

1,000,----------------------------------------,

100
3
(j)

10

C
<ll

u;

'0;

~
NO.1

.
o

0...
0.01L-------------------~---------L--------~

100

200

300

400

Immersion time, days

FIGURE 12.10. The change of polarization resistance of epoxy-coated steel with immersion time in 3 wI. %
sodium chloride solution. (1) and (3) Zinc-rich paint used as a primer; (2) and (4) without zinc-rich paint. After
Tanabe et al. [743 J.

attributed to the formation of passive zinc corrosion products caused by cathodically


produced hydroxyl ions, which flow from the center to the edge of the plate. An annular
band, the size of which depended on the flow rate, of thin gray corrosion product was
also formed on the steel adjacent to the surrounding coated surface. while a brown stain
was formed in the center of the bare steel area.

13
Corrosion in Concrete
13.1. INTRODUCTION
Steel-reinforced concrete is a widely used construction material. In the absence of
certain deleterious factors and agents, steel does not rust in the concrete environment.
However, in practice, the entrance of moisture, salts, and oxygen by diffusion or through
pores or hairline cracks in the concrete may cause the steel to corrode and may eventually
result in cracking and spalling of the concrete by the expansive force of the corrosion
products [269, 1219].
The problems resulting from corrosion of reinforcing steel in concrete are widespread and very serious. This is especially true for the bridge decks on highways where
deicing salts are used. Beginning at about the end of the 1960s, severe deterioration of
many reinforced concrete bridge decks in the "snow belt" states of the United States was
noted. Later, assessment of the bridges constructed in the 1960s and 1970s with black
steel reinforcement showed a drastic increase in the deterioration of the bridge decks and
substructures of highway bridges in the United States [1217]. The percentage of bridges
that were categorized as suffering moderate to extensive deterioration increased from
about 15% in 1986 to around 35% in 1988 and to over 50% in 1990 [1220]. Corrosion is
generally also a serious problem for steel-reinforced concrete structures in marine
environments.
The use of galvanized steel rebars as concrete reinforcement is one of the remedies
that have been suggested to alleviate the corrosion problems [268,271]. Galvanized steel
was used for concrete reinforcement as early as the 1930s and has been successfully used
in many concrete structures. The performance of galvanized steel inside concrete has been
investigated in a number of studies. These studies can be grouped into three categories
according to the experimental conditions applied in the corrosion tests: (1) studies of
rebars embedded in concrete and exposed to natural or field environments; (2) studies of
rebars embedded in concrete and exposed to accelerated-test environments; and (3)
studies of rebars exposed to simulated concrete solutions. In this chapter, only the
information obtained from tests using concrete samples, i.e., falling into categories 1 and
2, is discussed. The information obtained from the solution type of tests, which do not
have a real concrete environment and are only marginally relevant to the corrosion inside
concrete, is presented in Chapters 5 and 9.
351

352

CHAPTER 13

13.2. CONCRETE ENVIRONMENT

13.2.1. Formation of Concrete


Concrete is a two-phase material in which a mineral aggregate is dispersed in a matrix
of hardened cement paste [1211]. It is made by mixing together cement, sand, gravel, and
water, in proper proportions. Cement is a pulverized material which, when combined with
water, forms a paste and, after setting, eventually becomes a hard solid. Hardened cement
paste is the fundamental material as it provides the strength that allows concrete to be
used structurally. The aggregate, i.e., sand and gravel, is also essential in that it provides
rigidity and dimensional stability to the concrete.
The raw materials of the commonly used portland cements are limestone and clay.
These are fine-ground and sintered in special kilns at a temperature of 1400-1500C. The
product of the sintering process, called "clinker stone," is ground to dust together with a
few percent of gypsum (CaS0 4 2H20) and other additives to become cement. The typical
composition of portland cement is 23% Sial> 6.5% A1 20 3 , 3% Fe z0 3, 64% CaO, 0.6%
MgO, and 2.1 % SOz [272].
Upon contact with water, hydration occurs with simultaneous dissipation of heat.
Table 13.1 shows the reactions involved during hydration [272]. During hydration, the
water is absorbed at a high rate by the cement particles, especially in the initial phase. In
the later stages of hydration, the diffusion of water into the interior of the particles takes
place at a slower rate and less uniformly. The cement is hardened during the hydration
process in several stages, as schematically illustrated in Fig. 13.1 [1211]:
(a)

Immediately after mixing, the cement paste is in its most fluid state. The cement
grains are dispersed in the mixing water, the spacing being determined by the
water/cement ratio.
(b) After two hours, the cement paste is much less fluid but can still be worked.
Hydration on the surface of the cement grains occurs.

TABLE 13.1. Hydration of Portland Type Cements"


Hydrated clinker mineral

Clinker mineral

2CaSiO z(n)H 20 + Ca(OH)2


CaOSi0 2(m)H 20 + Ca(OH)2
3CaO Al 20 y 12H20 + CaO Fe20]
3CaOAI 23'6H20

3CaOSi0 2
2CaOSiOz
4CaO Al z0 3FeZ03
3CaOAI 20 3

Product of hydration

Accessory substance
Free CaO
CaS04 H20

Ca(OHh
3CaO A1 20 33CaS04' 31 H 20"
3CaO Al 20 3CaSO4.12H 2d
Mg(OHh c

MgO
"Ref. 272.
"With the help of CaOAI 20

3.

'The hydration of MgO present in crystalline form (periklase) is a very slow process;
consequently, hydration accompanied by swelling in the hardened concrete is detrimental.

353

CORROSION IN CONCRETE

~ . . . Unhydrated
~ _
"mom p.n,,'"

131

FIGURE 13.1. Schematic representation of the stages in the hydration


of cement particles and the formation
of a cement gel. (Sec text for explanation.) After Newman [ 12721.

(c)

(d)

-I bl

Cement gel

Capillary pores
and cavities
lei

Id l

After a day, the cement paste is set but has no real strength. The hydration has
penetrated further into the grains. The hydrates in the intergranular spaces have
grown and interconnected, forming a continuous gel and thus establishing a
solid skeletal structure.
After seven days, the hardened cement paste has achieved considerable
strength. The skeletal structure has been further developed by infilling between
the original hydration links to produce a denser gel structure.

The reactions and the hardening process do not terminate within the first few weeks of
solidification but continue for a considerable time. The time required to reach each stage
can be altered by adjustments to the proportions of the compounds taking place in the
chemical reactions.

13.2.2. Characteristics
13.2.2. J. Porosity and Permeability. The hardened cement paste contains small gel
pores and large capillary cavities. Gel pores are formed as a result of the shrinkage during
solidification, whereas capillary pores result from the evaporation of excess, unbound
water [272]. The gel pores are about 2.5 nm in size, and they constitute about 20-30%
of the volume of the hardened cement (also called cement stone). The sizes of the capillary
pores are in the range of 1-10 ,urn, and the number of pores increases with increasing
amount of mixing water but decreases with the progress of hydration. In addition, there
are also air pores, whose size ranges from 0.01 to 2 mm and whose volume may range
from 1 to 10% of the total volume of the concrete [272].
Concrete is thus permeable to water, air, and other agents owing to the presence of
pores. The gel pores have an insignificant effect on the permeability of a cement stone
owing to their small size and isolated nature. The permeability of concrete to water is
predominantly determined by the capillary pores, the amount of which increases with
increasing water/cement ratio. Figure 13.2 shows the relationship between the permeability and the water/cement (W /C) ratio [633]. When the concrete is submerged in water,
the capillary pores fill first. Water penetrates into the air pores only after the displacement
of air by diffusion from the capillary pores.
The air (and thus oxygen) permeability depends on many factors, particularly the
water content of the concrete [1221, 1274]. A dry concrete has a large fraction of pores
unfilled with water, and these pores are available for air transport. On the other hand, in

354

CHAPTER 13

1.2

a.

..c
~

0.8

~0.6

'0
~
'u

~o

0.4

0.2

0.3

0.5

0.4

0.6

0.7

W/C ratio of cement paste (95% hydrated)


FIGURE 13.2. Effect of water/cement (W/C) ratio on permeability of cement mortar specimens. After Power
et al. [1273].

water-saturated concrete, air has to diffuse through the liquid phase, which is a much
slower process. This can be noted in Fig. 13.3, which shows that the flux of oxygen
through concrete increases with decreasing thickness of the concrete cover and relative
humidity and with increasing water/cement ratio [1274].
The diffusion coefficient of chloride in concrete has been reported to be of the order
of 10- 7-10- 8 cm 2/s [634-638]. There are a number of factors that affect the diffusion of
chloride in concrete. The W/C ratio is particularly an important factor affecting the
diffusion of chloride in concrete, as shown in Fig. 13.4 [634]. Figure 13.5 shows that the
diffusion coefficient decreases with time [634].

ff"'
b

.'"
E

10,000

)(

()

1,000

'0

.S-

100

'"
~
0

10

15

w/c=0.40
100% rh

x
:>

u:::

-__

w/c'=0.60

100%rh

- - - -_ _ _ _ __

25

50

75

Concrete cover, mm

FIGURE 13.3. Effect of relative humidity (rh) and water/cement (W/C) ratio on the flux of oxygen through
concrete. Data are taken from Ref. 1274.

355

CORROSION IN CONCRETE

3.----------------------------------,

.
(J

+ 1 year

Cl)

E
(J

.. 6 months

"1

E
Cl)

'(3

i:

8'" 1
c
a

'iii

:J

:t:

(5

oL-________________

- L_ _ _ __ _ _ _

0,2

0.4

________

0.6

0 .8

Water to cement ratio

FIGURE 13.4. Chloride diffusion coefficient as a function of water!cement ratio. From Lin r634]. Copyright
by NACE International. All Rights Reserved by NACE; reprinted with permission.

13.2.2.2. Moisture Content and Resistivity. As one of the major components, water
is held in a number of different states in the hardened cement paste [12Il]:
(a)
(b)
(c)
(d)

Water vapor: The voids in hardened cement paste contain water vapor, exerting
a vapor pressure corresponding to the relative humidity and temperature.
Free water: The free water is mostly located in the capillaries and in larger gel
pores.
Adsorbed water: Several layers of water molecules are adsorbed on the surface
of the gel pores.
Interlayer water: Some water can penetrate the lattice layers of gel solids or into
the intercrystalline spaces.
12
<..>
Q)

V>

Eu
~

"E
Q)

'0

i:
Q)
0
<..>

c:
0

'iii

::l

:t::

(5

200

400

600

800

1,000

1,200

Exposure time, days

FIGURE 13.5. Time-dependent chloride diffusion coefficients for two different concretes: (A) Concrete with
ordinary portland cement; (B) concrete with ordinary portland cement mixed with fly ash. From Lin 1634].
Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.

356

CHAPTER 13

4r-----------------------------,
W /C rat io : 0.50

2cm

r.

o
..>:

3cm

OL-------~------~-------3~
00----~

100

200

Duration of test, days

(e)

FIGURE 13.6. Changes of resistivity at different concrete depths during drying under constant climatic
conditions (20C, 50% relative humidity). Data are taken from Ref.
637.

Chemically combined water: This is the water combining with the unhydrated
cement in the hydration reactions and forming an integral part of the solid.

The moisture content of portland cement concrete varies widely depending upon the
environment surrounding it. Fully saturated concrete contains 13-15% water by weight,
but the moisture content in air-dried concrete can be as low as 3-5%, depending upon the
atmospheric humidity level [632].
The resistivity of the concrete is directly related to the amount of water present in
the concrete. Figure 13.6 illustrates that the resistivity increases with time during drying;
the increase is faster near the surface than in the interior owing to the faster rate of drying
near the surface [637]. Resistivity increases with decreasing W/C ratio, and, for a given
W/C ratio, it increases with decreasing cement/aggregate ratio [647].
13.2.2.3. Cement Solution. The water trapped inside concrete pores contains the
soluble species leaching out from the solids in the cement. Table 13.2 shows the typical

TABLE 13.2. Typical Composition of Cement Solution


(One Part Cement, Th~ee Parts Watert
Element
Ca 2+
Na+
K+
Fe 3+
A1 3+
SiO~SO~CI-

Ml+

Concentration (mgll)

930
220
660
<I
<I
<I

310
<10
<I

aFrom Maahn and Sorensen [177J. Copyright by NACE International.


All Rights Reserved by NACE; reprinted with permission.

357

CORROSION IN CONCRETE

-....

..

~ro

rn

0.1

O.01L---------------~-----------------L--~

12

14

13

pH

FIGURE 13.7. Concentration ofC}T in saturated Ca(OHh solutions as a function of pH. After Macias and
Andrade [175].

composition of the cement solution (made with 1 part cement in 3 parts demineralized
water) [177]. The pH of the aqueous solution formed within the concrete pores may vary
from 12 to 14 depending on the nature of the cement and its degree of hydration. During
the hydration, part of the CaO combines with water to form Ca(OH)2' The solution
saturated with Ca(OH)2 has a pH of about 12.6 and a Ca 2+ concentration of about 0.9 gil
[175, 177]. The concentration of Ca2+ ions decreases with increasing pH, and it changes
also with the ionic strength of the solution. Figure 13.7 shows the concentration of Ca 2+
ions in Ca(OHh-saturated solutions as a function of pH [175].
The pH of the concrete solution can be affected by the addition of other salts. The
addition of CaCl 2 decreases the pH of the pore solution whereas addition of NaCl hardly
affects it. Also, the intake of chloride from the external environment during service has
little effect on the pH of the pore solution [268,624]. On the other hand, it has been found
that the presence of a high concentration of metal chlorides, such as FeCI 2 and ZnCI 2,
lowers the pH of a Ca(OH)2 solution [1106]. When cement contains a large amount of
Na+, K+, or SO~- ions, the pH of the pore solution can be as high as 13-14 [632, 1213}.
J3.2.2.4. Chloride Content. Chloride salts affect the corrosion of steel in concrete
in several ways. First, cr ions cause breakdown of the passive film on the steel. Secondly,
chlorides are hygroscopic, attracting water and maintaining a moist environment inside
the concrete. Also, CI- ions increase the concrete conductivity, which typically enhances
corrosion activity.
Chloride ions can diffuse through the concrete from deicing salts on highway bridges
or from seawater in marine structures. They may also be mixed with the concrete as part
of the water or aggregate. In some cases, chloride ions are added as calcium chloride to
serve as a set accelerator. After hydration, some of the chloride ions are bonded in
insoluble compounds such as 3CaOAI 20,CaCI 210H 20, and only a part of the chloride
can be dissolved in the pore water [1039}. For example, the available NaCI is 0.01 % for
0.04% NaCI added to the cement and 4.92% for 5.87% added.

358

CHAPTER 13

13.2.2.5. Carbonation. When in contact with air, calcium hydroxide in concrete


reacts with carbon dioxide and is transformed into calcium carbonate. This process is
referred as carbonation [272]:

Carbonation reduces the pH of concrete; a pH as low as 6.5 has been measured in


completely carbonated mortar samples [454]. The penetration rate of carbonation may
vary greatly, depending on concrete mix and environmental conditions [276J.
13.2.2.6. Stability. Deterioration of concrete in a medium can be caused by [272]:
(a) leaching of the free lime
(b) dissolution of compounds by aggressive species in water
(c) the expansion effect due to the formation of gypsum and calcium sulfoaluminate
hydrate crystals
The deterioration in soft water is essentially caused by dissolution and subsequent
leaching offree Ca(OH)2. Leaching is rapid at first but becomes slower later. Because the
solubility of calcium oxide is about 100 times that of calcium carbonate, the presence of
a carbonate layer formed by carbonation has a barrier effect to leaching of the free lime
when the concrete is exposed to soft water.
13.3. CORROSION OF STEEL REINFORCEMENT IN CONCRETE

13.3.1. Effect of Corrosion


For a concrete exposed in nonpolluted environments, the steel reinforcement is free
of corrosion. The alkaline concrete environment results in passivation of the steel surface.
However, in many applications, the protective environment may be offset by the presence
of aggressive agents that have entered the concrete from the external environment, and
significant corrosion of the steel may occur [632-646]. Chloride salts are particularly
aggressive in causing corrosion of steel in concrete. The chloride threshold level for
corrosion of steel in concrete has been considered to be in the range of 0.65-0.77 kg/m3
[268, 1219].

Concrete

Corrosion products
FIGURE 13.8. Schematic illustration of concrete spalling.

CORROSION IN CONCRETE

359

Severe corrosion of the steel reinforcement may result in weakening and failure of
a concrete structure. Analysis of deteriorated concrete structures indicates that the
corrosion of steel, which produces voluminous corrosion products, is responsible for the
cracking and spalling that can lead to failure of the structure [269,271,642]. A schematic
representation of the cracking and spalling phenomenon in reinforced concrete is presented in Fig. 13.8.
J3.3.2. Protection Methods

Many methods have been developed for protecting steel reinforcement in concrete.
They include concrete modifications [645, 1219], the use of inhibitors [275,644, 1039],
cathodic protection [270, 277, 643], and the use of epoxy coatings [269,290], zinc wire
[1225], and galvanized zinc coatings [271, 274] on the steel inside concrete. These
methods can be generally divided into three groups:
1. Methods based on impeding entrance and transport of deleterious materials
(water, oxygen, salts, carbon dioxide, etc.) in concrete.
2. Methods based on modifying the electrochemical processes through the use of
inhibitors or by cathodic protection.
3. Methods based on modifying the surface of the steel by an organic or inorganic
coating.
Detailed information on the performance and features of many protection methods can
be found in several review articles [645, 1219]. In the following discussion, only the
information related to galvanized steel will be covered.
J3.3.3. Galvanized Coatings

Galvanized coatings on reinforcing steel have been reported in a number of studies


to be beneficial in delaying the corrosion of the steel and the consequent deterioration of
the concrete structure [268, 271, 274, 290, 468].
Bonding between reinforcement and concrete is essential for reliable performance
of concrete structures. The bond strengths for reinforcements in concrete are usually
determined by either a pullout test or a bending test [274, 654]. Many factors, such as
concrete mix and additives, curing conditions, and age, may affect the bonding between
the galvanized steel and concrete. The bond strength of galvanized steel is largely similar
to that of black steel [380,654]. Particularly, for deformed rebars the bond strength for
black steel and galvanized steel is essentially the same because the strength is mainly
provided by the mechanical interlocking between the ridges of the deformed bars and the
concrete [290].
It has been noted that hydrogen evolution may occur during curing, resulting in a
more porous interface between the galvanized steel and the concrete; this may affect the
bond strength [963]. The addition of a small amount of chromate, 30-70 ppm, to the
concrete has been found to suppress the evolution of hydrogen and thus increase the bond
strength [379, 963]. It has also been reported that chromating of the galvanized rebar
before embedding in concrete could prevent hydrogen evolution [177].
Annealing a galvanized steel has been found to be harmful to its performance [174,
397]. Annealing may produce a structure in which zinc-iron alloy layers extend to the

360

CHAPTER 13

coating surface, and the rate of hydrogen evolution may be increased as a result of the
presence of the alloy layer on the surface. Also, owing to the lack of ductility, an annealed
galvanized steel coating that has zinc-iron alloy extending to the outer surface is
considered to be unsuitable for reinforcement in concrete [380, 397].
13.4. CORROSION OF GALVANIZED STEEL IN CONCRETE

13.4.1. Testing Methods


Corrosion inside concrete is a very complicated phenomenon involving a large
number of factors. Many of these factors are uncontrollable and change irregularly with
time. Because corrosion inside concrete is a rather slow process, typically taking years to
show its effect, laboratory-simulation types of tests are often used to accelerate the
corrosion process.
In general, the tests conducted on the performance of galvanized steel in concrete
fall into three broad categories: (1) tests conducted under natural exposure using concrete
samples embedded with rebars; (2) tests carried out under simulated environments using
concrete samples embedded with rebars; and (3) tests conducted under simulated concrete
solutions using bare rebar samples. While the amount of data from simulated tests is
enormous, the data from natural-exposure tests, which are the only reliable information
for predicting the life of galvanized reinforced concrete structures, are limited. Particularly, there is a lack of data from heavy-deicing-salt environments [1219].
The available field data suggest that galvanized steel reinforcement may provide
longer life compared to black steel [268, 173]; however, the data from studies using
simulated test conditions [1218] are often in disagreement with these field data. It needs
to be emphasized that laboratory tests designed to accelerate the corrosion process inside
concrete, although providing useful information for the understanding of the various
factors affecting the process, may not provide reliable information regarding the performance of the reinforced concrete structure in practice. The corrosion performance in the
field involves the synergism of a large number of factors, many of which cannot be
controlled, whereas only a limited number of predetermined factors are controlled and
enhanced in laboratory tests. In particular, it is known that in atmospheric environments
zinc generally corrodes 10 to 100 times slower than steel; it may, however, corrode at a
similar rate as steel in a salt spray test. Similarly, it is known that in natural environments
the under-paint corrosion rate is much slower for painted galvanized steel than for painted
cold-rolled steel, but the opposite may be true in a salt spray test [767, 827]. The
accelerating factor of each corrosion test can be very different for different materials.

13.4.2. Field Test Results


13.4.2.1. Corrosion Rate. When galvanized steel is covered with a good-quality
concrete free of chloride, the corrosion rate is very low. As reported by Treadaway et al.
[271] and shown in Fig. 13.9, the mean corrosion rate over an exposure period of 14 years
in an industrial environment is typically in the neighborhood of 0.1 pmJyr for concrete
with a WIC ratio of less than 0.6. The corrosion rate increases with increasing WIC ratio,
particularly with a thin concrete cover (1.3 cm). The higher corrosion rate with the thin
concrete cover was found to relate to carbonation.

CORROSION IN CONCRETE

361

0.8 - - - - - - - - - - - - - - - - - - - - ,

~>

1.3 em

0.6 -

i0.4~

- 2.5 em
""'3.8 em

" 5 em

.2

'"

~ 0.2~
o--------~--------~--------~--~

0.4

0.6

0.8

Water}:ement ratio
FIGURE 13.9. Mean corrosion rate of a zinc coating in concrete over 14 years in an industrial e nvironment as
a function of water/cement ratio and concrete cover thickness. Data are taken from Ref. 271.

Mercille [220J reported that a zinc coating suffered no long-term corrosion when
embedded in concrete and exposed in an urban atmospheric environment. It was found
that most of the corrosion experienced by embedded galvanized steel in concrete occurred
during curing of the concrete. As much as 5 flm was corroded in the first day during setting
of the concrete, but there was little change thereafter, less than 10 Jim being corroded
during exposure for 10 years.
High corrosion rates of a zinc coating may occur when the concrete is contaminated
by chloride salts. Treadaway et al. [271] reported that, after five years of industrial
atmospheric exposure, the corrosion rates for galvanized steel were low when the chloride
concentration in the cement was below 1% (equivalent to 1.7 kg/m' for concrete with a
25 ~------------------------------~-,

.,~

20

>

:1..15

0;

_ . 20 mm; W/C 0.6

-'- 10 mm; W /C 0.6

20mm; W /C 0.75

.. 10mm;W/CO.75

g 10

' Vi

5
O~0~:~-~-~&~~L---------2
L---------3~--Amount of chloride added in cement. %

FIGURE 13.10. Effec t of chloride added in cement on the mean corrosion rate of zinc coating exposed in an
indu strial atmosphere for three years for different combinations of cover thicknesses and water/cement (W/C)
ratios. Data are taken from Ref. 271.

362

CHAPTER 13

60
_ 50

-. shorlblasted steel

t'O

+ pre rusted steel

Q)

>

E40

+!

,,

galvanized steel

:::l...

105

,,

.. chromated galv.

oj

~ 30
c

.~ 20

(3

-- .. o

2
Amount of salt added in cement, %

FIGURE 13.11. Comparison of the mean corrosion rates for clean and prerusted steel and for galvanized steel and
chromated galvanized steel exposed in an industrial environment for three years. Data are taken from Ref. 271.

W/C value of 0.75 and 2.3 kg/m3 for a W/C value of 0.6) but increased significantly with
increasing chloride concentration over 1%, as shown in Fig. 13.10.
In the same study, Treadaway et ai. found that steel cleaned with shotblasting showed
somewhat lower corrosion rates than galvanized steel, but prerusted steel corroded several
times faster than galvanized steel at chloride concentrations higher than 2% (Fig. 13.11).
In the same test, the effect of chromating on the corrosion of galvanized steel was found
to be not significant.
High salinity does not necessarily result in an excessive corrosion of galvanized steel
inside concrete structures. Stark [273] did a field investigation on galvanized-steel-reinforced concrete structures in the tidal wave zone at four different marine locations for a
period of time ranging from 7 to 23 years. He reported that the corrosion rates in concrete
structures with high chloride content were generally below 1 l1m/yr, and in most cases
below 0.5 l1m/yr, as can be seen in Table 13.3.
TABLE 13.3. Average Corrosion Rates, R, of Zinc Coatings inside Various Concrete Structures
Exposed in Marine Environments in Bermuda"

Structure
Longbird Bridge
St. George Dock

Hamilton Dock

Age (years)

Cover (cm)

Sample location

CI - content of
concrete ~wt. %
(kg/m )]

23
12
10
7
10

10

Above HT
Above HT
AboveHT
Above HT
NearMT
Above HT
BelowHT

0.19(4.4)
0.27 (6.4)
0.22 (4.6)
0.14 (3.0)
0.08 (1.9)
0.14 (3.6)
0.16 (3.7)

10
Bermuda Yacht
Club
"Ref. 273.
"HT. High tide; MT, mean tide.

8
6
13
5
6
7

"

R (prnJyr)

0.2

1.1
0.5
0.29
0.5
0.8
0.0

CORROSION IN CONCRETE

363

TABLE 13.4. Potential Values of Galvanized Steel in Concrete Bridge Decks"

Location

Bridge

Age (years)

Cover (cm)

W/C ratioh

CI- content
of concrete
(kg/m J )

---"-~-"

Longbird'
Boca Chica
Seven Mile"
Flatts
Long Dick
Spring Street
Skokie"
a

Refs" 271 and 1274.

II

Water/cement ratio.

Bermuda
Key West, Fla.
Key West, Fla.
Bermuda
Ames, Iowa
Montpelier, Vt.
Skokie, III.

21
3
3
8
7
3
6

6.0
3.8
5.0

0.44

6.4
7.6
1.3

0.40
0.44
0.43

1.0
1.12
0.84
0.53
0.29
0.06
0.9

-"

Potential
(mVesE!
-~~--

-370 to -450
- \00 to - 3OC)
-280 to -3\0
-70 to -140
-100to-200
-200 to -300
-90 to -110

, With 5-em asphalt overlay"


d

With 2.'i-cm asphalt overlay"

'Test slab"

Arnold [1260] investigated the effect of mixing galvanized rebars with black rebars
in a bridge deck and found that connecting a galvanized rebar directly to a black rebar
did not cause excessive corrosion of the zinc coating near the contact points after several
years of field testing.
Zinc-aluminum alloy corrodes significantly faster than zinc in concrete owing to the
low stability of aluminum in an alkaline environment. Corrosion rates of about 10 prnJyr
for a 4.5% Al zinc alloy and about 20 prnJyr for a 12% Al alloy were observed when the
alloys were embedded in concrete exposed to a rural atmosphere for 10 years [22(n
13.4.2.2. Potential. The electrode potential is an important parameter indicating the
surface condition of galvanized steel inside concrete. Although the measurement of the
potential of rebar inside concrete is a simple practice, it is to be noted that several factors,
particular! y I iq uid j unction potentials at concrete/reference electrode interface, may cause
significant errors, as much as 100 m V [778]. Table 13 A lists the electrode potentials of
galvanized steel reinforcements in a number of bridge decks installed for 3-21 years in
different parts of the United States and in Bermuda r1279]. All the values are in the range
of -OA5 to -0.07 VesE , which is considerably more positive than the potential, about -I
VCSE' of an active zinc surface in aqueous solutions (see Chapter 5). These potential values
suggest, based on the electrochemical behavior of zinc in aqueous solutions, that the zinc
surfaces in these structures were probably in a passive state. Metallographic examination
of the cross-sectional coating structure indicated that the corrosion in all these structures,
except the Longbird bridge, was only superficial, and for the Longbird bridge it was
estimated that 60-70% of the coating remained after 20 years of service. Dugan et al.
[1226] also reported that the potential values determined for galvanized rebars in six
bridge decks located in Iowa, Florida, and Pennsylvania were in the range of -0.385 to
-0.07 VesE , and the corrosion of the zinc coatings was found to be rather superficial.
Arnold [1260[ also found that most of the potential values of galvanized rebars embedded
in concrete decks of several highway bridges in Detroit over a period of five years were
in the range of -OA to -0.03 VCSE '

364

CHAPTER 13

13.4.2.3. Effect of Corrosion on Concrete. The corrosion of concrete reinforcement


is a main cause of concrete cracking and spalling. Thus, cracking and spalling can be used
as criteria for assessing the effect of corrosion on the concrete structures.
A field survey by Stark [273] showed that the galvanized-steel-reinforced concrete
structures in the tidal wave zone at four different marine locations showed no sign of
concrete cracking over periods of time ranging from 7 to 23 years owing to the low
corrosion rate of the galvanized steel (Table 13.3).
The results of investigations on concrete structures or samples exposed in field
environments generally indicate that, compared to black steel, galvanized steel delays the
onset of concrete cracking. Baker et al. [1215] investigated the performance of concrete
80

A615 Steel

tao
.;

'"c

~ 40
.<3

Cracks

g.

Spaliing

20

123456789-

80

4340 Steel

.;

Spalling

Cracks

~,

I
E

I
7

I
B

No failure
Very slight crack
Slight crack
Moderate crack
Severe crack
Very slight spalling
Slight spalling
Moderate spalling
Severe spalling

80

Galvanized
steel

<II

~ao

.;

Spaliing

Cracks

I
4

Failure type

FIGURE 13.12. Comparison of the degree of concrete damage due to corrosion for galvanized steel (48
samples), 4340 steel (28 samples), and A-615 steel (140 samples) after exposure in marine environments for
II years. Data are taken from Ref. 1215.

CORROSION IN CONCRETE

365

slabs embedded with several steels exposed for II years in three different marine
environments: splash zone, tidal zone, and SO-ft marine atmosphere lot. The concrete
samples had a WIC ratio of either 0.4 or 0.5 and 1.3- or 3.S-cm cover thickness. The results
presented in Fig. 13.12 show that the galvanized steel clearly had better performance than
the black steel in terms of the extent of cracking and spalling.
Treadaway et al. [271 J found that for concrete embedded with galvanized steel with
10- or 20-mm cover thickness and a WIC ratio of 0.6 or 0.75 and exposed in an industrial
environment, the onset of corrosion-induced cracking of the concrete cover takes twice
as long as in the case of black steel.

13.4.3. Results from Simulated Tests


The information presented in this subsection is obtained from tests that use concrete
samples but are conducted in simulated environments. These tests include immersion of
concrete samples in water or in salt solutions. exposure of concrete samples to atmospheres having controlled relative humidities or controlled oxygen or carbon dioxide
levels. cyclic immersion and drying tests, salt spray tests, water or salt ponding tests. and
atmospheric exposure with periodic salt solution spraying. These tests. simulating the
various conditions in field environments, accelerate the corrosion processes to different
degrees. While it may take many years to obtain corrosion effects in field environments,
it only takes a few months to a few years to complete the tests conducted under simulated
environments.
The corrosion rate of galvanized steel in concrete in laboratory testing is often
measured with electrochemical techniques [745, 1214]. As discussed in Chapter 5, the
corrosion rates measured as corrosion currents, due to the many factors and assumptions
in the measurement technique and procedure, can have a wide range of deviations from
the real corrosion rates. In a concrete electrolyte, this can be a particularly serious problem
with polarization techniques due to the high resistivity of concrete [1096-1098] .

1 day

~
"'c""

10

,. 365 days

:;
()

'"

c
0

'in

2
0 0.1

'"
0.01

12

12.2

12.4

'"
'"

,. '"

12.6

'"

'"

'"
12.8

pH of cement suspensions

FIGURE 13.13. Relation between the corrosion rates of galvanized rebars inside mortar samples on days I and
365 of a partial immersion test and the pH values of the cement suspensions (waterlcement ratio = I). Data are
taken from Ref. 1214.

366

CHAPTER 13

TABLE 13.5. Effect of CaCl z Content in Cement, Carbonation, and Test Conditions on the
Corrosion Current and Potential of Galvanized Steel in Mortar"'!>
After 4 months
Test

50% relative humidity

100% relative humidity

Water immersion

After 12 months

CaCl z

Carbonation

E
(V SCE)

(pA/cm )

icorr 2

E
(V SCE)

No
2%
No
2%
No
2%
No
2%
No
2%
No
2%

No
No
Yes
Yes
No
No
Yes
Yes
No
No
Yes
Yes

-0.17
-0.84
-0.01
-0.58
-0.39
-0.45
-0.41
-0.82
-0.42
-0.56
-1.05
-1.00

0.065
0.14
0.004
0.023
0.14
1.9
0.25
1.0
0.15
0.046
0.2
0.2

-0.21
-0.85
0.01
-0.68
-0.43
-0.40
-0.37
-0.53
-0.35
-0.42
-0.99
-1.04

lcorr 2

(pA/cm )

0.03
0.12
0.0013
0.023
0.04
1.0
0.12
0.5
0.06
0.15
0.17
0.23

"Data from Ref. 468.


"Cover thickness, 7 mm; water/cement ratio, 0.5.

13.4.3.1. Corrosion Rate. Andrade and Macias [1214] studied the corrosion of
galvanized rebars inside mortar samples, made of 11 types of cements, as a function of
the pH value of the cement suspension. Figure 13.13 shows that the corrosion currents of
all the samples decreased by more than one order of magnitude over one year of a partial
immersion test.
Gonzalez and Andrade [468] investigated the effect of carbonation, CaCl z content in
cement, and test conditions on the corrosion current and potential of galvanized steel in
mortar samples. Table 13.5 shows that in most cases the corrosion current decreased with
time, The addition of CaCl z to cement significantly increased the corrosion current,
particularly in the test with 100% relative humidity. When compared to the effect ofCaCl 2,
the effect of carbonation on the corrosion current was found to be minor. However, data
obtained by Simm [454], shown in Table 13.6, indicate that carbonation is as important
as CaCl z in promoting the corrosion of zinc in concrete. The effect is the most significant
when carbonation and CaCl z are both present at the same time.
TABLE 13.6. Effect of Carbonation and CaCI 2 on the Corrosion of 0.15-mm-Thick Zinc Foil
in Concrete Exposed in 100% Relative Humidity Air at 25C for Up to 700 Days"
Corrosion rate (pm/yr)
Concrete condition"
High carbonation + 1.5% CaCI 2
Low carbonation + 1.5% CaCI 2
Low carbonation, no CaCl z
High carbonation, no CaCI 2
"Ref. 454.
"Cover thickness. 17 mm; water/cement ratio, 1.2.

Unchromated

10
0.5
<0.1
0.63

Chromated

1.2
0.25
0.25
<0.1

367

CORROSION IN CONCRETE
1 . 000 ~------------------------------------~

+
'0

Fe

Zn

100

)(

"'E
u

10

.r:
0

r:r;G.

o . 1 L-----------~------------~----------~10

0.01

0.1

Concrete conductiv ity. (ohm" em " ) x 10. 3

FIGURE 13.14. Relation between polarization resistance, Rp, for mild steel and zinc rods in concrete and the
conductivity of the concrete. After Feliu el al. [278].

Table 13.5 also shows the effect of test conditions. Of the three test conditions
investigated, exposure at 100% relative humidity appears to be the most aggressive in
terms of the corrosion current measured. In another study, Yeomans [388] found that the
corrosion rate of galvanized steel was considerably higher in a test involving a cyclic salt
solution immersion followed by oven drying than in a test consisting of continuous
exposure in a salt spray chamber.
Feliu et al. [278] found that the polarization resistance of zinc in concrete has a
logarithmically linear relationship with the conductivity of the concrete as shown in Fig.
13.14. As postulated by Feliu et ai., the logarithmic proportionality is not due to an ohmic
control of the corrosion process but may be due to the existence of a relation between the
polarization resistance and the fraction of the wetted metal surface, Which, in turn, is
related to the measured conductivity.
Table 13.6 shows that chromating reduces the corrosion rate of zinc, especially in
highly carbonated concrete with a high concentration of chloride [454]. In practice,
chromating is used to reduce the corrosion during the curing period, when the zinc surface
is active, but it does not seem to have a long-term effect on the corrosion of zinc in concrete
[271,275].
The corrosion of galvanized steel is, in general, more uniform than that of steel in
concrete [274]. Owing to the lower tendency for pitting, fewer and shallower pits were
found to develop on galvanized steel rebars than on black steel rebars. Localized corrosion
was found to occur in certain situations. Simm [454] observed pitting corrosion of zinc
coatings in carbonated concrete containing chloride. Macias and Andrade [197] found
that the corrosion on galvanized steel in Ca(OH)z solutions at pH < 11.5 was localized.
Treadaway et ai. [271] noted that in cracked concrete the corrosion tended to be restricted
to the vicinity of the cracks in the absence of added chloride in the concrete, whereas it
tended to occur at areas remote from the cracks when chloride was added to the concrete.
13.4.3.2. Potential. The potential of galvanized steel in concrete varies greatly with
the testing conditions. Table 13.7 lists the potential values reported in a number of studies.

CHAPTER 13

368

The values vary over a wide range, from -1.1 VSCE to -0.2 VSCE' It appears that the cyclic
wet/dry type of tests tends to give more negative values than single-mode tests.
It is indicated in Table 13.5 that, without the effect of carbonation and chloride salt,
the potentials of galvanized steel in all tests are more positive than -0.45 VSCE' The
presence of salt in concrete generally results in more negative potential values [l029].
Table 13.5 also shows that carbonation has little effect on the potential in the humiditycontrolled tests but results in more negative values in the water immersion test. The type
of cement paste and additions, such as pozzolanic materials, have also been found to affect
the potential of galvanized steel [745, 1249].
Figure 13.15 is a plot of corrosion potential versus corrosion current based on the
data in Table 13.5. There does not appear to be a clear correlation between the corrosion
current and potential except, perhaps, for the lower corrosion currents for the potential
values more positive than -0.4 VseE' Considering that the potential values measured in
most galvanized-steel-reinforced concrete structures in the field are also generally more
positive than -0.4 VCSE (Table 13.4), and that at the same time the corrosion activity in
these structures is also very low, it can perhaps be suggested that a large shift, in the
positive direction, of the potential from that of an active zinc surface may be taken as an
indication of a low corrosion rate of zinc in concrete.
13.4.3.3. Effect of Corrosion on Concrete. The results from different investigations
using laboratory accelerating testing methods are rather conflicting with respect to the
effect of corrosion of galvanized steel on the performance of concrete.

TABLE 13.7. Potential Values of Galvanized Steel inside Concrete Slabs in Accelerated Tests
Test"

WID
SID
SID
SS
SP
SP
RH

PWI
SID
SP
PSI
PSI
PWI
FSI
FWI
FWI
FSI
lab air

Time
20 cycles
20 cycles
10-140 days
10-142 days
Up to 5 years
Up to 5 years
12 months
12 months
10-100 days
1-5 years
300 days
300 days
300 days
I year
3 years
3 years
3 years
2 years

Cover (em)
1.0
1.0
1.2
1.2
2.5
2.5
0.8
0.8
1-3
1.2
2.5
2.5
2.5
2.5
2
2
2
3.1

W/C ratio b

0.5
0.5
0.8
0.8
0.4
0.5
0.5
0.5
0.8
0.41
0.63
0.51
0.56
0.5
0.68
0.68
0.55

E(V SCE )

-0.2
-1.0
-0.7 to -1.03
-0.7 to -1.02
-0.14 to -0.7
-0.2 to -0.6
-0.4
-0.35
-0.9 to -1.1
-0.44 to -0.76
-0.82 to -0.99
-0.68 to 0.98
-0.31 to -0.52
-1.02
-0.42
-0.5
-0.68
-0.15

Reference
177
177
388
388
1218
1218
468
468
1225
1260
1095
1095
1095
139
1029
1029
1029
745

"Abbreviations: WID. cyclic immersion in water and drying in oven; SID, cyclic immersion in salt solution and drying in
oven; SS, in salt spray chamber; SP, ponding with salt solution; RH, controlled relative humidity; PWI, partial water im
mersion; PSI, partial salt immersion; FSI, full salt immersion; FWI, full water immersion.
bWater/cement ratio.

369

CORROSION IN CONCRETE
10 r-----------------------------~

4 months
X 12 months
X

..;
<:
~

:;

X .

0.1

"<:o

oe

'0;

0.01

0.001 w_l---_0..J..8--_0'--.6---_--'
O.4----0-:-.-:2 ---0
~Corrosion potential, V'CE

FIGURE 13.15. Plot corrosion potential vs. corrosion current of galvanized steel in concrete. based on data in
Table 13.5.

In an investigation of the fatigue strength of reinforced concrete. Okamura and


Hisamatsu [508] reported that concrete with galvanized rebars could endure a larger crack
width than concrete with black rebars under a corrosive environment, as shown in Fig.
13.16. Swamy [519] studied the cracking of galvanized-steel-reinforced concrete in an
accelerated test (6 h in 60C seawater and 6 h of drying in air) and reported that the amount
of corrosion and the depth of pits on galvanized rebars were considerably less than on
plain rebars, as shown in Fig. 13.17. Also, there appeared to be fewer cracks in the concrete
embedded with galvanized steel samples. In another study comparing the performance
of galvanized steel and black steel reinforcements in concrete, Comet and Bresler [274]
found that under corrosive conditions concrete embedded with black steel cracked earlier,
and the cracks grew longer, as shown in Fig. 13.18.
~

240 r-------------------------------------------,

'V Black lower

a;

g> 220
~

'"
~
'":>

t>. Black upper

.. Galvanized upper
... Galvani zed lower

200

...

Q)

Ol

~ 180
Q)
u
ij"

c: 160

~
'E
N

140
0.2

0.3

0.4

Maximum crack width at the side of specimen, mm

FIGURE 13.16. Effects of the maximum crack width in concrete beams on the fatigue strength of reinforcing
bars (exposed to a solution of sodium chloride for a duration of 1 year). From Okamura and Hisamatsu [508].
Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.

1.5

2.0

Plain bar

0
c=tc~_~

0.5

....
~-~:;=;p

J.

2.73 mm

1.5

~ 0.5

0..

of.j

'0 '.0

's..

'-'

2.0

Galvanized bar

2.02 mm
!

FIGURE 13.17. Distribution of corrosion pits and of concrete cracks after a cyclic immersion and dry test. Concrete cover was 2 em. Reprinted
from Swamy [519]. with kind permission from Elsevier Science-NL. Sara Burgerhartstraat 25. lOSS KV Amsterdam. The Netherlands.

'"
Cl

~ 1.0

"'"0

'5s..

2.5

.....

'"'"

::0

""
;l

>-

::c

<:>

-.l

371

CORROSION IN CONCRETE

120
"

E
(,)

100

<Ii

Bla ck

a Galvanized

Q)
(,)

P lain

.l!1

80

,5

Deformed

II)

""ctI
(,)

60

t;

'0
L:

... - . . . . .. - ... - .. ..

c;, 40
c:

2!
"iii

E
E

Black

L:'

1.6

E
~
E

.~

E
Q)
en
~

Q)

.'

....... - .., .,

..

0.8

_.. _.. - rr - "

it

.'

- - - Deformed

0.4

...

__ Plain

(,)

"

6 Galvanized

""ctI
o 1.2

24

20

16

,,

20

-...-...'

~
I

"

,~

----....-~
Age at o bserv ation , mon ths

FIGURE 13,18, Effect of corrosion of black steel and galvanized steel reinforcements on concrete under stress
in a cyclic salt immersion and dry test (sample dimension 10 x 10 x 30 em). After Cornet and Bresler [274],

~100

'"

'0

ci
c:

'"t;

50

B
OJ

E
i=

10 L-----------~------------~--~~----~

0 ,2

0.4

0.6

Effective current densi ty, rn A /em'

FIGURE 13.19, Time to cracking as a function of effective anodic current density for concrete specimens
embedded with zinc samples immersed in natural seawater, Data are taken from Ref. 1106,

372

CHAPTER 13

TABLE 13.8. Time to Cracking of Concrete Cover a for


Concrete Slabs Embedded with Galvanized Steel and Black
Steel Tested by Partial Immersion in Saturated NaCI Solutionb
Steel

0.72
0.72
0.47
0.47

Black
Galvanized
Black
Galvanized

Days to crackinl
149
286
>622
468

"Cover thickness, 2.5 cm.


hRef. 1275.
C Water/cement ratio.
d Average of two samples.

Clear [ 1218] compared the corrosion performance of black steel and gal vanized steel
inside concrete slabs with a I-in. cover and water/cement ratios of 0.4 and 0.5. The slabs
were exposed outdoors with daily ponding to a 2-mm depth with 3% NaCI solution for
9 years. Based on the visual observation of cracking and spalling, Clear found that the
slabs embedded with galvanized rebars performed slightly better than those embedded
with black steel for a W /C ratio of 0.5 but slightly worse for a W /C ratio of 0.4. Similar
results were reported for W/C ratios of 0.72 and 0.47, as shown in Table 13.8 [645,1095].
Grimes et al. [1106] investigated the effect of anodic current on the time to cracking
of concrete slabs embedded with various metals, including iron and zinc. Figure 13.19
shows that the time to cracking decreases with increasing anodic current density. Grimes
et at. found that pH gradients, from 2 to 12, in the concrete pore water may be generated
by the anodic polarization among the local regions at the metal/concrete interface. They
also found that the rate of zinc dissolution and the solubility of zinc in the pore water
determine the rate of corrosion product formation and, therefore, the time of cracking.
The corrosion products have greater volumes than the metals from which they are
formed. This volume expansion is responsible for the cracking of concrete. The extent of
volume expansion depends on the degree of hydration as well as on the diffusion and
dissolution of the corrosion products inside the concrete. It is not clear how zinc corrosion
products form and diffuse in concrete under various conditions. It has been noted that,
compared to the corrosion products of iron, which are highly insoluble in concrete and
tend to remain at the metal/concrete interface, the zinc corrosion products are more soluble
and may, therefore, diffuse farther away from the metal/concrete interface [309, 645].

14
Corrosion in Batteries
14.1. INTRODUCTION
Zinc is one of the most commonly used battery electrode materials because of its
low equilibrium potential, reversibility, compatibility with aqueous electrolytes, low
equivalent weight, high specific energy, high volumetric energy density, abundance, low
cost, low toxicity, and ease of handling [685]. The use of zinc has mainly been in pnmary
(nonrechargeable) batteries, but developments have been made that allow zinc to be more
widely used in secondary (rechargeable) batteries.
The discussion in this chapter is limited to the corrosion of a zinc electrode. The
correlation between corrosion and battery performance is not within the scope of this
discussion. In order to specify the corrosion environments, the first section is used to
briefly describe the elemental features of each type of zinc battery. The many factors
affecting the corrosion behavior of zinc electrodes, including the roles of the electrolyte,
the zinc electrode, and operating conditions, are then discussed.
14.2. ZINC CELLS AND BATTERIES
There are a number of batteries that use zinc as the negative electrode material.
Zinc-manganese dioxide batteries, the most important zinc batteries, have been the
mainstay of primary-battery manufacturing for more than 100 years. Currently, there are
three commercial types of zinc-manganese dioxide batteries: Leclanche, zinc chloride,
and alkaline cells. The pH-potential diagram in Fig. 14.1 illustrates the stable chemical
species at each pH and potential for each type of battery. The cell potential for each type
can also be predicted from the diagram. Other zinc cells have been developed over the
years but are now of less commercial significance [221]. Besides the primary batteries,
various types of zinc secondary batteries have been developed. Currently, zinc-air and
zinc-nickel batteries are the two battery types under most active research and development.
14.2.1. LeclancheCell

The Leclanche cell, which is a zinc-manganese dioxide system, was first developed
by G. Leclanche in 1860 and has since been improved by many others [1149]. It is
typically manufactured in a paper-lined and paste-wall cylindrical configuration as shown
373

CHAPTER 14

374

2 .2
1.8

MnO~

1.4
1.0

(()

0 .6

Mn++

'"

0 .2

w -0,2

tl::l
G
c:

Zn

-0.6
- 1.0
Zn

- 1.4
-1.8 0

1O

14

12

16

pH
FIGURE 14.1. pH-potential diagram for the zinc-manganese dioxide system. assuming [Zn 2+J
[Mn2+] = IO-4M . After Brodd [1143].

= 1M and

in Fig. 14.2 [1143]. The zinc anode is a sheet metal, and the cathode is manganese dioxide
powder mixed with carbon powder. Carbon powder is used as the conducting media
because Mn0 2 is a poor conductor. The electrolyte which moistens this mixture is an
aqueous solution of ammonium chloride and zinc chloride. A typical electrolyte composition is, for example, 1M ZnCl 2 + 2M NH4 Cl [1143]. Although the battery can be used

Zinc can
Label rube
;, ~ .~... c.r.;;--Paper

separator

Positive active
material
Bottom washer
Bottom cap
FIGURE 14.2. Cross-sectional view of paper-lined construction of the Leclanche cell. After Brodd [1143].

375

CORROSION IN BATTERIES

in a wide pH range (4-9), the pH of the electrolyte is commonly set at slightly acidic
values, e.g., pH 5.2. Other additives may be added to increase efficiency, reduce corrosion,
and improve mechanical properties. The reaction at the zinc anode during discharge is
(14.1 )
This reaction is immediately followed by combination of the zinc ions with the chloride
ions of the electrolyte to yield a mixture of species, principally zinc tetrachloride:
(14.2)
At the manganese dioxide cathode, at least two reactions are observed. The initial reaction
is
(14.3)
The ammonia generated from this reaction usually reacts with the zinc chloride present
to form zinc ammonium chloride:
(14.4 )
The overall reaction during discharge is

Eo = 1.63 V

(14.5)

14.2.2. Zinc Chloride Cell


The zinc chloride cell is designed to provide higher drain capabilities than the
Leclanche cell [1143]. A more acidic environment in this cell results in a higher
open-circuit voltage, as noted in Fig. 14.l. The construction is similar to that of the
paper-lined ammonium Leclanche cell. A typical electrolyte is, for example, 27% ZnCl.
The zinc chloride electrolyte often contains a small amount of NH 4 CI to help control pH
and prevent zinc passivation. The overall cell reaction is

8MnOOH + ZnCI 2-4ZnO5H 20

Eo = 1.75 V 04.6)

14.2.3. Zinc Alkaline Cell


The alkaline cell possesses superior performance at higher current drains and has
longer shelf life and greater freedom from leakage than either the Leclanche or the zinc
chloride cell [1140]. While the alkaline cell employs the same active reactants (zinc and
Mn0 2) as the Leclanche and the zinc chloride cell, it differs in cell chemistry and
construction features.
As shown in Fig. 14.3 the cell is contained in a steel can which serves as current
collector for the cathode [1139]. The zinc anode is a high-sUlface-area zinc powder
suspended in a gelling agent such as carboxymethyl cellulose. A brass sheet or nail serves

376

CHAPTER 14

Can-Stee l

Positive Cover---+-""7"*+---///
Plated Steel
~~S'b=~~~

Meta lized
Plastic Film labet

ElectrOlytePotass ium
Hvdroxide
~~~r-+-

CathodeManganese Dioxide . ---1~H*f


Carbon

~---'~~I--+-

AnodePowdered Zinc

Current Collec t or B r ass

Sea l -Nylon

SeparatorNon - Woven Fabric--+---iI~"f-

Inner Ce ll Cove r Steel

~Negative Cover -

Me t al Washer- - - - - - - - - - '

Rivet-Brass

Pla t ed Steel

FIGURE 14.3. Cross-sectional view of a typical 'D' -size alkaline-manganese battery. After Schumm r1141J.

as the anode collector. The electrolyte is a concentrated KOH solution (25-35%, about
5-1OM).
The discharge reaction proceeds more slowly in an alkaline than in an acidic
electrolyte. The high surface area of the zinc powder makes up for the more sluggish
discharge reaction. The overall reaction in the alkaline cell is
Zn + 2Mn0 2 + H20

ZnO + 2MnOOH

Eo = l.55 V

(14.7)

14.2.4. Zinc-Air Battery


The zinc-air battery has a very high energy density because of its ability to use
oxygen from the air as the positive electrode reactant [1143]. It consists of a caustic
electrolyte, a zinc anode, and an air electrode on which oxygen is reduced. The electrolyte
is usually around 6N NaOH, and the overall reaction is
2Zn + O2 ~ 2ZnO

Eo = l.64 V

(14.8)

When the electrode is cast zinc, the battery is used for low-discharge applications since
the electrode can only sustain a maximum current of 30-40 mA/cm2 because passivation
occurs at higher current densities. Zinc powder is used as the anode material for higher
discharge rates.

377

CORROSION IN BATTERIES

14.2.5. Zinc-Nickel Battery


The zinc-nickel battery has been under development as a rechargeable system for
some time [221]. The overall reaction for this system may be written as
2NiOOH + Zn + 2H 20

2Ni(OHh + Zn(OH)2

Eo = 1.75 V

(14.9)

The electrolyte used is typically a solution of potassium hydroxide at a concentration of


6-7M. The main problems limiting the successful commercialization of this battery
system are (a) changes in the shape of the zinc electrode as a result of cycling; (b) disparity
in charging efficiency between positive and negative electrodes; (c) dendrite growth
during the charging process; and (d) zinc poisoning of the nickel electrode [221, 1207].
14.3. CORROSION
The corrosion of the zinc electrode in zinc cells and batteries is the main cause for
self-discharge, relatively short shelf life, and perforation of the zinc can in the case of
Leclanche cells, when the corrosion is localized. In sealed batteries, corrosion is also
responsible for pressure buildup by hydrogen resulting from the corrosion process.
The corrosion of zinc in a battery environment is extremely complicated because it
involves a large number of factors. These factors can be classified into three main groups:
1. Properties of the electrolyte
(a) Composition: Type of chemicaL concentration. pH. zinc concentration.
oxygen concentration, gelling agents. and additives
(b) Physical setting: Volume, tlUidity, and texture
2. Properties of the zinc electrode
(a) Form: Solid sheet (e.g., methods of production and surface treatments),
powder (e.g., shape and size of particles), porous structure (pore dimension
and porosity), or amalgamated material
(b) Composition: Impurities and alloying elements
3. Operating conditions
(a) Temperature
(b) Time
(c) Current collector
(d) Composition of cathode material
(e) Rate and depth of discharge
(f) Sealing method
(g) Cell geometry
The corrosion of zinc in batteries has been the subject of a number of investigations.
but the understanding of the effect of many of these factors has not progressed beyond an
empirical level. Although some agreement is found among the results of different
investigations, contlicting results are also found. The discrepancies among the results
from different investigations are usually due to the fact that it is not possible to control
all the factors affecting the corrosion of zinc in batteries. Moreover, the test conditions in
these investigations are practically always different.

CHAPTER 14

378

The main cathodic reaction accompanying the corrosion of zinc in a battery environment is hydrogen evolution. Thus, in Lec1anche cells the overall corrosion reaction is
(14.10)

and in alkaline cells it is


Zn + 2HP + 20~ ~ Zn(OH)~- + H2

(14.11)

The corrosion rate of zinc battery materials is most commonly measured by a gassing
test in which the volume of hydrogen evolved during the corrosion process is collected.
Figure 14.4 illustrates a typical setup for the gassing test [1146]. Measurements based on

10

FIGURE 14.4. Cell for the study of hydrogen evolution. After Riietschi [1146].

CORROSION IN BATTERIES

379

weight loss are less often used for determination of the corrosion rate. Electrochemical
techniques have also been used to determine the polarization characteristics of the
corroding electrode under various conditions [9,232,534,790]. The conversion between
the corrosion rates from a gassing test and from electrochemical measurement is based
on the equivalence of a gassing rate of l,ul!(cm 2h) to a corrosion current of 2.5 ,uA/cm2
131 I].
It is important to note that the gassing rates reported in different studies should be
compared with caution because the test conditions among the various studies differ in
temperature, form of the zinc electrode, amount of zinc material and electrolyte, electrolyte composition, and equipment design [1153].
The corrosion rates and basic electrochemical characteristics of zinc electrodes in
the corrosion process are presented in the following sections, which are organized
according to the major factors listed above. General information on the elemental
electrochemical reactions of a zinc electrode is discussed in Chapter 2.
14.3.1. Effect of Testing Time
The corrosion rate of zinc powder in a sealed cell or in a typical gassing test generally
decreases with time before reaching a steady value. Snyder and Lander [1144] reported
that the time required to reach a steady gassing rate varies, by several days, for sealed
"dry" cells under different conditions, as shown in Fig. 14.5. According to Gregory et al.
1234], who obtained similar results in a typical gassing experiment, the initial decrease
in gassing rate in a gassing experiment may be attributed to the fact that an average time
of about one day is required to achieve saturation of the electrolyte with hydrogen and
that formation of a surface coating of zinc oxide or hydroxide occurs, which may inhibit
the anodic dissolution.

14.3.2. Effect of Electrolyte


14.3.2.1. Concentration. The dependence of corrosion rate on KOH concentration
varies with the test conditions. As shown in Fig. 14.6, three types of relationships between
0.4,-----------------------------------------~

C 0.3

E
~

5% KOH, 1 % HgO

0.2

OJ

c
'iii

30% KOH, 1% HgO

(f)

'"

<.:J 0.1

_-=:::::::::::~=~~;;;;;5~%~K;O;H,

2% HgO

oL~============~==========~~~40~%~o:KO~H~,~2~%~H~g~O~

10

20

30

Time, days

FIGURE 14.5. Gassing rate in sealed "dry" cells vs. time. After Snyder and Lander [11441.

380

CHAPTER 14

8
Reference
-(1144)

Qi

+(234)

.~ 6
V

.(1146)

:>
(5
:>

-4

a::'"

C>

~ 2

'"

CJ

10

12

14

KOH Concentration (M)

FIGURE 14.6. Effect of KOH concentration on gassing rate.

corrosion rate and KOH concentration have been reported in the literature: (a) Corrosion
rate increases with increasing KOH concentration [1146]; (b) corrosion rate decreases
with increasing KOH concentration [231, 1144]; and (c) corrosion rate decreases with
increasing KOH concentration at low concentrations, reaches a minimum, and then
increases with concentration at high KOH concentrations [234].
The different dependencies of gassing rate on KOH concentration among various
studies indicate that different rate-determining reactions [Eqs. (14.12)-(14.16)] may be
involved in each case. Type (a) behavior was attributed by Rtietschi [1146] to the ability
of the electrolyte to dissolve interfacial Zn(OH)z formed as a result of discharging. Thus,
the reaction in Eq. (14.13) is the rate-determining step in the corrosion process. Type (a)
behavior may also be related to the increase in exchange current density of the zinc
electrode with increasing KOH concentration [1155]. The decrease in corrosion rate with
increasing concentration, type (b) behavior, was explained by Snyder and Lander [1144],
as well as by Dirkse and Timmer [231], as a result of reduced water activity, which means
that the reduction of water, Eq. (14.12), is the rate-determining reaction. As KOH
concentration increases, the water content is decreased. Type (c) behavior indicates a
change in the rate-determining reaction on going from low to high KOH concentrations.
According to Gregory et al. [234], in such situations water reduction may be the
rate-determining reaction at low KOH concentrations, whereas a step in the anodic
dissolution that requires the participation of hydroxyl ions, i.e., dissolution of the surface
oxide film, may be rate-determining at higher KOH concentrations.

Zn + 20W

Zn(OH)z + 2e-

(hydrogen evolution)

(14.12)

(zinc oxidation)

(14.13)

CORROSION IN BATTERIES

381

Zn(OHh + 20W ~ Zn(OH)~-

Zn(OH)~blllk

(formation of zincate)

(14.14)

(diffusion to bulk)

(14.15)

(formation of oxide film)

(14.16)

The differences in the rate-determining reaction for the corrosion of zinc electrodes
among the different studies must arise from the differences in experimental conditions,
such as electrolyte and electrode compositions, amount of electrode material versus
electrolyte, form of the electrode, etc. Curve a in Fig. 14.6 was obtained from tests that
lasted less than 2 days whereas curves band c were obtained from tests that lasted for
more than 20 days [232, 1144, 1146]. Also, curve c was obtained from a "dry" cell test,
in which the ratio of electrode material to the volume of electrolyte was small (about 100
gllOO ml) as compared to those (about 1 gllOO ml) for curves a and b. It seems that in the
"dry" cell the amount of water available for reactions is more likely to be limiting and
that the corrosion is limited by water reduction at all KOH concentrations. The effect of
the volume of electrolyte reflects the importance of the ratio of zinc powder to the amount
of electrolyte. Figure 14.7 shows that the gassing rate increases with a decrease in the
ratio of zinc powder to the amount of electrolyte [1152].
Differences in the relationship between gassing rate and KOH concentration may
also be caused by the differences in physical form of the zinc electrodes used in the various
studies; some have a porous structure whereas others have a solid structure. A porous
electrode is likely to differ from a solid sheet electrode not only in the reaction rate but
also in the limiting rates for diffusion, activation, film formation, and dissolution.

30

~ 5

1 0 9
C!>

25

+25 9

'5.

<Ii 20

E
::J

C!>

15

:a

'Vi

10

(!)

5
2

14

Time, days

FIGURE 14.7. Influence of quantity of zinc powder on the gassing rate of zinc powder (3% Hg, 0.02% In) in
100 mI electrolyte (480 g KOH, 60 g ZnO, 460 ml H 20) at 45C. Reprinted from Meesus el al. 111521. with
permission from International Power Sources.

382

CHAPTER 14

14.3.2.2. Zincate Ions. The presence of zincate ions in alkaline electrolytes is


generally found to reduce the corrosion rate of the zinc electrode [231, 232, 234, 1146].
The presence of zincate has the effect of reducing the rate of diffusion of corrosion
products away from the surface. Hence, for the same period of time, the surface coverage
by ZnO in a zincate-saturated electrolyte must be much larger than in electrolytes without
zincate, so that the corrosion must be correspondingly smaller. On the other hand, it has
been found that when the system is not controlled by an anodic reaction, addition of
zincate to the electrolyte can actually cause an increase in the corrosion rate. As reported
by Snyder and Lander [1144], the corrosion of zinc powder in a "dry" cell, where the
gassing rate increases with increasing amount of ZnO added to the electrolyte, is limited
by the activity of water in the electrolyte. According to Snyder and Lander, addition of
ZnO to the electrolyte increases the effective water activity. ZnO reacts with KOH in a
1:2 ratio to form a zincate ion, resulting in a lower effective concentration of KOH. It was
estimated that a 45% KOH solution saturated by ZnO has an effective concentration of
36%, and a 30% KOH solution has an effective concentration of 26% based on this
equivalence. A similar explanation can be given for the increase in gassing rate with the
addition of ZnO, as reported by Sato et al. [1147], for a zinc powder in 40% KOH gel.
Because of the high alkaline concentration and the gelling effect, the corrosion process
may be dependent on water activity. Amalgamation generally reduces the effect of ZnO
on the corrosion of the zinc electrode in alkaline electrolytes [1144].
14.3.3. Effect of Chemical Agents
14.3.3.1. Inorganic Species. Era et al. [115] investigated the effect of several ionic
species on the gassing rate of a solid zinc sheet of about 100 cm 2 in 500 ml of Leclanche
electrolyte (25% NH 40H + 12% ZnCl). Figure 14.8 shows that the presence ofCu 2+, Ni 2+,
As 3+, or Sb3+ at concentrations as low as 1 or 2 mg/l drastically increases the corrosion
rate of zinc. As 3+ and Sb3+ ions seem to be the most aggressive among these species in
promoting gassing. The presence of Fe 2+ also increases gassing, but Fe 2+ is much less
aggressive than the other species. Figure 14.9 shows that the gassing rate generally
decreases with addition of Pb 2+ to the electrolyte with or without the presence of the

Impurities

250

-Blank

+2mIJ/lCu"
-6'2 mg/l F.~

:::-200

.s.,
.,

+ 2 mgJ1 Ni ,.

*2mg/lAs'

0>150

+ \ mIJ/lSb"

'0
Ql

" 100

6
Time (days)

10

FIGURE 14.8. Effect of ionic impurities in Lec1anche electrolyte on the


gassing of a solid zinc sheet of about
100 cm 2 in 500 ml of Leclanche electrolyte (25% NH4 0H + 12% ZnCl).
Data are taken from Ref. 115.

383

CORROSION IN BATTERIES
Impurities

1,000

-Blank

+2mg/INi

+2 mg/I Cu2 +-fr2 mg/I Fe2 +


2+

3+

3+

*1 mg/IAs +1 mg/ISb

=E
(/)

ro

OJ

o
Q)

E
.2
o
FIGURE 14.9. Effect ofPh~+ concentration of Leclanche electrolyte, in the
presence and absence of other ionic
impurities, on gassing of a solid zinc
electrode. Data are taken from Ref.
115.

>

0.1L--------L----__- i_ _ _ _ _ _ _ _L -_ _ _ _ _ _
o
5
20
10
15

-L~

Pb concentration (mg/I)

aggressive ions. Mansfeld and Gilman [10] found that the presence of Sn 2+ has only a
slight influence on the gassing rate of a zinc rod in 6N KOH whereas Cu 2+ increases and
Pb 2+ decreases the gassing rate significantly. Addition of 1O-3M KCl or KBr has only a
marginal effect on the gassing of the zinc rod. The effects of these ions were found to
arise mainly from their alteration of the cathodic polarization resistance of the zinc
electrode.
Yoshizawa and Miura [12051 investigated the effect of various metal oxides and
hydroxides on the corrosion of zinc powder in alkaline battery electrolytes. They found
that addition of compounds of indium, yttrium, and zirconium resulted in significant
inhibition of the corrosion. Sato et al. [1147] reported that addition of a certain amount
of In 2 0 3 in zinc powder gel had a beneficial effect on reducing the gassing rate in an
alkaline solution (0.2 g of gel in 2 ml of KOH), as shown in Fig. 14.10. The gassinga:;
Ol
Ol
N

150

c::i

OJ

1
c
0

"-5

....

50

>

OJ

(f)

ctS

c.9

2
In,03 Concentration, wt%

FIGURE 14.10. Gas evolution rate from Zn alloy mixed with an equimolar amount ofZnO containing various
concentrations of In203 at 60 e. Reprinted from Sato et al. [11471, with kind permission from Elsevier
Science-NL, Sara Burgerhartstraat 25,1055 KV Amsterdam, The Netherlands.
0

384

CHAPTER 14

reducing effect resulting from the presence of In 20 3 in the electrode is attributed to the
inhibiting effect of metallic In, formed through the reaction:
(14.16)
The presence of metallic In was confirmed by X-ray diffraction. The reason for the
increase in gassing rate at higher In 20 3 content is not clear [1205].
The effect of electrolytes on the corrosion of a zinc anode has been investigated in a
number of studies. Baugh [111] found that the cathodic polarization curves of zinc are
similar in Br-, Ct, CIO;; and SO;- ammonium salt solutions with pH range 3.8-5.8. There
are, however, significant differences in the corresponding sodium salt solutions. This was
attributed to the dissociation of NH! ions, which produces hydrogen ions at the electrode
surface. Gregory et al. [234] noted that in NaOH electrolyte the rate of hydrogen evolution
is lower than that in KOH. Catino [311] measured the polarization behaviors of zinc
electrodes in LiOH, KOH, and NaOH and found that while the anodic polarization for
zinc dissolution decreases slightly in the order LiOH > NaOH > KOB, the cathodic
polarization for hydrogen reduction significantly increases in the order LiOH < KOH <
NaOH. According to Catino, the cathodic reaction at small overpotentials is mainly water
reduction; at large overpotentials, it involves alkali discharge. The mechanism of hydrogen evolution with alkali cation participation is discussed in Chapter 2.
14.3.3.2. Organic Additives. Organic additives are usually used in zinc batteries for
corrosion inhibition and to prevent passivation of the electrodes. They have acquired great
importance in recent years in the production of mercury-free zinc batteries. Also, addition
of organic inhibitors together with inorganic inhibitors has been found to produce
maximum corrosion inhibition [1205].
Narte.y et al. [1203] investigated 18 organic corrosion inhibitors for suitability in
rechargeable alkaline zinc batteries. The inhibitors were evaluated with a hydrogen gas
evolution test and polarization curve measurements and in a real battery situation. They
found that, except for two, all the organic species tested had various degrees of corrosioninhibiting effect in the out-of-cell hydrogen gas evolution test. Among the 18 additives
tested, a-diphenylglyoxime and polyethylene (600) were found to perform the best in the
real cell situation. Nartey et al. also found that in the case of the inhibitors performing
well in real cell situations, more positive corrosion potentials and lower anodic current
densities were obtained than in the standard condition, while the corrosion current
densities were similar to those in the standard condition. This means that at the corrosion
potential the rate of anodic dissolution is inhibited by to the presence of the inhibitors
while the rate of cathodic reactions is enhanced.
Brossa et al. [427] studied a number of organic agents as corrosion inhibitors in a
Leclanche cell electrolyte. Figure 14.11 shows that all but one have an effect of reducing
the corrosion of zinc in the electrolyte. In another study, the addition of perfluoroalkyl
polyethylene oxide as a surface activator was found to prevent zinc corrosion without a
loss of discharge performance [1205].
Organic inhibitors are considered to inhibit metal corrosion by adsorption on the
metal surface. The adsorption can be physical, electrostatic, or chemical [1203]. However,
there is still a lack of understanding on the detailed mechanisms of the corrosion inhibition
of battery electrodes by various organic inhibitors.

385

CORROSION IN BATTERIES

6 . -------------------------------------,
L-meth ionine

600
Time. minutes
FIGURE 14.11. Hydrogen evolution vs. time in pure and inhibitor-containing NH 4 CI solutions at 40C. NDA,
I-Decylaminc; TTH, 1,2,6-HexametrioI-trithiolgIycolate; PoeIe, Polyoxyethylene (23)-Laurylether; Triton,
Octylphenoxy-polyethoxyethanol; Forafac, Ethoxylated-polyfluoro-alcoho1. After Brossa et af. [4271.

14.3.4. Zinc Electrode


The composition and surface condition of the zinc electrode materials are two very
important factors for corrosion. Inhibitors are commonly added to electrode materials to
inhibit gassing. Inhibitors can be uniformly mixed within the total mass or applied only
on the surface. Gas evolution is a surface phenomenon and must be reduced at the surface
of the zinc electrode, whether it consists of powder particles or a solid sheet. Inhibitors
distributed uniformly in the bulk material can provide continuous inhibition during
discharge, while those deposited on the surface may disappear after the battery is put into
operation and surface material is dissolved. However, surface-only inhibition requires
much less inhibitor than mass alloying and is therefore more economical. In many
applications, the most important gassing inhibition is provided by inhibitors applied to
the original electrode surface before the battery is put into operation. For effective
surface-only inhibition during battery operation, the inhibitors must be sufficiently
mobile to remain on the receding surface or, if they dissolve in the electrolyte, they must
have the tendency to coat fresh zinc surface.
14.3.4.1. Amalgamation. Amalgamation is the most effective way of reducing the
corrosion rate of zinc in zinc cells and batteries. As an example, Fig. 14.12 shows that the
gassing rate for a sealed "dry" cell is significantly reduced with increasing amount of
HgO in the electrode material [1144]. However, the use of Hg is now very limited owing
to environmental concerns. Through alloying with other elements, the amount of mercury
currently used in batteries for corrosion inhibition has been reduced to very low levels.
Low gas evolution is achieved for surface-amalgamated special zinc alloy powder
containing as little as 0.05% Hg [1204]. With alloying and the use of organic inhibitors,
batteries containing no Hg have recently been produced in Western countries 11147].

CHAPTER 14

386
0.12

'E

0.08

ai

OJ

'0;

'"

'"

0.04

(9

1%

:045%

4%

2%

HgO Concentration,

wt%

FIGURE 14.12. Variation of the steady-state gassing rate for a sealed "dry" cell with HgO content in the anode
material at 38C. After Snyder and Lander [1144].

Traditionally, two techniques have been employed for amalgamation of zinc battery
material [1152]:
Surface amalgamation by mixing zinc powder and mercury metal or salts
Mass amalgamation by atomizing a homogeneous Zn-Hg alloy
In surface amalgamation, Hg is initially deposited at the outer surface of zinc particles.
The mercury then diffuses through the grain boundaries into the grains. A combination
of surface amalgamation and low mass amalgamation is also used to obtain optimum
corrosion inhibition.
Electrochemical measurements indicate that amalgamation affords corrosion inhibition mainly through reducing the cathodic reaction. Baugh et al. [1150] investigated the
electrochemical characteristics of amalgamated zinc electrodes. Various amounts of
mercury (10-1000 /1g/cm2) were plated onto the zinc surface. At low mercury levels, the
mercury deposits were localized. A complete coverage of the surface was obtained at the
I-mg/cm2 level. Diffusion into the zinc occurred in grains, particularly at grain boundaries. Figure 14.13 illustrates the anodic and cathodic polarization curves in 30% KOH
electrolyte for the zinc electrodes having different levels of amalgamation. It can be seen
that while the dissolution of zinc is slightly facilitated by amalgamation, the hydrogen
evolution current is greatly diminished. In another study, Baugh et al. [534] found that
amalgamation has little effect on the anodic behavior of the zinc electrode in NaCI solution
but has a significant effect in NH4CI solutions. The hydrogen-inhibiting effect of amalgamation is less pronounced in NaCI solution than in KOH and NH 4CI solutions; this was
attributed to the diffusion-limited hydrogen evolution in the NaCI solution. According to
Dirkse et at. [1154], amalgamation also lowers the overvoitage for zinc dissolution in
alkaline electrolytes. The overvoltage for the dissolution reaction at amalgamated electrodes is for the charge transfer while at non amalgamated electrodes it is for adatom
diffusion.

CORROSION IN BATTERIES

387

~
.;;;

c:

FIGURE 14.13. Polari zation characteristic s for Zn in 6.8 molldm 3 (30%


w/w) KOH as a function of amalgama1
tion level: 0, Pure zinc; e, 5 Ilg/cm-;
,-I, 10 pg/cm 2; . , 20 Ilg/cm", <>, 75
jl g/cm 2, x, 100 jlg/cm2, +, 250
jl g/cm 2; 6., 500 Ilglc m2; .... 1000
pg/cm 2 Reprinted from Baugh e! al.
11150], with permissi on from International Power Sources .

.g

100

:::l

10

Potential, V (Hg/ HgO)

J4.3.4.2. Alloying Elements. Variation of zinc electrode composition is perh aps the
most effective way to control gassing in zinc batteries. In an extensive investigation, Miura
et al. [1151] found that Cd, Pb, In, Bi, and TI inhibit the corrosion of the zinc electrode
in 40% KOH while Ag, Cu, Ga, Sn, and Sb promote the corrosion. Al and Ni inhibit
corrosion when they are present in very small quantities but promote corrosion when the
concentration is increased beyond certain values, as shown in Fig. 14.14. Figure 14.15
shows that when AI, In, Pb, and Hg are present together, AI and In inhibit corrosion until
their concentrations reach certain values, beyond which they promote corrosion . The
optimum composition obtained in Miura's study, in terms of the effectiveness of corrosion
inhibition, was 0.02% In, 0.05 % Pb, 0.05 % AI, and 1 %Hg. This alloy is equivalent in its
effect to that containing 9% Hg. Glaeser [1153] also reported that, with multielement
alloying, zinc powder containing less than 0.1 % Hg is made equivalent to 3% Hg powder
in terms of gassillg rate.

45
Waighl ""
. 0.05
(l0.1

35

I
0)

0 1

25

.,
'"
CJ

'iii

15

FIGURE 14.14. Amount of gassing


for various zinc all oys amalgamated
with 1% Hg (tested with 10 of Zn
powder in 1.9 ml of 40 WI. % KOHl.
Data are taken from Ref. I 151.

Ag

AI

rn~
Bi

Cd Cu Ga

Il1J
In

Ni

Pb Sb Sn

Alloying elements

TI Base
Zn
(1 % Hg)

CHAPTER 14

388

%4

"0

In +
X

0,02
0,02

Pb + Al + 1% Hg

0,02

X
0.02

0,05
0,05

In

Ol
C

'"'"
'"
t:)

1
Pb
0
0,001

0.Q1

0.1

Content of elements, X, (wt%)


FIGURE 14.15. Effect of alloying elements on gassing. Data are taken from Ref. 1151.

The roles of various alloying elements in the corrosion of amalgamated zinc have
been summarized by Miura et al. [1151] as follows:
In, TI, Pb, and Cd increase H2 overpotential, maintain Hg at surface, and inhibit
Hg diffusion.
AI, Bi, Ca, Mg, and Sr smooth surface and inhibit Hg diffusion.
Ni and Ag maintain Hg at surface and inhibit Hg diffusion.
Lee [9, 1121] measured the hydrogen overpotential for zinc contammg small
amounts of alloying elements and found that the presence of 0.05% Cd or Ca increases
the overpotential while the presence of 0.05% Fe or Mn reduces the overpotential, as
shown in Table 2.9. Bhatt and Udhayan [1206] found that a Zn-0.27Pb-0.09Cd alloy
electrode has a slightly higher corrosion rate than a pure zinc electrode in 5.5N ammonium
chloride but less internal resistance. One effect of Cd and Pb as inhibitive alloying
elements is to reduce the grain size; it is generally agreed that a fine-grain structure is less
sensitive to corrosion than a coarse-grain structure [1155]. Also, Cd and Pb were reported
to inhibit corrosion perforation of zinc cans. Figure 14.16 shows that pitting tendency is
reduced with the addition of Cd and Pb to the zinc electrodes for Leclanche cells [1276] .
In a recent study on mercury-free batteries, Yoshizawa and Miura [1205] reported
that zinc alloys including Pb, In, and Bi [i.e., Zn(Pb + a), a = In + Bi] as well as Zn(ln
+ Bi + Ca) alloys as mercury-free electrode materials show the highest corrosion
resistance. Zinc alloys containing 0.00 I-I % In, 0.005-1 % Mn, 0.005-1 % Pb, 0.005-1 %
AI, and 0.0005-0.1 % rare-earth metals have also been developed as mercury-free
electrode materials [1208].
14.3.4.3. Surface Treatment. Catino [311] found that dipping a zinc wire in 0.05M
metal-ion solutions produces various degrees of precipitates on the zinc surface. Immersion in HgCI2 solution produces a dark surface that is removed when immersed in 9N

389

CORROSION IN BATTERIES
100,-----------------------------------------,

75
~

Pb

cO

~
(;

50

0.15 Cd

+ Pb

Cd

0-

25

OL-______L -______L -______L -____


o
0.25
0.5
0.75

~~

____

1.25

Metal addition, %

FIGURE 14.16. Influence of Pb and Cd additions on pitting of zinc battery cans. After Aufenast [12761

KOH. Immersion in Pb(C 2H 30 2)2 yields a deposit with a dense, fine, needlelike structure
that does not change after immersion in 9N KOH. Immersion in InCl) does not change
surface morphology, whereas immersion in TICI yields a uniform distribution of TI-rich
particles. With metal-ion treatment, both the anodic and cathodic reactions show a higher
polarization than in the case of nontreated zinc. Table 14.1 shows the corrosion rates of
zinc samples treated with these metal-ion solutions.
14.3.4.4. Physical Forms. Zinc powder with a larger particle size is found to have
smaller gassing rates than that having a smaller particle size [1148, I 152]. The difference
between the gassing rates of zinc powders having different particle sizes decreases with
amalgamation [1148]. Cachet et al. [843] reported that the corrosion of zinc powder in a
Leclanche cell electrolyte causes an increase in the reactive surface area owing to surface
roughening but does not significantly change the shape and size of the zinc particles.
According to Gregory et al. [234] and Catino [311], the formation of zinc hydroxide and
zinc oxide on different areas of the electrode surface may also occur, leading to nonuniform corrosion of the surface because zinc hydroxide can be more easily dissolved than
zinc oxide.

TABLE 14.1. Initial and Final Corrosion Rates for Metal-Ian-Treated Zn Wires in 9N KOH
Based on R" Measurements and Gas Volumes Produced on Metal-Ian-Treated Zn Powder after
24 Hours at 71C in 45% KOH"
~------~--

Metal-ion treatment
None

HgCI2
Ph(C 2H 30 2 )2
PhCl 2

TICI
InCI,]
"Ref. 311.

i", initial (j1A/cm 2 )


13.8
5.1
3.4
10.7
16.6
12.1

ie' final (pAlcm 2 )


21.7
4.0
1.5
4.3
12.8
22.1

Gas volume (ml)


0.18
0.05
0.07
0.06
0.22
0.6

390

CHAPTER 14

14.3.5. Operating Conditions


14.3.5.1. Temperature. The rate of gassing from corrosion in a zinc cell increases
with increasing temperature. Snyder and Lander [1144] found that the steady-state
gassing rate for sealed "dry" alkaline cells increased by a factor of about 4-7 as the
temperature was increased from 25 to 50C. Gregory et at. [234] observed a lO-fold
increase in the gassing rate in gassing tests when the temperature was increased from 25
to 60C. The enthalpy of activation for the gassing reaction has been reported to be about
19 kcal. A lO-fold increase in the gassing rate was also observed by Riietschi [1146] for
zinc powder immersed in electrolytes containing zincate with a temperature increase from
60 to 95C. The effect of temperature on gassing rate depends sensitively on the
composition of the electrode material [1153].
14.3.5.2. Discharging. In real situations, a battery may operate under a certain
discharging rate for a length of time, or it may stand idle after a certain depth of discharge.
The corrosion that occurred on the electrode during or after discharging may be changed
by the discharging process, because discharging can cause changes in electrolyte composition and texture, changes in the surface area and composition, formation of surface films,
and development of non uniformity across the surface. The effects of most of these changes
on the corrosion are little known.
Sato et at. [1147] investigated the effect of discharging on the corrosion of a zinc
powder gel. Figure 14.17 shows that the gassing rate significantly increases with depth
of discharge. It also increases with increasing discharge current, as illustrated in Fig.
14.18. The increase of gassing rate with depth of discharge and current density was
explained by Sato et al. as a result of the buildup of zincate ions in the gel, since an increase
in the gassing rate with addition of zinc oxide to the gel was observed in a separate
experiment. According to Yoshizawa and Miura [1205], in an alkaline manganese battery,
more hydrogen is produced in the case of partial discharge than in the case of nondischarge

~ 12
Cl
C\I

ci

>-

"'"

:i..
cD

c:

S
(5

>
Q>

'"'"

20

40

60

Depth of discharge, %

FIGURE 14.17. Gas evolution rate from Zn alloy with 2% Hg at 60C at various discharge depths after
discharge at 30 rnA for 0.2 g of gel. Reprinted from Sato et at. [1147], with kind permission from Elsevier
Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.

391

CORROSION IN BATTERIES

a:;

OJ

~400

o
>-

.g

300

:i..
Q)

" 200
c

o
~

~
100
Q)
(/)

ro
c.9

O~------~---------L--------~------~~

10

20

30

40

Discharge current, mA

FIGURE 14.IS. Gas evolution rate from 50% discharged Zn alloy at various cUlTents for 0.2 g of gel at 60C.
After Sato et al. 111471.

owing to (1) roughening of the zinc surface and (2) decreasing hydrogen overvoltage at
contacting faces with ZnO particles.
14.3.5.3. Current Collector. The function of a grid is to conduct electric current and
to support the active material. Grid materials with electrode potentials more positive than
that of zinc may form a local galvanic cell to cause extra corrosion of the zinc electrode.
It has been established that silver and copper grids promote gassing of the zinc electrode
l234, 1144]. The grid can be treated to reduce gassing. In one study, it was found that
amalgamation of a silver grid reduced the gassing rate in a "dry" alkaline cell. In another
study, amalgamation of silver grid material was found to have little or no effect on the
gassing rate whereas amalgamation of copper grid material significantly reduced the
gassing rate of a porous zinc electrode immersed in alkaline electrolytes [2341. The
difference between the silver and copper grids was attributed to the difference in the
concentration of mercury on the surface of the two types of grids.

15
Corrosion in Other Environments
15.1. INTRODUCTION
This chapter deals with the corrosion situations that are not covered in other chapters.
More specifically, it includes information on corrosion in organic solvents and gases and
on zinc anodes.
15.2. ORGANIC SOLVENTS

15.2.1. Classification
The reactivity of metals in organic solvents depends on the type and structure of the
organic compound. Organic solvents can be classified as nonpolar aprotic, dipolar aprotic,
and protic, as summarized in Table 15.1 [424]. The classification of a solvent as protic or
aprotic is based on whether it has an ability to provide protons. Protic media contain acidic
hydrogen atoms, and aprotic media do not. Dipolar aprotic solvents display electrostatic
forces due to ion-dipole and dipole-dipole interactions.
An important role of a solvent in a corrosion process is solvation, a process whereby
solvent molecules form shells around each dissolved ion to compensate the Coulombic
forces between the oppositely charged ions in the solution. Changes in the composition
of a solvent alter its solvation properties. The solubility of a metal significantly varies
from solvent to solvent owing to differences in solvation properties. Figure 15.1, as an
example, shows that the solubility of zinc chloride in different organic solvents varies
over several orders of magnitude [500].
Usually, an oxidizing group within the molecular structure of a solvent is responsible
for the corrosion process [424]. For example, for carboxylic acids such as acetic acid,
CH)COOH, the corrosive group is COOH- and/or H;o'v, and for alcohols such as ethanol,
C 2HsOH, the corrosive group is H2COH- and/or H;olv. In addition, as in aqueous solutions,
the presence of chemical species such as oxygen and halogens may significantly increase
the corrosivity of the organic solvents.
The corrosion reactions in organic solvents can be grouped into two types: electrochemical and chemical [424]. In the electrochemical type of reaction, the anodic partial
reaction for zinc corrosion can be expressed as
393

394

CHAPTER IS

TABLE 15.1. Organic Solvent Systems with Protic and Aprotic Propertiesa
Kind of bonding

Species dissolved

Solvent
Hydrocarbons
Cyclopentadiene
Propylene carbonate

o~ HX, halogens
Fe +, Co2+, AgCI04
HCI

Acetone
Carboxylic acids

FeCI 3
Hydrohalogens

Alcohols

Inorganic and organic


acids

Water

Inorganic and organic


acids

van der Waals


n-Complexes
Ion-dipole forces, dipoledipole forces
Covalent complexes
Dipole-dipole forces, hydrogen
bridges
Dipole-dipole forces, iondipole forces, hydrogen
bridges
Dipole-dipole forces, iondipole forces, hydrogen
bridges

Type of solvent
Nonpolar aprotic
Nonpolar aprotic
Dipolar aprotic
Dipolar aprotic
Protic
Protic

Protic

"Ref. 424.

Zn + 2X-

ZnX 2 + 2e-

(15.1)

where X- is a halogen ion, organic acid anion, etc. The cathodic partial reaction can be
represented as

HA + e- ~ -} H2 + A-

(15.2)

where A- = carboxylic acid anion, hydroxyl ion. alcoholate ion, etc.


gil
Water
Methanol
Ethanol
}
Acetone
Diethyl ether
Ethyl acetate
I-Butanol
I-Propanol

- - - --

1000
____ _
-

100

lA-Dioxane

10

FIGURE 15.1. Solubility of zinc chloride in solvents containing


0.05MHCl at 20e. From Hronsky [500]. Copyright by NACE
International. All Rights Reserved by NACE; reprinted with
permission.

395

CORROSION IN OTHER ENVIRONMENTS

The chemical type of reaction can be represented in a general form, for example, for
solvents with halogen ligands:
(15.3)
15.2.2. Corrosion
Hronsky [500] investigated the corrosion of zinc in various organic solvents. Figure
15.2 shows that the corrosion rate varies greatly with the type of solvent. It also shows
that the corrosion rate of zinc in some solvents is much higher than in water, while in
others it is much lower. Figure 15.3 illustrates that the viscosity of the solvent is a
predominant factor determining the corrosion rate of zinc in organic solvents containing
a small amount of acid. Hronsky found that the corrosion rate decreases slightly with
increasing molecular weight of the sol vent and with decreasing dielectric constant. Also,
it may be noted by comparing Figs. 15.1 and 15.2 that the corrosion rate does not appear
to be a direct function of the solubility of zinc in the solvents. Furthermore, according to
Hronsky [500], neither the dissociation level of the dissolved acid nor the electrolytic
conductivity has a major effect on the corrosion rate. He thus concluded that the
rate-determining step in the corrosion of zinc in these solvents is probably the transport
of the oxidizing agent, which is a function of the viscosity.
For a given solvent, the corrosivity is affected significantly by the presence of other
chemical agents. Lechner-Knoblauch and Heitz [842] investigated the corrosion of zinc

____ _

(4)l'I01olcn P'OO!JC.I Borrltf Film rormtd

,1

90

Dltth,.. cl/"e'

80

10

60

50
1,4

40

-Oio~o t'lt

Wattt

20

FIGURE 15.2. Zinc corrosion in various organic media containing 0.05 mol HCI per liter.
From Hronsky [5001. Copyright by NACE
International. All Rights Reserved by NACE;
reprinted with permission.

I - P,oponol

t -8ul0nl)l
_ _ _ _ _ <1

l/iA}'lidtM
c.h\l)I't~t

Time (hours)

396

CHAPTER 15

50,--------------------------------------,
Acetone.

"Diethyl
ether

40
Ethyl acetate

Methanol
1 ,4-Dioxane

Benzene

Ethanol- Water
1-Propanol

Vinylidene
chloride

0~__~t-~B=uta=n=ol____L __ _ _ _ _ _ _ _ _ _~2------------~3~

Reciprocal kinematic viscosity (s/mm2j

FIGURE 15.3. Corrosion rate of


zinc vs. reciprocal kinematic viscosity of corrosive media containing
0.05 mol HCI per liter at 20e. From
Hronsky [500]. Copyright by
NACE International. All Rights Reserved by NACE; reprinted with permission.

in alcohols containing various gaseous and solid contaminants. Table 15.2 shows that the
addition of other solvents such as formic acid, sodium formate, acetic acid, and sodium
acetate results in increases in the corrosion rates.
Jaenicke and Schweitzer [1275] determined the apparent exchange current density
of the Zn 2+/Zn amalgam electrode in mixtures of organic solvents and water. Figure 15.4
shows the dependence of io on the molar fraction of tetrahydrofuran, acetone, acetonitrile,
ethanol, and dioxane in mixtures with water. It can be seen that io decreases with the
addition of water to the solvents up to a certain proportion and then rapidly increases with
further addition when the water content reaches more than 95%. According to Jaenicke
and Schweitzer, in the water-solvent mixture the metal surface is preferentially solvated
by organic molecules, while the solvation shells of the dissolved metal ions consist mainly
of water molecules, which are the most polar complexing agents present in the medium.
At a certain solvent/water ratio, this difference in the solvation of the metal surface and
the metal ions in the solution reaches a maximum. Consequently, the energy barrier for
the metal atom-ion transition through the double layer at the electrode is maximal, which
is reflected by a minimum in io, as shown in Fig. 15.4. In general, according to Miles and
Gerischer [1189], the variation in electrode kinetics with different mixtures of solvent
and water results from a combination of (a) specific adsorption of the sol vent, (b) changes
in the potential across the double layer, and (c) changes in the charge-transfer activated
complex as the solvation sheath changes.
The results from a number of studies indicate that the corrosion processes in solvents
containing small amounts of contaminants are usually diffusion-controlled [424, 488,
500, 749]. This is supported by the fact that viscosity, which determines the diffusion
coefficient, is a predominant factor in determining the corrosion rate in a number of
solvents. Similarly, for zinc in primary alcohols containing O.01N HCl, the corrosion rate
decreases with increasing chain length of the alcohol, as illustrated in Fig. 15.5 [749].
Also, the polarization curves of a zinc electrode in methanol and octanol, shown in Fig.
15.6, indicate that the rate-limiting process in the corrosion is the diffusion of oxidizing
agents to the surface.
Electrode processes other than diffusion can affect the corrosion rate in some
solvents. Biallozor and Bandura [463] postulated that the charge-discharge of Zn 2+ in
dimethyl sulfoxide, N,N-dimethylformamide, and acetonitrile proceeds by a two-step

N2

1.29

0.02

0.18

O2

.. -

H 2O

~,-------

0.01

..

0.01
0.14

0.09

3.9

0.46

4.2
0.87
0.62

0.38

N2

Air

S.S

0.3

7.0
3.93
0.28
0.16
0.13

0.67

0'
0.01

0,

CH 30H

9.7

11.7

Air

0.17
0.24
0.20
0.29
0.07

0.01

0.17
1.06
0.43
0.21
0.03

0.03

0
0

N2

0.08

2.8
2.8
0.62
0.29
0.02

0.03

0
0

O2

C 2H s OH

0.13

3.51

Air

0.14

0.48

0.43

0.73

0.12

0.56

6.28

1.60

CH 30H

SOC

0.14

0.01

0.1

0.37

'0 corresponds to weight loss rates of <0.01 g/(m'h).

"Reprinted from Lechner-Knoblauch and Heitz [842J. with kind permission from Elsevier Science Ltd. The Boulevard. Langford Lane. Kidlington OXS 1GB. United Kingdom.

dCorrosion rates in grams per square meter per hour nearly correspond to millimeters per year.

O.IM
O.OSM
O.OIM
O.OOSM
O.OOIM

HCOOH

O.IM

HCOONa

O.IM
0.05M
O.OIM
O.OOSM
O.OOIM

CH 3COOH

O.IM

None
50 ppm CICH 3COONa

Contaminant
mol I-I

Room temperature

0.18

0.02

6.19

0.01

C 2H sOH

TABLE 15.2. Weight Loss Rates (in Grams per Square Meter per Hour)" of Zn in Alcohols Containing Different Gases and Contaminants at Room
Temperature and 50C"

....

C/l

-l

tIl

3:

:;:;
o
z

~
<

;>:l

tIl

:t

oZ

oC/l

:xl
:xl

398

CHAPTER 15
100,---------------________________--,
(4 ) H2 0 +CHaCN
(5)
)CH2)2"
(1 ) H20 + CH -CH OH
Hz<:> +0"
3
Z
(2) H20+CHa-TI-CHa
(CH2)2

(3 )

10

.~

0.1

0.01

L - - L____L __ _~L__ _~_ _ _ _- L____L _ _ I

0.4

0.6

0.8

1.0

Organic solvent

Water

FIGURE 15.4. Exchange current of zinc (2mM ZnCI04 ) in binary water mixtures as a function of mole fraction
of organic solvent. (Concentration of supporting electrolyte, NaCI04 : in tetrahydrofuran and acetone, 0.7 M; in
acetonitrile and ethanol, 0.6M; in dioxane, 0.2M). After Jaenicke and Schweitzer [1275].
2.5,---------------------------------------,

~ 1.5
<C

0.5

C1

C2

C3

C4

C5

C6

C7

C8

C-Atoms
FIGURE 15.5. Corrosion rate ofzine in primary alcohols of different ehain lengths with addition of 0.0 IN HCI

+ IN LiCI at 25C (with hydrogen bubbling). After Heitz et at. [749].

CORROSION IN OTHER ENVIRONMENTS

399

100,-------- - - -- - - -- -- - - -- - - - - -- - - - - - - ---,

Eu

10
Methanot

:E.u;
cQ)
u

0.1

Octanol

:;

0.01

0.001

-1.15

-0.95

-1.05

E.

-0 .85

-0.75

Vs ... e

FIGURE 15.6. Polarization curves of zinc in methanol and octanol solutions containing O.OIN HCI and IN
LiCI at 25C. After Heitz et al. [749].

reaction. Singh and co-workers [386,453] found that the corrosion in dimethylformamide
is determined by an activation controlled cathodic reaction. The addition of water to the
solvent reduces the cathodic polarization and thus increases the corrosion rate. Hronsky
[500] noted that in some solvents a poorly adhering, permeable, black layer, identified as
consisting of amorphous zinc compounds and zinc oxide, is formed during corrosion. The
formation of the surface film appears to have an effect of hindering the further corrosion
of zinc in the solvents.
Babaqi et al. [1013] determined the corrosion current for zinc in methanol solutions
with various water contents. The corrosion current was found to increase linearly with
increasing water content at different temperatures. The activation energy values determined experimentally indicate that the rate-determining step in the corrosion is more
related to the adsorption of a participating species and less to a diffusion process, which
is also supported by the fact that solution stirring has no effect on the corrosion current.
Their results seem to suggest that when the solvent is pure and the corrosion rate is low,
the corrosion is less likely to be controlled by diffusion and more likely to be controlled
by other processes.
It should be noted that although some general rules may be applied to estimate the
corrosiveness of an organic solvent solution, reliable values for the corrosion rate in a
specific solvent solution still need to be experimentally determined, since the corrosion
rate of zinc in organic solvents varies greatly with solvent composition. The corrosion
rates of zinc and its alloys in many specific organic solutions can be found in the book
Corrosion Resistance of Zinc and Zinc Alloys by Porter [1190].
15.3. GASEOUS ENVIRONMENTS
The gaseous environments discussed in this section are not related to the normal
atmospheric environments that were considered in Chapter 8. A gaseous environment is
defined here as an environment confined in a container or package in which gases other
than those in normal air are also present. These gases may be present originally, or they

400

CHAPTER 15

Specimens
inHCI

E
<.l

vapor

0)

c:

.iij

0)

:g,

in wat er

vapor
20

30

40

50

60

Exposure, days

FIGURE 15.7. Gain in weight of galvanized sheets exposed to the atmosphere in a sealed tank containing 20%
hydrochloric acid. After Gilbert and Hadden [437].

may be released from liquids or solids, particularly plastic materials, also present in the
confined space.
Corrosion in gaseous environments is governed by principles similar to those
elucidated for atmospheric corrosion although it has its special features [480]. As in
normal atmospheric corrosion, the amount of moisture in the environment plays an
important role in gaseous corrosion. Depending on the kind of gases and their concentrations, the critical relative humidity required for corrosion may vary. Also, depending on
whether an electrolyte is or is not formed, the corrosion can be electrochemical or
chemical in nature.
The corrosion of zinc in HCI vapors has been studied by several investigators. Gilbert
and Hadden [437] found that the corrosion rate of galvanized sheets hanging in a sealed
tank above a 20% HCI solution was many times higher than that of sheets placed above
water, as shown in Fig. 15.7. Askey et al. [811] observed that the corrosion rate of zinc
in HCl vapor increases approximately linearly with HCl concentration. According to
them, at normal atmospheric levels HCl reacts with preexisting Zn(OH}z to form insoluble

TABLE 15.3. Corrosion Rate of Zinc in a Moving


Atmosphere Containing S02 and Water Vapor
(Average S02 Content 1.8%)"
Time (days)

Corrosion rate (J1.mlyr)

7
50
100
200
500

627
238
145
130
127

"From McLeod and Rogers [499]. Copyright by NACE International.


All Rights Reserved by NACE; reprinted with permission.

CORROSION IN OTHER ENVIRONMENTS

401

NO.8

E
-2.
OJ
E_ 0.6
Q)
rJ)

ro
~
()

.~

..,

0.4

..c

OJ
Q)

~ 0.2

Mg

Pb

Cr

AI

Fe Avialite Zn Brass Ag

Ni

Cu

18-8
FIGURE 15.8. Corrosion stability of metals in dry hydrogen sulfide, as indicated by change in weight in 180
days. Data are taken from Ref. 556.

basic zinc chloride but at higher levels transformation of Zn(OH)2 to more soluble ZnCl 2
can proceed.
Table 15.3 shows that the corrosion rate of zinc in a moving mixture of S02, water
vapor, and air is rather high and that it decreases with exposure time [499]. In hydrogen
sulfide gas, the corrosion rate of zinc is low compared to those of other common metals
and alloys, as shown in Fig. 15.8 [556]. Variation of the relative humidity of the H 2S-air
mix has little effect on the corrosion rate of zinc, although it greatly changes the corrosion
rates of some other metals, as shown in Fig. 8.24 in Chapter 8. It is noted that although
Fe corrodes little in dry H2S, it corrodes severely in H 2S-water vapor mixtures. This
indicates that zinc can be a corrosion-resistant coating material for protection of steel in
H 2S environments [556].
Knotkova-Cermakova and Vlckova r480] investigated the corrosion of several metals
placed in sealed compartments containing various plastics, rubber, and various types of
wood. Table 15.4 shows the corrosive effect of these materials on zinc and iron, which is
due to their tendency to produce acetic acid and, in lesser amounts, formic acid by
hydrolysis. In general, direct contact causes more corrosion than mere exposure to the
vapor released by these materials. Table 15.5 shows the corrosion rates of zinc in a sealed
chamber containing different types of wood at 35C. Oak appears to be distinctively more
corrosive to zinc compared to other woods.
Table 15.6 contains the corrosion rates reported by Helwig and Bird [69] for
galvanized steel over sealed aqueous solutions of several organic acids. The corrosion
rate is greatly increased due to the presence of these acids. Table 15.6 also shows that
chromate surface treatment reduces the amount of corrosion in these environments. It also
shows that chromating the zinc surface can significantly reduce the amount of corrosion
under the test condition.

402

CHAPTER 15

TABLE 15.4. Corrosive Effects of Organic High-Molecular-Weight Materials on Steel and


Zinc in Sealed Enclosures or by Contact"
Degree of corrosive effect on metal"
In sealed enclosure
Material
Phenolic resins
Amino plastics
Polyformaldehyde
Polycarbonates
Poly(vinyl chloride}
Polyamide alkali
Polyamide hydro
Glass-reinforced polyesters
Epoxides
Pol yethy 1ene
Poly(methyl methacrylate)
Pol ystyrene
Poly( vinyl acetate)
Cellulose acetate
c
Rubber
Paints oil
Epoxide'
Melamine alkyd'"
Wood
Oak
Beech
Chestnut
Birch
Fir
Walnut
Elm

Steel
2
1

0
0
0-1
0
0
1
0
0
0
1
1
0
0
1

3
3
2
1-2
0-1
0
0

By contact

Zn

Steel

3
2
I
0
0-1
0
0
2
2
0
0
0

Zn

3
3

3
2
I
1
I
0
0
I
I
0
0
0
I
1-2

1-2
0
0
2

I
I
I
0
2
2
0
0
0
I
1-2
I
2

3
3

3
3
2
2
1-2
0
0

3
3
2
2
1-2
0
0

2
1-2
0-1
0
0

"Data from Ref. 480.


"Corrosive effect: O. None; I. slightly corrosive; 2. corrosive; 3. highly corrosive
'The corrosive effect is most intense during the curing period.

TABLE 15.5. Corrosion Losses of Iron and Zinc Enclosed with Wood at 35C for 32 Days"
Material
(veneer)
Oak
Beech
Ash
Maple
Spruce
Poplar

"Ref. 480.

Relative humidity

Corrosion loss of metal (urn)

(%)

Iron

100
84
100
84
100
84
100
84
100
84
100
84

10.57
1.37
3.02
0.53
3.16
0.53
1.04
0.48
0.58
0.03
1.12
0.07

Zinc
15.42
10.47
3.00
4.33
3.00
4.33
1.53
4.22
0.40
0.46
5.20
6.37

403

CORROSION IN OTHER ENVIRONMENTS


TABLE 15.6. Zinc Loss for Specimens Stored for 30 Days at Ambient Temperature over
Various Solutions".!'
Zinc loss (mg/dm 2 ) for specimens stored over:
Cr in surface
film (mg/dm 2 )
0
0.2
1.3
2.2

Water

1.8% Ammonia

2.5% Formic
acid

0
0
2
4

185
11
4
4

690
204
131
69

2.1 % Acetic acid 5.1 % Acetic acid


795
400
161
90

303
137
35

"Reprinted from HelwIg and Bird [69J, with kind permission from Elsevier Science Inc., 655 Avenue of the Americas, New
York.
hOriginai coating weight was 1.12 oz/fr~ of sheet.

15.4. ZINC ANODES

15.4.1. Sacrificial Anodes


Zinc is a commonly used material for making sacrificial anodes owing to its low
position in the electromotive series. Zinc anodes were first used to provide galvanic
protection to the copper-sheathed hulls of warships a century and half ago [1193].
Historically, zinc sacrificial anodes have been used most extensively in seawater applications. They are also widely used for the cathodic protection of hot water tanks [447], fuel
storage tanks [281], underground steel structures [471,473], and steel-reinforced concrete
structures [329,1232,1284,1285,1288].
Compared to other metals with low electrode potentials, such as aluminum and
magnesium, zinc has distinct advantages as an anodic material in many common environments. It does not usually passivate, in contrast to aluminum, and it has a low
self-corrosion rate, and thus high efficiency, in contrast to magnesium.
The performance of sacrificial zinc anodes in providing galvanic protection depends
on a number of factors, many of which are discussed in detail in Chapter 7 . In the following
discussion, only the information related to material properties is presented. Anode design
considerations, in terms of shape, size, and position, regarding specific applications can
be found elsewhere [469,471, 1192, 1193J.
15.4.1.1. Definitions. Anode potential, current output, and anode efficiency are the
three most important properties of sacrificial anode materials, The anode potential
determines whether there is a sufficient driving force for the cathodic protection in a given
environment. Zinc sacrificial anodes are, in most cases, used for protection of steel
structures: for complete protection, the potential of steel needs to be cathodically
polarized to about -0.75 V SCE' The corrosion potential of an active zinc surface in neutral
electrolytes is about -1.05 V SCE' This allows about a 0,3-V potential drop in the electrolyte
caused by the galvanic current flow between the anode and steel cathode, assuming that
very little polarization is associated with the zinc anode, The actual potential drop in the
electrolyte depends also on the shape and geometry of the anode and cathode as well as
the resistivity of the electrolyte.
Current output is the galvanic current available for cathodic protection under a given
set of conditions for an anode-electrolyte system, It can be defined as 1 = (E I - E 2 )IR,

404

CHAPTER 15

where E1 is the anode potential, E2 is the potential of steel under cathodic polarization,
and R is the resistance in the electrolyte [1192]. In an active state the dissolution of a zinc
anode generally has a low overpotential, and thus E1 is close to the open-circuit value of
-1.05 V SCE
Anode efficiency is the percentage of the anodic dissolution due to galvanic action.
The total anode consumption also includes self-corrosion (nongalvanic action). The
efficiency can be determined by comparing the total electrical charge passing through
from anode to cathode to the charge calculated from Faraday's law for the weight loss of
the anode.
15.4.1.2. Effect of Anode Composition. Anode composition has long been recognized as critical in determining the performance of the anode. There are many patents on
alloy composition, in which claims are made for the beneficial effects of specific alloying
additions [1192]. In general, the composition of the alloy is designed to result in an anode
possessing the following properties:
1. An electrode potential that is sufficiently negative, but not too negative, for a
specific application.
2. A large current output at small polarizations, so that most of the potential
difference between the anode and cathode can be used to polarize the cathode.
3. An electrode that remains active and does not become passivated throughout the
anode life.
4. A high efficiency: a high galvanic corrosion rate versus a low self-corrosion rate.
Specifications of alloy compositions for anodes and their basic perfonnance characteristics can be found in the literature [471,1192].
It has been well established that the presence of Fe in zinc anodes, even in very small
amounts, e.g., 0.001 %, is harmful to the performance of the anode [230,470, 1194]. Fe
in a zinc anode generally causes a reduction of current output and ennoblement of the
anode potential, due to the formation of an insulating dissolution product film on the
surface of the anode. Addition of a certain amount of Al can effectively remove the
harmful effect of Fe.
Carson et al. [470] found that the effects of alloy composition, time, and current
density on anode perfonnance in seawater are interrelated. Figure 15.9 shows the potential
of zinc and zinc-aluminum alloy anodes as a function of Fe content current density. In
the case of the pure zinc anodes, the presence of Fe in the anode results in an ennoblement
of the potential. This effect is very small for Fe contents less than 0.0003% but significantly increases for Fe contents higher than 0.001 %. Carson et al. also found that the
extent of potential ennoblement increases with time. The effect of Fe disappears when
small amounts of Al are present, as shown in Fig. 15.9b, c.
Waldron and Peterson [230] investigated the effect of Fe, AI, and Cd on the
perfonnance of zinc anodes in seawater. They found that the anodes containing small
amounts of Al and Cd give higher total current outputs, as shown in Figs. 15.10 and 15.11.
The effects of other common elements on the perfonnance of zinc anodes are
relatively less significant compared to those of Fe and Al [469]. Crennel and Wheeler
[1194] found that the addition of Sn has little effect whereas surface amalgamation is
beneficial.

,.,,,;>

-r""

(a) 0% AI

1;;)':>
~e

(b) 0.3% AI

(c) 1.2% AI

Ht
j

FIGURE 15 .9. Potential as a function of current density and iron content for zinc-aluminum alloy anodes. From Carson
el al. [470]. Copyright by NACE International. All Rights Reserved by NACE; reprinted with permission.

Po.

'0

.,c

lJ j //\

til

"'"

(/)

-l

::::
m

;;0

<

:r:
m
;;0
m

-l

0
z
Z

(/)

;;0
;;0

406

CHAPTER 15

6,-----------------------------------------,

-;;5
til

:;
o

s::.

3~--

____L __ _ _ _ _ _L __ _ _ _ _ _L __ _ _ _ _ _L __ _ _ _
0.1

0.2

0.3

0.4

0.5

% Aluminum
FIGURE 15.10. Effect of small amounts of aluminum in zinc anodes on current output in seawater. From
Waldron and Peterson [230]. Copyright by NACE International. All Rights Reserved by NACE; reprinted
with permission.

Perkins and Bornholdt [464] identified the dissolution products formed on zinc
anode surfaces in seawater as zinc oxide. The oxide film was porous and consisted of
many discrete single-crystal plates on the order of 10-100 Jlm in diameter. The growth
of the dissolution products was largely by plate broadening with limited thickening. The
morphology of dissolution products of zinc anode has been found to be a function of
current density and flow rate of electrolyte [1069].
15.4.1.3. Efficiency and Effect oJ Temperature. Zinc anodes normally have a high
efficiency of around 95% or higher [1192]. Waldron and Peterson [230] tested zinc anodes
in seawater and reported that all showed an efficiency of nearly 100%, whether containing
small amounts of Fe, AI, and Cd or not.
0.5% AI

------ ....

~~

0.1% AI

~
)(

til

:;
0

s::.

'"Q;

a.
E

4L-----------~------------~------------~

0.08

0.04

0.12

% Cadmium
FIGURE 15.11. Effect of small amounts of cadmium in zinc anodes on current output in seawater. From
Waldron and Peterson [230). Copyright by NACE International. All Rights Reserved by NACE; reprinted
with permission.

407

CORROSION IN OTHER ENVIRONMENTS

TABLE 15.7. Effect of Temperature on Efficiency of a Zinc Alloy


Tested for 28 Days with Impressed Current in Synthetic Seawater"
(DC)

Galvanic efficiency
(%)

Anodic potential (V seE)

21
38
43
54
60
69
74

95
91
89
86
84
76
70

-1.04
-1.03
-1.02
-1.01
-1.01
-1.00
-0.97

Ambient temperature

"From Kurr 1471J. Copyright by NACE International. All Rights Reserved by NACE:
reprinted with pennission.

The efficiency of a zinc anode may change with temperature. Kurr [471] reported
that the efficiency of an aluminum-containing zinc alloy anode (0.1-0.5% AI) decreased
with increasing temperature as shown in Table 15.7. At room temperature the efficiency
was 95%, but at 74C it was only 70%. The drop in efficiency is due to the increased
self-corrosion associated with the higher intergranular corrosion of AI-containing zinc
alloys at higher temperatures. Ahmed et al. [154] found that a Zn-0.3AI-0.03 Cd anode
failed prematurely at 70C in seabed mud owing to intergranular corrosion of the anode.
Another phenomenon that may be associated with the performance of zinc anodes
at higher temperatures is polarity reversal (see Chapter 7). It has been found that the
predominant factor in polarity reversal in hot water is the interaction between zinc and
the water and that the presence of a small amount of Fe in the zinc anode does not alter
the extent ofreversal [447].
15.4.2. Anodes for Impressed Current Cathodic Protection

In addition to its application as a sacrificial anode material, zinc has also been used
as an anode material for impressed current cathodic protection of steel-reinforced concrete structures [267, 310]. In this application, zinc is thermally sprayed onto the surface
of the concrete to form a continuous coating. During the operation of such a system, an
anodic current, sufficient for the protection of the steel inside concrete, is impressed
through the zinc coating via a rectifier. Good adhesion of the coating to the concrete
surface and a small driving voltage under a given current density are required to ensure
an effective performance of such a cathodic protection system [1286].
Brousseau et al. [520, 1286, 1287] investigated the factors affecting the adhesion
and driving voltage. They found that the adhesion of sprayed zinc coatings on concrete
can be affected by moisture content, temperature, current density, time, coating thickness,
and concrete surface texture. It was concluded that, in order to maximize the adhesion,
the concrete should be dry, warm, and grit-blasted at low air pressure without exposing
too much aggregate. These authors also found that the driving voltage tends to increase
with the applied current density and polarization time owing to the formation of zinc
dissolution products at the coating/concrete interface. Sprayed Zn-15%AI is not suitable
for using as an impressed current anode; severe disbonding and blistering of the coating
was observed during the impressed current test with this alloy.

References
I. Pourbaix, M.: Alias of Eleclrochemical Equilibria in Aqueous Solutions, 2nd ed., National Association
of Corrosion Engineers, Houston, pp. 406-413,1974.
2. Nevison, D. C. H.: Corrosion of zinc, in ASM Mewls Handbook, 9th ed., Vol. 13, 755-769, ASM
International, Materials Park, Ohio, 1987.
3. Kannangara, D. C. W., and Conway, B. E.: Zinc oxidation and redeposition processes in aqueous alkali
and carbonate solutions, 1. Electrochem. Soc. 134,894-918,1987.
4. Mattsson, E.: The atmospheric corrosion properties of some common structural metals-a comparative
study, Mater. Peiform. 21,9-19,1982.
5. Frydrych, D. 1.: Corrosion mechanisms of zinc-rich organic coatings on steel, Ph. D. Thesis, University
of Pennsylvania, 1986.
6. Kita, H.: Periodic variation of exchange current density of hydrogen electrode reaction with atomic
number and reaction mechanism, 1. Electrochem. Soc. 113, 1095-1111, 1966.
7. Lee, T. S.: Hydrogen overpotentlal on pure metals in alkaline solution, 1. Electrochem. Soc. 118,
1278-1282, 1971.
8. Bockris, J. 0' M., and Reddy, A. K. N.: Modern Electrochemistry, Vol. l, Plenum Press, New York.
1970.
9. Lee, T. S.: Hydrogen overpotential on zinc alloys in alkaline solution, 1. Electrochem. Soc. 122,
171-173,1975.
10. Mansfeld, F., and Gilman, S.: The effect of several electrode and electrolyte additives on the corrosion
and polarization behavior of the alkaline zinc electrode, 1. Electrochem. Soc. 117, 1328-1333, 1970.
II. Bard, A. J. (ed.): Encyclopedia of Electrochemislry of the Elements, Vol. IX, Part A, Marcel Dekker,
New York, 1982.
12. Bockris, J. O. M., Nagy, Z., and Danjanovic, A.: On the deposition and dissolution of zinc in alkaline
solutions, 1. Eleclrochem. Soc. 119,285-295, 1972.
13. Uhlig, H. H., and Revie, R. w.: Corrosion and Corrosion Control-An Introduction to Corrosion
Science and Engineering, 3rd ed., John Wiley & Sons, New York, 1985.
14. Berke, N. S., and Friel, J. J.: Applications of electrochemical techniques in screening metallic-coated
steels for allTIosphenc use, In Laborarory CorrosIOn Tests and Standards, STP 866, pp. 143-158,
American Society for Testing and Materials, Philadelphia, 1985.
15. Gad Allah, A. G., Hefny, M. M., Salih, S. A., and EI-Basiouny, M. S.: Corrosion inhibition of zinc in
HCI solution by several pyrazole derivatives, Corrosion 45, 574-578, 1989.
16. Augustynski, J., Dalard, F., and Sohm, 1. c.: Oxydation anodique du zinc en milieu faiblement basique,
Corros. Sci. 12,713-724, 1972.
17. Augustynski, J.: Etude de la rupture de passivite de certains metaux electrochimiquement actifs, Corros.
Sci. 13, 955-965, 1973.
18. Liu, M., Cook, G. M., and Yao, N. P.: Passivation of zinc anodes in KOH electrolytes, 1. Electrochem.
Soc. 128, 1663-1668, 1981.
19. Bocchi, N., da Cunha, M. R., and D' Alkaine, C. V: The passivating films of zinc in alkaline solution,
Key Eng. Maler. 20-28, 3941-3946, 1988.

409

410

REFERENCES

20. Aurian-Blajeni, B., and Tomkiewicz, M.: The passivation of zinc in alkaline solutions, J. Electrochem.
Soc. 132, 1511-1515, 1985.
21. De Pinto, I. S., De Pauli, C. P., Herrera, H .. Mishima, H., and Lopez, B. A: Effect of arsenate anion on
the dissolution and passivation of zinc electrode in slightly alkaline solutions, Electrochim. ACTa 31,
527-533, 1986.
22. Bocchi, N., and D' Alkaine, C. Y.: Zinc behavior in slightly alkaline solutions. The reduction processes,
Key Eng. Mater. 20-28,417-423, 1988.
23. D' Alkaine, C. Y., and da Cunha, M. R.: Influence of the coj- on the passivation of zinc in slightly
alkaline solutions, International Congress on Metallic Corrosion, Toronto, Ontario, pp. 122-125, 1986.
24. Chang, Y, and Prentice, G.: A model for the anodic dissolution of the zinc electrode in the prepassive
region, J. Electrochem. Soc. 136,3398-3403,1989.
25. Elder, 1. P.: The electrochemical behavior of zinc in alkaline media, J. Electrochem. Soc. 116,757-762,
1969.
26. Powers, R. W., and Breiter, M. W.: The anodic dissolution and passivation of zinc in concentrated
potassium hydroxide solutions, J. Electrochem. Soc. 116,719-729, 1969.
27. Powers, R. W.: Film formation and hydrogen evolution on the alkaline zinc electrode, J. Electrochem.
Soc. 118,685-695, 1971.
28. Szpak, S., and Gabriel, C. 1.: The Zn-KOH system: The solution-precipitation path for anodic ZnO
formation,J. Electrochem. Soc. 126, 1914-1923, 1979.
29. Powers, R. W.: Anodic films on zinc and the formation of cobwebs, 1. Electrochem. Soc. 116,
1652-1659,1969.
30. Burleigh, T. D.: Anodic photocurrents and corrosion currents on passive and active-passive metals,
Corrosion 45, 464-472,1989.
31. van Ooij, W. 1., and Sabata, A: Under-vehicle corrosion testing of primed zinc and zinc alloy-coated
steels, Corrosion 46, 162-171, 1990.
32. Darwish, N. A: Contribution to the electrochemical behaviour of some Zn-Ni systems in halide media,
J. Electrochem. Soc. India 32, 259-267, 1983.
33. Dattilo, M.: Polarization and corrosion of electrogalvanized steel-evaluation of zinc coatings obtained
from waste-derived zinc electrolytes, J. Electrochem. Soc. 132,2557-2561,1985.
34. Wu, 1. K., and Hsu, Y S.: Electrochemical corrosion of zinc in sodium chlorite, J. Mater. Sci. 21,
3475-3478, 1986.
35. Lee, H. H., and Hiam, D.: Corrosion resistance of galvannealed steel, Corrosion 45,852-856, 1989.
36. Budinski, M. K., and Wilde, B. E.: An electrochemical criterion for the development of galvanic coating
alloys for steel, Corrosion 43, 60-62, 1987.
37. Abd EI Haleem, S. M.: Environmental factors affecting the pitting corrosion potential of a zinc-titanium
alloy in sodium hydroxide solutions, Br. Corros. J. 14, 171-175. 1979.
38. Shams EI Din, AM., Abd EI Kader, 1. M., and Badran, M. M.: Galvanic corrosion in the copperlzinc
system. Part i. Potential distribution in relation to the nature of the cathode and to the type of anion in
solution, Br. Corros. 1. 16,32-37,1981.
39. Hutchison, P. E, and Turner, 1.: Some aspects of the electrochemical behavior of zinc in the presence
of acrylate and methacrylate anions, J. Electrochem. SOC. 123, 183-186, 1976.
40. Hefny, M. M., Gad-Allah, A. G., Salih, S. A, and EI-Basiouny, M. S.: Zinc passivation in oxalate
solution, Corrosion 44, 691-695,1988.
41. LeRoy, R. L.: Evaluation of corrosion rates from nonlinear polarization data, J. Electrochem. Soc. 124,
1006-1012,1977.
42. Kapali, V., Srinivasan, K. N., Venkataraman, B., and Balakrishnan, K: Development of a series of
AI-Zn-In ternary alloy anodes for the cathodic protection of submerged structures, Key Eng. Mater.
20-28, 955-962, 1988.
43. Lambert, M. R, Hart, R. G., and Townsend, H. E.: Corrosion mechanism ofZn-Ni alloyelectrodeposited
coatings, at The 2nd Automotive Corrosion Prevention Conference, Dearborn, Michigan, 1983, Paper
No. 831817.
44. Short, N. R., Abibsi, A, and Dennis, 1. K.: Corrosion resistance of electroplated zinc alloy coatings,
Trans. Inst. Met. Finish. 67 (Part 3), 73-77, 1989.
45. Alvarez, M. G., and Galvele, J. R.: Pitting of high purity zinc and pitting potential significance,
Corrosion 32, 285-294, 1976.

411

REFERENCES

46. Abd El Haleem. S. M.: Dissolution current and pitting potential of zinc in KOH solutions in relation to
the concentration of aggressive ions, Br. Corros. 1.11,215-218,1976.
47. Szklarska-Smialowska, Z.: Pitting Corrosion of Metals, National Association of Corrosion Engineers,
Houston, 1986.
48. Devillers, L. P.: The mechanism of aqueous intergranular corrosion in zinc-aluminium alloys, Ph. D.
Thesis, University of Waterloo, 1974.
49. Devillers, L. P., and Niessen, P.: The mechanism of intergranular corrosion of dilute zinc-aluminium
alloys in hot water, Corros. Sci. 16,243-252, 1976.
50. Devillers, L. P., and Niessen, P.: The mechanism of intergranular corrosion of dilute zinc-aluminium
alloys in hot water, Corros. Sci. 16,243-252,1976.
51. American Hot Dip Galvanizers Association, Wet Storage Stain, Brochure, 1984.
52. Helwig, L. E.: Comparison of humid-storage tests for chromated galvanized sheet, Met. Finish. 75(5),
33-38, 1977.
53. Bird, C. E., and Strauss, F. J.: Effect of wet storage staining on subsequent atmospheric corrosion of
galvanized iron sheets. Mater. Perform. 15(11),27-29, 1976.
54. Cominco PTC Project No. 83-5-11, August 1990.
55. EI-Mallah, A. T.. and Magd, M. R. G. A: Chromate passivation of zinc, Met. Finish. 82(2),49-54, 1984.
56. Helwig, L. E.: Chromate treatment of galvanized sheet, Met. Finish. 68(7), 54-59. 1970.
57. Williams, L. F. G.: Chromate conversion coatings on zinc, Plating 1972, 931-938.
58. Williams, L. F. G.: The formation and performance of chromate conversion coatings on zinc, Surf.
Technol. 5,105-117,1977.
59. Williams, L. F. G.: The mechanism of formation of chromate conversion films on zinc, Surf Technol.
4,355-366,1976.
60. Paatsch. w.: Quality testing of coatings produced by chromating zinc, Met. Finish. 76(9), 15-17, 1978.
61. Epelboin, I., Ksouri, M., and Wiart, R.: On a model for the electrocrystallization of zinc involving an
autocatalytic step, 1. Electrochem. Soc. 122, 1206-1214, 1975.
62. McBreen, J., and Cairns, E. J.: The zinc electrode, in Advances in Electrochemistry and Electrochemical
Engineering, Vol. II, H. Gerischer and C. W. Tobias (eds.), pp. 273-352, John Wiley & Sons, New York.
1978.
63. Abdel Aal, M. S" Radwan, S .. and EI-Saied, A: Phenothiazines as corrosion inhibitors for zinc in NH4 CI
solution, Br. Corros. 1.18, 102-106, 1983.
64. Fiaud, c., Bensarsa, S., des Aulnois, I. D., and Tzinmann, M.: Inhibiting properties of phosphines against
zinc corrosion in acidic media. Br. Corros. 1. 22,109-112,1987.
65. Ostrander. C. W.: Chromate conversion coatings, in Electroplating Engineering Handbook, 3rd ed., A.
K. Graham (ed.), pp. 437-447, Van Nostrand Reinhold, Toronto, 1971.
66. Krishnamurthy, S.: Chromate conversion coatings on zinc plating, 1. Electrochem. Soc. India 30, 47 -51,
1981.
67. Van de Leest, R. E.: Yellow chromate conversion coatings on zinc: chemical composition and kinetics,
Trans. Inst. Met. Finish. 56, 51-54,1978.
68. Lazauskas, v., Sarmaitis, R., Butkevicius. J" and Matulis, J.: The role of sulfate ions in chromating zinc
and cadmium, Plating Surf. Finish. 66(1),58-60, 1979.
69. Helwig, L. E., and Bird, J. E.: Protection of galvanized coatings from attack by organic acid vapors,
Met. Finish. 7l( II), 45-48, 52. 1973.
70. Helwig, L. E., and Swan, J. D.: Use of spot-test reagents for detecting chromates on hot-dip galvanized
surfaces, Met. Finish. 80(9),37-40, 1982.
71. Duncan, J. R.: Electron spectroscopy of chromated galvanized steel sheet after heating, immersion in
water or outdoor weathering, Surf. Techno!. 17,265-276, 1982.
72. Lengyel, B., Baroti-Labar, Z., and Kahan, R.: Study of chromate coatings by means of electrode
impedance measurements, Mater. Chem. 7,183-194,1982.
73. Efimov, E. A, Tok, L. D., and Tverdynina, T. B.: Mechanism of the reduction of chromic acid anions
to trivalent chromium ions, SOy. Electrochem. 25,1249-1251,1989 [in English].
Chromate passivation protection of Zn- and Al-Zn-coated steel
74. Townsend, H. E., and Zoccola, J.
sheet against wet-storage stain, 1. Electrochem. Soc. 125, 1290-1292, 1978.
75. Van de Leest, R. E.: Yellow chromate conversion coatings on zinc: Electrochemical measurements and
corrosion tests, Werkst. Korros. 29.648-653, 1978 [in English].

c.:

412

REFERENCES

76. Pearlstein, E: Selection and application of inorganic finishes. Part II. Chromate coatings, Plating Suif.
Finish. 66(1), 30-34,1979.
77. Treverton, J. A., and Amor, M. P.: The structures and surface composition of chromate conversion
coatings: An XPS and SEM study, Trans. Inst. Met. Finish. 60, 92-96, 1982.
78. Van de Leest, R E., and Wessels, J.: Stabilization of black chromate conversion coatings on zinc, Plating
Surf Finish. 67(5), 86-89, 1980.
79. Spring, S., and Woods, K.: Prepaint treatments-chromate coatings, Met. Finish. 79(6),49-55, 1981.
80. Choosing chromate and phosphate processes for corrosion resistance, Corros. Prevo Control 1981
(February), 16-18,32-33.
81. Rozovsky, V. G., and Sarmaitis, R R: Corrosion resistance of black chromate conversion films on zinc,
Plating Suif. Finish. 68(10), 54-56,1981.
82. Sarmaitis, R: Corrosion resistance of chromate films on zinc, Plating Suif. Finish. 68(9), 83-86, 1981.
83. Perfetti, B. M.: A comparison of pretreatments for prepainted galvanized sheets, Met. Finish. 81(5),
57-60, 1983.
84. Timmins, E D.: Adhesion of paint to zinc surfaces, Ind. Carras. 1(3), 15-17, 1983.
85. Murphy, M.: Technical developments in 1983. Inoraganic metallic finishes, processes and equipment,
Met. Finish. 82(2), 17-32, 1984.
86. Gallaccio, A., Pearlstein, E, and D' Ambrosio, M. R: Effects of heating chromate conversion coatings,
Met. Finish. 64(8),50-54, 1966.
87. Bardin, P. C: Conversion coatings-prepare galvanized steel for painting, Ind. Finish., 1967 (July),
58-61.
88. Pearlstein, E, and D'Ambrosio, M. R: Heat resistant chromate conversion coatings, Plating 55.
345-346,1968.
89. Helwig, L. E.: Heavy chromate conversion treatments on zinc-coated steel, Met. Finish. 70(9),51-55,
1972.
90. Chromate conversion coatings for zinc & cadmium, Met. Finish. Plants Processes 7(1),15-24, 1971.
91. Bridger, R. D. E. L.: Pretreatment of galvanized surfaces prior to powder coating, Trans. Inst. Met.
Finish. 61, 13-18, 1983.
92. Centre de Recherches Metallurgiques (Belgium): White Rust Inhibition for Galvanized Steel. International Lead Zinc Research Organization Project ZE-348, Final Report, February 1989.
93. Jardy, A., Rosset, R, and Wi art, R: Diphosphate coatings for protection of galvanized steel: quality
control by impedance measurements, 1. Appl. Electrochem. 14,537-545, 1984.
94. Kargol, J. A., and Jordan, D. L.: The influence of phosphorous alloy additions on the zinc phosphate
coating formation on cold rolled steels, Corrosion 38(4), 201-206, 1982.
95. Nancollas, G. H.: Phosphate precipitation in corrosion protection: Reaction mechanisms, Corrosion
39(3),77-82, 1983.
96. EI-Mallah, A. T., Morsy, S. M., and Ibrahim, I. G.: Electrophosphate coatings on zinc, Met. Finish. 82(8),
31-72,1984.
97. Leidheiser, H., and Suzuki, I.: Cobalt and nickel cations as corrosion inhibitors for galvanized steel, 1.
Electrochem. Soc. 128,242-249, 1981.
98. Bijimi, D., and Gabe, D. R.: Passivation studies using group VIA anions. Part III. Anodic treatment of
zinc, Br. Carras. 1. 18, 138-141, 1983.
99. Kargol, J. A., Jordan, D. L., and Palermo, A. R.: The influence of high strength cold rolled steel and
zinc coated steel surface characteristics on phosphate pretreatment, Corrosion 39, 213-218, 1983.
100. Wrubl, C, and Mor, E. D.: Zinc gluconate as an inhibitor of the corrosion of copper and zinc in sea
water, Br. Carras. 1.18,142-147,1983.
101. Kuznetsov, Y. I., and Podgornova, L. P.: Corrosion and electrochemical behavior of zinc in phosphate
solutions, Zashch. Met. 19,98-102,1983 [in English).
102. Wilcox, G. D., and Gabe, D. R: Chemical molybdate conversion treatments for zinc, Met. Finish. 86(9),
71-74,1988.
103. Leidheiser, H., and Suzuki, I.: Towards a more corrosion resistant galvanized steel, Corrosion 36,
701-702, 1980.
104. Macias, A., and Andrade, C.: Corrosion of galvanized steel reinforcements in alkaline solutions. Part I.
Electrochemical results, Br. Carras. 1. 22,113-118,1987.

REFERENCES

413

105. Macias. A .. and Andrade. c.: Corrosion of galvanized steel reinforcements in alkaline solutions. Part
II. SEM study and identification of corrosion products, Br. Corros . .I. 22. 119~ 129. 1987.
106. Leidheiser. H., Simmons. G. W, Nagy. S., and Music, S.: Mossbauerspectroscopic study of the corrosion
inhibition of zinc by cobalt ions,./. Electrochem. Soc. 129. 16S8~1662. 1982.
107. Leidheiser. H., Momose. Y, and Granata, R. D.: Inhibition of the cathodic reaction on zinc in aerated
3% NaCI solution, Corrosion 38, 178~ 179, 1982.
108. LeRoy, R. L.: Organic wet storage stain inhibitor-a substitute for chromates, Mate/: Pofilnn. 19(8).
54~55, 1980.
109. Shams EI Din. A. M., Abd EI Haleem, S. M .. and Abd EI Kader, 1. M.: Studies on the pitting corrosion
of zinc in aqueous solutions. Part II. Measurement of pitting corrosion currents operating under natural
conditions, .l. Electroanal. Chem. 65, 335~349, 1975.
110. Baugh, L. M.: Corrosion and polarization characteristics of zinc in neutral~acid media. Part I. Pure zinc
in solutions of various sodium salts, Elcctrochim. Acta. 24, 657~667, 1979.
III. Baugh, L. M.: Corrosion and polarization characteristics of zinc in neutral~acid media. Part II. Effect
of NH! ions and the role of battery zinc alloying constituents. Electrochim. Acta 24. 669~677, 1979.
I 12. Yamashita, T.: Electro-deposition and electro-dissolution of zinc in the vicinity of equilibrium potential
in aqueous zinc sulfate solutions . .I. Electroanal. Chel1l. 106, 95~ 102, 1980.
113. Deslouis. c.. Duprat. M .. and Tuiet-Tournillon, c.: The cathodic mass transport process during zinc
cOlTosion in neutral aeratcd sodium sulphate solutions,.I. Electmanal. Chem. 181, 119~136. 1984.
114. Boto. K. G .. and Williams. L. F. G.: The determination of the cOlTosion rate of zinc in solution by the
differential pulse method . ./. Electrochem. Soc. 124. 6S6~661, 1977.
liS. Era, A .. Takehara. Z .. and Yoshizawa. S.: Influence of impurities especially lead contained in manganese
dioxide upon the self-discharge of the leclanche dry cell, Electrochim. Acta 13. 183~396. 196X
116. Zembura. Z .. and Burzynska. L.: The corrosIOn of zinc III de-aerated O. I M NaCIIn the pH range trom
I. 6to 13.3. Corms. Sci. 17. 871~878. 1477.
117. Fujishima. A .. Kato. T., Maekawa. E.. and Honda, K.: Mechanism of the currem doubling effect. Part
I. The ZnO photaanode in aqueous solution of sodium formate, Bull. Chern. SoC, .lpn. 54. 1671 ~ 1674.
1984.
118. Walter, G. W: Corrosion rates of zinc, zinc coatings and steel in aerated slightly acidic chloride solutions
calculated from low polarization data. Corms. Sci. 16, S73~586. 1976.
119. Helwig. L. E.: Premature darkening of galvanized roofing sheets. Part L Mel. Finish. 82(4). 41~46,
1484.
120. Helwig. L. E.: Premature darkening of galvanized roofing sheets. Part II, Met. Finish. 82(5). 61~64.
1984.
121. Mansfeld, F., and Kenkel, J. V: Electrochemical monitoring of atmospheric corrosion phenomena.
Corms. Sci. 16. III ~ 122, 1976.
122. Yan, B., Meilink. S. L., Warren. G. W. and Wynblatt. P.: IEEE Trans. Componellls, Hybrids Mallufac
turing Technoi .. CHMT-IO. 247, 1987.
123. Stratmann, M.: The investigation of the corrosion properties of metals, covered with adsorbed electrolyte
layers-a new experimental technique, Corros. Sci. 27, 869~872, 1987.
124. Stratmann, M., and Streckel, H.: On the atmospheric corrosion of metals which are covered with thin
electrolyte layers. Part I. Verification of the experimemal technique, Corms. Sci. 30, 681 ~696. 1990.
125. Stratmann, M .. and Streckel. H.: On the atmospheric corrosion of metals which are covered with thin
electrolyte layers. Part II. Experimental results, Corros. Sci. 30, 697~ 714, 1990.
126. Stratmann, M .. Streckel, H .. Kim, K. T., and Crockett, S.: On the atmospheric corrosion of metals which
are covered with thin electrolyte layers. Part III. The measurement of polarisation curves on metal
surfaces which are covered by thin electrolyte layers, Carras. Sci. 30, 715~ 734, 1990.
127. Kaesche, H.: The passivity of zinc in aqueous solutions of sodium carbonate and sodium bicarbonate,
Electrochim. Acta 9, 383~394, 1964.
128. Bota, K. G., and Williams, L. F. G.: Rotating disc electrode studies of zinc corrosion,./. Electroanal.
Cizem. 77. 1~20, 1977.
129. Gerischer. H.: Electrochemical behavior of semiconductors under illumination,.I. Eieelrochem. Soc.
113, 1174~ I 182, 1966.
130. Epelboin, I.. Ksouri, M., and Wi art, R.: On a model for the electrocrystallization of zinc involving an
autocatalytic step,.I. Electmehem. Soc. 122, 1206~ 1214, 1975.

414

REFERENCES

131. Isaacson, M. J., McLarnon, E R, and Cairns. E. J.: Zinc electrode rest potentials in concentrated
KOH-K2Zn(OH)4 electrolytes. 1. Electrochem. Soc. 137,2361-2364,1990.
132. St-Pierre, J . and Piron. D. L.: Mechanism of cathodic potential oscillations of the zinc electrode in
alkaline solutions. 1. Electrochem. Soc. 137,2491-2498, 1990.
133. Bard, A. J .. and Faulkner, L. R: Electrochemical Methods-Fundamentals and Applications, John Wiley
& Sons. New York. 1980.
134. Mansfeld, E, Kendig, M. w., and Tsai. S.: Recording and analysis of AC impedance data for corrosion
studies. Part II. Experimental approach and results, Corrosion 38. 570-580. 1982.
135. Mansfeld. E, and Kendig, M.: Evaluation of protective coatings with impedance measurements,
Proceedings of the International Congress on Metallic Corrosion, Toronto. Ontario. 1984. Vol. 3. pp.
74-84.
136. Mansfeld, E: Recording and analysis of AC impedance data for corrosion studies. Part I. Background
and methods of analysis. Corrosion 36. 301-307,1981.
137. Macdonald, D. D.: Theoretical analysis of electrochemical impedance, National Association of Corrosion Engineers Corrosion '87 Conference, San Francisco, March 9-13,1987, Paper No. 479.
138. Macdonald, D. D.: Some advantages and pitfalls of electrochemical impedance spectroscopy, Corrosion
46.229-242 (1990).
139. Brown, R P., and Kessler. R. J.: One year laboratory testing of plain and galvanized rebars in chloride
containing concrete, Dept. of Transportation, Florida Corrosion Report No. 78-2, 1978.
140. Silverman, D. c.: Rapid corrosion screening in poorly defined systems by electrochemical impedance
technique. Corrosion 46. 589-598, 1990.
141. Mansfeld, E. Kendig, M. W., and Tsai, S.: Evaluation of corrosion behavior of coated metals with AC
impedance measurements. Corrosion 38, 478-485,1982.
142. Sagues, A A: Corrosion measurements of reinforcing steel in concrete exposed to various aqueous
environments, National Association of Corrosion Engineers Corrosion' 87 Conference, San Francisco,
March 9-13,1987, Paper No. 118.
143. Wheat, H. G., and Eliezer, Z.: Comments on the identification of a chloride threshold in the corrosion
of steel in concrete, Corrosion 43,126-128,1987.
144. Kendig, M., and Mansfeld, E: Corrosion rates from impedance measurements: An improved approach
for rapid automatic analysis, Corrosion 39, 466-467, 1983.
145. Matsuoka, K.: Monitoring of corrosion of reinforcing bar in concrete, National Association of Corrosion
Engineers Corrosion '87 Conference. San Francisco, March 9-13, 1987, Paper No. 121.
146. Berthier, E, Diard. J. P.,Jussiaume, A, and Rameau, J. J.: The instantaneous impedance of non-stationary
electrochemical systems: Application to a corroding zinc electrode, Corros. Sci. 30. 239-247, 1990.
147. Hendrikx, J., Visscher, W., and Barendrecht, E.: Impedance measurements of zinc and amalgamated
zinc electrodes in alkaline electrolyte, Electrochim. Acta 30, 999-1006, 1985.
148. Rodgers, R .. Rothstein, M. L.. and Palus, A E: Instrumentation for the automated A.C. impedance
analysis of electrochemical systems, Proceedings of the National Association of Corrosion Engineers
Metallography and Corrosion Conference. July 1986, pp. 203-220.
149. Castle, J. E.: Corrosion test methods, Br. Corros. 1. 22, 77-82,1987.
I SO. Siebert, O. W.: Laboratory electrochemical test methods, Laboratory Corrosion Tests and Standards,
STP 866, American Society for Testing and Materials, Philadelphia, 65-90, 1985.
151. American Society for Testing and Materials: Standard reference test method for making potentiostatic
and potentiodynamic anodic polarization measurements. in Annual Book ofASTM Standards, Vol. 03.02,
No. G5-87, pp. 79-85,1989.
152. American Society for Testing and Materials: Standard practice for conventions applicable to electrochemical measurements in corrosion testing, in Annual Book of ASTM Standards, Vol. 03.02, No. G3,
pp.62-67, 1989.
153. American Society for Testing and Materials: Standard practice for conducting potentiodynamic polarization resistance measurements, in Annual Book of ASTM Standards, Vol. 03.02. No. G59-78, pp.
221-224, 1989.
154. Ahmed, D. S., Ashworth, v., Scantlebury, J. D., and Wyatt, B. S.: Mechanism of intergranular attack in
Zn-AI-Cd sacrificial anodes at elevated temperatures, Br. Corros. 1. 24, 149-152, 1989.
ISS. Clarke, S. G., and Andrew, J. E: The chromate passivation of zinc, 1. Electrodepos. Tech. Soc. 20,
119-138,1945.

REFERENCES

415

156. Stratmann, M., Streckel, H., Kim, K. T., and Crockett. S.: On the atmospheric corrosion of metals which
are covered with thin electrolyte layers. Part III. The measurement of polarisation curves on metal
surfaces which are covered by thin electrolyte layers, Corros. Sci. 30, 715-734, 1990.
157. Stratmann, M., and Streckel, H.: On the atmospheric corrosion of metals which are covered with thin
electrolyte layers. Part II. Experimental results, Corros. Sci. 30, 697-714. 1990.
158. Gonzalez, J. A., Otero. E .. and Cabanas. c.: Recording polarisation curves in thin electrolyte layers by
electrochemical atmospheric corrosion monitors. Br. Corros. 1. 25.125-130,1990.
159. White, R. T.: A review of corrosion inhibitors. Part I, Corrosion and Coatings South Africa 1986(8), 2.
4,6,8.
160. White, R T.: A review of corrosion inhibitors. Part II, Corros. Coatings S. Afr. 1986( 10), 4, 6, 10.
161. Leidheiser, H.: A review of proposed mechanisms for corrosion inhibition and passivation by metallic
cations, Corrosion 36, 339-345, 1980.
162. Leroy, R L.: Polythoiglycolate passivation of zinc, Corrosion 34. 113-119, 1978.
163. Rausch, w., and Mueller, G., (MetallgesellschaftA. G., Germany): Process for phosphating metals, U.S.
Patent 4637838. January 20, 1987.
164. Abdel Aal, M. S., Abdel Wahab, A. A., and EI Saied, A.: A study of the inhibiting action of benzene
thiols and related compounds on the corrosion of zinc in acidic media, Corrosion 37, 557-563, 1981.
165. Penninger, J. (Henkel Kommanditgesellschaft auf Aktien, Germany): Method for inhibiting corrosion
of zinc using bis-triazoles, U.S. Patent 4636359, January 13, 1987.
166. Abd EI Haleem, S. M., and Abdel Fattah, A. A.: The role of some organic anions in promoting or
inhibiting the corrosion of zinc, Surf Coatings Techno/. 29, 41-49. 1986.
167. Massinon, D., Dauchelle. D., and Charbonnier, 1. c.: On the contribution of electrochemical methods
in the study of corrosion mechanisms in automotive body steel sheets, Mater. Sci. Forum, 44-45,
461-476, 1989.
168. Shastry, C R, and Townsend. H. E.: Mechanisms of cosmetic corrosion in painted zinc and zinc-alloycoated sheet stcels, Corrosion 45, 103-118. 1989.
169. Shastry, C. R., and Townsend, H. E.: Failure mechanisms of painted sheet steel in a chipping corrosion
test, Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo. September 5-7, 1989, pp. 511-518.
170. Serra, E. T., and de L. Fragata, E: Surface treatments of galvanized steel for painting, International
Congress on Metallic Corrosion. Toronto, Ontario, 1984, pp. 384-389.
171. Lengyel, 8., Meszaros, L., Kahan, R., and Jeney, I.: Study of chromate coatings by impedance technique,
Proceedings of the Symposium on Electroplating, Budapest, Hungary, April 16-19, 1985.
172. Abu Zahra, R. H .. and Shams EI Din, A. M.: A radiotracer study on the reaction between zinc and
chromate, Corros. Sci. 5, 517-524.1965.
173. Friel. J. J.: Atmospheric corrosion products on AI, Zn. and AIZn metallic coatings. Corrosion 42,
422-426, 1986.
174. Sergi, G., Short. N. R., and Page. C L.: Corrosion of galvanized and galvannealed steel in solutions of
pH 9. to 14.0, Corrosion 41. 618-624, 1985.
175. Macias, A., and Andrade, c.: Corrosion rate of galvanized steel immersed in saturated solutions of
Ca(OH}z in the pH range 12-13. 8, Br. Corros. 1.18,82-87,1983.
176. Jones, D. A .. and Nair, N. R.: Electrochemical corrosion studies on zinc-coated steel, Corrosion 41,
357-362, 1985.
177. Maahn, E .. and Sorensen, B.: The influence of microstructure on the corrosion properties of hot-dip
galvanized reinforcement in concrete, Corrosion 42,187-196,1986.
178. Kiss, K., and Coll-Palagos, M.: Cyclic voltammetric and scanning electron microscopic evaluation of
the corrosion resistance ofZn-phosphated steel, Corrosion 43,8-14.1987.
179. Mor, E. D., and Beccaria, A. M.: Effect of temperature on the corrodibility of copper and zinc in synthetic
sea water, Corrosion 31, 275-279, 1975.
180. Batrakov, Y Y, Bogdanov, N. Y, and Kocherov, A. V.: The application of alternating current method
for adsorption and corrosion measurement on iron and zinc, Proceedings of the International Congress
on Metallic Corrosion, Toronto, Ontario, June 3-7,1984, Vol. I, pp. 287-290.
181. Johnson, 1. W., Sun, Y. C, and James, W. 1.: Anodic dissolution ofZn in aqueous salt solutions, Corros.
Sci. 11, 153-159. 1971.

416

REFERENCES

182. James, W. J., Straumanis, M. E., and Johnson, J. W: Anodic disintegration of metals undergoing
electrolysis in aqueous salt solutions, Corrosion 23,15-23,1967.
183. Stiles, D. C, and Edney, E. 0.: Dissolution of zinc into thin aqueous films as a function of residence
time, acidic species, and pH, Corrosion 45, 896-901. 1989.
184. Yee, S., Stratmann, M., and Oriani, R. A: Application of a Kelvin microprobe to the corrosion of metals
in humid atmospheres, 1. Electrochem. Soc. 138, 55-61, 1991.
185. Haynie, F. H., Spence, J. W, and Upham, 1. B.: Effects of air pollutants on weathering steel and
galvanized steel: A chamber study, in Atmospheric Factors Affecting the Corrosion of Engineering
Metals, STP 646, pp. 30-47, American Society for Testing and Materials, Philadelphia, 1978.
186. German, G.: Behavior of zinc-coated steel in highway environments, in Atmospheric Factors Affecting
the Corrosion of Engineering Metals, STP 646, pp. 74-82, American Society for Testing and Materials,
Philadelphia, 1978.
187. LegaUlt, R. A, and Pearson, Y. P.: Kinetics of the atmospheric corrosion of galvanized steel, in
Atmospheric Factors Affecting the Corrosion of Engineering Metals, STP 646, pp. 83-96, American
Society for Testing and Materials, Philadelphia, 1978.
188. Payer, J. H., and Safranek, W H.: Atmospheric corrosion of electroplated zinc alloy die castings, in
Atmospheric Factors Affecting the Corrosion of Engineering Metals, STP 646, pp. 115-128, American
Society for Testing and Materials, Philadelphia, 1978.
189. Zoccola,1. C, Townsend, H. E., Borzillo, A. R., and Horton, J. B.: Atmospheric corrosion behavIOr of
aluminum-zinc alloy-coated steel, in Atmospheric Factors Affecting the Corrosion of Engineering
Metals, STP 646, pp. 165-184, American Society for Testing and Materials, Philadelphia, 1978.
190. Guttman, H.: Effects of atmospheric factors on the corrosion of rolled ZInC, in Mewl Corrosion in the
Atmosphere, STP 435, pp. 223-239, American Society for Testing and Materials, Philadelphia, 1968.
191. Guttman, H., and Sereda, P. J.: Measurement of atmospheric factors affecting the corrosion of metals,
in Metal Corrosion in the Atmosphere, STP 435, pp. 326-359, American Society for Testing and
Materials, Philadelphia, 1968.
192. Guttman, H.: Atmospheric and weather factors in corrosion testing, in Atmospheric Corrosion, W. H.
Ailor (ed.), pp. 51-68, John Wiley & Sons, New York, 1982.
193. Mansfeld, E, and Kenkel, J. Y.: Electrochemical studies of atmospheric corrosion, Corrosion '75,
Symposium on the Protection and Performance of Materials, Toronto, Ontario, April 14-18, 1975, Paper
No. 161.
194. Glass, G. K., and Ashworth, Y.: The corrosion behaviour of the zinc-mild steel galvanic cell in hot
sodium bicarbonate solution, Corros. Sci. 25, 971-983,1985.
195. Mansfeld, E, Jeanjaquet, S. L., Roe, D. K., and Carey, R.: A new corrosion rate monitor and data logging
system for atmospheric corrosion studies, National Association of Corrosion Engineers Corrosion' 87
Conference, San Francisco, March 9- 13, 1987, Paper No.4 I I.
196. Hubbard, D. J., and Shanahan, C E. A.: Corrosion of zinc and steel in dilute aqueous solutions, Br.
Corros.l. 8, 270-274,1973.
197. Macias, A., and Andrade, C: Corrosion of galvanized steel in dilute Ca(OHh solutions (pH I I. I - I 2.6),
Br. Corros. 1.22,162-171,1987.
198. Calvo, E A., Otero, E., and Pardo, A.: Corrosion protection of galvanized steel pipe in acidic media by
surface amalgamation, Br. Corros. 1. 22, 185-189, 1987.
199. Wilcox, G. D., and Gabe, D. R.: Passivation studies using group VIA anions. Part Y. Cathodic treatment
of zinc, Br. Corros. 1. 22, 254-258, 1987.
200. Andrade, C, and Castelo, Y.: Practical measurement of the AC impedance of steel bars embedded in
concrete by means of a spectrum analyser (fast Fourier transform), Br. Corros. 1. 19, 98-100, 1984.
201. Gonzalez, J. A, Otero, E., Cabanas, C, and Bastidas, 1. M.: Electrochemical sensors for atmospheric
corrosion rates. A new design, Br. Corros. 1. 19, 89-94, 1984.
202. Blanco. M. T., Andrade, C, and Macias, A.: SEM study of the corrosion products of galvanized
reinforcements immersed in solutions in the pH range 12.6 to 13.6, Br. Corras. 1. 19,41-48,1984.
203. Defrancq, J. N.: Zinc and zinc-lead alloys in domestic water, Br. Corros. 1.17, 125-130, 1982.
204. Saleh, R. M., Ismail, A A., and El Hosary, A. A.: Corrosion inhibition by naturally occurring substances,
Part VII. The effect of aqueous extracts of some leaves and fruit-peels on the corrosion of steel, AI, Zn
and Cu in acids, Br. Corros. 1.17,131-135,1982.

REFERENCES

417

205. Shams EI Din, A. M., AbdEI Kader,J. M., and Kuhn, A. T.: Pitting corrosion of zinc in aqueous solutions.
Part III. Effect of anode to cathode surface area ratio on pitting corrosion currents, Br. Corros. 1. 15,
208-210,1980.
206. Muralidharan, V. S., and Rajagopalan, K. S.: Kinetics and mechanism of passivation of zinc in dilute
sodium hydroxide solutions, Br. Corros. 1. 14, 231-234, 1979.
207. Rudresh, H. B., and Mayanna, S. M.: n-Decylamine as a corrosion inhibitor for zinc in acidic solution,
Br. Corros. 1.12,54-56,1977.
208. Mor, E. D., and Beccaria, A. M.: Inhibitory action of acrylonitrile on the corrosion of zinc in sea water,
Br. Corros. 1.8,25-27,1973.
209. Mor, E. D., Beccaria, A. M., and Poggi, G.: Behaviour of zinc in sea water in the presence of sulphides.
Br. Corros. 1. (Quarterly) 1974(1). 53-56.
210. Mansfeld, E. and Tsai. S.: Laboratory studies of atmospheric corrosion. Part I. Weight loss and
electrochemical measurements, Corros. Sci. 20, 853-872. 1980.
211. Kesternich, W.: Corrosion testing of metallic and non-metallic coatings in an artificial industrial
atmosphere, Stahl Eisen 71( II), 1951.
212. Townsend, H. E.: Status of a cooperative effort by the automotive and steel industries to develop a
standard accelerated corrosion test, SAE Technical Paper No. 892569. Society of Automative Engineers
1989 Automotive Corrosion & Prevention Conference & Exposition, Dearborn, Michigan, December
4-6, 1989.
213. Lyon, S. B., Thompson, G. E., Johnson, 1. B., Wood, G. c., and Ferguson, 1. M.: Accelerated atmospheric
corrosion testing using a cyclic wet/dry exposure test: Aluminum, galvanized steel, and steel, Corrosion
43,719-726, 1987.
214. Lambert, M. R., Townsend, H. E., Hart, R. G., and Frydrych, D. J.: Accelerated corrosion tests of
precoated sheet steels for automobiles, Ind. Eng. Chem. Prod. Res. Dev. 24, 378-384. 1985.
215. Abbas, M. I.. Ahmed, M. S., and Shahin, M. A.: Effect of galvannealing heat treatment on the corrosion
characteristics of galvanized steel, CIM Bull. 82, 69-73, 1989.
216. Wilde, B. E.: An electrochemical criterion for galvanic protection of steel exposed to marine environments, C1M Bull. 83, /62-164, 1990.
217. Slunder, C. 1., and Boyd, W. K.: Zinc: Its Corrosion Resistance, 2nd ed., International Lead Zinc
Research Organization, Inc . New York, 1986.
218. Morgan. S. W. K.: Zinc and Its Alloys and Compounds, John Wiley & Sons, Toronto, 1985.
219. Ho, M. S.: Atmospheric Corrosion Results of ZM-1I3 Galvanized Samples after Fifteen Years of
Exposure, Cominco PTC Project No. 84-5-511, Report No. I. June 4. 1985.
220. Mercille, G. P.: Corrosion Evaluation of Zinc, No.3 Alloy and ZA-12 Alloy Embedded in Concrete for
10 Years, Cominco PTC Project No. 83-5-511. Report No.3, 1987.
221. McBreen. J . and Cairns, E. J.: The zinc electrode. in Advances in Electrochemistry and Electrochemical
Engineering, Vol. II, H. Gerischer and C. W. Tobias (eds.). pp. 273-352. John Wiley & Sons, New York.
1978.
222. Ashton, R. E. and Hepworth, M. T.: Effect of crystal orientation on the anodic polarization and passivity
of zinc. Corrosion 24, 50-53, 1968.
223. Goodrich. 1. D . and Schmid, G. M.: The dissolution of zinc in the NaNO r NH4CI-NH r H 20 system.
Corrosion 24. 252-254. 1968.
224. Devillers, L. P. and Niessen. P.: Accelerated grain boundary corrosIOn of dilute zinc-aluminum alloys
under high applied cathodic currents, Corrosion 32, 152-154, 1976.
225. Melton. K. N .. and Edington, J. W.: Effect of increasing aluminum content on the boiling water corrosion
resistance of superplastic Zn-Al alloys, Corrosion 31, 146-147, 1975.
226. Weast. R. c., and Shulman, S. L.: Corrosion of zinc in pressurized dilute aqueous solution systems,
Corrosion 18, 417-420,1962.
227. Steinbicker, R. N., and Fountoulakis, S. G.: Production of zinc-nickel electoplated coatings, Iron Steel
Eng. 1989(7), 28-31.
228. Dargis, R. G.: Zinc-alloy electroplates as functional finishes, Prod. Finish. 1990(10), 43-54.
229. Sharples. T. E.: Zn-Co: Fighting corrosion in the 90's, Prod. Finish. 1990(3),38-44.
230. Waldron, L. J., and Peterson, M. H.: Effect of iron, aluminum and cadmium additions on the performance
of zinc anodes in sea water, Corrosion 16, 81-85, 1960.

418

REFERENCES

231. Dirkse, T. P., and Timmer, R.: The corrosion of zinc in KOH solutions, 1. Electmchem. Soc. 116,
162-165,1969.
232. Vorkapic, L. Z., Drazic, D. M., and Despic, A R.: Corrosion of pure and amalgamated zinc In
concentrated alkali hydroxide solutions, J. Electrochem. Soc. 121, 1385-1392,1974.
233. Beard, R. B., Carim, H. M., Dubin, S. E., DeRosa, J, E. and Miller. A S.: Corrosion and histopathological
studies on anode materials for implantable power sources. J. Electrochem. Soc. 121. 1129-1133. 1974.
234. Gregory, D. P., Jones, P. c., and Redfearn. D. P.: The corrosion of zinc anodes in aqueous alkaline
electrolytes, J. Electrochem. Soc. 119, 1288-1292, 1972.
235. Williams, D. N., Koehl, B. G., Berry. W. E., and Bartlett, E. S.: The corrosion behavior of Zn-22AI
alloy sheet, J. Electrochem. Soc. 118, 1684-1688, 1971.
236. Townsend, H. E., and Meitzner, C. E: Corrosion resistance of zinc/4% aluminum and zincl7% aluminum
alloy coatings compared to zinc and zinc/54% aluminum alloy coatings, Mater. Perform. 22( 1),54, 1983.
237. Irving, R. R.: Galfan the zinc-aluminum coating, Iron Age 1986(Aug. I)' 47. 49.
238. Witmer, D. A: New markets for corrosion-resistant galvalume steel sheets. ASM Technical Paper No.
8201-035, 1982 American Society for Metals Metals Congress. SI. Louis. Missouri. October 23-28.
1982.
239. Lynch, R. E, and Goodwin, E E.: Galfan coated steel for automotive applications. SAE Technical Paper
No. 860658, Society of Automotive Engineers International Congress and Exposition, Detroit, February
24-28. 1986.
240. Baumgartner, M., and Kaesche, H.: Intercrystalline corrosion and stress corrosion cracking of AIZnMg
alloys, Corrosion 44, 231-239, 1988.
241. Centre de Recherches Metallurgiques (Belgium): Galfan: Its Corrosion Resistance in an Industrial
Atmosphere, International Lead Zinc Research Organization Project ZM 285, Progress Report No. 14A
July I-Dec. 31,1985.
242. Centre de Recherches Metallurgiques (Belgium): Atmospheric Exposure Testing of Galfan, International Lead Zinc Research Organization Project ZM 285, Progress Report No. 15B, Jan. I-June 30,
1986.
243. Centre de Recherches Metallurgiques (Belgium): Electrochemical Study of Coil Coated Galfan,
International Lead Zinc Research Organization Project ZM 285, Progress Report No. 14C. July I-Dec.
31,1985.
244. Hubrecht, 1., Skenazi, A E, Vereecken, J .. and Coutsouradis, D.: Corrosion resistance of Galfan and
coil coated Galfan in chloride containing atmospheres. ATB Metall. 25(3), 243-252, 1985.
245. Herrschaft, D. c., Radtke. S. E, Coutsouradis, D., and Pelerin, J.: Galfan-A New Zinc-Alloy Coated
Steel for Automotive Body Use. SAE Technical Paper No. 830517, Society of Automotive Engineers
International Congress and Exposition, Detroit, February 28-March 4, 1983.
246. Sharples, T. E.: Zinclzinc alloy plating, Prod. Finish. 1988(4),50-56.
247. Loto, C. A: Corrosion evaluation of cast iron, aluminium alloy, galvanized and stainless steels in cassava
juice, Corros. Prevo Control 1990(October). 13] -] 38.
248. Townsend, H. E., and Borzillo, A. R.: Twenty-year atmospheric corrosion tests of hot-dip coated sheet
steel, National Association of Corrosion Engineers Corrosion' 87 Conference. San Francisco, March
'9-13, 1987, Paper No. 421.
249. Vi art, G.: Galvanised steel wire with high corrosion resistance, Wire Ind. 1984( I), 76-78, 83.
250. Allegra, L., Berke, N. S., and Townsend, H. E.: Resistance of galvanized, aluminum-coated, and 55%
AI-Zn-coated steel sheet to atmospheric corrosion involving standing water, in Atmospheric Corrosion,
W. H. Ailor (ed.), pp. 595-606, John Wiley & Sons, New York, 1982.
251. Dreulle, N., Frappa, A, Limare, A, Piessen, P., and Sokolowski, R.: Effect of aluminum on the protection
of steel by zinc against atmospheric corrosion, Mater. Tech. 1985(Feb.-Mar.), 77-83.
252. Nuenninghoff, R., and Fischer, H.: Corrosion behaviour of zinc-aluminum alloys on steel wire, Stahl
Eisen 105(9), 517-522,1985.
253. Leroy, V., and Schmitz, B.: Surface chemistry of Zn and ZnlAI hot-dipped steels: Influence of some
processing parameters, Scand. J. Metall. 17, 17-23, 1988.
254. Matthews, P.: Impact of Galvalume on zinc consumption, Cominco Ltd., Metals Market Research
Report, October 1981.
255. Townsend, H. E., and Zoccola, J. c.: A comparison of the corrosion resistances of skyward- and
groundward-exposed surfaces of 55% aluminum-zinc alloy-coated and zinc-coated steel sheet, Pro-

REFERENCES

256.

257.
258.

259.
260.

261.

262.
263.

264.

265.

266.

267.

268.

269.

270.
271.

272.
273.

419

ceedings of the International Congress on Metallic Corrosion, Toronto, Ontario, Vol. 3, pp. 216-219,
1984.
Dreulle, P, Rosset, R., and Jardy, A.: Corrorsion properties of steel sheet coated with Zn-AI alloys,
Proceedings of the 12th International Galvanizing Conference, INTERGALVA '79, Paris, France, May
17-23, 1979.
Campbell, H. S.: Zinc-aluminium coatings for hot water tanks: Summary report. The British NonFerrous Metals Research Association, Report No, A, 1382, March 1962.
Pourbaix, M., Van Muylder, J., Pourbaix, A., and Kissel, J.: An electrochemical wet and dry method for
atmospheric corrosion testing, in Almospheric Corrosion, W. H. Ailor (ed.), pp. 167-177, John Wiley
& Sons, New York, 1982.
American Society for Testing and Materials: Standard test method of salt spray (fog) tcsting, in Annllal
Book ofASTM Standards, Vol. 02.05, No. BI17-90, pp. 19-25, 1990.
Warnecke, W., Bode, R., Kothe, R., and Meyer, L.: Modern hot dip coated sheet steel-processing,
coating characteristics and fabricating properties, in Zinc-Based Sleet Coaling Srslems: Mela/lurgy and
Per/(Jrlnance, G. Krauss and D. K. Matlock (cds.), pp. 3-17, The Minerals, Metals and Materials Society,
Warrendale, Pennsylvania, 1990.
Marder, A. R.: Microstructural characterization of zinc coatings, in Zinc-Based Sleel Coaling :Svslems:
Metallurgv and Performance, G. Krauss and D. K. Matlock (eds.), pp. 55-82, The Minerals. Metals and
Materials Society. Warrendale, Pennsylvania, 1990.
Mackowiak, J., and Short. N. R.: Metallurgy of galvanized coatings, Int. Mel. Rev. No. I, 1979, Review
No. 237.
Takahashi, A., Miyoshi, Y, and Hada, T.: A study on microstructure ofZn-Ni-Si0 2 composite platings
by transmission electron microscope, in Zinc-Based Sleel Coating Systems: Meta/lll~"r and Per/rlr/nance, G. Krauss and D. K. Matlock (eds.), pp. 83-94, The Minerals. Metals and Materials Society.
WaITendale. Pennsylvania, 1990.
Chen, Y L., and Snyder, D. D.: Microstructural investigation of electroplated Zn-Ni coatings on
cold-rolled steel, in Zinc-Based Steel Coating Systems: Melallurgyand Performance, G. Krauss and D.
K. Matlock (cds.). pp. 95-107, The Minerals, Metals and Materials Society, Warrendale, Pennsylvania,
1990.
Lin, Y, Pak, S., and Meshii, M.: Investigation of pure zinc electrogalvanized coatings by electron
microscopy, in Zinc-Based Steel Coating SYstems: Metallurgy and Per/rlrlnance, G. Krauss and D. K.
Matlock (cds.), pp. 109-119, The Minerals. Metals and Materials Society, Warrendale, Pennsylvania,
1990.
Shaffer, S. 1., Morris, 1. W., and Wenk, H. R.: Textural characterization and its applications on zinc
electrogalvanized steels, in Zinc-Based Sleel Coating Systems: Metallurgy and Peljormance, G. Krauss
and D. K. Matlock (eds.), pp. 129-140, The Minerals, Metals and Materials Society, Warrendale,
Pennsyl vania, 1990.
Apostolos. J. A.: Cathodic protection of reinforced concrete using /lame sprayed zinc, National
Association of Corrosi on Engineers Corrosion' 83 Conference, Anaheim, California, April I R-22, 19R3.
Paper No. 180.
Cook, A. R., and Radtke, S. E: Recent research on galvanized steel for reinforcement of concretc, in
Chloride Corrosion of'Steel in Concrete, STP 629, pp. 51-60, American Society for Testing and
Materials. Philadelphia, 1977.
Kilareski, W. P.: Epoxy coatings for corrosion protection of reinforcement steel, in Chloride Corrosion
of Sleet in Concl'Cle, STP 629, pp. 82-88, American Society for Testing and Materials, Philadelphia,
1977.
Vrable, 1. B.: Cathodic protection for reinforcing steel in concrete, in Chloride Corrosion of Steel in
Concrete, STP 629, pp. 124-149, American Society for Testing and Materials, Philadelphia, 1977.
Treadaway, K. W. J., Brown. B. L., and Cox, R. N.: Durability of galvanized steel in concrete. in
Corrosion olReill/orcillg Sleel in Concrele, STP 713, pp. 102-131, American Society for Testing and
Materials, Philadelphia, 1980.
Biczok, I.: COllclde Corrosion and Concrete Prolection, Chemical Publishing Co., New York, 1967.
Stark, D.: Measurement techniques and evaluation of galvanized reinforcing steel in concrete structures
in Bermuda, in Corrosion ofReil!/orcing Steel in Concrete, STP 713, pp. 132-141, American Society
for Testing and Materials, Philadelphia, 1980.

420

REFERENCES

274. Cornet, I., and Bresler, B.: Critique of testing procedures related to measuring the performance of
galvanized steel reinforcement in concrete, in Corrosion ofReinforcing Steel in Concrete, STP 713, pp.
160-195, American Society for Testing and Materials, Philadelphia, 1980.
275. Corderoy, D. 1. H., and Herzog, H.: Passivation of galvanized reinforcement by inhibitor anions. in
Corrosion of Reinforcing Steel in Concrete, STP 713, pp. 142-159, American Society for Testing and
Materials, Philadelphia, 1980.
276. Wyatt, B. S., and Irvine. D. 1.: A review of cathodic protection of reinforced concrete, Mater. Perform.
26(12),12-21, 1987.
277. Mudd, C. 1., Mussinelli, G. L., Tettamanti, M., and Pedeferri, P.: Cathodic protection steel in concrete,
Mater. Perform. 27(9), 18-24, 1988.
278. Feliu, S., Gonzalez, 1. A., Feliu, S., Jr., and Andrade,
Relationship between conductivity of concrete
and corrosion of reinforcing bars, Br. Corros. 1. 24, 196-198, 1989.
Developments in the continuous galvanizing of steel, 1. Metals 41(8),34-36, 1989.
279. Bush, G.
280. Gustafson, D. P.: Steel reinforcement: Purpose, types, and uses, ASTM Standardization News 1990(12),
38-42.
281. Bogner, B.: Review of internal corrosion of underground fuel storage tanks, Mater. Perform. 29(10),
30-32, 1990.
282. Irie, T.: Development of zinc-based coatings for automotive sheet steel in Japan, in Zinc-Based Steel
Coating Systems: Metallurgy and Performance, G. Krauss and D. K. Matlock (eds.), pp. 143-155, The
Minerals, Metals and Materials Society, Warrendale, Pennsylvania, 1990.
283. Lee, H. H.: The corrosion resistance of electrodeposited Zn-Ni alloy coatings, in Zinc-Based Steel
Coating Systems: Metallurgy and Performance, G. Krauss and D. K. Matlock (eds.), pp. 157-169, The
Minerals, Metals and Materials Society, Warrendale, Pennsylvania, 1990.
284. Abe, M., Shiohara, Y., and Okado, A.: Coating characteristics and performance of electrodeposited
Zn-Si02 composite coating, in Zinc-Based Steel Coating Systems: Metallurgy and Performance, G.
Krauss and D. K. Matlock (eds.), pp. 171-181, The Minerals, Metals and Materials Society, Warrendale,
Pennsylvania, 1990.
285. Goodwin, F. E.: Mechanisms of corrosion of zinc and zinc-5% aluminum steel sheet coatings, in
Zinc-Based Steel Coating Systems: Metallurgy and Performance, G. Krauss and D. K. Matlock (eds.),
pp. 183-193, The Minerals, Metals and Materials Society, Warrendale, Pennsylvania. 1990.
286. Jordan, D. L.: Influence of iron corrosion products on the underfilm corrosion of painted steel and
galvanized steel, in Zinc-Based Steel Coating Systems: Metallurgy and Performance, G. Krauss and D.
K. Matlock (eds.), pp. 195-205, The Minerals, Metals and Materials Society, Warrendale, Pennsyl vania,
1990.
287. Mabuchi, M., Takao, K., Ogishi, H., Kimura, H., and Ichida, T.: Effect of organic composite coating on
corrosion resistance ofZn-Ni alloy coated sheet steel, in Zinc-Based Steel Coating Systems: Metallurgy
and Performance, G. Krauss and D. K. Matlock (eds.), pp. 207-218, The Minerals, Metals and Materials
Society, Warrendale, Pennsylvania, 1990.
288. Claus, G., Houbaert, Y., Dilewijns, 1., Meers, U., Vanthournout, M., and Verhoeven, G.: Behaviour of
prephosphated Zn-coated steel sheet during and after final phosphating, in Zinc-Based Steel Coating
Systems: Metallurgy and Performance, G. Krauss and D. K. Matlock (eds.), pp. 219-230, The Minerals,
Metals and Materials Society, Warrendale, Pennsylvania, 1990.
289. Yau, Y., and Fountoulakis, S. G.: Characterization of the coating-to-steel interface strength in ZnNi
coated sheet steel, in Zinc-Based Steel Coating Systems: Metallurgy and Performance, G. Krauss and
D. K. Matlock (eds.), pp. 371-386, The Minerals. Metals and Materials Society, Warrendale, Pennsylvania, 1990.
290. Yeomans, S. R., and Novak, M. P.: Further studies of the comparative properties and behaviour of
galvanized and epoxy coated steel reinforcement, International Lead Zinc Research Organization
Project ZE-341, Progress Report No.4, Research Triangle Park, North Carolina, July 1990.
291. Fiaud, C.: Electrochemical behavior of atmospheric pollutants in thin liquid layers related to atmospheric corrosion, in Atmospheric Corrosion, W. H. Ailor (ed.), pp. 161-166, John Wiley & Sons, New
York,1982.
292. Warwick, M. E., and Hampshire, W. B.: Atmospheric corrosion of tin and tin alloys, in Atmospheric
Corrosion, W. H. Ailor (ed.), pp. 509-527, John Wiley & Sons, New York, 1982.

c.:

w.:

REFERENCES

421

293. Kucera. Y. and Mattsson, E.: Atmospheric corrosion of bimetallic structures. in Atmospheric Corrosion,
W. H. Ailor (ed.), pp. 561-574. John Wiley & Sons, New York. 1982.
294. Dunbar. S. R., and Showak. W.: Atmospheric corrosion of zinc and its alloys, in Atlllospheric Corrosion,
W. H. Ailor (ed.), pp. 529-552, John Wiley & Sons, New York, 1982.
295. Legault, R. A.: Atmospheric corrosion of galvanized steel, in Atmospheric Corrosion, W. H. Ailor (ed.),
pp. 607-613, John Wiley & Sons, New York, 1982.
296. Leidheiser, H., and Suzuki, I.: Cobalt and nickel cations as corrosion inhibitors for galvanized steel. in
Atmospheric Corrosion, W. H. Ailor (ed.), pp. 615-629, John Wiley & Sons, New York, 1982.
297. Biestek, T.: Atmospheric corrosion testing of electrodeposited zinc and cadmium coatings, in Atmospheric Corrosion, W. H. Ailor (cd.), pp. 63 I -643, John Wiley & Sons, New York, 1982.
298. Altorfer. K. J.: Service evaluation of zinc thermal spraying systems, in Atmospheric Corrosion. W. H.
Ailor (cd.), pp. 645-650, John Wiley & Sons, New York, 1982.
299. Talati, J. D., and Patel. B. M.: Atmospheric corrosion of metals under moving conditions, in Atmospheric
Corrosion, W. H. Ailor (ed.), pp. 695-703, John Wiley & Sons, New York, 1982.
300. Moresby. J. E, Reeves, E M .. and Spedding, D. J.: Atmospheric corrosion testing in Australasia. in
Atmospheric Corrosioll, W. H. Ailor (ed.). pp. 745-754, John Wiley & Sons, New York, 1982.
30 I. Dutra, A. C, and Vianna, R. de 0.: Atmospheric corrosion testing in Brazil, in Atlllospheric Corrosion,
W. H. Ailor (ed.). pp. 755-774. John Wiley & Sons, New York, 1982.
302. Biestck, T: Testing clectrodeposited coatings in tropical China, in Atmospheric Corrosion, W. H. Ailor
(ed.), pp. 775-785, John Wiley & Sons. New York, 1982.
303. Hakkarainen, T, and Ylasaari. S.: Atmospheric corrosion testing in Finland, in Atlllospheric Corrosion,
W. H. Ailor (ed.), pp. 787-795, John Wiley & Sons, New York, 1982.
304. Atteraas. L., and Haagenrud, S.: Atmospheric corrosion in Norway. in Atmwpheric Corrosion. W. H.
Ailor (cd.), pp. 873-891. John Wiley & Sons, New York, 1982.
305. Callaghan, B. G.: Atmospheric corrosion testing in southern Africa, At11lospheric Corrosion, W. H. Ailor
(ed.). pp. 893-912. John Wiley & Sons, New York, 1982.
306. Feliu, S .. and Morcillo. M.: Atmospheric corrosion testing in Spain, in Atmospheric Corrosion. W. H.
Ailor (cd.), pp. 9 I 3-92 I, John Wiley & Sons. New York, 1982.
307. Mikhailovski, Y. N., and Strekalov. P Y: Atmospheric corrosion tests in the USSR, in Atmospheric
Corrosion, W. H. Ailor (cd.). pp. 923-942. John Wiley & Sons. New York. 1982.
308. Knotkova, D., Barton. K., and Cerny, M.: Atmospheric corrosion testing in Czechoslovakia, in Atmospheric Corrosion. W. H. Ailor (cd.), pp. 991- 1014, John Wiley & Sons. New York. 1982.
309. Tomlinson, W. 1., and Marsh, P. 1.: Corrosion of iron, steel, zinc, nickel andelectroless nickel in concrete.
J. Matn: Sci. Lett. 7. 808-810. 1988.
3 I O. Apostolos. J. A., Parks, D. M .. and CareJlo. R. A.: Cathodic protection of reinforced concrete using
metallized zinc. Mater. Perform. 26(2), 22-28,1987.
3 I I. Catino. J. W.: The polarization and corrosion characterization of Zn and modified Zn in concentrated
alkali metal hydroxide electrolyte: The effect of a Zn oxide/hydroxide surface, Ph. D. Thesis, Lehigh
University, 1988.
3 I 2. Suzuki. I.: Corrosion-Resistant Coatings Technology, Corrosion Technology Series, PA. Schweitzer
(ed.), Marcel Dekker. New York, 1989.
3 I 3. Serra, E. T, and Mannheimer, W. A.: On the estimation of the corrosion rates of metals in soils by
electrochemical measurements, in Underground Corrosion, STP 741, pp. 1 I 1-122, American Society
for Testing and Materials. Philadelphia, 198 I.
314. Miller. F. P., Foss. J. E., and Wolf, D. C: Soil surveys: Their synthesis, confidence limits, and utilization
for corrosion assessment of soil, in Underground Corrosion, STP 74 I, pp. 3-23, American Society for
Testing and Materials, Philadelphia, 198 I.
3 I 5. Guttman, H.: Effects of atmospheric factors on the corrosion of rolled zinc, in Metal Corrosion in the
Atmosphere, STP 435, pp. 223-239, American Society for Testing and Materials, Philadelphia, 1968.
316. Dunbar. S. R.: Effect of one per cent copper addition on the atmospheric corrosion of rolled zinc, in
Metal Corrosion in the Atmosphere, STP435. pp. 308-325, American Society for Testing and Materials,
Philadelphia, 1968.
317. Baboian, R. (cd.): Automotive Corrosion by Deicing Salts. National Association of Corrosion Engineers,
Houston, 1981.
3 I 8. Geduld. H.: Zinc Plming, ASM International. Metals Park, Ohio, 1988.

422

REFERENCES

319. International Lead Zinc Research Organization: A Comparison between Hot Dip Galvanized and Fusion
Bonded Epoxy Coatings for Steel Reinforcement in Concrete. ILZRO Project No. ZE-341. Annual
Report, Research Triangle Park, North Carolina, 1988.
320. Tomashov, N. D., and Chernova, G. P.: Passivity and Protection of Metals against Corrosion. Plenum
Press, New York, 1967.
321. Waber,1. T.: Analysis of size effects in corrosion processes, in Localized Corrosion, pp. 221-237,
National Association of Corrosion Engineers, Houston, 1974.
322. Chen, Z. W, Sharp, R M., and Gregory, 1. T.: Fe-AI-Zn ternary phase diagram at 450 C, Mater. Sci.
Techno!. 6, 1173-1176, 1990.
323. Albalat, R, Gomez, E., Muller, J., Pregonas, J . Sarret, M., and Valles, E.: Zinc-nickel coatings:
Relationship between additives and deposit properties, 1. Appl. Electrochem. 21, 44-49, 1991.
324. Djoufac Woumfo, E., and Vittori, 0.: Electrochemical behaviour of a zinc electrode in 8 M KOH under
pulsed potential loading, 1. Appl. Electrochem. 21,77-83, 1991.
325. Beccaria, A. M.: Zinc layer characterization on galvanized steel by chemical methods, Corrosion 46,
906-912, 1990.
Velinov. v.. Riesenkampf. A., and Pawlik, B.: Factors influencing the
326. Tomov, I., Cvetkova,
preferential orientations in zinc coatings electrodeposited from chloride baths. 1. Appl. Electrochem. 19,
377-382, 1989.
327. Faita, G., Mazza, E, and Bianchi, G.: Role of water and ionic solvation in localized corrosion
phenomena, in Localized Corrosion, pp. 34-44, National Association of Corrosion Engineers, Houston,
1974.
328. Newman, J.: Mass transport and potential distribution in the geometries of localized corrosion, in
Localized Corrosion, pp. 45-61, National Association of Corrosion Engineers, Houston, 1974.
329. Fujii, C. T.: A review: Macro-cells for the study of solution chemistry and electrochemistry in localized
corrosion, in Localized Corrosion, pp. 144-150, National Association of Corrosion Engineers, Houston,
1974.
330. Selvam, M., and Guruviah, S.: Corrosion of electrodeposited zinc-manganese alloys, Bull. Electrochem.
6(5),485-486. 1990.
331. Graedel, T. E.: Corrosion mechanisms for zinc exposed to the atmosphere. 1. Electrochem. Soc. 136,
193-203,1989.
332. Townsend, H. E.: On the effects of magnesium on the atmospheric corrosion resistance of galvanized
sheet steel, Corrosion 44. 229-230, 1988.
333. Evans, U. R: Factors deciding between corrosion and protective film formation. The 1st International
Congress on Metallic Corrosion. London, April 1961, pp. 3-9.
334. Lorking, K E. and Mayne, J. E. 0.: Some observations on the corrosion of zinc. The 1st International
Congress on Metallic Corrosion, London, April 1961, pp. 144-146.
335. Cismaru, D., and Cismaru, G. D.: Zinc oxidation at high temperatures. Effect of electric field on zinc
oxidation, The 1st International Congress on Metallic Corrosion, London, April 1961. pp. 237-239.
336. Rosenfeld, I. L.: Atmospheric corrosion of metals. Some questions of theory, The 1st International
Congress on Metallic Corrosion, London, April 1961, pp. 243-253.
337. Neville, R J., and de Souza, K M.: Electrogalvanized or hot dip galvanized-results of five years
undervehicle corrosion testing, SAE Technical Paper No. 862010, Society of Automotive Engineers
Automotive Corrosion and Prevention Conference, Dearborn, Michigan, December 8- 10, 1986.
338. Stephens, M. L., Davidson, D. D., Soreide, L. E., and Shaffer, R 1.: Developing the chipping corrosion
test, SAE Technical Paper No. 862013, Society of Automotive Engineers Automotive Corrosion and
Prevention Conference, Dearborn, Michigan, December 8- 10, 1986.
339. Satoh, H., Shimogori, K, Nishimoto, H., Miki, K, Ikeda, K, Iwai, M., Sakai, H., and Nomura, S.:
Analysis of perforation corrosion of cold-rolled and galvanized steel sheets by extreme-value statistics,
SAE Technical Paper No. Technical Paper No. 862018, Society of Automotive Engineers Automotive
Corrosion and Prevention Conference, Dearborn, Michigan, 8-10, 1986.
340. Memmi, M., Ferrari, v., Arrigoni, G., Sarracino, M., Moelzer, H., and Ulivieri, L.: A qualitative and
quantitative evaluation of Zn + Cr-CrOx multilayer coating compared to other coated steel sheets, SAE
Technical Paper No. 862028, Society of Automotive Engineers Automotive Corrosion and Prevention
Conference. Dearborn, Michigan, December 8-10, 1986.

c..

REFERENCES

423

341. Mohri, T, Tsugawa, S., Kobayashi, S .. Ichida, T, and Kurosawa, M.: Newly developed organic
composite-coated steel sheet with bake-hardenability, SAE Technical Paper No. 862030. Society of
Automotive Engineers Automotive Conosion and Prevention Conference, Dearborn. Michigan December 8-10, 1986.
342. Tator, K. B.: How coatings protect and why they fail, Mater. Perfimn. 17(11),41 -44, 1978.
343. Leidheiser, H.: Conosion of painted metals-a review. Corrosion 38, 374-383. 1982.
344. Barrett, L. D.: Preparation of metals by phosphating, Mater. Prot. 3(7),20-23,1964.
345. Munger, C. G.: Inorganic zinc coatings past, present, and future, Mater. PetjiJrm. 14(5),25-29, 1975.
346. Robbins, D. 1., Smith, D. M., and Roberts, T R.: Corrosion testing and behavior of zinc and zinc alloy
coated autobody sheet steels. ASM Corrosion-Resistant Automotive Sheet Steels Conference, Chicago,
September 24-30, 1988, pp. 1-14.
347. Lin, K. L., Yang, C. E, and Lee, J. T: Correlation of microstructure with eonosion and electrochemical
behavior of the batch-type hot-dip AI-Zn coatings. Part I. Zn and 5% AI-Zn coatings, Corrosion 47,
9-16,1991.
348. Lin, K. L., Yang, C. E, and Lee, J. T.: OllTelation of microstructure with conosion and electrochemical
behaviors of the batch-type hot-dip AI-Zn coatings. Part II. 55% AI-Zn coating, Corrosion 47, 17-23,
1991.
349. Watanabe, T, Kubota, T, Yamashita, M .. Urakawa, T, and Sagiyama, M.: Con'osion-resistant precoated
steel sheets for automotive body panels, ASM Conosion-Resistant Automotive Sheet Steels Conference,
Chicago, September 24-30, 1988, pp. 15-20.
350. Neville, R. L and de Souza, K. M.: Corrosion performance of zinc and zinc alloy coated steels exposed
to deicing salt spray under vehicles, ASM Conosion-Resistant Automotive Sheet Steels Conference,
Chicago, September 24-30, 1988, pp. 2 I -30.
351. Yasuda, A., Umino, S., Kyono, K., and Yamato, K.: Cosmetic conosion ofZn alloy coated sheet steels,
ASM Conosion-Resistant Automotive Sheet Steels Conference, Chicago, September 24-30,1988, pp.
31-37.
352. Yasuda, A., Koumura, H., Yamato, K., Onizawa, K., and Ota, H.: Development of high performance
galvannealed steel sheet for outer panels of automotives, ASM Conosion-Resistam Automotive Sheet
Steels Conference, Chicago. September 24-30, 1988, pp. 45-53.
353. Lynch, R. E: Hot-dip galvanizing alloys, 1. Met. 39(8), 39-41. 1987.
354. Scholl, p, and Prentice, G.: Photocurrcnt response of the anodic film on zinc electrodes in alkaline
elcctrolyte, Electrochemical Society Extended Abstracts, Vol. 91-1, Abstract No. 75, p. I J 3. 1991.
355. Davies, D. E., and Lotlikar, M. M.: Passivation and pilling characteristics of zinc, Br. Corros. 1. L
149-155, 1966.
356. Mansfeld, E. and Kenkel, J. V: Galvanic conosion of Al alloys. Part III. The effect of area ratio, Corm.l.
Sci. 15, 239-250, 1975.
357. Romanoff, M.: Underground Conosion, U.S. National Bureau of Standards Circular 579. 1957.
358. Rachlot, B.: Sacrificial anode aJloys-Their specifications and uses, Proceedings of the 2nd International Congress on Metallic Conosion, New York, March 11-15, 1963, pp. 285-289.
359. Bianchi, G., Mazza, E, and Trasatti, S.: Mechanism offormation of black oxide films on zinc, aluminum
and titanium in alkaline aqueous solutions, Proceedings of the 2nd International Congress on Metallic
Conosion, New York, March 11-15, 1963, pp. 905-915
360. Chou, K. c., and Jagannathan, V: Corrosion and powdering resistance of galvanneal for automotive
applications, National Association of Corrosion Engineers Corrosion '91 Conference, Cincinnati, Ohio,
March II-IS, 1991,PaperNo.375.
361. Mori, K., Miyawaki, T' and Matsushima, Y.: Effect of zinc composite coating on conosion resistance
at high temperature, National Association of Con os ion Engineers Conosion '91 Conference, Cincinnati,
Ohio, March I 1-15, 1991, Paper No. 40 I.
362. Wang, Y. M., and Radovic, D.: Porosity measuremcnts of phosphate coatings by electrochemical
impedance spectroscopy (EIS) method, National Association of Corrosion Engineers Corrosion '91
Conference, Cincinnati, Ohio, March I J -15, 1991, Paper No. 419.
363. Granata, R. D.: Mechanisms of coatings disbondment, National Association of Conosion Engineers
Corrosion '91 Conference, Cincinnati, Ohio, March 11-15, 1991, Paper No. 382.

424

REFERENCES

364. Massinon, D., and Thierry, D.: Rate controlling factors in the cosmetic corrosion of coated steels,
National Association of Corrosion Engineers Corrosion '91 Conference, Cincinnati, Ohio, March
11-15,1991, Paper No. 574.
365. Chang, J. c., and Wei, H. H.: Electrochemical and Mossbauer studies of the corrosion behavior of
electrodeposited fe-Zn alloys on steel, Corros. Sci. 30, 831-837,1990.
366. Wippermann, K, Schultze, 1. W, Kessel, R., and Penninger, 1.: The inhibition of zinc corrosion by
bisaminotriazole and other triazole derivatives, Corros. Sci. 32, 205-230, 1991.
367. Kirihara, S., and Ohshima, K (Kobe Steel, Ltd., Japan): Method for anticorrosive treatment of
galvanized steel, U.S. Patent 4385940, May 31, 1983.
368. Sriveeraraghaven, S., Krishnan, R. M., and Natarajan, S. R.: Corrosion behavior of zinc and cadmium
deposits obtained from different baths, Met. Finish. 89(8),51-53, 1991.
369. Cremaschi, P., Tantardini, G. E, Muilu, 1., and Pakkanen, T. A.: Adsorption of hydrogen on zinc-calculation of potential energy surface and dynamics, Vacuum 41( 1-3), 260-264, 1990.
370. Girin, O. B., and Panasenko, S. A.: Effect of the texture of electrolytic zinc coatings on their corrosion
resistance, Zashch. Met. 25(3),480-482, 1989 [in English].
371. Novoa, X. R., Izquierdo, M., Merino, P., and Espada, L.: Electrochemical impedance spectroscopy and
zero resistance ammeters (ZRA) as tools for studying the behaviour of zinc-rich inorganic coatings,
Mater. Sci. Forum 44-45, 223-234, 1989.
372. Fiaud, c., Vastra, I., and Tribollet, B.: Application of the electrohydrodynamical impedance method to
the study of corrosion inhibitors, Mater. Sci. Forum 44-45, 205-212, 1989.
373. Juttner, K., and Lorenz, W. J.: Electrochemical impedance spectropscopy (EIS) of corrosion processes
on inhomogeneous surfaces, Mater. Sci. Forum 44-45, 191-204, 1989.
374. Guillaume, R., Keddam, M., and Takenouti, H.: In-situ electrochemical investigations of a scratched
galvanized steel during a climatic test, Mater. Sci. Forum 44-45, 109-122, 1989.
375. Lay, D. E., and Eckles, W E.: The fundamentals of zinc/cobalt, Plating Suif. Finish. 77(10), 10-14,
1990.
376. Sjoukes, E: Chemical reactions in fluxes for hot dip galvanizing, Anti-Corrosion 1990(April), 12-14.
377. Gu, M., and Marder, A. R.: A new method to fabricate single-phase Zn-Fe coating, Plating Suif. Finish.
78(1),77-79,1991.
378. Arup, H.: Galvanized steel in concrete, Mater. Perform. 18(4),41-44,1979.
379. Hofsoy, A., and Gukild, I.: Bond studies on hot dip galvanized reinforcement in concrete, Amer. Conc.
Inst. 1. 66(3), 174-184, 1969.
380. Cornet, I., and Bresler, 8.: Corrosion of steel in prestressed concrete, Mater: Prot. 4(11), 35-37, 1965.
381. O'Donnell, A. 1.: Comparison of the sacrificial protection afforded by continuous galvanised coatings,
Corras. Coatings S. Afr. 16(10),2,4-6, 1990.
382. Imagawa, H., Matuno, K, and Konishi, T.: Zinc spraying and corrosion cracking prevention for liquid
ammonia tanks, Corros. Eng. 38, 353-361, 1989.
383. Mourad, M. Y., Kandell, A. Y., and Elgaber, A. S.: Effect of surface preparation on the corrosion
behaviour of zinc in sodium sulphate solution, Indian 1. Technol. 28, 612-614, 1990.
384. Akhter, R., Watkins, K G., and Steen, W. M.: Electrochemical characterisation of laser welded zinc
coated steel, Mater: Lett. 9, 550-556, 1990.
385. Kothari, N. C.: Corrosion ofzinc coated steel in simulated industrial environment, Corrosion-Air, Sea,
Soil Conference, Auckland, New Zealand, Nov. 19-23, 1990, Paper No. 25.
386. Singh, W. R., and Singh, R. N.: Corrosion of Zn in propylene carbonate: Role of water, H+ and O2 , 1.
Electrochem. Soc. India 39, 104-106, 1990.
387. Helwig, L. E., and Swan, 1. D.: Use of spot-test reagents for detecting chromates on hot-dip galvanized
surfaces, Met. Finish. 80(9), 37-40, 1982.
388. Yeomans, S. R.: Comparative studies of galvanized and epoxy coated steel reinforcement in concrete,
The 2nd CANMET/ACI International Conference on Durability of Concrete, Montreal, Quebec, August
1991.
389. Ravindran, V., and Muralidharan, V. S.: Corrosion of zinc in NH4CI solution, Bull. Electrochem. 6(2),
237-238, 1990.
390. Wilde, B. E., and Budinski, M. K (Ohio State University): Galvanic protection of steel with zinc alloys,
U.S.Patent4917966,AprilI7,1990.

REFERENCES

425

391. Verberne, W. M. 1. C. (OMI International Corporation, Michigan): Passivation. U.S. Patent 4776898,
October II, 1988.
392. Kressler, R 1., Powers, R G.:and Lasa, I. R: Cathodic protection using scrap and recycled materials,
Mater. Pelform. 30(6),29-31, 1991.
393. Cheng, T P, Lee, 1. T, Lin. K. L., and Tsai, W. T: Electrochemical behavior of galvanized AI-Zn
coatings in saturated Ca(OHh solution, Corrosion 47. 436-442, 1991.
394. Sjoukes, F: Brown staining of hot dip galvanised coatings in atmospheric cOlTosion, Br. Corros. l. 26,
103-107,1991.
395. Allen,
Corrosion of galvanised steel in synthetic mine waters, Br. Corros. l. 26, 93-101, 1991.
396. Smith, C. 1. E.: Research into the control of corrosion on aircrafts, Aircr. EnK. 1990(August), 15.
397. Bird, C. E.: Bond of galvanized steel reinforcement in concrete, Nature (London) 194, 798. 1962.
398. Sebisty, 1. 1.: Continuous-strip galvanized coatings at elevated temperatures. Electrochem. Technol. 6.
330-336,1968.
399. Cook, T.: Myths and truths in hot dip galvanizing, Met. Finish. 89(6),107-108,1991.
400. Kenworthy, L.. and Smith, M. D.: Corrosion of galvanized coatings and zinc by waters containing frcc
carbon dioxide. l. Illst. Mel. 1944,463-489.
401. Evans, U. R.: Pitting and cracking, Chem. Ind. 1956, 1291-1297.
402. Evans, U. R .. and Davies, D. E.: The pitting of zinc by distilled water and dilute solutions. Corrosio/l
8(5),165-170,1952.
403. Thornhill, R S., and Evans, U. R.: The electrochemistry of the corrosion of partly immersed zinc, 1.
Chem. Soc. 1938,2109-2114.
404. Feitknecht, W.: Studies on the influence of chemical factors on the corrosion of metals, Chem.lnd. 1959,
1102-1109.
405. Davydov, A. D., Zhukova, T B., and Engel'gardt, G. R.: Concentration changes in the solution layer
next to the anode, and the maximum currents of electrochemical zinc dissolution in alkalies, Eleklrokhimiva 26, 1207-1209, 1990 [in Englishl.
406. Munier, G. B .. Psota, L. A., Reagor, B. T.. Russiello, B., and Sinclair, 1. D.: Contamination of electronic
equipment after an extended urban exposure, 1. Electrochem. Soc. 127,265-2 fL, 1980.
407. Pushpavanam. M., Natarajan. S. R, Balakrishnan, K., and Sharma, L. R.: Corrosion behaviour of
electrodeposited zinc-nickel alloys, l. Appl. Electrochem. 21,642-645,1991.
408. Townsend, H. E., Allegra, L., Dutton. R. J., and Kriner, S. A.: Hot-dip coated sheet steels-a review.
Maler. PerjcJI"In. 25(8), 36-46, 1986.
409. Hoxeng, R B.: Electrochemical behavior of zinc and steel in aqueous media. Part II, Cormsion 6(9).
308-312,1950.
410. Hoxeng, R. B., and Prutton, C. F: Electrochemical behavior of zinc and steel in aqueous media,
Corrosion 5(10), 330-338,1949.
411. Kobayashi, N., Yamasaki, Y., and Inakazu, N.: Effects of cold rolling on corrosion rate of zinc in acid
solution, 1. lpn.lnst. Mel. 53,1255-1262,1989.
412. Cox, G. L.: Effect of temperature on the cotTosion of zinc, Ind. Eng. Chem. 23,902-904. 1931.
4\3. Trabanelli. G.: Inhibitors-An old remedy for a new challenge, Corrosion 47, 410-419. 1991.
414. Sizelove. R R: Developments in alkaline zinc-nickel alloy plating, PlatinK SUli Finish. 78(3),26-30.
1991.
41S. Munn. R S., and Devereux. O. F: Numerical modeling and solution of galvanic corrosion systems. Part
II. Finite-element formulation and descriptive examples, Corrosion 47, 61 ~-634. 1991.
416. Hinton, B. R. W.: Corrosion prevention and chromates: the end of an era?, Mel. Finish. 89( 10). IS-20.
1991.
417. Yoshioka, K .. Watanabe, K., and Kato. K. (Kawasaki Steel Corp., Japan): Zn-coated stainless steel
welded pipe. U.S. Patent 4885215. December 5, 1989.
418. Rocha, L. A .. and Barbosa, M. A.: Microstructure, growth kinetics, and corrosion resistance of hot-dip
galvanize\! Zn-S% AI coatings, Corrosion 47, 536-541, 1991.
419. Dean, S. W.: Corrosion testing of metals under natural atmospheric conditions, in Corrosion Testing and
Evaluation: Silver Anniversary Volume. STP 1000, pp. 163-176, American Society for Testing and
Materials, Philadelphia, 1990.
420. Hinton, B. R. W.: Corrosion prevention and chromates. The end of an era?, Mel. Finish. 89(9),55-61.
1991.

c.:

426

REFERENCES

421. Richards, C. E.: The atmospheric corrosion of galvanised iron wire and its bearing on specification
testing, J.lron Steellnst. 137, l27p-176p, 1938.
422. Scholl, P., Shan, X., Bonham, D., and Prentice, G. A.: Photoelectrochemical characterization of the
anodic film on zinc in KOH solution, J. Electrochem. Soc. 138,895-899, 1991.
423. Burleigh, T. D.: Anodic photocurrents and corrosion currents on passive and active-passive metals,
Corrosion 45, 464-472, 1989.
424. Heitz, E.: Corrosion of metals in organic solvents, in Advances in Corrosion Science and Technology,
Vol. 4, M. G. Fontana and R W. Stachle (eds.), pp. 149-243, Plenum, New York, 1977.
425. Adaniya, T.: The development of corrosion-resistant composite zinc-coated steel sheet, Sheet Met.
Ind. Int. 1978(December) 73-79.
426. Foster, G. R, Lee, S. G., and Myers, 1. R: Stress corrosion of eta-phase, hot-dip-zinc alloy in potable-hot
water, Corrosion 30, 239-241,1974.
427. Brossa, A., Maja, M., Penazzi, N., Farnia, G., Sandona, G., and Marcuzzi, E: Inhibition of zinc corrosion
in chloride media: Green piles, Proceedings of the 7th European Symposium on Corrosion Inhibitors,
Ann. Univ. Ferrara. N. S. Sez. V, 1990(Suppl. No.9), 711-724.
428. Gronostajski. J., Ali, W. 1., and Ghattas, M. S.: The effect of deformation on the corrosion resistance of
coated steel sheets, J. Mater. Process. Technol. 23, 21-28, 1990.
429. Brown, W. N., and Mackowiak, 1.: A study of the kinetics of interaction between Fe(s) and Zn(l) in the
temperature range 570-740 C. Corros. Sci. 5, 779-785. 1965.
430. Mackowiak, 1., and Short. N. R: The effect of pressure on the reaction between Fe(s)-Zn(l) at 501 and
554 c, Corros. Sci. 16,519-528, 1976.
431. Short, N. R., and Mackowiak, 1.: The effect of pressure on the reactions between Fe(s)-Zn: 1.5% AI(I)
at 501 C, Corros. Sci. 17.397-404,1977.
432. Pereira. D., Scantlebury, 1. D., Ferreira, M. G. S., and Almeida, M. E.: The application of electrochemical
measurements to the study and behaviour of zinc-rich coatings, Corros. Sci. 30. 1135-1147, 1990.
433. Hadden, S. E.: Effect of annealing on the resistance of galvanized steel to atmospheric corrosion, J.lron
Steellnst. 1952(June), 121-127.
434. Campbell, H. S., Stanners, 1. E, and Watkins, K. 0.: Effect of heat treatment on the protective properties
of zinc coatings on steel, J. Iron Steellnst. 1965(March), 248-251.
435. Rajagopalan, K. S., and Ramaseshan. G.: Relative corrodibility of zinc and steel in unpolluted
atmospheres, J. Appl. Chern. 10, 493-496, 1960.
436. Anderson, E. A.: Zinc in marine environments, Corrosion 15, 409t-4121. 1959.
437. Gilbert, P. T., and Hadden, S. E.: White rust formation on zinc, J. Inst. Met. 78(Part 1),47-70, 1950.
438. Mansfeld, E: The effect of temperature on the corrosion rate of Zn of different purity in 6N KOH. Corros.
Sci. 11,557-559, 1971.
439. Blickwede, D.l.: 55% AI-Zn-alloy-coated sheet steel, Iron Steel 1980(7), 821-834.
440. Mathieu, S., and Fenaille, B.: Characterization of zinc-iron and tin-iron coatings by coulometric method
associated with x-ray diffraction and glow discharge optical spectrometry, Methodes d'Investigation des
Metaux Conference, France, 1980.
441. Katz, W.: The structure of zinc coatings on the basis of electrochemical dissolutions, Arch. Eisenhiittenwes. 25(7/8), 307-314, 1954 [in GermanJ.
442. Chang, S.: Characterization and development of Zn-Fe alloyed coating layer, in Zinc-Based Steel
Coating Systems: Metallurgy and Performance. G. Krauss and D. K. Matlock (eds.), pp. 319-330, The
Minerals, Metals and Materials Society, Warrendale, Pennsylvania, 1990.
443. Landriault, 1. P., and Harrison, E w.: A comparison of thermally alloyed Zn-Fe coatings produced by
mechanical wiping to those produced by post-coating annealing, Proceedings of the 25th Canadian
Institute of Mining, Metallurgy, and Petroleum Conference, Toronto, Ontario, August 1986, pp. 71-78.
444. Imagawa, H., Matuno, K., and Konishi, T.: Prevention of corrosion cracking in liquid ammonia tanks
by zinc spraying, Boshoku Gijutsu 38(6), 321-326, 1989.
445. Gmytryk, M., and Sedzimir, 1.: Corrosion of Zn in deaerated sulphate solutions at different pH values,
Corros. Sci. 7,683-695, 1967.
446. Abayarathna, D., Hale, E. B., O'Keefe, T. J., Wang, Y. M., and Radovic, D.: Effects of sample orientation
on the corrosion of zinc in ammonium sulfate and sodium hydroxide solutions, Corros. Sci. 32, 755-768,
1991.

REFERENCES

427

447. Weast, R. C: Zinc anodes for corrosion protection of domestic hot water storage tanks. Proceedings of
the 7th International Conference on Hot Dip Galvanizing, Paris, June 1964, pp. 487-507.
448. TOITI~s- Villasenor, G .. Ugalde, A.. Hernandez. L.. and Singer. I. L.: Water vapour corrosion of lamellar.
superplastic and cast dendritic Zn-2IAI alloy. Corms. Sci. 24. 159-166. 1984.
449. Trabanelli. G .. Zucchi. E. Brunoro. G .. and Gilli. G.: Characterization of the corrosion or anodic
oxidation products on zinc. Electrodeposition SUlj: Treat. 3. 129-138. 1975.
450. Stuart. N . and Evans. U. R: The effect of zinc on the corrosion-fatigue life of steel. 1. [ron Steel [n.H.
147.131-144.1943.
451. Stamatakis. P. Palmer. 8. R. Salzman. G. C. Bohren. C F., and Allen. T 8.: Optimum particle size of
titanium dioxide and zinc oxide for attenuation of ultraviolet radiation. 1. Coatings Teclmol. 62. 95-98.
1990.
452. Solorzano. I. G .. Purdy. G. R .. and Weatherly, G. C: Studies of the initiation. growth and dissolution of
the discontinuous precipitation product in aluminum-zinc alloys. Acta Metall. 32. 1709-1717. 1984.
453. Singh. R. N .. Bahadur. L .. and Singh. P.: The corrosion and electrochemical behaviour of zmc in binary
mixtures of dimethylformamide and water. Corms. Sci. 27.561-566.1987.
454. Simm. D. W.: The corrosion of zinc in mortar and its relevance to wall tic COITosion. Corms. Prevo
ContmI1984(December). 5-10.
455. Pettinger. B .. Schoppel. H.-R .. and Gerischer. H.: Tunnelling processes at highly doped ZnO electrodes
in contact with aqueous electrolytes. Part I. Electron exchange with the conduction band. Ber. BUIlsellges. 78. 450-455. 1974.
456. Van Craen. M .. Haemers. G . and Adams. E: Sims analysis of the atmospheric corrosion of a 55% AI-Zn
coated steel wire. [nt. 1. Mass Spectrol1l. [on Phvs. 46,531-534, 1983.
457. von Fraunhofer. 1. A .. and Lubinski. A. T: Polarity reversal in the zinc-mild steel couple. Corms. Sci.
14. 225-232. 1974.
458. Gilbert. P T.: The nature of zinc corrosion products. 1. Electrochem. Soc. 99. 16-21. 1952.
459. Girin. O. B .. and Panasenko. S. A.: Effect of the texture of electrolytic zinc coatings on their corrosion
resistance. Zashch. Met. 25(3). 480-482. 1989 Iin English].
460. Gomes, W. P.. Freund. T. and Morrison. S. R: Chemical reactions involving holes at the zinc oxide
single crystal anode. 1. Electrochem. Soc. 115. 818-823. 1968.
461. Shih. Y. R.: Investigation of the zinc-iron potential reversal and its application to cOITosion in domestic
hot water systems. Ph.D. Thesis, San Jose State University. 1989.
462. Kotnik. L. J.: An investigation of the properties of zinc COlTosion products. Ph.D. Thesis. Casc Institute
of Technology. 1960.
463. Biallozor, S .. and Bandura, E. T: On electrochemical behaviour of zinc in organic solvents. Elcctrochim.
Acta 32. 891-894. 1987.
464. Perkins, 1.. and Bornholdt. R. A.: The corrosion product morphology found on sacrificial zinc anodes.
Corros. Sci. 17.377-384. 1977.
465. McDonald. R. D.: Steel embrittlement problems associated with hot dip galvanizing-causes. mechanisms. controls. and selected references. Mate!: Perfonn. 14( 1). 31-37. 1975.
466. Keogh. T J.: The influence of pickling on the embrittlement of mild steel. Met. AlislI: 1972(August).
239-244.
467. Johnson. W. H.: On some remarkable changes produced in iron and steel by the action of hydrogen and
acids. Pmc. R. Soc. London 23,1875.
468. Gonzalez. J. A .. and Andrade. C: Effect of carbonation. chlorides and relative ambient humidity on the
corrosion of galvanized rebars embedded in concrete. Br. Corros. 1. 17,21-28. 1982.
469. Carson. J. A. H.: Zinc as a self-regulating galvanic anode for ship hulls. Corrosion 16(10).99-104.
1960.
470. Carson. 1. A. H .. Phillips. W. L. M . and Wellington. 1. R.: A laboratory evaluation of zinc anodes in sea
water, CorrosioI116(4). 107-113. 1960.
471. Kurr. G. W.: Zinc anodes-underground uses for cathodic protection and grounding. Mater. Perform.
18(4),34-41.1979.
472. Onishi. M., Wakamatsu. Y, and Miura. H.: Formation and growth kinetics of intermediate phases in
Fe-Zn diffusion couples, Trans. lpn. Inst. Met. 15,331-337, 1974
473. Zinc Institute. Inc .. and Zinc Development Association: Cathodic protection of pipelines with zinc
anodes. Pipei' Pipelines [Ill. 1975(August). 23-27.

428

REFERENCES

474. Matsumoto, Y., Yoshikawa, T., and Sato, E.: Dependence of the band bending of the oxide semiconductors on pH, J. Electrochem. Soc. 136,1389-1391,1989.
475. Justice, D. D., and Hurd, R. M.: Electrochemical dissolution of ZnO single crystals, J. Electrochem.
Soc. 118,1417-1420,1971.
476. Craig, B. D.: Fundamental Aspects of Corrosion Films in Corrosion Science, Plenum Press, New York,
1991.
477. Bindra, P., Gerischer, H., and Kolb, D. M.: Electrolytic deposition of thin metal films on semiconductor
substrates, J. Electrochem. Soc. 124, 1012-1018, 1977.
478. Dirkse, T. P., and Shoemaker, R: Double layer capacitance of zinc electrodes in KOH solutions, J.
Electrochem. Soc. 115, 784-786, 1968
479. Gomes, W. P., Freund, T., and Morrison, S. R.: Chemical reactions involving holes at the zinc oxide
single crystal anode, J. Electrochem. Soc. 115, 818-823,1968.
480. Knotkova-Cermakova, D., and V1ckova. J.: Corrosive effect of plastics. rubber and wood on metals in
confined spaces, Br. Corros. J. 6. 17-22. 1971
481. De Pauli, C. P.. Derosa. O. A. H., and Giordano, M. c.: Zinc dissolution and passivation in buffered
phosphate solutions. Part I. A comparative study with sodium hydroxide solutions. J. EleCfroallal. Chem.
86.335-348, 1978.
482. Pirnat, A., Meszaros, L.. and Lengyel, B.: Study of the formation ofchromate layer on zinc by impedance
technique, Electrochim. Acta 35, 515-522. 1990.
483. Juttner. K.: Electochemical impedance spectroscopy (ElS) of corrosion processes on inhomogeneous
surfaces. Electrochim. Acta 35.1501-1508,1990.
484. Vi1che. J. R. Juttner. K., Lorenz, W. 1., Kautek, w.. Paatsch, w., Dean, M. H., and Stimming, U.:
Semiconductor properties of passive films on Zn, Zn-Co, and Zn-Ni substrates, J. Electrochem. Soc.
136,3773-3778,1989.
485. Allegra, L., and Townsend, H. E.: Undervehicle corrosion resistance of 55% AI-Zn coated steel sheet,
Met. Prog. 119(4),33-35, 1981.
486. Uchiyama, Y.: Corrosion behaviour of zinc-aluminium alloys, Corros. Eng. 39, 297-307,1990.
487. Srivastava, S. K., Srivastava, S., Shukla, R, and Srivastava, S. c.: Effect of addition agents on Zn-Ni-Ti
alloy deposition from a sulphate bath, Indian J. Technol. 27. 564-568, 1989.
488. Menzies, I. A., Marshall, G. w., Gearey, D., and Griffin, G. B.: Some aspects of the mechanism of the
anodic dissolution of zinc in sulfamic acid-formamide solutions, J. Electrochem. Soc. 117, 849-854,
1970.
489. Radeker, w., Peters, F. K., and Friehe, w.: The effect of alloy additions on the properties of hot galvanized
coatings, Proceedings of the 6th International Conference on Hot Dip Galvanizing. Zinc Development
Association, London, England, June 1961.238-264.
490. Arsenault, B., Champagne, B., Lambert, P., and Dallaire. S.: Zinc-nickel coatings for improved
adherence and corrosion resistance, Surf. Coatings Technol. 37, 369-378, 1989.
491. Kim, D. K., and Leidheiser, H.: Chemistry of the surface of hot-dipped galvanized steel and its relation
to paint adherence, Surf. Technol. 5, 379-396, 1977.
492. Graedel, T. E.: Corrosion-related aspects of the chemistry and frequency of occurrence of precipitation,
J. Electrochem. Soc. 133,2476-2482.1986.
493. MacEwan, J. R: Chemical and electrochemical finishes foq:,i.I)c. Consolidated Mining and Smelting
Company of Canada Ltd., Research and Development DiviSIOn, August 1956.
494. Whitaker, M., and Fry, H.: The anodic oxidation of zinc and zinc-aluminium alloys in a sodium
hyroxide-sodium carbonate solution, The British Non-Ferrous Metals Research Association, Report
No. A. 1172, February 1958.
495. Lange. N. A.: Lange's Handbook of Chemistry, J. A. Dean (ed.), 13th ed., McGraw-Hill, New York,
1985.
496. Gubbins, K. E., and Walker, R D.: The solubility and diffusivity of oxygen in electrolytic solutions, J.
Electrochem. Soc. 112,469-471,1965.
497. Roetheli, B. E., Cox, G. L., and Littreal, W. B.: Effect of pH on the corrosion products and corrosion
rate of zinc in oxygenated aqueous solutions, Met. Alloys 3(3), 73-76, 1932.
498. Engelbrecht, G. J.: A fundamental study of the influence of aluminium on the white rusting of galvanized
steel, Ph.D. Thesis, Delft Technical College, Republic of South Africa, 1963.

REFERENCES

429

499. McLeod, W.. and Rogers. R. R.: The nature of corrosion of zinc by sulfurous aCid at ordmary
temperatures, CorrositJl1 25. 74-76, 1969.
500. Hronsky, P.: Corrosion behavior of metallic materials in organic media containing hydrogen chloride,
Corrosion 37. 161-170. J 981.
50 I. Porter. F.: Zinc Handbook, Marcel Dekker, New York. 1991.
502. Dean. S. W.: Planning. instrumentation, and evaluation of atmospheric corrosion tests and a review of
ASTM testing. in Atmospheric Corrosion. W. H. Ailor (ed.), pp. 195-216. John Wiley & Sons. New
York, 1982.
503. American Society for Testing and Materials: Standard practice for conducting atmospheric corrosion
test on metals, in Annual Book of'ASTM Standards, Vol. 03.02. No. G50-76. pp. 192-196, 1989.
504. American Society for Testing and Materials: Standard practice for laboratory immersion corrosion
testing of metals. in Anllual Book of ASTM Standards. Vol. 03.02, No. G3 J -72, pp. \08-115. 1989.
505. Nakasa. K .. and Suzawa, M.: Liquid-zinc-induced crack propagation in high-strength steels in Proceedings oFlhe 1st Inlernational Conference 011 Environment-Induced Cracking of'Metals. October 2-7,
1988, Kohler, Wisconson, National Association of Corrosion Engineers, Houston, 537-540, 1988.
506. Anon.: Panel discussion of inorganic zinc primers at corrosionl75, Male!: Peifonn. 15(5),35-38, 1976.
507. Hinden, E. R: How zinc rich epoxy coating protect structures, Mater. ProI7(4), 38-40, 1968.
508. Okamura, H., and Hisamatsu. Y: Effect of use of galvanized steel on the durability of reinforced
concrete, Mala Peiform. 15(7),43-47, 1976.
509. Haynie. F. H., and Upham. 1. B.: Effects of atmospheric sulfur dioxide on the corrosion of zinc, Male!:
Prot. and Perform. 9(8),35-40. 1970.
510. Bird, C. E.: Corrosion behavior of galvanized sheet in relation to variation in coating thickness, Mate!:
Perform. 16(4), 14-16, 1977.
511. Townsend, H. E .. and Zoccola, J. c.: Atmospheric corrosion resistance of 55% AI-Zn coated sheet steel:
I3 year test results, Mater. Prol. 18(10). 13-20, 1979.
512. Vickers. R.: Commercial hot dip galvanizing of fabricated items, Maler. Pmt. 1( 1).30-39. 1962.
513. Hendry. C. M.: Inorganic zinc rich primers-fact and fancy, Mate!: Perf'orm. 17(5). 19-27. 1978.
514. Dewald. 1. E: The charge and potential distributions at the zinc oxide electrode, Bell S\"-'I. Tech . .1. 39.
615-639. 1960.
515. Doyle. D. P, and Wright. T. E.: Quantitative assessment of atmospheric galvanic corrosion. in Galvanic
Corrosion, STP 978, pp. 161-171, American Society for Testing and Materials, Philadelphia. 1988.
516. Scully. J. R .. and Hack, H. P: Prediction of tube-tubesheet galvanic corrosion using finite clement and
wagner number analyses, in Galvanic Corrosioll. STP 978. pp. 136-157. American Society for Testing
and Materials. Philadelphia. 1988.
517. Escalante. E.: The effect of soil resistivity and soil temperature on the corrosion of galvanically coupled
metals in soil. in Galvanic Corrosion. STP 978. pp. 193-202. American Society for Testing and
Materials. Philadelphia. 1988.
518. Schick. G.: Avoiding galvanic corrosion problems in the telephone cable plant, in Galvanic Corrosion.
STP 9n, pp. 283-290, American Society for Testing and Materials, Philadelphia. 1988.
519. Swamy. R. N.: Resistance to chlorides of galvanized rebars, in Corrosion o/Rein/iJrcCmelll in COllcrete,
C. L. Page. K. W. J. Treadaway. and P B. Bamforth (eds.), pp. 586-600. Elsevier Science Publishing
Co., New York. 1990.
520. Brousseau, R .. Arnott. M .. Dallaire, S .. and Feldman. R: Factors affecting the adhesion on concrete of
arc sprayed zinc, Corrosion 48. 947-952,1992.
521. Geduld. H.: Zinc Plaling, Finishing Publications Ltd., Teddington, Middlesex, England. 1988.
522. Zhang. X. G., and Valeriote. E. M.: Galvanic protection of steel and galvanic corrosion of zinc under
thin layer electrolytes, Corros. Sci. 34, 1957-1972. 1993.
523. Chen. Z. W., Kennon, N. E, Sec, J. B., and Barter. M. A.: Technigalva and other developments in batch
hot-dip galvanizing, .1. MelLlls 44(1),22-26, 1992.
524. Fontana. M. G .. and Greene, N. D.: Corrosion Engineering, 2nd cd., McGraw-Hill. New York. 1978.
525. Cheng. H. B.: Hot dip galvanizing, Galvaniz:ers f(JlInd. Taiwan 20(4),19-30,1992 [in Chinesel.
526. Bothe, A. E., Vilche, 1. R, Juttner, K., Lorenz, W. J., Kautek, W., and Paatsch, W.: An electrochemical
impedance spectroscopy study of passive zinc and low alloyed zinc electrodes in alkaline and neutral
aqueous solutions, Corros. Sci. 32, 621-633, 1991.

430

REFERENCES

527. Morrow, H.: Encyclopedia of Materials Science and Engineering, Vol. 7, M. B. Bever (ed.), MIT Press,
Cambridge, Massachusetts, 1986.
528. National Research Council: Zinc, University Park Press, Baltimore, 1979.
529. Mongeon, L., and Barnhurst, R. 1.: Zinc and zinc alloys, in ASM Metals Handbook, 9th ed .. Vol. 9, pp.
488-496, American Society for Metals, Materials Park, Ohio, 1985.
530. Barnhurst, R. 1.: Zinc and zinc alloys, in ASM Metals Handbook, 10th ed., Vol. 2, American Society for
Metals, Materials Park, Ohio, pp. 527-542,1990.
531. Zhang, X. G.: Electrochemical stripping of galvannealed coatings on steel, Corrosion 50, 308-317,
1994.
532. Brodd, R. J., and Leger, V. E.: Zinc, in Encyclopedia of Electrochemistry of the Elements, Vol. V, A. 1.
Bard (ed.), pp. 2-60, Marcel Dekker, New York, 1976.
533. Naybour, R. D.: The effect of electrolyte flow on the morphology of zinc electrodeposited from aqueous
alkaline solution containing zincate ions, J. Electrochem. Soc. 116,520-524, 1969.
534. Baugh, L. M., Tye, E L., and White, N. c.: Corrosion and polarization characteristics of zinc in battery
electrolyte analogues and the effect of amalgamation, J. Appl. Electrochem. 13, 623-635, 1983.
535. Roberts, C. W: Intercrystalline corrosion in cast zinc-aluminium alloys, J. Inst. Met. 81,301-309,
1952-53.
536. Kolberg, T., Rausch, W, Schubach, P., and Wendel, T. (Metallgesellschaft A. G., Germany): Formation
of conversion coatings on surfaces of zinc or zinc alloys, CA Patent 2041892, November 30, 1991.
537. Sereda, P. 1., Croll, S. G., and Slade, H. E: Measurement of the time-of-wetness by moisture sensors
and their calibration, in Atmospheric Corrosion of Metals, STP 767, pp. 267-285, American Society
for Testing and Materials, Philadelphia, 1982.
538. American Society for Testing and Materials: Standard practice for measurement of time-of-wetness on
surfaces exposed to wetting conditions as in atmospheric corrOSIOn testing, in Annual Book of ASTM
Standards, Vol. 03.02, No. G84-89, pp. 353-358,1989.
539. Corrosiveness of various atmospheric test sites as measured by specimens of steel and zinc, in Metal
Corrosion in the Atmosphere, STP 435, pp. 360-391, American Society for Testing and Materials,
Philadelphia, 1968.
540. Evans, E. L. 1.: Chapter 2.1 in Corrosion, Vol. 1, Corrosion of Metals and Alloys, L. L. Shreir (ed.),
George Newnes Ltd., London, 1963.
541. Sereda, P. J.: Weather factors affecting corrosion of metals, in Corrosion in Natural Environments, STP
558, pp. 7-22, American Society for Testing and Materials, Philadelphia, 1974.
542. Brown, P. W., and Masters, L. W.: Factors affecting the corrosion of metals in the atmosphere, in
Atmospheric Corrosion, W H. Ailor (ed.), pp. 31-49, John Wiley & Sons, New York, 1982.
543. Copson, H. R.: Report of subcommittee VI, of Committee B-3, on Atmospheric Corrosion, ASTM 58th
Annual Meeting, Symposium on Atmospheric Corrosion of Non-Ferrous Metals, STP 175, pp. 3-19,
American Society for Testing and Materials, Philadelphia, 1955.
544. Teeple, H. 0.: Atmospheric galvanic corrosion of magnesium coupled to other metals, ASTM 58th
Annual Meeting, Symposium on Atmospheric Corrosion of Non-Ferrous Metals, STP 175, pp. 89-115,
American Society for Testing and Materials, Philadelphia, 1955.
545. Compton, K. G., and Mendizza, A.: Galvanic couple corrosion studies by means of the threaded bolt
and wire test, ASTM 58th Annual Meeting, Symposium on Atmospheric Corrosion of Non-Ferrous
Metals, STP 175, pp. 116-125, American Society for Testing and Materials, Philadelphia, 1955.
546. Anderson, E. A.: The atmospheric corrosion of rolled zinc, ASTM 58th Annual Meeting, Symposium
on Atmospheric Corrosion of Non-Ferrous Metals, STP 175, pp. 126-134, American Society for Testing
and Materials, Philadelphia, 1955.
547. Baker, E. A., and Lee, T. S.: Calibration of atmospheric corrosion test sites, in Atmospheric Corrosion
of Metals, ST.P 767, pp. 250-266, American Society for Testing and Materials, Philadelphia, 1982.
548. Haynie, E H.: Evaluation of the effects of microclimate differences on corrosion, in Atmospheric
Corrosion of Metals, STP 767, pp. 286-308, American Society for Testing and Materials, Philadelphia,
1982.
549. Showak, W, and Dunbar, S. R.: Effect of I percent copper addition on atmospheric corrosion of rolled
zinc after 20 years' exposure, in Atmospheric Corrosion of Metals, STP 767, pp. 135-162, American
Society for Testing and Materials, Philadelphia, 1982.

REFERENCES

431

550. Tonini, D. E.: Atmospheric corrosion test results for metallic-coated steel panels exposed in 1960, in
Atmospheric Corrosion of Metals, STP 767, pp. 163~ 185, American Society for Testing and Materials,
Philadelphia, 1982.
551. Baboian, R.: Final report on the ASTM study: atmospheric galvanic corrosion of magnesium coupled
to other metals, in Atmospheric Factors Affecting the Corrosion of Engineering Melllls, STP 646, pp.
17~29, American Society for Testing and Materials, Philadelphia, 1978.
552. Longo, F N., and Durrnann, G. 1.: Corrosion prevention with thermal-sprayed ZInc and aluminum
coatings, in Atmospheric Factors Affecting the Corrosion of Engineering Metals, STP 646, pp. 97 ~ I 14,
American Society for Testing and Matenals. Philadelphia. 1978.
553. Schikorr, G.: Behavior of zinc in the atmosphere, in Corrosion 8ehavior o{Zinc, Vol. I, American Zinc
Institute and the Zinc Development Association, 1965; cited on p. 30 in Zinc: Its Cormsion Resistance,
2nd ed .. International Lead Zinc Research Organization. New York, 1986.
554. Barton, K .. and Beranek. E.: Reaction mechanism of the atmospheric corrosion of metals in moist air

555.
556.
557.
558.
559.
560.
561.
562.

563.
564.
565.
566.

contaminated with sulfur dioxide, Werkst. Korros. 10, 377~383, 1959; cited on p. 27 in Zinc: Its
Corrosion Resistance. 2nd ed .. International Lead Zinc Research Organization, New York, 1986.
Schikorr, G., and Schikorr, I.: The resistance of zinc to atmospheric corrosion. Z. Metallkd. 35(9).
175~181. 1943 [TranslationJ.
Rozenfeld, 1. L.: Atlllospheric Corrosion of Metals, National Association of Corrosion Engineers.
Houston, 1972.
Hudson. J. c.: TrailS. Faradav Soc. 25. 177. 129.
Drane, C. W.: Chapter 2.2 in Corrosion, Vol. 1, CorrosionofMetalsandAllo.\'s, L. L. Shreir(ed.), George
Newnes Ltd., London, 1963.
Anderson, E. A .. and Reinhard. C. E.: Zinc, in The Corrosion Handbook, H. H. Uhlig (ed.), pp. 331 ~346.
John Wiley & Sons. New York. 1948.
Schikon', G.: Z. Metallkd. 32. 314.1940; cited on pp. 4 and 138 in Corrosion, Vol. I. Corrosion o{Metals
and Allo,vs, L. L. Shrcir (ed.). George Newnes Ltd .. London. 1963.
Fujii, T.: Rust Prevention (80sei Kanri) 27, 85,1983; cited on p. 97 in Corrosion-Resistant Coatings
Technology, P. A. Schweitzer (ed.), Marcel Dekker, New York, 1989.
Suzuki, N.: Safety Engineering of Corrosion Protection-Protective Coatings, The High Pressure Gas
Safety Institute of Japan, 1986, p. 13; cited on p. 104 in Corrosion-Resistant Coatings Technologv, P.
A. Schweitzer (ed.). Marcel Dekker, New York, 1989.
Crennell. J. T.: Chapter 2.3 in Corrosion, Vol. I, Corrosion of Metals and Alloys, L. L. Shreir (ed.),
George Newnes Ltd., London, 1963.
Tsukuda, T., and Kawabe, M.: Therm. Nucl. Power Eng. Soc. Jpn. 25, 985, 1974: cited on p. 103 in
Corrosion-ResisIallt Coatings Technologr. P. A. Schweitzer (ed.), Marcel Dekker, New York, 1989.
Logan. K. H.: Corrosion by soils, in The Corrosion Handbook. H. H. Uhlig (ed.), pp. 446~466. John
Wiley & Sons, New York, 1948.
Redfield. A. c.: Characteristics of sea water, in The Corrosion Handhook, H. H. Uhlig (cd.), pp.

III1 ~ 1122. John Wiley & Sons, New York. 1948.


567. Kim, D. K., and Leidheiser, H.: Metall. Trans. 98. 581. 1978; cited on p. 218 in Corrosion-Resistant
Coatings Technology. P. A. Schweitzer (ed.). Marcel Dekker. New York. 1989.
568. Standish, J. V: Atmospheric corrosion mechanisms of painted steel. Society of Automotive Engineers
International Congress & Exposition, Detroit, February 28~March 4, 1983, SP-538. pp. 1~lJ.
569. Hafford, B. c.: Zinc-containing protective coatings, in Encvc/opedia of'Materials Science and Engineering. Vol. 7, M. B. Bever (ed.), pp. 5521~5528, MIT Press. Cambridge, Massachusetts, 1986.
570. Yan, M. F: Zinc oxide, in Encyclopedia of Materials Science and Engineering, Vol. 7, M. B. Bever
(ed.), pp. 5528~5530, MIT Press. Cambridge. Massachusetts, 1986.
571. Baboian. R.: The automotive environment. in Automotive Corrosion b,' Deicing Salts, R. Baboian (ed.),
pp. 3~ 12, National Association of Corrosion Engineers, Houston, 1981.
572. Neville, R. J.: A test for under vehicle corrosion resistance, in Automotive Corrosioll by Deicing Salts,
R. Baboian (ed,), pp. 182~218, National Association of Corrosion Engineers. Houston, 1981.
573. Rynewicz,1. F: Evaluation of paint coatings tested in the deep Atlantic and Pacific oceans, in Corrosion
in Natural Environments. STP 558, pp. 209~235, American Society for Testing and Materials, Philadelphia. 1974.

432

REFERENCES

574. Huot, J. Y: The effects of silicate ion on the corrosion of zinc powder in alkaline solutions, J. Appl.
Electrochem. 22,443-447,1992.
575. Reinhart, EM.: Corrosion of metals and alloys in the deep ocean, Naval Facilities Engineering
Command, Alexandria, Virginia, Technical Report R834. February 1976.
576. Natorski, T. J.: Zinc and zinc alloy plating in the '90s, Met. Finish. 90(3),15-17.1992.
577. Pearlstein, E: Selection and application of inorganic finishes. Part 1. Black oxide and phosphate coatings,
Plating Suif. Finish. 65(] 2), 22-29, 1978.
578. Phosphate coating, in ASM Metals Handbook, 9th ed., Vol. 5, pp. 434-456, American Society for Metals,
Materials Park, Ohio. 1982.
579. Cheever, G. D.: Formation and growth of zinc phosphate coatings, J. Paint Techno!. 39. 1-13, 1967.
Photoelectrochemistry of semiconductor ZnO particulate films, J.
580. Hotchandani, S., and Kamat, P.
Electrochem. Soc. 139. 1630-1634, 1992.
581. Tegehall, P. E., and Vannerberg, N.-G.: Nucleation and formation of zinc phosphate conversion coating
on cold-rolled steel, Corros. Sci. 32, 635-652, 1991.
582. International Organization for Standardization: Chromate conversion coatings on zinc and cadmiumtest methods, International Standard, ISO 3613-1980 (E), pp. 243-246.
583. American Society for Testing and Materials: Standard practice for testing chromate coatings on zinc
and cadmium surfaces, Annual Book of ASTM Standards, Vol. 02.05, No. B201-80, pp. 58-60, 1980.
584. Centre de Recherches Metallurgiques (Belgium): Continuous gal fan coating of steel sheet and wire,
International Lead Zinc Research Organization Project ZM-285, Progress Report No. 20, November
1988-ApriI1989.
585. Johnsson, T., and Kucera,
Eight years' field exposure of coatings of zinc, aluminium and their alloys,
Proceedings of the 2nd International Conference on Zinc Coated Steel Sheet, Rome, Italy, June 9-10,
1988, pp. SA6/I-SA6/l1.
586. Davin, A., and Goodwin, E E.: Atmospheric corrosion behaviour of gal fan coated steels, Proceedings
of the 2nd International Conference on Zinc Coated Steel Sheet, Rome, Italy, June 9-10, 1988, pp.
SA2/l-SA2/31.
587. Belisle, S., and Dufresne, R.: Corrosion resistance of ZA alloys, Canadian Institute of Mining,
Metallurgy, and Petroleum 25th Annual Conference of Metallurgists, International Symposium on
Zinc-Aluminum (ZA) Casting Alloys, Toronto, Ontario, August 17-20, 1986, pp. 109-126.
588. Vrable, J. B.: Effect of forming deformation on corrosion test performance of electroplated zinc and
iron-zinc alloy coatings for automotive sheet, SAE Technical Paper No. 862012, Society of Automotive
Engineers Automotive Corrosion and Prevention Conference, Dearborn, Michigan, December 8-10,
1986.
589. Gray, B. M., and Shaffer, R. 1.: An evaluation of cosmetic corrosion performance of zinc and zinc alloy
coated steels, SAE Technical Paper No. 862014, Society of Automotive Engineers Automotive Corrosion and Prevention Conference, Dearborn, Michigan, December 8-10,1986.
590. Lay, D. E., and Eckles, W. E.: The fundamentals of zinc/cobalt, Plating Suif. Finish. 77(10), 10-14.
1990.
59!. Anderson, E. A.: Chromate coatings on zinc, in The Corrosion Handbook, H. H. Uhlig (ed.), pp.
862-864, John Wiley & Sons, New York, 1948.
592. Taylor, E: Chromate passivation, Met.lnd, 1944(September), 149-150.
593. Anderson, E. A.: The Cronak process, Proc. Am. Electroplaters Soc. 31, 6-13, 1943.
594. Coppins. W. C: Chemical colouring, Met. Ind. 1948(June),482.
595. Stareck, 1. E.: Anozinc and other new coatings on zinc, Proc. Am. Electroplaters Soc. 32, 235-247,
1944.
596. Pearlstein, E: Selection and application of inorganic finishes. Part IV. Anodic coatings for other metals,
Plating Suif. Finish. 66, 41-46, 1979.
597. Wilcox, G. D., and Gabe, D. R.: Passivation studies using group via anions. Part iv. Cathodic redox
reactions and film formation, Br. Corros. J. 19, 196-200, 1984.
598. Barnes, C, Ward, J. J. B., Sehmbhi, T. S., and Carter, V. E.: Non-chromate passivation treatments for
zinc, Trans. Inst. Met. Finish. 60,45-48, 1982.
599. Shalaby, L. A., and Abbas, H.: The behaviour of zinc and zinc-silver alloys in the presence of some
inhibitors, Carras. Sci. 13, 545-552, 1973.

v.:

v.:

REFERENCES

433

600. Abdul Azim, A. A., Shalaby, L. A., and Abbas, H.: Mechanism of the corrosion inhibition ofZn anode
in NaOH by gelatine and some inorganic anions, Corros. Sci. 14,21-24, 1974.
601. LeRoy. R. L., and Zavorsky, Z.: Cathodic pretreatment of zinc and zinc-coated surfaces, Corms. Sci.
17, 943-944, 1977.
602. Rudresh, H. B., and Mayanna, S. M.: The synergistic effect of halide ions on the corrosion inhibition
of zinc by n-decylamine, Corros. Sci. 19,361-370,1979.
603. De Pauli, C P., Giordano, M. C, Lopez, B. A., and Manzur, M. E.: The passivation of zinc in slightly
acidic buffered solutions in the presence of phosphate ions, Corros. Sci. 28, 769-785, 1988.
604. Bagger, N. B.: Colouring of metals, Mater. Methods 1949(June), 67-82.
605. Hinton, B. R. W.: New approaches to corrosion inhibition with rare earth metal salts, National
Association of Corrosion Engineers Corrosion '89 Conference, New Orleans, Louisiana, April 17-21,
Paper No. 170.
606. Hosny, A. Y., El-Rafei, M. E., Ramadam, T A., and El-Gafari, B. A.: Corrosion resistance ofzinc coatings
produced from a sulphate bath, Met. Finish. 1995(November), 55-59.
607. Suzuki, l.: The behavior of corrosion products on zinc in sodium chloride solution, Corros. Sci. 25,
1029- 1034, 1985.
608. Keddam, M., Hugot-Le-Goff, A., Takenouti, H., Thierry, D., and Arevalo, M. C: The influence of a thin
electrolyte layer on the corrosion process of zinc in chloride-containing solutions, Corms. Sci. 33,
1243-1252,1992.
609. Dante, J. F., and Kelly, R. G.: The evolution of the adsorbed solution layer during atmospheric corrosion,
Symposium on Corrosion and Reliability of Electronic Materials and Devices, Electrochemical Society
Meeting, Toronto, Ontario, October 1992.
610. Kawafuku, J., Katoh, J., Toyama, M., Nishimoto, H., Ikeda, K., and Satoh, H.: Structure and corrosion
resistance of zinc alloy coated steel sheets obtained by continuous vapor deposition apparatus, 1. Iron
Steel lilst. Jpn. 77, 995-1002,1991.
611. Sagiyama, M., Hiraya, A., and Watanabe, T.: Electrochemical behavior of electrodeposited zinc-iron
alloys in 5'70 NaCI solution, J.lmn Steel Inst. lpn. 77,244-250, 1991.
612. Sagiyama, M., Hiraya, A., and Watanabe, T.: Electrochemical behavior of electrode posited zinc-iron
alloys in alkaline solutions, J. Iron Steel blSt. Jpn. 77,251-257, 1991.
613. Schubert. R.: The interaction of sulfuric acid vapors and zinc surfaces, 1. p;fectrochem. Soc. 139,
3450-3453, 1992.
614. Zhang, X. G.: Effect of weather pattern on corrosion of zinc and steel, unpublished results.
615. Grossman, P. R.: Investigation of atmospheric exposure factors that determine time-of-wetness of
outdoor structures, in Atmospheric Factors Affecting the Corrosion of Engineering Metals, STP 646,
pp. 5-16, American Society for Testing and Materials, Philadelphia, 1978.
616. Baedecker, P. A., Edney, E. 0., Moran, P. J., Simpson, T C, and Williams, R. S.: Effects of acidic
deposition on materials, State-of-Science/Technology, National Acid Precipitation Assessment Program, Report No. 19, May 1990.
617. Daescn, J. R.: Corrosion resistance of galvanized coatings, Proceedings of the 2nd International
Congress on Metallic Con'osion, New York, March 11-15, 1963, pp, 686-705.
618. Ellis, O. B.: Effect of weather on the initial corrosion rate of sheet zinc, Proc. ASTM COllf 49, 152-170,
1949.
619. Salt, F. W., Stanners, J. F., and Watkins, K. 0.: Atmospheric exposure trials of electrodeposited iron-zinc
alloy coatings, Br. Corros. 1. 1,5-8, 1965.
620. Dey, A. K.. Sinha Mahapatra, A. K., Khan, D. K., Mukherjee, A. N., Narain, R., Mukherjee, K. P., and
Baneljee. T: Interim report on atmospheric corrosion studies under marine atmosphere, NML (Natl.
Mew/!. Lab., Jmnshedpur. India) Tech. J. 8, 11-16, 1966.
621. Cooke, F., Cosslett, J. K., Scott, R. W., and Shanahan, C E. A.: The atmospheric cOlTosion resistance
of elcctrodeposited tin-cadmium and tin-zinc coatings on steel, Br. Corros. l. 1. 283-289, 1966.
622. Payer, J. H .. and Safranek, W. H.: Atmospheric corrosion of electroplated zinc alloy die casting, in
Atmospheric Factors Affecting the Corrosion of Engineering Metals, STP 646, pp. 115-128, American
Society for Testing and Materials, Philadelphia, 1978.
623. Fukushima, T, Sato, N., Hisamatsu, y, Matsushima, T, and Aoyama, Y.: Atmospheric corrosion testing
in Japan, in Atmospheric Corrosion, W. H. Ailor (ed.), pp. 841-872, John Wiley & Sons, New York,
1982.

434

REFERENCES

624. Longo, EN., and Durmann, G. 1.: Corrosion prevention with thermal-sprayed zinc and aluminum
coatings, in Atmospheric Factors Affecting the Corrosion of Engineering Metals, STP 646, pp. 97-114,
American Society for Testing Materials, Philadelphia, 1978.
625. Spedding, D. 1.: Some investigations into hydrogen sulphide adsorption by zinc, Corros. Sci. 17,
173-178, 1977.
626. Spence, 1. W, Edney, E. 0., Haynie, E H., Stiles, D. c., Corse, E. W, Wheeler. M. S., and Cheek, S. E:
Advanced laboratory and field exposure systems for testing materials, in Corrosion Testing and
Evaluation: Silver Anniversary Volume. STP 1000, pp. 191-207, American Society for Testing and
Materials, Philadelphia, 1990.
627. Spence, 1. W, and Haynie, E H.: Derivation of a damage function for galvanized steel structures:
Corrosion kinetics and thermodynamic considerations, in Corrosion Testing and Evaluation: Silver
Anniversary Volume, STP 1000. pp. 208-224. American Society for Testing and Materials. Philadelphia.
1990.
628. Haynie, E H . Spence, 1. W, Lipfert, E W, Cramer, S. D., and McDonald. L. G.: Evaluation of an
atmospheric damage function for galvanized steel, in Corrosion Testing and Evaluation: Silver Anniversary Volume, STP 1000, pp. 225-240. American Society for Testing and Materials, Philadelphia,
1990.
629. Cramer, S. D., and McDonald, L. G.: Atmospheric factors affecting the corrosion of zinc. galvanized
steel and copper, in Corrosion Testing and Evaluation: Silver Anniversary Volume, STP 1000. pp.
241-259, American Society for Testing and Materials, Philadelphia. 1990.
630. Zhang. X. G.: Corrosion of zinc in salt spray test. Cominco Report 651-008, 1992.
631. Spence. J. W, Haynie, E H., Lipfert, E w., Cramer, S. D . and McDonald, L. G.: Atmospheric corrosion
model for galvanized steel structures, Corrosion 48.1009-1019, 1992.
632. Locke, C. E.: Solving rebar corrosion problems in concrete, National Association of Corrosion
Engineers. Chicago. September 1982, reprint #2.
633. Locke, C. E.: Solving rebar corrosion problems in concrete. National Association of Corrosion
Engineers. September. Chicago. September 1982, preprint #5.
634. Lin, S. H.: Chloride diffusion in a porous concrete slab, Corrosion 46. 964-967. 1990.
635. lang. 1. W. and Iwasaki, I.: Rebar corrosion under simulated concrete conditions using galvanic current
measurements, Corrosion 47. 875-884, 1991.
636. Gau, Y.. and Cornet, I.: Penetration of hardened concrete by seawater chlorides with and without
impressed current. Corrosion 41, 93-100. 1985.
637. Tritthart. J.: Pore solution composition and other factors influencing the corrosion risk of reinforcement
in concrete. in Corrosion of Reinforcement in Concrete. C. L. Page, K. W. J. Treadaway, and P. 8.
Bamforth (eds.), pp. 96-106, Elsevier Science Publishing Co. New York. 1990.
638. Bamforth, P. 8.. and Pocock, D. c.: Minimising the risk of chloride induced corrosion by selection of
concreting materials, in Corrosion of Reinforcement in Concrete, C. L. Page, K. W. J. Treadaway, and
P. B. Bamforth (eds.), pp. 119-131. Elsevier Science Publishing Co . New York, 1990.
639. Dehghanian, c., and Locke, C. E.: Electrochemical behavior of steel in salt contaminated concrete: Part
I, Corrosion 39. 299-305. 1983.
640. American Society for Testing and Materials: Standard test method for half-cell potentials of uncoated
reinforcing steel in concrete, in ASTM Standards for Corrosion Testing of Metals, No. C876-87. pp.
94-98. ASTM, Philadelphia, 1990.
641. Elsener, 8., Muller, S., Suter, M., and Bohnl, H.: Corrosion monitoring of steel in concrete-theory and
practice, in Corrosion of Reinforcement in Concrete, C. L. Page, K. W. J. Treadaway, and P. 8. Bamforth
(eds.), pp. 348-357, Elsevier Science Publishing Co., New York, 1990.
642. Borgard, 8., Ramirez, c., Somayaji, S., Jones, D., Keeling, D., and Heidersbach, R.: Failure analysis
in concrete structures: A comparison of field data with results from laboratory exposure, Corrosion 47,
758-769, 1991.
643. Funahashi, M., and Bushman, J. 8.: Technical review of 100 mv polarization shift criterion for
reinforcing steel in concrete, Corrosioll47, 376-386,1991.
644. Rosenberg, A. M., Gaidis, 1. M., Kossivas, T. G., and Previte, R. W: A corrosion inhibitor formulated
with calcium nitrite for use in reinforced concrete, in Chloride Corrosion of Steel in Concrete, STP 629,
pp. 89-99, American Society for Testing and Materials, Philadelphia, 1977.

REFERENCES

435

645. Cornet, I.. and Bresler. B.: Galvanized steel in concrete: Literature review and assessment of perform
ance, in Galvanized Reinforcement for Concrete~lI. pp. I-56. International Lead Zinc Research
Organization. New York. 1981.
646. Gonzalez, 1. A .. Algaba, S., and Andrade. c.: Corrosion of reinforcing bars in carbonated concrete. Br.
Corms . .I. 15, 135-139. 1980.
647. Clemena, G. G.:Electrically conductive portland cement concrete, Proceedings of the National Asso
ciation of Corrosion Engineers Corrosion '87 Symposium on Corrosion of Metals in Concrete. pp.
49-64.
648. Krepski. R. P., Zinc silicate thermal spray coating for corrosion control of steel in concrete, Proceedings
of Fourth National Thermal Spray Conference. Pittsburgh, Pennsylvania, 1991. p. 433.
649. Yeomans. S. R.: Corrosion testing of black, galvanized and epoxy coated reinforcing steel in concrete,
University of New South Wales, internal report Australia, 1993.
650. Piercy, R. c.: Cominco Project No. 676747, Report No.5. 1967.
65 I. Watson. T. w.: Cominco Project No. 676749. Report No.3. 1967.
652. Keitelman. A. D., Gravano. S. M . and Galvele. J. R.: Localized acidification as the cause of passivity
breakdown of high purity zinc. Corms. Sci. 24, 535-545, 1984.
653. Li, S., He. 1.. and Li. S.: The effect of rare earth elements on corrosion resistance of electrodeposited
zinc coating. Proceedings of the 7th APCCC Conference on Corrosion Control, Beijing China. 1991.
Vol. 2. pp. 1034-1038.
654. Bengough, G. D., and Hudson, O. F.: Fourth report to the corrosion committee of the institute of metals.
Presented at the Annual General Meeting. London. March 25, 1919.
655. Featherly. R. L.: Practical experiences in corrosion of hot water tanks. Plumbing Hearing .I.
1952(ApriI/May).
656. Newell, I. L.: The corrosion of domestic galvanized hot water storage tanks, Corrosio/1 9, 46-51. 1953.
657. Duval R.. and Arliguie. M. G.: Passivation du zinc dans I'hydroxyde de calcium. eu, egard au
comportement de I'acier galvanise dans Ie beton. Mem. Sci. Rev. Metal/urg. LXXI. 719. 1974.
658. Stein, A.: The prevention of corrosion in galvanized iron water tanks-the results of 12 year field trials.
Proceedings of the 5th Annual Conference of the Australasian Corrosion Association. Newcastle. New
South Wales. Australia. November 16-20,1964.
659. Chandler. K. A.: Marine and Offshore Corrosion, Butterworth & Co . London, 1985.
660. Fink. F. W.. and Boyd. W. K.: The corrosion of metals in marine environments, Defense Metals
Information Center. DMIC Report No. 245, Columbus, Ohio, May. 1970.
661. Darbin M .. Jailloux, J. M .. and Montuelle, 1.: Durability of reinforced earth structure: The results of a
longterm study conducted on galvanized steel. Proc. Inst. Civ. Eng. Part I, 1988, 1029.
662. Khan, D. K .. Mukherjee. K. P., and Banerjee, T.: Correlation of data in laboratory and field studies on
sea water corrosion. NML (Natl. Metall. Lab., Jamshedpur, India) Tech. J. 8(4).17-21. 1966.
663. Southwell. C. R .. and Alexander, L. L.: Marine corrosion of cast and wrought nonferrous metals,
National Association of Corrosion Engineers Conference, Houston. 1969, Paper No. 63.
664. Sedzimir. 1. A., and Gmytryk, M. W.: Corrosion of zinc in neutral sodium chloride solutions at constant
pH, Proceedings of the 2nd International Congress on Metallic Corrosion. New York, March 11-15,
1963, pp. 972-975.
665. Memmi, M.: Improvement in the corrosion resistance of Zn-Al coatings by addition of magnesium.
Proceedings of the 14th International Galvanizing Conference, INTERGALVA '85. Munich. Germany,
1985. pp. 8/1-8/2.
666. Frappe. A. L.. Limare, A .. Piessen. P.. and Vacher, 1. c.: Investigation of the influence of impurities or
alloying clements on the atmospheric corrosion of gal vanized steel, Proceedings of the 14th International
Galvanizing Conference, INTERGALVA '85. Munich, Germany, 1985. pp. 8/3-8/18.
667. Dreulle. P. Rosset. R .. and Jardy. A.: Hydrophosphate coating, Proceedings of the 12th International
Galvanizing Conference, INTERGALVA '79, Paris, May 17-23, 1979, pp. 92-96.
668. Hissel, 1.: Corrosion of galvanized steel related to water composition, Proceedings of the 12th
International Galvanizing Conference, INTERGALVA '79. Paris, May 17-23, 1979. pp. 84-91.
669. Rosset, R., Jardy. A., and Parthasarathi, M. N.: Protection of galvanized steel against cold and hot water
corrosion by a pyrophosphate coating, Proceedings of the 13th International Galvanizing Conference.
INTERGALVA '82, England. 1982.

436

REFERENCES

670. Mihailov, G., and Iovchev, M.: Corrosion studies of galvanised steel and zinc in soft water at various
temperatures, Proceedings of the 8th International Congress on Metallic Corrosion. Frankfurt. Germany,
1981,pp.1754-1759.
671. Leclercq, M.: Preweathered zinc for roofing, Proceedings of the 8th International Congress on Metallic
Corrosion, Frankfurt, Germany, 1981, pp. 299-303.
672. Abdel Aal, M. S., Abdel Wahab, A A., and EI Saied, A: A study of the inhibiting action of benzene
thiols and related compounds on the corrosion of zinc in acidic media, Proceedings of the 8th
International Congress on Metallic Corrosion, Frankfurt, Germany, 1981, pp. 1212-1216.
673. Sagiyama, M., Inagaki, 1.-1., and Morita, M.: Fe-Zn alloying behavior and the coating microstructure
of galvannealed steel sheets, NKK Technical Report, No. 135, 1991, pp. 49-56.
674. Takimoto, K-I., Suzuki, K.-I., Nishizaka, K, and Ohtsubo, T.: Quantitative analysis of zinc-iron alloy
galvanized coatings by glow discharge spectrometry and secondary ion mass spectrometry, Nippon Steel
Technical Report, No. 33,1987, pp. 28-35.
675. Hayashi, K, Ito, y', and Miyoshi, Y.: Effect of coating weight and corrosive environment on under-film
corrosion of Zn and Zn-Fe alloy coated steel sheets, l. Iron Steellnst. lpn. 78, 127-133. 1992.
676. Odashima, H.: Characteristics of electrolytic chromate films formed on zinc or zinc alloy plated steel
sheet in low concentration Cr0 3-Heavy Metal Ions-Halogen Bath, 1. Iron Steellnst. lpn. 78, 121-126,
1992.
677. Saito, T., Wake, R., Oka, J., and Kitayama, M.: Development and properties ofZn-Ni alloy electroplated
steel sheet (DURZINKLITE), Nippon Steel Technical Report, No. 25. 1985, pp. 1-10.
678. Hada, T., Kanamaru, T., Nakayama, M., Arai, K, Fujiwara, T., Suemitsu, Y., Sato, M., and Ogawa, Y.:
Development and Properties of Zn-Fe Alloy Electroplated Sheet Steel DUREXCELITE, Nippon Steel
Technical Report, No. 25,1985, pp. 11-18.
679. Shindo, y', Yoneno, M., Yamada, A., Oka, 1., and Ejima, M.: Development and properties of organic
composite-coated steel sheet, Nippon Steel Technical Report, No. 25, 1985, pp. 19-27.
680. Tano, K, and Higuchi, S.: Development and properties of zinc-aluminum alloy coated steel sheet with
high corrosion resistance (SUPER ZINC), Nippon Steel Technical Report, No. 25,1985, pp. 29-37.
681. Cabot, P. L., Cortes, M., Centellas, E, and Perez, E.: Potentiostatic passivation of zinc in alkaline
solutions, l. Appl. Electrochem. 23, 371-378, 1993.
682. Schikorr, G.: The cathodic behavior of zinc versus iron in hot tap water, Trans. Electrochem. Soc. 76,
247-258, 1939.
683. Cachet, c., and Wiart, R: Zinc deposition and passivated hydrogen evolution in highly acidic sulphate
electrolytes, depassivation by nickel impurities, 1. Appl. Electrochem. 20. 1009-1014, 1990.
684. Edney, E. 0., Stiles, D. c., Corse. E. W., Wheeler, M. L., Spence, 1. w., Haynie, E H., and Wilson, W.
E.: Controlled field study to determine the impact of dry and wet deposition of air pollutants on the
corrosion rate of galvanized steel, National Association of Corrosion Engineers Corrosion '87 Conference, San Francisco, March 9-13, 1987, Paper No. 410.
685. McLarnon, E R, and Cairns, E. J.: The secondary alkaline zinc electrode, l. Electrochem. Soc. 138,
645-664, 1991.
686. Bonilla, C. E: Pipe service tests on Baltimore water-pitting of black and galvanized wrought iron and
steel pipe carrying cold water, hot water, and returning condensate, Trans. Electrochem. Soc. 87,
237-254,1945.
687. Aurian-Blajeni, B., and Tomkiewicz, M.: Space charge effects at passive zinc electrodes, 1. Electrochem.
Soc. 133, 1766-1769, 1986.
688. Morral, E R: X-ray analysis of corrosion products from galvanized sheets, Trans. Electrochem. Soc.

77n8,279-288.1940.

689. Southwell, C. R, Hummer, C. w., Alexander, A. L, and Forgeson, B. W.: Corrosion of metals in tropical
environments. Part VI-galvanic couples, National Association of Corrosion Engineers, 19th Annual
Conference, New York, March 11-15, 1963.
690. Alexander, A L., Forgeson, B.
Mundt, H.
Southwell, C. R, and Thompson, L. J.: Corrosion of
Metals in Tropical Environments-Part I-Test Methods Used and Results Obtained for Pure Metals
and a Structural Steel. Naval Research Laboratory, Report No. 4929, U.S. Dept. of Commerce,
Washington, D.C., June 19,1957.
691. Kweon, Y. G., and Coddet, C.: Behavior in seawater of zinc-base coatings on aluminum alloy 5086,
Corrosion 48, 97-102, 1992.

w.,

w.,

REFERENCES

437

692. Suzuki, I., and Enjuzi, M.: The development of the corrosion resistance of an Fe-Zn alloy coating on
the basis of the behaviour of the corrosion product, Corms. Sci. 26. 349-355, 1986.
693. Stern, M.: A method for determining corrosion rates from linear polarization data, Corrosion 14,
440t-444t, 1958.
694. Stern, M., and Geary, A. L.: Electrochemical polarization-Part I. A theoretical analysis of the shape
of polarization curves, .I. Eleclrochelll. Soc. 104. 56-63, 1957.
695. Kautek, W.: The galvanic corrosion of steel coatings: Aluminum in comparison to cadmium and zinc,
Corros. Sci. 28, 173-199. 1988.
696. Campbell, H. S.: Sprayed aluminium and zinc protect against corrosion by aqueous solutions.
Anti-Corrosioll 1982(August), 10-12.
697. Rangel, C Moo De Damborenea, 1., De Sa. A. I.. and Simplicio. M. H.: Zinc and polyphosphates as
COlTosion inhibitors for zinc in ncar neutral waters. Br. Corros . .I. 27, 207-212, 1992.
698. De Pauli, C Poo Giordano. M. C, and Mishima. H. T.: Zinc dissolution and passivation in buffered
phosphate solutions. Part II. About the nature of the passivating film,.!. Eleclmallal. Chelll. 103,95-102,
1979.
699. Belisle. S .. and Dufresne, R.: Performance of galvanized steel in municipal wastewater treatment plants.
Maler. Perform. 31(6). 64-67. 1992.
700. Deslouis. C, Duprat. Moo and Tourni II on, C: The kinetics of zinc dissolution in aerated sodium sulphate
solutions. A measurement of the corrosion rate by impedance techniques. Corms. Sci. 29. 13-30. 1989.
701. Fedrizzi. L.. Ciaghi, L., Bonora, P. L.. Fratesi, Roo and Roventi, G.: Corrosion behaviour of electrogalvanized steel in sodium chloride and ammonium sulphate solutions, a study by E.I.Soo J. Appl.
Eleclrochelll. 22.247-254.1992.
702. Cachet, C. De Pauli, C P, and Wiart, R.: The passivation of zinc in slightly alkaline solutions. Corms.
Sci. 25.493-502, 1985.
703. Abdel-Aal, M. S., Ahmed, Z. A., and Hassan. M. S.: Inhibiting and accelerating effects of some
quinolines on the corrosion of zinc and some binary zinc alloys in HCI solution, .J. AI)I)I. fJeClmchcm.
22,1104-1109,1992.
704. Rangel, C Moo and Cruz, L. E: Zinc dissolution in Lisbon tap water. Corros. Sci. 33.1479-1493, 1992.
705. Tsuru, Y, Takenouchi, K.. and Hosokawa. K.: Scanning micro-electrode study of zinc corrosion in
hydrochloric acid solution. Asahi Glass Foundation for Industrial Technology, Report, Vol. 57. pp.
193-199, 19901 in Japanese I.
706. Chiu, S. L.. and Selman, 1. R.: Determination of electrode kinetics by corrosion potential measurements:
Zinc corrosion by bromine . .I. Apl'l. Eleclrochem. 22, 28-37, 1992.
707. Grubitsch, H., and !ilL 0.: The hot water corrosion of zinc. Part II. Korros. Melallsclillr: 16. 197-203,
1940 1in German I.
708. Bianchi, G.: Corrosion of zinc by differential aeration, Corrosion 14, 245t-248t, 1958.
709. Gilbert, P. T.: An investigation into the corrosion of zinc and zinc-coated steel in hot waters, Sheer Mel.
Ind. 1948( October-December).
710. Lorking. K. E: The corrosion of zinc, Department of Supply, Australian Defence Scientific Service,
Metallurgy Report No. 67, Melbourne, Australia, May 1967.
711. Pryor. M. J., and Keir, D. S.: Galvanic con-osion,.I. Eleclrochem. Soc. 104.269-275, 1957.
712. Shatfer, R. 1., and Belisle, S.: An assessment of the performance of coated sheet steels in vehicle and
atmospheric environments, NatIOnal Association of Corrosion Engineers Corrosion '91 Conference,
Cincinnati. Ohio, March 11-15. 1991. Paper No. 411.
713. Subramanian, G .. Ananth, v., Chandrasekaran, P, Pairaj, S., and Guruviah, S.: Con-os ion and surface
modifications of zinc at mandapam, Bull. Eleclrochem. 6, 494-496, 1990.
714. Arimura, M .. I wai, M., Sakai, H., and Nishimoto. H.: Field test of Zn-Ni alloy electroplated steel sheet
formed into automotive door in Okinawa for 5 years, National Association of Corrosion Engineers
Con'osion '91 Conference, Cincinnati, Ohio. March 11-15. 1991. Paper No. 370.
715. Fedrizzi, L., and Bonora, P. L.: Effects of mechanical deformation on electrochemical behaviour of
galvanised steel, Br. Corms . .I. 28, 37-42. 1993.
716. Zheng. Y, Wang. Z., and Zhang, E: Influence of acid deposition on atmospheric corrosion of linc,
Proceedings of the 7th APCCC Conference on Corrosion Control, Beijing China. 1991, Vol. I. pp.
487-490.

438

REFERENCES

717. Nagy, Z., and Wesson. 1. M.: Error analysis of the polarization-resistance technique for corrosion-rate
measurements.J. Electrochem. Soc. 139. 1261-1266. 1992.
718. Mansfeld. E: Polarization resistance measurements-today's status. Electrochemical Techniquesfor
Corrosion Engineering. pp. 67-71, National Association of Corrosion Engineers. Houston. 1986.
719. Silverman, D.
Primer on the AC impedance technique. in Electrochemical Techniquesfor Corrosion
Engineering. pp. 73-79. National Association of Corrosion Engineers. Houston. 1986.
720. Nielsen, K.. and Yding. E: Influence of pipe quality on corrosion of galvanized steel pipes for domestic
water supply, Werkst. Karras. 34, 547-556,1983.
721. Abayarathna. D.: Corrosion and polarization characteristics of single crystal and polycrystalline zinc
surfaces. Ph.D. Thesis. University of Missouri-Rolla, 1989.
722. Roberts, C. W.: An investigation into the causes of intercrystalline corrosion in zinc-aluminium alloys.
Metallurgia. 1961(9).57-66.
723. von Riecke. E.: Investigation on the influence of zinc on the corrosion behavior of high strength steels.
Werkst. Korros. 30, 619-630.1979.
724. Chen. C. L, Lee. P. Y. Wu. 1. K.. Chiou. D. 1.. Chu. C. y, and Lin, 1. Y: The use of zinc and tin coatings
and chemical additives for preventing hydrogen embrittlement in steel. Carras. Prevo Control 40(3),
71-74.1993.
725. Biber, H. E.: Scanning auger microprobe study of hot-dipped regular-spangle galvanized steel. Part I.
Surface composition of as-produced sheet. Metall. Trans. 19A. 1603-1608, 1988.
726. Ross, T. K., and Wolstenholme. 1.: Anti-corrosion properties of zinc dust paints. Carras. Sci. 17,
341-351.1977.
727. Johnson. B. V. and Ross, T. K.: The protection of mild steel by zinc-rich paint in flowing aerated 0.5
M NaCI solution. Part II. The effect of exposed area ratio, Corros. Sci. 18.511-518. 1978.
728. Chua, H. H., Johnson, B. V., and Ross, T. K.: The protection of mild steel by zinc-rich paint in flowing
aerated 0.5 M NaCI solutions. Part I. The effect of zinc particle size. Carras. Sci. 18, 505-510, 1978.
729. Chua, H. H., and Ross, T. K.: The protection of mild steel by zinc-rich paint in flowing aerated 0.5 M
NaCI solution. Part IV. The effect of zinc oxide addition, Corros. Sci. 18, 523-525, 1978.
730. Yakimenko, G. Ya .. Andryushchenko. E K., Kharchenko. E. P.. and Valenya, E. P.: Colored molybdate
films on zinc and their protective properties. Zashch. Met. 13. 123-125, 1977.
731. Pedram, R., and Ross, T. K.: The protection of mild steel by zinc-rich paint in flowing aerated 0.5 M
NaCI solution. Part III. The effect of zinc content, Carras. Sci. 18, 519-522, 1978.
732. Vannerberg, N.-G., and Sydberger, T.: Reaction between S02 and wet metal surfaces, Corros. Sci. 10,
43-49, 1970.
733. Strekalov, P. v.. Agafonov, V. V., and Mikhailovskii. Yu. N.: Effect of temperature on the adsorption of
moisture and the corrosion rate of zinc under atmospheric conditions. Zashch. Met. 8(5).577-580. 1972.
734. Miki. K.. Shimogori, K., Satoh, H., Ikeda, K., M, and Terada.: Corrosion behaviour of cold rolled and
zinc alloy plated steel sheets in various kinds of corrosion tests, Tetsu To Hagane 72, 1090-1097. 1986.
735. Fukuda, Y., Tsuchiya. Y. Terasaka. M., Nakaoka. K., and Hara. T.: Analysis of corrosion layers on
electrodeposited Zn-Ni (I3wt%) alloys using AES, SAM, XPS, SEM and XRD, Tetsu Ta Hagane 72,
1782-1789,1986.
736. Mattei, I. V., Martorano, R.. and Johnson. E. A.: Formulating stable latex paints with zinc oxide, Jo.
Coatings Techno!. 63. 39-45. 1991.
737. Kenworthy. L.: The problem of copper and galvanized iron in the same water system. J. Inst. Met. 69,
67-90. 1943.
738. Tanabe. H .. Shinohara. T. Hoshino, M., Itami, K.. and Sato, Y.: Effect of dissolved oxygen on under-film
corrosion. Boshoku Gijutsu 30. 622-626. 1981.
739. Westberg, J., and Borjesson, L-G.: Influence of sheet metal surface conditions on the corrosion
properties of automotive paint systems, National Association of Corrosion Engineers Corrosion '80,
Chicago, March 3-7, 1980, Paper No. 278.
740. Jordan, D. L.: Galvanic interactions between corrosion products and their bare metal precursors: A
contribution to the theory of underfilm corrosion, Proceedings of the Symposium on Advances in
Corrosion Protection by Organic Coatings, 1989, pp. 30-43.
741. Kurokawa, S., Ban, T., Yamato, K., and Ichida, T.: A study on cosmetic and perforation corrosion test
for automotive steel sheets, J. Iron Steellnst. Jpn. 72, 1111-1118, 1986.

c.:

REFERENCES

439

742. Shibuya, A .. Kurimoto, T.. Korekawa, K., and NojI, K.: Corrosion-resistance of electroplated Ni-Zn
alloy steel sheet, Tetsu To Hagane 66. 771-778, 1980.
743. Tanabe, H .. Shinohara. T.. Sato. Y. Nii. H .. Hoshino, M., and Itami, K.: Effect of zinc rich paints on the
durability of epoxy top coats in sodium chloride solution. Boshoku Gijutsu 30, 390-395, 1981.
744. Ramasamy, N .. Sarangapani, S., Nityanandan, J. P, Venkatesan, S .. and Aravamuthan. V: Gas evolution
studies with powder compact Zn anodes, Corros. Sci. 11, 873-884, 1971.
745. Moreno, E .. and Sagues, A. A.: Performance of plain and galvanized reinforcing steel during the initial
stage of corrosion in concrete with pozzolanic addition, National Association of Corrosion Engineers.
Corrosion '96, Paper No. 326, 1996.
746. Vernon, W. H. J.: Second experimental report to the Atmospheric Corrosion Research Committee
(British Non-FelTous Metals Research Association), TrailS. Faraday Soc. 23, 113-204, 1927.
747. Vernon, W. H. J.: A laboratory study of the atmospheric corrosion of metals. Part II. Iron: The primary
oxide film. Part III. The secondary product or rust (influence of sulphur dioxide, carbon dioxide and
suspended particles on the rusting of iron), Trans. Faradav Soc. 31, 1668-1700, 1935.
748. Vernon, W. H. J.: A laboratory study of the atmospheric corrosion of metals. Part I. The corrosion of
copper in certain synthetic atmospheres, with particular reference to the influence of sulphur dioxide in
air of various relative humidities, Trans. Faraday Soc. 27,255-277, 1931.
749. Heitz, E., Hukovic, M., and Maier. K. H.: Basic process of corrosion in organic solvents II, Werkst.
Korros. 21. 457-462,1970 lin Germani
750. Fujita, S., Shimizu, Y, and Matsushima, I., Corrosion resistance of painted zinc plated steel in sodium
solution, Tetsu To Hagane 72, 1897-1902, 1986.
751. Wagner, C: Contribution to the theory of cathodic protection, l. Electrochem. Soc. 99, 1-12, 1952.
752. Myers, J. R., and Obrecht. M. E: Potable water systems-recognition of cause vital in minimizing
cotTosion, Mater. Prot. Perfimn. 11(4),41-46, 1972.
753. Fedot'ev, N. P.. Grilikhes, S. Ya., and Orekhova, E. E: Effect of light on the corrosion of zinc and
cadmium, Zh. Prikl. Khilll. 32,1428-1430,1959.
754. Miyoshi, Y: Evaluation technology of corrosion behavior for automotive steel sheet, ISIJ Int. 31(2),
122-133,199L
755. LaQue. E L., and Boylan, 1. A.: Effect of composition of steel on the performance of organic coatings
in atmospheric exposure, Corrosion 9(8), 237-241,1953.
756. Ikeda, K., and Satoh, H.: An analysis of corrosion cracking phenomena of electrode posited Zn-Ni alloy
layers, J.lron Steel Inst. lpn. 77,1162-1168,1991.
757. Arsalane, M .. and Tosser, A.: Interband photoconduction edge in zinc sulphide thin films, NOliI'. Rev.
Opt. 7(5), 323-327, 1976 lin French].
758. Suzuki. N .. Bando, S .. and Sugisawa. S.: Phosphotability of Zn-Ni alloy electroplated steel sheets, l.
lroll Steellnsf. lpn. 77, 1058-1065, 1991.
759. Margulies. P. H.: Peroxygen compounds hold important place in treating metal surfaces, lrollAge 175(4),
71-74,1955.
760. Mukherjee, D., Venkataraman, B., Palaniswamy, N., Natesan, M., and Balakrishnan, K.: COlTosion and
oxidation resistance of zinc composite materials, The International Congress on Metallic Corrosion,
Madras, India, 1987, Vol. 4, Sessions 14-19, pp. 3689-3694.
761. Fernandez, A., del Carmen Leiro, M., Rosales, B. M., Ayllon, E. S., Varela, E E., Gervasi, C A .. and
Vikhe, J. R.: Techniques applied to the analysis of the atmospheric corrosion of low carbon steel, zinc,
copper and aluminum, in Proceedings of the 12th International Corrosion Congress on Corrosion
Control for Low-Cost Reliability, September 19-24, 1993, Vol. 2, pp. 574-589, 1993. National
Association of Corrosion Engineers, Houston.
762. Wakano, S., Shibuya, A., Hobo, Y, and Nishihara, M.: Corrosion performance of Zn alloy precoated
steels for automotive body, Trans. Iron Steel In.51. lpn. 23, 967-973, 1983.
763. Cachet, C, Stroder, U., and Wi art, R.: The kinetics of zinc electrode in alkaline zincate electrolytes,
Electrochim. Acta 27, 903-908, 1982.
764. Shibuya, A., Kurimoto, T., Hoboh, Y. and Usuki, N.: Development of Ni-Zn alloy plated steel sheet,
Trans. lroll Steellnst. ll'". 23,923-929, 1983.
765. Svensson, 1.-E., and Johansson, L.-G.: A laboratory study of the initial stages of the atmospheric
corrosion of zinc in the presence of NaCI, influence of S02 and N0 2 , Corros. Sci. 34,721-740, 1993.

440

REFERENCES

766. Miura, N., Saito, T., Kanamaru, T., Shindo, Y., and Kitazawa, Y.: Development of new corrosion-resistant
steel sheets for automobiles, Trans. Iron Steel Inst. lpn. 23,913-922, 1983.
767. Standish, J. v., Whelan, G. w., and Roberts, T. R: The corrosion behavior of galvanized and cold rolled
steels, SAE Technical Paper No. 831810, Society of Automotive Engineers International Congress and
Exposition, Detroit, February 28-March 4, 1983.
768. Yamashita, M., Kubota, T., and Adaniya, T.: Organic-silicate composite coated steel sheet for automobile
body panel, SAE Technical Paper No. 8620 17. Society of Automotive Engineers International Congress
and Exposition, Detroit, February 24-28, 1986.
769. Lambert, M. R, and Hart, R G.: Corrosion resistance of 0-15% Ni-Zn alloy electroplated coatings,
SAE Technical Paper No. 860266, Society of Automotive Engineers International Congress and
Exposition, Detroit, February 24-28, 1986.
770. Tsugawa, S.-I., Mohri, T., Kobayashi, S., and Ichida, T.: Newly-developed organic composite-coated
steel sheet with high perforation resistance, SAE Technical Paper No. 850004, Society of Automotive
Engineers International Congress and Exposition, Detroit, February 25-March I, 1985.
771. Watanabe, T., Ohmura, M., Honma, T., and Adaniya, T.: Iron-zinc alloy electroplated steel for
automotive body panels, SAE Technical Paper No. 820424, Society of Automotive Engineers International Congress and Exposition, Detroit, February 22-26, 1982.
772. Miyoshi, Y., Kitayama, M., Nishimura, K, and Naito, S.: Cosmetic corrosion mechanism of zinc and
zinc alloy coated steel sheet for automobiles, SAE Technical Paper No. 850007, Society of Automotive
Engineers International Congress and Exposition, Detroit, February 25-March I, 1985.
773. Bassi, G., and Hugo, J. P.: Etch pits in zinc, 1. Inst. Met. 87, 376-379, 1958-1959.
774. Obrecht, M. E: Potable water systems in buildings: Deposit and corrosion problems, Heat.lPipinglAir
Condo 1973(May), 77-83.
775. Khairy, E. M., and Khater, M. M.: Contribution to the electrochemical behaviour of zinc electrode in
pyridine media, Corros. Sci. 12, 121-132, 1972.
776. Shams EI Din, A. M., and Khedr, M. G. A.: Role of anions in the corrosion and corrosion inhibition of
Zn in aqueous solutions. Part II. Behaviour in deaerated solutions and the determination of critical
currents for passivation, Corros. Sci. 12, 393-404, 1972.
777. Grauer, R, and Kaesche, H.: Electronmicrospic and electrochemical investigations on the passivation
of zinc in O.IM NaOH, Corros. Sci. 12,617-624, 1972.
778. Myrdal, R.: Phenomena that disturb the measurement of potentials in concrete, National Association of
Corrosion Engineers, Corrosion '96, Paper No. 339, 1996.
779. Williams, L. E G.: The initiation of corrosion of chromated zinc electroplate on steel, Corros. Sci. 13,
865-868, 1973.
780. Theiler, E: The rust preventing mechanism of zinc dust paints, Corros. Sci. 14,405-414,1974.
781. Dirkse, T. P.: The solubility product constant ofZnO, 1. Electrochem. Soc. 133, 1656-1657, 1986.
782. Caswell, P., Hampson, N. A., and Larkin, D.: The differential capacitance of zinc in aqueous solution-a
further note, 1. Electroanal. Chem. 20, 335-338,1969.
783. Cescon, P., Marassi, R., Bartocci, v., and Fiorani, M.: Standard electrode potentials of silver, cadmium,
cobalt, indium, titanium, zinc in molten alkali thiocyanates, 1. Electroanal. Chem. 23, 255-259, 1969.
784. Awad, S. A., and Kamel, Kh. M.: Mechanism of corrosion-inhibition and corrosion-promotion of zinc
by phosphate ions. 1. Electroanal. Chem. 24,217-225,1970.
785. Hampson. N. A., Herdman, G. A., and Taylor, R: Some kinetic and thermodynamic studies of the system
Zn, OH, 1. Electroanal. Chem. 25. 9-18, 1970.
786. Armstrong, R D., and Bell, M. E: The active dissolution ofzinc in alkaline solution, Electroanal. Chem.
Interfacial Electrochem. 55, 201-211,1974.
787. Drury, J. S., Hampson, N. A., and Marshall, A.: The anodic behaviour of zinc in silicate-containing
alkaline solutions, Electroanal. Chem. 1I1Ierfaciai Electrochem. 50, 292-294, 1974.
788. Bonasewicz, P.. Hirschwald, W., and Neumann, G.: Intluence of surface processes on electrical
photochemical and thermodynamical properties of zinc oxide films, 1. Electrochem. Soc. 133, 22702278, 1986.
789. Hurien, T., and Fischer, K P.: Kinetics of zinc reactions in acidified solutions of potassium chloride,
Electroanal. Chem. Interfacial Electrochem. 61, 165-173, 1975.
790. Muralidharan, V. S., and Rajagopalan, K S.: Kinetics and mechanism of corrosion of zinc in sodium
hydroxide solutions by steady-state and transient methods, 1. Electroanal. Chem. 94, 21-36,1978.

REFERENCES

441

791. Lemasson. P., and Gautron. J.: On the anodic decomposition of the zinc selenide electrode. l.
Electmanal. Chern. 119,289-299. 1981.
792. Masuda. H., Morishita, S . Fujishima. A., and Honda, K.: Study of photoelectrochemical reactions at a
n-type polycrystalline ZnO electrode using photoacoustic detection, 1. Electmanal. Chern. 121. 363368. 1981.
793. Gerischer, H.: The role of semiconductor structure and surface properties in photoelectrochemical
processes. l. Electmanal. Chern. 150,553-569. 1983.
794. Cabot, P. L.. Cortes, M .. Centellas. F. A., Garrido. 1. A., and Perez, E.: Potentiodynamic passivation of
zinc in aqueous KOH solutions, l. Electmanal. Chern. 201. 85-100,1986.
795. Hamnett. A . and Mortimer. R. 1.: The electrochemical oxidation of zinc studied by ellipsometry. 1.
Electroanal. Chern. 234.185-192.1987.
796. Novak. M .. and Szucs.: Study of electropolishing of a Zn anode in acidic medium. l. Electroanal. Chern.
266. 157-172. 1989.
797. Wroblowa. H. S .. and Qaderi. S. 8.: The mechanism of oxygen reduction on zinc. l. Electroanal. Chern.
295.153-161.1990.
798. Kouyate, D .. Ronfard-Haret. J.-c.. and Kossanyi. 1.: Electroluminescent properties of polycrystalline
zinc oxide electrodes doped with erbium ions, 1. Electroanal. Chern. 319, 145-160, 1991.
799. Cachet. c.. and Wiart, R: The kinetics of zinc dissolution in chloride electrolytes: Impedance measurements and electrode morphology. l. Electroanal. Chern. 111,235-246. 1980.
800. De Pauli. C .. Derosa, O. A. H.. and Giordano, M. C.: Effect of phosphate ions on zinc dissolution in
alkaline solutions, l. Electroanal. Chern. 73. 105-108, 1976.
801. Elsdale. R N., Hampson. N. A., Jones, P. c.. and Strachan, A. N.: The anodic behaviour of porous zinc
electrodes, l. Appl. Electrochern. 1, 213-217,1971.
802. Huber, K.: Anodic formation of coatings on magnesium, zinc, and cadmium, 1. Electrochern. Soc. 100.
376-382.1953.
803. Bruno. R, and Memmi, M.: The development of new galvanized coatings for use in aggressive
environments, Proceedings of the II th International Galvanizing Conference. INTERGALVA '76. Madrid, pp. 213-221.
804. Steinsmo. U .. and Bardal, E.: Use of DC electrochemical methods for evaluation of paint films on steel,
aluminum and zinc, Corrosion 48, 910-917, 1992.
80S. Armas, R A., Gervasi, C. A., Di Sarli. A . Real, S. G .. and Vilche. J. R.: Zinc-rich paints on steels in
artificial seawater by electrochemical impedance spectroscopy, Corrosion 48. 379-383, 1992.
806. Eger, D., Many, A.. and Goldstein. Y.: Very strong accumulation layers on ZnO surfaces, Phys. Lett.
55A(3). 197-198, 1975.
807. Cunningham, J., and Zainal, H.: Reactions involving electron transfer at semiconductor surfaces. Part
IV. Zinc oxide-promoted photoreductions in aqueous solutions at neutral pH. l. Phys. Chern. 76.
2362-2374, 1972.
808. Edwards. 1. D.: Anodic and surface conversion coatings on metals, Trans. Electrochern. Soc. 81/82,
341-357, 1942.
809. Ohsawa, S., and Takeda. M.: Studies of inhibition effect of crown ether on Zn-4% Al alloy. lbaraki
Daigaku Kogakubu Kenkyu Shuh. 37. 37-45, 1989.
810. Baugh, L. M., and Lee. J. A.: Differential capacitance of polycrystalline zinc. Part I. Effect of pH and
ammonium ions, Electroana/. Chern. Interfacial Electrochern. 48,55-61, 1973.
81 I. Askey. A .. Lyon, S. 8., Thompson, G. E., Johnson, J. 8.. Wood, G. c., Cooke. M . and Sage, P.: The
corrosion of iron and zinc by atmospheric hydrogen chloride. Corros. Sci. 34, 233-247, 1993.
812. Giridhar. J., and van Ooij, W. J.: Study of Zn-Ni and Zn-Co alloy coatings electrodeposited on steel
strips. Part II. Corrosion, dezincification and sulfidation of the alloy coatings, Surf. Coatings Techno/.
53.35-47, 1992.
813. Fragata. F. L., and Mussoi. C. R S.: Influence of extender pigments on the performance of ethyl silicate
zinc-rich paints, l. Coatings Technol. 65, 103-109, 1993.
814. Miyoshi. Y.. Oka, 1., and Maeda, S.: Fundamental research on corrosion resistance of precoated steel
sheets for automobiles, Trans. Iron Steellnst. lpn. 23, 974-983. 1983.
815. Shastry. C. R., and Townsend, H. E.: Effect of pretreatment on performance of zinc and zinc alloy
electroplated sheet in a cyclic corrosion test. SAE Technical Paper No. 892556. Society of Automotive

442

816.

817.

818.
819.
820.
821.
822.
823.

824.
825.
826.

827.

828.

829.
830.
831.
832.

REFERENCES
Engineers 1989 Automotive Corrosion & Prevention Conference & Exposition, Dearborn, Michigan,
December 4-6, 1989.
May, T. P., and Alexander, A. L.: Spray testing with natural and synthetic sea water. Part I. Corrosion
characteristics in the testing of metals, American Society for Testi ng and Materials 53rd Annual Meeting,
June 26-30,1950, pp. 1131-1141.
Cunningham, J., and Corkery, S.: Reactions involving electron transfer at semiconductor sUlfaces. Part
VI. Electron spin resonance studies on dark and illuminated aqueous suspensions of zinc oxides, 1. Phys.
Chem. 79,933-941,1975.
Shaw, B. A., and Moran, P. J.: Characterization of the corrosion behavior of zinc-Aluminum thermal
spray coatings, Corrosion '85, Boston, March 25-29,1985, Paper No. 212.
Anicai, L., Siteavu, M., and Grunwald, E.: Corrosion behaviour of zinc and zinc alloy depositions,
Carras. Prevo Control 1992(August), 89-93.
Townsend, H. E.: Effects of zinc coatings on the stress corrosion cracking and hydrogen embrittlement
of low-alloy steel, Metall. Trans. 6A, 877-883, 1975.
Riedl, G.: Anodes-dissolution and change of shape, Mater. Perform. 20( 12), 34-40. 1981.
Montes, E.: Influence of particle size distribution of zinc dust in water-based, inorganic, zinc-rich
coatings, 1. Coatings Techno!. 65, 79-82, 1993.
Sagiyama, M .. Kawabe, M., and Watanabe, T.: Effects of electrolysis and bath conditions on the surface
roughness. morphology and crystal orientation of zinc electrodeposit. Tetsu To Hagane 76. 1301-1308,
1990.
Petrovich, I.. and Galvan. J. c.: The application of electrochemical techniques to the study of duplex
systems. Galvanotecnica 36(2).29-34, 1985.
Sehgal, A.. Li, D . Kho, Y. T. Osseo-Asare, K, and Pickering, H. w.: Reproducibility of polarization
resistance measurements in steel-in-concrete systems. Corrosion 48. 706-714. 1992.
Edney, E. 0., and Stiles. D. c.: Laboratory investigations of the impact of dry deposition of S02 and
wet deposition of acidic species on the atmospheric corrosion of galvanized steel, Atmos. Environ. 20,
541-548, 1986.
Kurokawa, S .. Yamato. K, and Ichida, T.: A study on cosmetic and perforation corrosion test procedures
for automotive steel sheets, National Association of Corrosion Engineers Corrosion '91 Conference,
Cincinnati, Ohio, March 11-15, 1991, Paper No. 396.
Koda, M., Uchida, y', Suzuki, M., Deguchi, T, and Hirose, Y.: Inhibition mechanism of black patina
development on a hot dip Zn-AI alloy coated steel sheet by spraying aqueous solution of Co salt, Tetsu
To Hagane 76,383-390, 1990.
Grunberg, L.: The formation of hydrogen peroxide on fresh metal surfaces, Proc. Phys. Soc. 866,
153-161,1953.
Kaneko, F., and Nagasaka, H.: Evaluation of corrosion resistance of metal sprayed and painted steel,
Surf. Eng. 6, 125-128, 1990.
Chance, D. A., and Nobe, K: Polarization and photopotentials of zinc single crystals in Na2S, Corrosion
28, 170-171, 1972.
Satoh, H., Shimogori, K., Nishimoto, H .. Miki, K, Ikeda, K, Iwai, M., Sakai. H., and Nomura, S.:
Analysis of perforation corrosion of cold rolled and galvanized steel sheets by extreme-value statistics,

Tetsu To Hagane 72, 1098-1105, 1986.


833. Danilov, F. I., Sukhomlin. D. A., Gerasimov. V. v., and Popovich, V. A.: Electrodeposition of zincmanganese alloys, Sov. Electrochem. 28, 217-220. 1992.
834. Zhdanov, B. B., and Sukharev, N. P.: Corrosion-electrochemical behavior of alloys having a eutectic
structure: Zn-Cd, Zn-Pb and Cd-Pb, Zh. Prikl. Khim. 62. 2361-2363, 1989.
835. Hope, G. A . and Ritchie. I. M.: The effect of gold on the oxidation of zinc in moist oxygen. Corros.
Sci. 16. 177-181, 1976.
836. Dorfman, M . Vargas, J., Clayton, C. R .. and Herman, H.: Cavitation behavior of active thermal-sprayed
coatings. Thin Solid Films 64, 351-357, 1979.
837. Hanneman. R. E., and Aust. K T: Solute clustering and intergranular corrosion, Metallurgica 2,
235-238, 1968.
838. Hayashi, K, Ito, Y., Kato, C, and Miyoshi. Y.: Under-film corrosion behavior of zinc and zinc alloy
coated steel sheets for automobiles, Tetsu To Hagane 76. 1317-1324, 1990.

REFERENCES

443

839. Johnson. W. C: French process zinc oxide in primers for hand-cleaned structural steel, J. Coatings
Technol. 57,67-75,1985.
840. Krishnan. R. M., and Muralidharan. V. S.: Electrochemical and salt spray testing of zinc electrodeposits,
Br. Corros. l. 27. 317-318.1992.
841. Takazawa. H .. Saito, H .. and Nakai, H.: Corrosion resistance of hot-dipped Zn-AI alloy coatings in
coastal areas. Zaino To Kankyo 40(7), 470-477.1991.
842. Lechner-Knoblauch. U., and Heitz, E.: Corrosion of zinc, copper and iron in contaminated non-aqueous
alcohols. Electrochirn. Acla 32. 901-907,1987.
843. Cachet. C, De Pauli, C P. and Wiart. R.: The pore texture of zinc electrodes corroded in acidic
electrolytes, Electrochim. Acta 30. 719-723. 1985.
844. Garrels. R. M.: Mineral species as functions of pH and oxidation-reduction potentials. with special
reference to the zone of oxidation and secondary enrichment of sulphide ore deposits. Geochim.
Cosrnochim. Acta 5, 153-168, 1954.
845. Kuffel, K.: Comparative testing of AI and Zn hot-dip coatings in a fertilizer environment. Ochr. Korol.
29(5).116-117.1986 [in Polishl.
846. Hayashi. K .. Ito. Y. Kato. C. and Miyoshi. Y: Electrochemical study on under-film corrosion mechanism ofZn and Zn-Fe alloy coatings, J.lron Steeilnsl. lpn77. 1122-1129, 1991.
847. Fruhwirth, 0 . Herzog, G. W., and Poulios, 1.: ZnO photoeffects under galvanostatic and potentiostatic
conditions. Suif. Technol. 11.259-267. 1980.
848. Lin. C, Wen. L., and Qian. S.: Corrosion failure analysis of flame sprayed Zn-A115 coating in sea water,
Acla Metal!. Sin. 26(2), B 110-BI13. 1990.
849. Cachet. C. and Wiart. R.: Reaction mechanism for zinc dissolution in chloride electrolytes, l.

Electroallal. Chem. 129, 103-114. 1981.


850. Ito. Y. Hayashi. K., and Miyoshi, Y: The influence of drying process in cyclic corrosion test on the
underfilm corrosion of galvanized steel sheets. J. Iron Steeilnsl. lpn. 77. 1138-1145, 1991.
851. De Pauli, C P. Derosa, O. A. H., and Giordano. M. C: Zinc dissolution and passivation in buffered
phosphate solutions. Part I. A comparative study with sodium hydroxide solutions. l. Electmwwl. Chem.
86, 335-348, 1978.
852. EI-Sharif. M. R .. Suo Y 1., Watson, A., and Chisholm, CU.: Studies of chemical conversion treatments
of electrodeposited zinc-chromium and zinc-nickel-chromium alloys, in Proceedings of the 121h
International Corrosion Congress on Corrosion Control for Low-Cost Reliabilitv, Septemher 19-24,
1993, Vol. I, pp. 329-336, National Association of Corrosion Engineers, Houston, 1993.
853. Sarmaitis, R., andJuzeliunas, E.: Protective action of chromate conversion coatings, Plaling Surf: Finish.
78(1). 72-76, 1991.
854. Abdel Haleem, S. M., and Abdel Fattah, A. A.: The role of some organic anions in promoting or inhibiting
the corrosion of zinc, Suif. Coatings Technol. 29,41-49, 1986.
855. Fratesi, R., and Roventi, G.: Electrodeposition of zinc alloys in chloride baths containing cobalt ions.
Mater. Chern. Phys. 23, 529-540, 1989.
856. Arshadi, M. A., Johnson, J. B., and Wood, G. C: The influence of an isobutane-S02 pollutant system
on the earlier stages of the atmospheric corrosion of metals, Corros. Sci. 23, 763-776, 1983.
857. Sakurai, S., and Naruse, T.: Conversion treatment on zinc using titanium chelate, l. Met. Finish. Soc.
lpn. 37, 674-676,1986.
858. Ohsawa. S., Takeda, M., and Yamamura, T.: Study on corrosion inhibition of zinc-aluminum alloy by
organic compounds. J. Met. Finish. Soc. lpn. 39, 310-314. 1988.
859. Nikolova. M .. Raichevski. G .. Banova, R.. Rashkov, S. T., and Klaus. M.: Corrosion hehaviour of
electrodeposited zinc-cobalt alloy coatings, Bull. Electrochern. 5, 314-318, 1989.
860. Addison, W. E.: Slruclural Principles in Inorganic Compounds, John Wiley & Sons, New York. \953,
p.57.
861. Almeida. E., Pereira, D .. and Figueiredo, 0.: The degradation of zinc coatings in salty atmospheres,
Prog. Org. Coat. 17, 175-189, 1989.
862. Fukuda, Y, and Fukushima, T.: Electrodcposition of nickel and zinc into the pores of anodic oxide film
on aluminum, Trans. NaIl. Res.lnsl. Mel. 26,315-321,1984.
863. Hauffe, K., and Range, J.: On the anodic dissolution of zinc-oxide single crystals in the presence of
light, Phys. Chern. 71, 690-697, 1967.

444

REFERENCES

864. Sydberger, T, and Vannerberg, N.-G.: The influence of the relative humidity and corrosion products on
the adsorption of sulfur dioxide on metal surfaces, Corros. Sci. 12, 775-784, 1972.
865. Zeller, R. L., and Savinell, R. E: Interpretation of a.c. impedance response of chromated electrogalvanized steel, Corros. Sci. 26, 389-399,1986.
866. Wilcox, G. D., Gabe, D. R., and Warwick, M. E.: The development of passivation coatings by cathodic
reduction in sodium molybdate solutions, Corros. Sci. 28, 577-587,1988.
867. Felloni, L., Fratesi, R., Roventi, G., and Fedrizzi, L.: Assessment of the corrosion resistance of Zn-Ni
electrodeposits, Proceedings of the 11th International Corrosion Congress, Florence, Italy, April 2-6,
1990, Vol. 2, pp. 365-372.
868. Biestek, T, Drys, M., Sokolov, N., Knotkova, D., Ramishvili, R., Kozhukharov, v., and Zeldel, M.:
Atmospheric corrosion of metallic systems. Part V. Identification of the chemical compounds in the
corrosion products of zinc, Zashch. Met. 19(5), 750-753, 1983.
869. Hayashi, K, Ito, Y, Kato, c., and Miyoshi, Y: Electrochemical study on under-film corrosion mechanism of Zn and Zn-Fe alloy coatings, Tetsu To Hagane 77, 1122-1129, 1991.
870. Calvert, J. G., Lazrus, A., Kok, G. L., Heikes, B. G., Walega, J. Goo Lind, 1., and Cantrell, C. A: Chemical
mechanisms of acid generation in the troposphere, Nature (London) 317, 27-35, 1985.
871. Alimov, V. I., Berezin, A. v., Marchuk, S. I., and Lavrent'eva, G. A.: Corrosional resistance of zinc
coating on steel after annealing, fzv. Akad. Nauk SSSR Met. 1988(3), pp. 160-163.
872. Ohsawa, S., Takeda, M., and Yoshinuma, K: Studies on corrosion inhibition of zinc single crystals by
organic inhibitors, fbaraki Daigaku Kogakubu Kenkyu Shuho, 36(12),141-148, 1988.
873. Aoki, T, Sumiya, 1., Uchida, y, and Hirose, Y: Hydrogen Permeation Behavior in the Coating Layers
ofZn-AI Alloy Coated Sheets, Nissin Steel Technical Report, Vol. 52, June 10, 1985.
874. Beranek, E., Barton, K, Sekerka, I., and Smrcek, K: Reactions in aqueous suspensions of zinc oxide.
Part II. Reaction mechanism for the formation of hydrogen peroxide and its influence on the photoelectric properties of an iron electrode kept in the zinc oxide suspension, Collect. Czech. Chem. Commun.
25,369-374,1960.
875. Rausch. W.: Chemical surface treatment of steel coated with zinc and zinc alloys, Proceedings of the
International Conference on Zinc and Zinc Alloy Coated Steel Sheet. GALVATECH '89. Tokyo. September 5-7, 1989. pp. 197-205.
876. Johnsson. T, and Nordhag. L.: Corrosion resistance of coatings of aluminium. zinc and their alloys:
Results offouryears' exposure, United Nations Seminar, Steel Committee, Economic and Technological
Aspects of the Protection of Steel Against Corrosion, May 7-11, 1984. Geneva.
877. Akerdida, M. T, and Lyon, S. B.: Comparative corrosion resistance of galvanized and galvannealed
steel, in Surface Engineering Practice-Processes, Fundamentals and Applications in Corrosion and
Wear. K N. Strafford, P. K Datta, and J. S. Gray (eds.), Ellis Horwood, New York, pp. 543-552. 1990.
878. Condensation in Insulated Domestic Roofs, Department of the Environment, Building Research
Establishment. Digest No. 270, Glasgow, Scotland, February 1983.
879. Common Defects in Low-Rise Traditional Housing, Department of the Environment, Building Research
Establishment, Digest No. 268, Garston, Watford, England, December 1982.
880. Notowidjojo, B. D., Kennon, N. E, and Wingrove, A L.: Hot dip galvanizing: The process, the
economics, the future, Proceedings of the Conference Step into the 90's, August 27-31, 1989,
Queensland, Australia, Vol. 2.
881. Sakoda, A, Wakano, S., and Nishihara, M.: Effect of coating weight on cosmetic corrosion performance
ofZn andZn-Alloy electroplated steel sheets in automotive body, l.fron Steelfnsf. lpn. 72, 1106-1110,
1986.
882. Bohe, A E., Vilche, J. R., Juttner, K, Lorenz, W. J., and Paatsch, w.: Investigations of the semiconductor
properties of anodically formed passive layers on Zn and of ZnO single crystals in different aqueous
electrolytes by EIS, Electrochim. Acta 34, pp. 1443-1448, 1989.
883. Stern, E: Iteration methods for calculating self-consistent fields in semiconductor inversion layers, l.
Comput. Phys. 6, 56-67, 1970.
884. Baugh, L. M., Gidney, P. M., and Lee, J. A.: Pitting ofzinc in leclanche and related electrolytes. Part II.
Mode of growth of pits and rate of propagation, Electrochim. Acta 25,765-777, 1980.
885. Baugh, L. M., Field, R. J., and Lee, 1. A.: Pitting of zinc in leclanche and related electrolytes. Part I.
Pitting characteristics on battery alloy and single crystal electrodes, Electrochim. Acta 25, 751-763,
1980.

REFERENCES

445

886. Dirkse. T P. and Hampson. N. A.: The anodic behaviour of zinc in aqueous solution. Part III. Passivation
in mixed KF~KOH solutions. Electrochim. Acta 17, 813~818, 1972.
887. Dirkse, T P.. and Hampson. N. A.: The zinc exchange reaction in KOH solution. Part II. Exchange
current density measurements using the double-impulse method, Electrochim. Acta 17, 383~386. 1972.
888. Schwabe, K.. and Luck. H.-B.: The anodic behaviour of zinc in saturated solution of zinc sulphate.
Electrochim. Acta 17. 99~IOS. 1972.
889. Dirkse, T P, and Hampson. N. A.: The anodic behaviour of zinc in aqueous KOH solution. Part I.
Passivation experiments at very high current densities, Electrochim. Acta 16. 2049~20S6. 1971.
890. Breiter. M. W.: Dissolution and passivation of vertical porous zinc electrodes in alkaline solution.
Electrochim. Acta IS, 1297-1304. 1970.
891. Krug, H., and Borchers. H.: Results of corrosion experiments with zinc alloys. Electrochim. Acta 13.
2203~2205. 1968.
892. Morrison. S. R., and Freund. T.: Chemical reactions of electrons and holes at the ZnO electrolytesolution interface. Electrochim. Acta 13, 1343-1349. 1968.
893. Bonnemay. M .. Hamelin. A.. and Sanchez de Haro. M.: Etude de la dechargc du zinc et du cadmium
sur electrodes monocristallines. Electrochim. Acta 8, S09-S19. 1963.
894. Hurlen. T.: Kinetics of metal/metal-ion electrodes: Iron, copper. zinc, Electrochilll. Acta 7. 6S3~668.
1962.
895. Shoji, T, Hishinuma. M .. and Yamamoto, T.: Zinc~manganese dioxide galvanic cell using zinc sulphate
as electrolyte: Rechargeability of the cell, J. Appl. Electrochem. 18, S21-526, 1988.
896. Randin, l-P: Determination of state-of-discharge ofzinc~silver oxide button cells. Part I. Galvanostatic
measurements, 1. Apl'l. Electrochem. IS, 293~306, 1985.
897. Thomas, B. K., and Fray. D. 1.: The conductivity of aqueous zinc chloride solutions. 1. Appl. Electmchem. 12. 1-5. 1982.
898. Popov, K. I., Pavlovic, M. G., Spasojevic, M. D., and Nakic, V. M.: The critical overpotential for zinc
dendrite formation, J. Apl'l. Electrochem. 9, S33~S36, 1979.
899. Poa, S.-P., and Wu, C. H.: The current distribution and shape change of zinc electrodes in secondary
silver-zinc cells, J. Appl. Electmchem. 8. 491~SOI. 1978.
900. Poa, S.-P.. and Wu, C. H.: A quantitative study of shape change in zinc secondary electrodes. 1. Appl.
Electrochem. 8, 427 ~436. 1978.
901. Coates, G., Hampson, N. A., Marshall, A., and Porter. D. F: The anodic behaviour of porous zinc
electrodes. Part II. The effects orspecific surface area of the zinc compact material. 1. Appl. Electrochem.
4, 7S~80, 1974.
902. Dirkse, T P, and Kroon. D. 1.: Effect of ionic strength on the passivation of zinc electrodes in KOH
solutions, 1. Appl. Electrochelll. I, 293~296. 1971.
903. Bushrod, C. J., and Hampson. N. A.: The anodic behaviour of zinc in KOH solution. Part V. Galvanostatic
polarization with an interruption. J. Appl. Electrochem. 1, 99~ 101, 1971.
904. Dirkse, T P.: Voltage decay at passivated zinc anodes, J. Appl. Electrochel1l. 1. 27~33. 1971.
90S. Schuldiner, S" and White. R. E.: Studies of time~potential chang~s on an electrode surface during
current interruption. Part I. Zinc~steel couple in synthetic sea water. J. Electrochem. Soc. 97. 433~447.
1950.
906. Delahay,P, Pourbaix, M., and Van Rysselberghe, P.: Potential-pH diagram orzinc and its applications
to the study of zinc corrosion. 1. Electrochem. Soc. 98, 101 ~ lOS, 195 I.
907. EI Wakkad. S. E. S., Shams EI Din. A. M .. and Kotb, H.: The anodic oxidation of zinc and zinc~tin
alloys at very low current density. J. Electrochem. Soc. 105, 47~51. 1958.
908. Gulbransen. E. A .. and McMillan. W. R.: Electron diffraction studies on the oxidation of pure copper
and pure zinc between 200 and 500 C, J. Electrochem. Soc. 99, 393~40 1, 1952.
909. Granese. S. L., Fernandez. A., and Rosales. B. M.: Chemical characterization of the corrosion products
formed on plain C steel, zinc, copper and aluminum, in Proceedings ()fthe 12th International Corrosion
Congress 011 Corrosion COlltrolfor Low-Cost Reliability, September 19~24, 1993, Vo\. 2. pp. 652~661,
National Association of Corrosion Engineers, Houston.

910. Fry, H., and Whitaker. M.: The anodic oxidation of zinc and a method of altering the characteristics of
the anodic films. 1. Electrochell1. Soc. 106,606-611, 1959.
911. Papazian, H. A., Flinn. P A., and Trivich, D.: Influence of impurities on the photoconductance of zinc
oxide, J. Electrochem. Soc. 104, 84~92, 1957.

REFERENCES

446

912. Gulbransen, E. A., and McMillan, W. R: Electron diffraction studies on the oxidation of pure copper
and pure zinc between 200 and 500 C, 1. Electrochem. Soc. 100, 292-293, 1953.
913. Jorgensen, P. 1.: Effect of an electric field on the oxidation of zinc, 1. Electrochem. Soc. 110,461-462,
1963.
914. Harrison, D. E., and Hummel, E A: Phase equilibria and fluorescence in the system zinc oxide-boric
oxide, 1. Electrochem. Soc. 103,491-498, 1956.
915. Rudresh, H. B., and Mayanna, S. M.: Adsorption of n-decylamine on zinc from acidc chloride solution,
1. Electrochem. Soc. 124,340-342, 1977.
916. Chuang, S.-E, Collins, S. D., and Smith, R. L.: Porous silicon microstructure as studied by transmission
electron microscopy, Appl. Phys. Lett. 55, 1540-1542, 1989.
917. Mukherjee, D., Venkataraman, B., Palaniswamy, N., and Balakrishnan, K.: Effect of alloying elements
on the corrosion resistance properties of zinc, 1. Electrochem. Soc. India 37, 233-236, 1988.
918. Kobayashi, T., Yoneyama, H., and Tamura, H.: Effect of illumination intensity on stabilization ofZnO
photoanodes in halide solutions, Chem. Lett. 1979,457-460.
919. Kohl, P. A., and Bard, A. 1.: Semiconductor electrodes. Part 13. Characterization and behavior of n-type
ZnO, CdS, and GaP electrodes in acetonitrile solutions, 1. Am. Chem. Soc. 99, 7531-7539, 1977.
920. Sakoda, A: Corrosion mechanism of zinc-iron alloy mechanical coating, Bosei Kanri 34(4), 131-134,
1990.
921. Morrison, S. R, and Freund, T.: Chemical role of holes and electrons in ZnO photocatalysis, 1. Chem.
Phys.47, 1543-1551, 1967.
922. Yamato, K., Umino, S., Yasuda, A . and Ichida, T.: Corrosion behavior of zinc alloy and finely alumina
dispersion coating, Hyomen Gijutsu 40, 162-163, 1989.
923. Diaz, A., Gonzalez, C. L., Gonzalez, S., and Arevalo, A: Activating or inhibiting action of some
compounds on zinc corrosion, Bull. Electrochem. 3, 289-294, 1987.
924. Hinton,8. R
and Wilson, L.: The corrosion inhibition of zinc with cerous chloride, Corros. Sci. 29,
967-985, 1989.
925. Shuldener, H. L., and Lehrmen, L.: Effect of increasing the area of iron to zinc on the rate of reversal
of potential in a zinc-iron system, Corrosion 14(12), 17, 1958.
926. Phibbs, M. K., and Giguere, P. A: Hydrogen peroxide and its analogues. Part III. Absorption spectrum
of hydrogen and deuterium peroxides in the near ultraviolet, C. 1. Chem. 29, 490-493, 1951.
927. Jepson, S., Meecham, S., and Salt, E W.: The electrodeposition of iron-zinc alloys, Trans. Inst. Met.
Finish. 32, 160-185, 1955.
928. Ruckert, J., Neubauer, E, and Zietelmann, c.: Influence of bitumineous roof coverings on corrosion
behaviour of roof draining systems made of zinc and galvanized steel, Werkst. Korros. 34, 355-364,
1983.
929. Obrecht, M. E: The forms of corrosion in potable water piping. Part II., Heat./Piping/Air Condo
1973(July),33-37.
930. Obrecht, M. E: The forms of corrosion in potable water piping, Heat./Piping/Air Condo 1973(June),
57-62.
931. Case, 8.: The diffusivity of oxygen in dilute alkaline solution from 0 to 65 C, Electrochim. Acta 18,
293-299, 1973.
932. Morrison, S. R: Measurement of surface state energy levels of one-equivalent adsorbates on ZnO, Surf.
Sci. 27, 586-604, 1971.
933. Dirkse, T. P., Passivation studies on the zinc electrode, Power Source 3, Proceedings of the 7th
International Symposium, Brighton, September 1970, p. 485.
934. Bech-Nielsen, G., Larsen, M. K., Jensen, K. A., and Ulrich, D.: The corrosion and dissolution valence
of a chromated Zn-Fe alloy studied by CMT and EC measurements, Corros. Sci. 36, 1907-1919, 1994.
935. Leidheiser, H.: Applications of mossbauer spectroscopy to studies of electrodeposits and the chemistry
of metal surfaces, 1. Electrochem. Soc. 135, 5-11, 1988.
936. Asami, K., Hashimoto, K, Masumoto, T., and Shimodaira, S.: ESCA study of the passive film on an
extremely corrosion-resistant amorphous iron alloy, Carras. Sci. 16,909-914, 1976.
937. Johansson, E., and Linder, M.: The influence of environmental acidification on the atmospheric
corrosion of zinc, in Proceedings of the 12th International Corrosion Congress on Corrosion Control
for Low-Cost Reliability, September 19-24, 1993, Vol. 2, pp. 549-560, National Association of
Corrosion Engineers, Houston, 1993.

w.,

REFERENCES

447

938. Persson, D .. and Leygraf, C: Metal carboxylate formation during indoor atmospheric con'osion of Cu.
Zn. and Ni, 1. Electrochern. Soc. 142, 1468-1477, 1995.
939. Gomes. W. P., and Cardon, F.: Study of noise associated with oxidation reactions at the illuminated
single crystal zinc oxide anode, 1. Solid State Chern. 3, 125-130, 1971.
940. Svensson, I.-E., and Johansson, L.-G.: Initial stages of the SOrinduced atmospheric corrosion of zinc
investigated by in-situ IR spectroscopy and time-resolved trace gas analysis. synergistic effects of N0 2
and 0 3 , in Proceedings of the 12th International Corrosion Congress on Corrosion Controlfor Low-Cost
Reliabilitv, September 19-24. 1993. Vol. 2, pp. 662-675, National Association of Corrosion Engineers.
Houston, 1993.
941. Gerischer, H.: On the stability of semiconductor electrodes against photodecomposition,./. Electmanal.

Chern. 82. 133-143. 1977.


942. Micka. K .. and Gerischer, H.: Anodic photooxidation of formic acid and methanol at a zinc oxide
electrode and the influence of anions, 1. Electroanal. Cheln. 38, 397-402, 1972.
943. Rosales, B. M., and Granese, S. L.: Establishment of the Zn corrOSIOn rate in NaHS01 aqueous solutions,

./. Electrochern. Soc. 132, 1281-1284, 1985.


944. Hada. H., Tanemura. H., and Yonezawa, Y: The photoreduction of the silver ion in a zinc oxide
suspension, Bull. Chern. Soc. lpn. 51, 3154-3160.1978.
945. Valencia. A .. Perez. R .. Arroyave, C, and Mesa, S.: Environmental effects in the atmospheric corrosion
of zinc: An immersion-drying study, in Proceedinf?s of the 12th International Corrosion Congress on
Corrosion Controlfor Low-Cost Reliability, September 19-24, 1993, Vol. 2, p. 748, National Association
of Corrosion Engineers, Houston, 1993.
946. Pettinger, B., Schoppel, H.-R. Yokoyama, T, and Gerischer, H.: Tunnelling processes at highly doped
ZnO-electrodes in aqueous electrolytes. Part II. Electron exchange with the valence band, Ber. Bunsenges. Phvs. Chern. 78, 1024-1030, 1974.
947. Graedel, T E.: Dissolution and precipitation phenomena in atmospheric corrosion, in Proceedings of
the 12th International Corrosion Congress on Corrosion Control for Low-Cost Reliabilitv, September
19-24. 1993, Vol. 2. pp. 711-721, National Association of Corrosion Engineers, Houston, 1993.
948. Sengupta, T K., and Chatterjee, S. D.: Studies on the transport properties and surface states of anodic
ZnO film on Zn, Proceedings of the 4th International Symposium on Frontiers of Electrochemistry,
November 14-16,1989, Madras, India.
949. Freund, T.: Less-than-bandgap photosensitivity of ZnO and dye sensitization by ferrocyanide ion, SUfi
Sci. 33, 295-305. 1972.
950. Freund, T, and Morrison, S. R: Mechanism of cathodic processes on the semiconductor zinc oxide,
Surf Sci. 9,119-132,1968.
951. Morrison, S. R.: Electron capture by ions at the ZnO/solution interface, Sutf. Sci. 15,363-379,1969.
952. Schwerdtfeger W. J., and McDorman, O. N.: Potential and current requirements for the cathodic
protection of steel in soils, 1. Res. Natl. Bur. Stand. 47, Research Paper RP2233, 1951.
953. Denison, I. A., and Romanoff, M.: Behavior of experimental zinc-steel couples underground, ./. Res.
Natl. Bur. Stand. 40, Research Paper RP1876, 1948.
954. Yamashita, M., Enatsu, A., Adaniya, T.. and Hara, T.: Organic-silicate composite coated Ni-Zn plated
steel sheet with high corrosion resistance for automobile bodies, 1. Iron Steellnst. lpn. 72, 1038-1043,
1986.
955. Ritter. I. J.: Ellipsometric studies on the cathodic delamination of organic coatings on iron and steel. 1.
Coatings Techno!. 54, 51-57, 1982.
956. Leidheiser, H., and Wang, W.: Some substrate and environmental influences on the cathodic delamination of organic coatings, 1. Coatings Techno!. 53, 77-84, 1981.
957. Schmetzer, K., Schnorrer-Kohler, G., and Medenbach, 0.: Wulfingite and simonkolleite, two new
minerals from Richelsdorf, Hesse, F. R G., Nelles lahrb. Mineral Monatsh. 1985(4), 145-154.
958. King, R A., Miller, I. D. A., and Stott, J. F. D.: Subsea pipelines: Internal and external biological
corrosion, The National Association of Corrosion Engineers International Corrosion Conference,
Houston, 1986, pp. 268-274.
959. Chan, L., and Griffin, G. L.: Temperature programmed desorption studies of hydrogen on Zn(OOOI)
surfaces, Sutf. Sci. 145, 185-196, 1984.

448

REFERENCES

960. Hudson, J. C.: Atmospheric corrosion of metals. Third (experimental) report to the Atmospheric
Corrosion Research Committee (British Non-Frrous Metals Research Association), Trans. Faraday Soc.
25, 177-252, 1929.
961. Sumita, M., Maruyama, N., and Uchiyama, I.: Fatigue crack growth rates and their thresholds of high
strength steels in sea water at the zinc potential, Tetsu To Hagane 69( 11), 113-120, 1983.
962. Zinc-Coated Reinforcement for Concrete, Department of the Environment, Building Research Establishment, Digest No. 109, Garston, Watford, England, September 1969.
963. Bird. C. E.: The influence of minor constituents in portland cement on the behaviour of galvanized steel
in concrete, Corros. Prevo Control 1964(July), 17-21.
964. Miyoshi, Y., Oie, Y., Amano, M., and Koyahara, H.: Corrosion behavior of electrophoretically coated
cold rolled, galvanized and galvannealed steel sheet for automobiles-adaptability of cataphoretic
primer to zinc plated steel, SAE Technical Paper No. 820334, Society of Automotive Engineers
International Congress and Exposition, Detroit, February 22-26, 1982.
965. Lindqvist, S. A., Meszaros, L., and Svenson, L.: Aspects of galvanic action of zinc-rich paints.
Electrochemical investigation of eight commercial primers,J. Oil Colour Chern. Assoc. 1985( I). 10-14.
966. Costa, 1. M., and Vilarrasa, M.: Effect of air pollution on atmospheric corrosion of zinc, Br. Carras. J.
28,117-120,1993.
967. Ahmed, D. S.: The intergranular attack in zinc alloy sacrificial anodes at elevated temperatures, Ph. D.
Thesis, University of Manchester, 1984.
968. Justo, M. J., and Ferreira, M. G. S.: The corrosion of zinc in simulated SOrcontaining indoor
atmospheres, Corros. Sci. 34, 533-545, 1993.
969. Wenz, R. P., and Claus, J. J.: Influence of abrasive cleaning on the corrosion resistance of coated steel,
National Association of Corrosion Engineers Corrosion '81 Conference, Toronto, Ontario, April 6-10,
1981, Paper No. 82.
970. Ito, K., Oohashi, H., and Yoshii, T.: Corrosion behavior of zinc plated stainless steel, Proceedings of the
International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7,1989, pp. 569-576.
971. Leclercq, M.: The electrochemical determination of the proportions of metal and oxide in zinc-base
coatings obtained by hot spraying, Anti-Corrosion 1978(September), 5-11.
972. Oohashi, H., and Adachi, T.: Atmospheric corrosion resistance of zinc plated stainless steel, Proceedings
of International Conference on Stainless Steels, Chiba, Japan, 1991.
973. Lieberman, E. S., Clayton, C. R., and Herman, H.: Thermally-sprayed active metal coatings for corrosion
protection in marine environments, Naval Sea Systems Command, David Taylor Naval Ship R&D
Center, Washington, D.C., Final Report, January 1984, Report No. SUSB-84-1.
974. Odnevall, I., and Westdahl, M.: Zinc chlorohydroxosulfates: Newly-discovered corrosion products on
zinc, Carras. Sci. 34, 1231-1242, 1993.
975. Swamy, R. N.: In-situ behaviour of galvanized reinforcement, Galvanized Rebar Seminar, Structural
Integrity Research Institute, University of Sheffield, April 12, 1991.
976. Gage, N., and Wright, D. A.: Corrosion testing of zinc and zinc alloy electrodeposited coatings,
Proceedings of the Asia Pacific Interfinish 90: Growth Opportunities in the 1990's Conference,
Singapore, November 19-22, 1990.
977. Yoshihara, R., Wake, R., and Yamamoto, M.: Characteristics of thermal diffused Zn/SnlNi triple coated
steel sheets, Proceedings of the Asia Pacific Interfinish 90: Growth Opportunities in the 1990's
Conference, Singapore, November 19-22, 1990.
978. van Ooij, W. J., and Sabata, A.: Effect of paint adhesion on the underfilm corrosion of painted precoated
steels, National Association of Corrosion Engineers Corrosion '91 Conference, Cincinnati, Ohio, March
11-15,1991, Paper No. 417.
979. Baboian, R.: Chemistry and corrosivity of the automotive environment, National Association of
Corrosion Engineers Corrosion '91 Conference, Cincinnati, Ohio, March 11-15, 1991, Paper No. 371.
980. Skerry, B. S., and Simpson, C. H.: Combined corrosion/weathering accelerated testing of coatings for
corrosion control, National Association of Corrosion Engineers Corrosion '91 Conference, Cincinnati,
Ohio, March 11-15, 1991, Paper No. 412.
981. Giridhar, J., and van Ooij, W. J.: Adhesion and corrosion properties of a new NiZnlZnCo-coated steel
tire cord, Suif. Coatings Technol. 53, 243-255, 1992.

REFERENCES

449

982. van Ooij. W. J., and Sabata. A.: Chemical and thermal stability of phosphate layers on c()ld~rollcd and
elcctrogalvanized steels . .I. Coatinfis Technol. 61. 51-65. 1989.
983. van Ooij. W. J . Sabata, A .. Loison, D .. Jossic. 1., and Charbonnier. J.~c.: Paint delamination from
electrocoated automotive stcels during atmosphenc corrosion. Part II. Zinc alloy~coatcd steeb. J
Adhesion Sci. Technol. 3(2). 79-97.1989.
984. van Ooij. W. J., Sabata, A .. Loison. D., Jossic. T.. and Charbonnier. J.~c.: Paint delamination from
electrocoated automotive steels during atmospheric corrosion. Part I. Hot-dip galvanized and electro~
galvanized steel. J. Adhesion Sci. Techno!. 3( I). 1-27. 1989.
985. van Ooij. W. J., and Sabata. A.: Developments towards corrosion~resistant sheet steels in the building
and transportation industries. Scand. J. Meta/!. 21. 32-48, 1992.
986. van Ooij. W. 1., Edwards. R. A., and Neihciscl. G. L.: Cosmetic corrosion of welded hot dip galvanized
steel panels, SAE Technical Paper No. 912287. Proceedings of the 5th Society of Automotive Engineers
Automotive Corrosion & Prevention Conference, Dearborn, Michigan. October 21- 23. 1991.
987. Chaker. V: Corrosion testing in soils-past, present, and future, in Corrosion Testini!, and Evaluation:
Silver Anniversary Volume, STP 1000, pp. 95-111, American Society for Testing and Materials,
Philadelphia, 1990.
988. Kanda. K.: Properties of black electrogalvanized steel sheet, CAMP-/SIJ 4,614-617, 1991.
989. Molina, A., Blanco, M. 1., and Andrade, c.: Corrosion rate of three different types of galvanized coatings
of steel reinforcements in contact with mortar, Proceedings of the 9th International Congress on Metallic
Corrosion Conference, Toronto, Ontario, June 3-7,1984.
990. Walter, G. W.: Laboratory simulation of atmospheric corrosion by S02' Part I. Apparatus. e1ectrochemi~
cal techniques, example results. Corros. Sci. 32, 1331-1352, 1991.
991. Radeker, w., and Friehe, w.: Effect of alloying elements on the properties of hot dip galvanized coatings,
Proceedings of the 7th International Galvanizing Conference on Hot Dip Galvanizing. Paris. June 1964,
pp.167-178.
992. Askey, A., Lyon, S. B .. Thompson, G. E., Johnson, J. B., Wood, G. c.. Sage, P. w.. and Cooke, M. 1.:
The effcct of f1y~ash particulates on the atmospheric corrosion of zinc and mild steel. Corms. Sci. 34,
1055-1081. 1993.
993. van Ooij. W. J., and Sabata, A.: Chemical stability of phosphate conversion coatings on cold~rolled and
clcctrogalvanized steels. SurF Coatings Technol. 39/40,667-674. 1989.
994. van Ooij. W. 1., Sabata, A .. and Strom, G.: Corrosion of primed and fully painted prccoated automotive
steels in atmospheric and scab tests, Prog. (Jig Coat. 18, 147-172,1990.
995. Baboian, R.: Corrosion behavior of painted galvanized auto~body steel in contact with exterior
automotive trim, Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel
Sheet. GALVATECH '89. Tokyo. September 5-7. 1989, pp. 297-305.
996. Thomas. D. G., Cox. R. T.. and Richardson. B. A.: The role of metallic zinc in zinc rich paints.
Proceedings of the 5th Annual Conference of the Australasian Corrosion Association, Newcastle, New
South Wales, Australia, November 16-20.1964.
997. Zhu. Y, Li. S., Wang. J., Wang, J., and Qi, Y: The effect of titanium on improving the corrosion resistance
of hot dip galvanizing zinc coating on malleable cast iron, Matet: Mech. Eng. (Chino) 15(3)' 1-4. 1991
I in Chinese].
998. Ayoub, c.. Davin, A., Goodwin, F. E., and Coutsouradis, D.: Atmospheric corrosion behaviour of galfan
coated steels, Proceedings of the 15th International Galvanizing Conference. INTERGAl.VA ' 88, Rome.
June 1988.
999. Odnevall. I.. and Leygraf. c.: Formation of NaZn4CI(OH)6S046H20 in a marine atmosphere, Corros.
Sci. 34,1213-1229,1993.
1000. Guttman. H., and Redden, 1. J.: Handling and storing of galvanized steel, Cominco PTC Project No.
71 ~5~412, Progress Report. 1972.
1001. Jagannathan, Y.: Emcrging technologies in the hot~dip coating of automotive sheet steel. 1. Metals 45(8),
48-51. 1993.
1002. Strom, M., van Ooij, W. J., Sabata, A., Edwards, R. A., Strom, G., and Ramamurthy, A. c.: A statistically
designed study of atmospheric corrosion simulating automotive field conditions using a high perform~
ance climate chamber-status report of work in progress, SAE Technical Paper No. 912282, Proceed~
ings of the 5th Society of Automotive Engineers Automotive Corrosion & Prevention Conference,
Dearborn, Michigan, October 21-23,1991.

450

REFERENCES

1003. Otero, E., and Royuela, J. J.: The assessment of short term data of pipe corrosion in drinking water. Part
I. Galvanized steel, Carras. Sci. 34, 1581-1593, 1993.
1004. Sebisty, J. J., and Palmer, R. H.: Hot dip galvanizing with less common bath additions, Proceedings of
the 7th International Conference on Hot Dip Galvanizing, Paris, June 1964, pp. 235-265.
1005. Abu Zahra, R. H.: A radiotracer study on the reaction between zinc and chromate. Part II. The corrosion
of zinc in dilute chromate solutions, Corros. Sci. 6,349-355, 1966.
1006. Johansson. E., and Rendahl, B.: A swedish field exposure program of precoated sheet steels intended
for the automotive industry. National Association of Corrosion Engineers Corrosion '91 Conference.
Cincinnati. Ohio, March 11-15, 1991, Paper No. 390.
1007. Townsend, H. E.: Coated steel sheets for corrosion-resistant automobiles. National Association of
Corrosion Engineers Corrosion '91 Conference, Cincinnati. Ohio, March 11-15. 1991. Paper No. 416.
1008. Loar. G. w.. Romer. K R., and Aoe. T. J.: Zinc-alloy electrodeposits for improved corrosion protection,
Plating Surf. Finish. 78(3),74-79.1991.
1009. Lindsay. J. H.: Electrogalvanized automotive sheet steel and the manufacturing system. Plating Surf.
Finish. 79(5).129-135.1992.
1010. Fawcett. N. C. Stearns. C. E .. and Bufkin. B. G.: Zinc consumption in zinc-rich primers co-pigmented
with di-iron phosphide. 1. Coatings Technol. 56. 31-34. 1984.
1011. Lee. H. H.: Galvanized steel with improved resistance to intergranular corrosion, Proceedings of the
Galvanizers Committee '77, 1977, pp. 17-23.
1012. Stareck, J. E.. and Cibulskis. W. S.: Recent developments in the use of conversion coatings on zinc.
Proc. Am. Electroplaters Soc. 1947, 171-179.
1013. Babaqi. A. S .. Hefny. M. M .. Abdulla, R. M . and EI-Basiouny. M. S.: Cathodic behaviour of zinc in
aqueous methanol. 1. Electrochem. Soc. India 39.198-200.1990.
1014. Ghali, E. L., and Potvin. R. J. A: The mechanism of phosphating of steel, Corros. Sci. 12.583-594.
1972.
10 15. Townsend. H. E.: Twenty-five-year corrosion tests of 55% AI-Zn alloy coated steel sheet. Mater.
Petform. 32(4). 68-71. 1993.
1016. Shaldaev, V. S .. Tomashova. N. N .. and Davydov. A D.: A combined protection of zinc in the alkaline
current sources. Zashch. Met. 28. 516-519. 1992 [in Russian].
1017. Notowidjojo. B. D . Kennon. N. E. and Wingrove. A L.: Zinc-nickel coating: A new galvanizing
technology, Proceedings of the Step into the 90's Conference, August 27-31. 1989, Queensland.
Australia. pp. 623-632.
1018. Sommer, A. J .. and Leidheiser. H.: Effect of alkali metal hydroxides on the dissolution behavior of a
zinc phosphate conversion coating on steel and pertinence to cathodic delamination. Corrosion 43.
661-665.1987.
1019. Leidheiser, H.: Electrochemical methods for appraising corrosion protective coatings, 1. Coatings
Techno!. 63, 21-31. 1991.
1020. Jones, D. A. Blitz. R. K, and Hodjati. I.: Alternate immersion testing of coated sheet steel. Corrosion
42,255-262. 1986.
1021. Guttman, H., and Niessen, P.: Reactivity of silicon steels in hot-dip galvanizing, Can. Metall. Q. 11.
609-615. 1972.
1022. Harvey. G. 1.: Structure and cOrrosion resistance of zincalume coatings. BHP Tech. Bull. 25(2). 63-67.
1981.
1023. Kaneko, K, Mochizuki, A., Nakajima, H .. Ikeda, K . and Toyose, K: Development of galvanized
aluminum alloy sheet for body panels with an excellent filiform corrosion resistance, SAE Technical
Paper No. 930703. Society of Automotive Engineers International Congress and Exposition, Detroit.
March 1-5, 1993.
1024. Cornet, I.. and Bresler, B.: Some recent developments in the use of galvanized steel reinforcement in
concrete. Proceedings of the 9th International Conference on Hot Dip Galvanizing, Dusseldorf.
Germany, June 1970. pp. 414-426.
1025. Grenon, P.: Steel structures coating with zinc metallizing, National Association of Corrosion Engineers
Corrosion '87 Conference, San Francisco, March 9-13. 1987. Paper No. 427.
1026. Fawcett, N. C, Stearns, C E., and Bufkin, B. G.: Effect of zinc oxidation on the conductivity and
performance of di-iron phosphide augmented zinc-rich primers. 1. Coatings Technol. 56. 49-52. 1984.

REFERENCES

451

1027. Horstmann, D., and Riecke, E.: Fracture characteristics of hot dip galvanized steels in prestressed
concrete in cases of corrosion in neutral and alkaline media, Proceedings of the 12th International
Galvanizing Conference, INTERGALVA '79, Paris, May 17-23,1979.
1028. Hampson, N. A., Tarbox, M. J.: The behavior of zinc in potasium hydroxide solution, 1. Electrochem.
Soc. 110,95-98,1963.
1029. Thangavel, K., Rengaswamy, N. S .. and Balakrishnan, K.: Corrosion resistance of galvanized steel in
concrete. Mater. Perform. 34, 59-63, 1995.
1030. Cachet, C, Wiart, R., Ivanov, I., and Rashkov, S.: Mechanism of the reverse dissolution of zinc in the
presence of nickel. Part I. Influence of anodic products and substrate purity, 1. Appl. Electrochem. 23,
1011-1016,1993.
1031. Raichevskii, G., and Vitkova, SI.: The effect of the phase composition and the structure of electrolytic
cobalt deposits on their corrosional electrochemical behavior in an acid medium, Zashch. Met. 19,
418-424, 1973 lin Russian].
1032. Cardon, E, and Gomes, W. P.: Experiments on the zinc oxide/electrolyte interface during anodic
oxidation reactions under pulsed illumination, Surf. Sci. 27, 286-296, 1971.
1033. Vanden Berghe, R. A. L., Cardon, E, and Gomes, W. P.: On the electrochemical reactivity of the redox
couple Fe(CN)i-lFe(CN)~- at the single crystal zinc oxide electrode, Surf. Sci. 39, 368-384, 1973.
1034. Pyun, S., Bae, J., Park,s., Kim, J., and Lee, Z.: The anodic behavior of hot-galavnized zinc layer in
alkaline solution, Corros. Sci. 36, 827-835, 1994.
1035. Grabner, E. W.: Electrochemiluminescence ofrubrene at a zinc oxide electrode, Electrochim. Acta 20,
7-12, 1975.
1036. Pettinger, B., Schoppel, H.-R., and Gerischer, H.: Electroluminescence at semiconductor electrodes
caused by hole injection from electrolytes, Ber. Bunsenges. SO, 849-855, 1976.
1037. Hampson, N. A., Shaw, P. E., and Taylor, R.: Anodic behavior of zinc in potassium hydroxide solution,
Br. Corros. 1. 4, 207-211. 1969.
1038. Fichou, D., and Kossanyi, J.: Electroluminescence at the n-ZnO/electrolyte interface: A comparative
study of the emission processes under cathodic and anodic pulsed polarization, .I. Electrochem. Soc.
133,1607-1617.1986.
1039. Lewis, D. A.: Some aspects of the corrosion of steel in concrete, Proceedings of the First International
Congress on Metallic Corrosion, London, April 10-15, 1961, pp. 547-555.
1040. Hair, W. J., and Wall, R.: Metal spraying and painting for the protection of steel structures in polluted
railway atmospheres, Proceedings of the First International Congress on Metallic Corrosion, London.
April 10-15, 1961, pp. 556-566.
1041. Weast, R. C, Kotnik, L. J., and Geehan, D. M.: Corrosion of zinc in dilute aqueous solutions, Proceedings
of the 6th International Conference on Hot Dip Galvanizing, Interlaken, Switzerland, June 1961, pp.
155-176.
1042. Stricker, E: Comparison of various methods for determining the thickness of galvanized coatings, of
the iron-zinc alloy layers and of the pure zinc layer, Proceedings of the 7th International Conference
on Hot Dip Galvanizing, Paris, June 1964, pp. 277-288.
1043. Johnsson, T., and Kucera, V: Possibilities of improving the corrosion resistance of zinc coatings by
alloying, Proceedings of the 13th International Galvanizing Conference, INTERGALVA '82, London,
1982, pp. 4711-47/6.
1044. Iovchev, M., and Mihailov. G.: Restraining the corrosion of galvanized steel by the use of neutralite-a
case study, Proceedings of the 15th International Galvanizing Conference, INTERGALVA '88, Rome,
June 1988, pp. GH311-GH3/5.
1045. Yang, Z .. Li, J., and Jia, L.: Development and application of Zn-Fe alloy-coated (galvannealed) steel
sheet, Proceedings of the 1st Asian-Pacific General Galvanizing Conference, September 15-18,1992,
pp.208-216.
1046. Hauffe, K.: Oxidation (if Metals, Plenum Press, New York, 1965, pp. 201-228.
1047. Glayman, J.: Anodization of zinc, Galvallo 1970(February), 147-149.
1048. Terada, K.: Solution of zinc oxide in water charged with CO2 , Bull. It/st. Phys. Chem. Res. Tokyo 10,
864-871, 1931.
1049. Reinhart, F. M.: Twenty-year atmospheric corrosion investigation of zinc-coated and uncoated wire and
wire products, STP 290, pp. 1-141, American Society for Testing and Materials, Philadelphia, 1961.

452

REFERENCES

1050. Townsend. H. E . Borzillo, A R, and Barker, W. D.: Performance of Al-Zn alloy coated steel sheet after
twenty-two years of atmospheric corrosion testing, Proceedings of the 2nd International Conference on
Zinc Coated Steel Sheet, Rome, June 9-10,1988, pp. SAIII-SAllIl.
1051. Bonaretti, A, Capoccia, A., Giardetti, G., and Mucchino, U.: Process and use properties of lavegal,
Proceedings of the 2nd International Conference on Zinc Coated Steel Sheet, Rome, June 9-10, 1988,
pp. SC611-SC61l3.
1052. Applegate, L. M.: Cathodic Protection, McGraw-Hill, New York, 1960, pp. 115-117.
1053. Lucas, R W., and Eborall, R: Effects of additions to the metal and atmosphere on the surface oxidation
of zinc, The British Non-Ferrous Metals Research Association, Research Report No. A.1463, December
1963.
1054. Mayne, J. E. 0.: Effect of rust and environment on inhibition by zinc-rich paints, 1. Iron Steelln.H. 176,
140-143, 1954; Atmospheric exposure tests with zinc-rich polystyrene paints, 176, 143-146, 1954.
lOSS. Mayne, J. E. 0.: Cementiferous paints. Part I & II. Zinc phosphate cements, 1. Iron Steellnsf. 219-228,
1948;289-293,1950
1056. Ryabov, A K., and Gratsansku, N. N.: Influence of small additions of metals on the physicochemical
properties of zinc coatings, Ukr. Khim. Zh. (Russ. Ed.) 30, 883-886, 1964.
1057. Pourbaix, A.: Corrosion of galvanized steel pipes used for water supply, CELBELCOR, Brussels,
Belgium, Technical Report No. 180, July 1970.
1058. Hisamatsu, Y: Science and technology of zinc and zinc alloy coated steel sheet, Proceedings of the
International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September5-7,1989,pp.3-12.
1059. Tsuzuki, Y, Suganuma, A, Yamamoto, T, Kozuka, N., and Mase, K.: Development of new zinc
phosphating for automobile, Proceedings of the International Conference on Zinc and Zinc Alloy Coated
Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 206-213.
1060. Sato, N.: Characterization of zinc phosphate films on zinc alloy coated steels, Proceedings of the
International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7,1989, pp. 214-221.
1061. Suzuki, M., Hayashi, H., Kiyomatsu, 1., Miyawaki, T, and Matsushima, Y: A study on the hopeite crystal
deposited on galvanized steel, Proceedings of the International Conference on Zinc and Zinc Alloy
Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 222-229.
1062. Nakazawa, M., and Yoneno, M.: Effect of phosphate and silica additives on the performance and structure
of chromate conversion coatings, Proceedings of the International Conference on Zinc and Zinc Alloy
Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7,1989, pp. 238-245.
1063. Koizumi, S., Shima, S., and Matsushima, Y: A development of black chromate-oxide finishes by baking
process for galvanized steel, Proceedings of the International Conference on Zinc and Zinc Alloy Coated
Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 246-253.
1064. Fountoulakis, S. G., and Steinbicker, R. N.: Influence of coating thickness on the cosmetic-corrosion
performance of zinc, zinc alloy and multilayered electroplated sheet steels, Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7,
1989, pp. 306-313.
1065. Petrey, L. D., Shaffer, R 1., Berndt, T, and Macciocco, J.: The effect of automotive plant painting
operations on the paint adhesion and cosmetic corrosion resistance of precoated steels, Proceedings of
the International" Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo,
September 5-7,1989, pp. 314-320.
1066. Soshiki, Y, Ishida, w., Iida, A, Kanamaru, T, and Morita, 1.: Application of Fe-Zn alloy electroplated
galvannealed steel sheet to automotive bodies, Proceedings of the International Conference on Zinc and
Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7,1989, pp. 329-337.
1067. Tanaka, S., and Miyosawa, Y: Outdoor exposuring of formed specimens of precoated metal sheets,
Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH
'89, Tokyo, September 5-7,1989, pp. 381-388.
1068. Mizuno, K., Suzuki, K., Ohtsubo, T, and Suzuki, M.: Characterization of phosphate conversion coatings,
Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH
'89, Tokyo, September 5-7,1989, pp. 428-434.

REFERENCES

453

1069. Jixun, W, Yanping, L., Zhiyuan, Y, Yinhua, L., and Yingting, X.: A study ofthe structure ofZn-Fe alloy
plating, Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet.
GALVATECH '89, Tokyo, September 5-7,1989, pp. 619-624.
1070. Gu. Moo Notis, M. R., and Marder, A. R.: Structure of as-plated electrodeposited zinc-iron coatings,
Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH
'89. Tokyo. September 5-7,1989, pp. 462-469.
1071. Schedin, E . Engberg, G .. Karlsson, S .. Kiusalaas, R., and Klang, H.: Plasticity of pure-zinc hot-dip
galvanized coatings, Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel
Sheet, GALVATECH '89, Tokyo, September 5-7,1989, pp. 493-499.
1072. Ito, Y, Hayashi, K., Kato, c., and Miyoshi, Y: A study on simulated and accelarated corrosion test
methods for automotive precoated steel sheet, Proceedings of the International Conference on Zinc and
Zinc Alloy Coated Steeel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 503-510.
1073. Davidson, D. D., Soreide, L. E., and Stephens, M. L.: Evalution of the cosmetic corrosion resistance of
zinc coated sheet steels-using the chrysler chipping corrosion test and long term outdoor exposure
testing, Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet,
GALVATECH '89, Tokyo, September 5-7, 1989. pp. 519-527.
1074. Sakauchi, T., and Kunimi, H.: Evaluation of the corrosion resistance of zinc-alloy-plated steel sheet
applied to automobile body, Proceedings of the International Conference on Zinc and Zinc Alloy Coated
Steel Sheet. GALVATECH '89, Tokyo. September 5-7,1989, pp. 528-535.
1075. Davidson, D. D., Soreide, L. E., and Stephens, M. L.: Assessment of the chrysler chipping corrosion
test using laboratory scab and long term outdoor exposure testing, Proceedings of the International
Conference on Zinc and Zinc Alloy Coated Steel Sheet. GALVATECH '89, Tokyo, September 5-7, 1989,
pp. 536-542.
1076. Uchiyama, Y, Hasaka, M., and Koga, H.: Effect of structure and mischmetal addition on the corrosion
behaviour ofZn-5 mass% AI alloy, Proceedings of the International Conference on Zinc and Zinc Alloy
Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 545-552.
1077. Tajiri. Y.. Shimada, S., Yamaji, T., and Adaniya, T.: Effect of coating structure on corrosion resistance
of hot dipped Zn-5%AI + Mg alloy coated steel sheet, Proceedings of the International Conference on
Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 553-559.
1078. Walter. G. w.: Corrosion rates of zinc based coatings calculated by a galvanostatic non-linear method
from low polarization data, Proceedings of the International Conference on Zinc and Zinc Alloy Coated
Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989. pp. 560-564.
1079. Walter, G. W. Use of electrochemical sensors to measure corrosion characteristics of Zn and Zn-55%
AI alloy coated steel under simulated atmospheric conditions, Proceedings of the International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp.
565-568.
1080. Fukumoto, H .. Mizuki. H., and Masuhara, K.: Application of scanning vibrating electrode technique for
investigation of chemical surface treatment on zinc coated steel sheets, Proceedings of the International
Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo. September 5-7. 1989.
pp.577-584.
1081. Massinon, D .. and Dauchelle, D.: Recent progress towards the understanding of underfilm corrosion of
coated steels used in the automotive industry, Proceedings of the International Conference on Zinc and
Zinc Alloy Coated Steel Sheet. GALVATECH '89, Tokyo, September 5-7, 1989, pp. 585-595.
1082. Masuhara. Koo and Kumon, E: Corrosion resistance of prepainted Zn-4%AI-0.I % Mg-misch metal
alloy coated steel sheets, Proceedings of the International Conference on Zinc and Zinc Alloy Coated
Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 596-602.
j 083. Matsumoto. M., Shinohara, T., and Tsujikawa. S.: Application of scanning vibrating electrode technique
to detect corrosion of zinc and zinc alloy layer under organic coating, Proceedings of the International
Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989,
pp.603-61O.
1084. AI-Hashem, A., and Thomas, D.: The effect of corona discharge treatment on the corrosion behaviour
of metallic zinc spots contained in polymeric coatings, Proceedings of the International Conference on
Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, September 5-7, 1989, pp. 611-618.

454

REFERENCES

1085. Camitz, G., and Vinka, T.: Corrosion of steel and metal-coated steel in swedish soils-effects of soil
parameters, in Effects of Soil Characteristics on Corrosion, STP 1013, pp. 37-53, American Society for
Testing and Materials, Philadelphia, 1989.
1086. RabeJer, R. c.: Soil corrosion evaluation of screw anchors, in Effects of Soil Characteristics on
Corrosion, STP 1013, pp. 54-80, American Society for Testing and Materials, Philadelphia, 1989.
1087. Escalante, E.: Concepts of underground corrosion, in Effects of Soil Characteristics on Corrosion, STP
1013, pp. 81-94, American Society for Testing and Materials, Philadelphia, 1989.
1088. Edgar, T. V.: In-service corrosion of galvanized culvert pipe, in Effects of Soil Characteristics on
Corrosion, STP 1013, pp. 133-143, American Society for Testing and Materials, Philadelphia, 1989.
1089. International Lead Zinc Research Organization: Corrosion Resistance of Unpainted Galfan Steel,
ILZRO Project ZM-285, Progress Report No. 24, Research Triangle Park, North Carolina, 1992.
1090. Linch, R. E, and Goodwin, E E.: Galfan coated steel for automotive applications, SAE Technical Paper
No. 860658, Society of Automotive Engineers Automotive Corrosion and Prevention Conference,
Dearborn, Michigan, December 8-10, 1986.
1091. Guttman, H.: Corroded Highway Guard Rails, Cominco PTC Internal Report No. TSF 6715, 1967.
1092. Guttman, H.: A Study of Corroded Highway Guard Rail, Cominco Project No. TSF 5-410,1978.
1093. Pryor, M. J.: Bimetallic corrosion, in Corrosion, Vol. I, Shreir, L. L. (ed.), Chapter 1.69, George Newnes
Ltd., London, 1963.
1094. Zhang, X. G., and Tran, H. Q.: Effect of cyclic wetting and drying on corrosion of zinc and steel, in
Cyclic Cabinet Corrosion Testing, G. S. Haynes (ed.), ASTM STP 1238, pp. 125-135, American Society
for Testing and Materials, Philadelphia, 1995.
1095. Hill, G. A., Spellman, D. L., and Stratfull, R. E: Laboratory corrosion tests of galvamzed steel in
concrete, Transp. Res. Rec. No. 604, pp. 25-31,1976.
\096. Sagues, A. A., and Kranc, S. c.: On the determination of polarization diagrams of reinforcing steel in
concrete, Corrosion 48, 624-633, 1992.
\097. Sagues, A. A.: Corrosion measurement techniques for steel in concrete, National Association of
Corrosion Engineers, Corrosion '93, Paper No. 353, 1993.
1098. Videm, K., and Myrdal, R.: The electrochemical behavior of steel in concrete and how to evaluate the
corrosion rate, National Association of Corrosion Engineers, Corrosion '96, Paper No. 348, 1996.
1099. Anderson, E. A.: Zinc and zinc coatings, in: Corrosion Resistance of Metals and Alloys, 2nd ed., E L.
LaQue and H. R. Copson (eds.), pp. 223-247, Reinhold Publishing Corp., New York, 1963.
1100. Lauer, G., and Mansfeld, E: Measurement of galvanic corrosion current at zero external impedance,
Corrosion 26, 504-506,1970.
1101. Mansfeld, E, and Kenkel, J. Y.: Laboratory studies of galvanic corrosion: Effect of velocity in NaCI and
substitute ocean water, Corrosion 33, 236-240, 1977.
1102. Mansfeld, E, Hengstenberg, D. H., and Kenkel, J. Y.: Gal vanic corrosion of AI alloys: Effect of dissimilar
metal, Corrosion 30. 343-353, 1974.
1103. Mansfeld, E, and Kenkel, J. Y.: Laboratory studies of galvanic corrosion: I. Two-metal couples.
Corrosion 31, 298-302,1975.
1104. Zhang, X. G., and Valeriote, E. M.: Galvanic protection of steel by zinc under thin layer electrolytes, in
Atmospheric Corrosion, W. W. Kirk and H. H. Lawson (eds.), ASTM STP 1239, pp. 230-239, American
Society for Testing and Materials, Philadelphia, 1995.
1105. Mansfeld, E, and Kenkel, 1. Y.: Laboratory studies of galvanic corrosion: Two-metal couples, Corrosion
31,298-302,1975.
1106. Grimes W. D., Hartt, W. H., and Turner, D. H.: Cracking of concrete in sea water due to embedded metal
corrosion, Corrosion 35. 309-316,1979.
1107. O'Reilly, M. W.: Chapter 15:2 in Corrosion, 2nd ed., Vol. 2, L. L. Shreir (ed.), Newnes-Butterworths,
London, 1976.
1108. Mayne J. E. 0.: Chapter 15: 10 in Corrosion, 2nd ed., Vol. 2, L. L. Shreir (ed.), Newnes-Butterworths,
London, 1976.
1109. Leidheiser, H.: Corrosion of painted metals-a review, Corrosion 38, 374-383, 1982.
1110. Ito, K., Oohashi, H., and Yoshii, T.: Corrosion behaviour of zinc plated stainless steel, Proceedings of
the International Conference on Zinc and Zinc Alloy Coated Steel Sheet, GALVATECH '89, Tokyo, 1989,
pp. 569-576.

REFERENCES

455

1111. Hess, J. c., and Soreide, L. E.: The effect of phosphate chemistry and rinsing conditions on the corrosion
resistance and the paint adhesion of zinc coated and cold rolled sheet steels, Society of Automotive
Engineers Automotive Corrosion and Prevention Conference Proceedings, Dearborn, Michigan. 1989,
Paper No. 892557, pp. 17-38.
1112. Townsend, H. E.: Status of a cooperative effort by the automotive and steel industries to develop a
standard accelerated corrosion test, Society of Automotive Engineers Automotive Corrosion and
Prevention Conference Proceedings, Dearborn, Michigan. 1989, Paper No. 892569, pp. 133-145.
lin. Ito, Y, and Miyoshi, Y: Corrosion protection of galvanized steel sheet-corrosion investigation offield
vehicle and its laboratory evaluation methods, Society of Automotive Engineers Automotive CorrosIon
and Prevention Conference Proceedings, Dearborn, Michigan, 1989, Paper No. H92580, pp. 217-227.
1114. Franks, L. L.: An eight-year atmospheric and on-vehicle corrosion test of painted and zinc coated steels,
Society of Automotive Engineers Automotive Corrosion and Prevention Conference Proceedings.
Dearborn, Michigan, 1989, Paper No. 892581, pp. 229-238.
1115. Kokubo. Z., Nomura, S., Sakai, H., Sakaguchi, M., and Zwai, M.: Newly developed Zn-Fe/Zn-Ni
double-layer electroplated steel sheet, Society of Automotive Engineers Automotive Corrosion and
Prevention Conference Proceedings, Dearborn, Michigan. 1983, Paper No. 830518, pp. 75-82.
1116. Catanzano, A., Arrigoni, G., Palladino, M., and Sarracino, M.: Multilayer electrogalvanized (Zn-CrCrOx) steel sheet for optimum corrosion protection of car bodies, Society of Automotive Engineers
Automotive Corrosion and Prevention Conference Proceedings, Dearborn, Michigan. 1983, Paper No.
830583, pp. 107-124.
1117. Zhang, X. G .. and Tran, H. Q.: Effect of cyclic wetting and drying on corrosion of zinc and steel, in STP
1238, Cyclic Cahinet Corrosion Testing. G. S. Haynes (cd.), pp. 125-135, American Society for Testing
and Materials, Philadelphia, 1995.
IIIH. Despic. A. R.: Deposition and dissolution of metals and alloys, Part B: Mechanisms, kinetics. texture,
and morphology, in Comprehensive Treatise ()( Electrochemistry, Vol. 7, B. E. Conway et al. (cds.),
Plenum Press, New York, pp. 451-528.
1119. Straumanis, M. E., and Wang, Y: Surface disintegration of zinc mono- and polycrystals dissolving
anodically in sodium bromate solutions and the apparent valency of zinc ions, Corrosion 22, 132-136,
1966.
1120. Kim, J. T, and Jorne, 1.: The kinetics and mass transfer of zinc electrode in acidic zinc-chloride solution,
1. Electrochem. Soc. 127,8-15.1980.
1121. Lee, T S.: Hydrogen overpotential on zinc containing small amounts of impurities in concentrated
alkaline solution, 1. Electrochem. Soc. 120,707-709, 1973.
1122. Hoare, J. P.: The Electrochemistrv o(Oxvr;ell, Intersciencc Publishers, John Wiley & Sons. New York,
1968.
1123. Tarasevich, M. R., Sadkowski, A., and Yeager, E.: Oxygen electrochemistry, in Comprehmsil'e 7imtise
ofElectrochemistrv, Vol. 7, B. E. Conway et al. (eds.), Plenum Press, New York, pp. 301-398.
1124. Stratmann, M., and MUller, J.: The mechanism of the oxygen reduction on rust-covered metal substrates.
Corms. Sci. 36, 327-359, 1994.
1125. Holler, H. D.: Studies on galvanic couples, 1. Electrochem. Soc. 97(12),271, 277, 453,1950.
1126. Sato, N.: The passivity of metals and passivating films, in Passivity of Metals, R. Frankenthal and J.
Kruger (eds.), pp. 29-58, The Electrochemical Society, Princeton, New Jersery, 1978.
1127. Vetter, K. J.: EleClrochim. Acta 16,1923,1971.
1128. Hull, M. N., Ellison, J. E., and Toni, J. E.: The anodic behaviour of zinc electrodes in potassium
hydroxide electrolytes, J. Electrochem. Soc. 117, 192-198, 1970.
1129. Zhang, X. G., and Hwang, 1.: Cominco Internal Report 651-008, 1994.
1130. Eisenberg, M., Baumann, H. F, and Brettner, D. M.: Gravity field effects on zinc anode discharge in
alkaline media, 1. Electrochem. Soc. 108,909-915, 1961.
1131. Hampson, N. A., Shaw, P. E., and Taylor, R.: Anodic behaviour of zinc in potassium hydroxide solution,
Br. Corrosion 1. 4, 207-211,1969.
1132. Dirkse, T P.: The behaviour of the zinc electrode in alkaline solutions, 1. FJeclrochem. SoC, 125,
1591-1594, 1978.
1133. van der Lcij, M.: The possibility of black zinc oxide as spectrally selective coating for low temperature
solar collectors, 1. Electrochem. Soc. 125, 1361-1364, 1978.

456

REFERENCES

1134. Aurian-Blajeni, B., and Tomkiewicz, M.: Passive zinc electrodes: Application of the effective medium
theory, J. Electrochem. Soc. 132,869-870, 1985.
1135. Loftus, D., Roberts, J., Weaver, R, Leach, S., and Nanis, L.: Diffusivity in zinc chloride-potassium
chloride electrolyte, J. Electrochem. Soc. 130,332-334, 1983.
1136. Bonasewicz, P., Hirschwald, W., and Neumann, G.: Influence of surface processes on electrical,
photochemical, and thermodynamical properties of zinc oxide films, J. Electrochem. Soc. 133, 22702278,1986.
1137. Sato, Y., Niki, H., and Takamura, T.: Effects of carbonate on the anodic dissolution and the passivation
of zinc electrode in concentrated solution of potassium hydroxide, 1. Electrochem. Soc. 118, 1269-1272,
1971.
1138. O'Brien, R N., Leja, J., and Beer, E. A: Interferometric study of the ZnlZnS041Zn system, J.
Electrochem. Soc. 121,370-375, 1974.
1139. Gouda, V. K., Khedr, M. G. A., and Shams EI Din, A M.: Role of anions in the corrosion and
corrosion-inhibition of zinc in aqueous solutions, Corros. Sci. 7,221-230,1967.
1140. Kordesch, K. V.: Alkaline manganese dioxide zinc batteries, in Batteries, K. V. Kordesch (ed.), Vol. I,
pp. 241-384, Marcel Dekker, New York, 1974.
1141. Schumm, B.: Zinc-carbon batteries, in Modern Battery Technology, C. D. S. Tuck (ed.), pp. 87-111,
Ellis Horwood, Chichester, England, 1991.
1142. Hunter, 1. C.: Alkaline-manganese batteries, in Modern Battery Technology, C. D. S. Tuck (ed.), pp.
111-125, Ellis Horwood, Chichester, England, 1991.
1143. Brodd, R J.: Batteriesfor Cordless Appliances, Research Studies Press, Letchworth, England, 1987.
1144. Snyder, R N., and Lander, 1. 1.: Rate of hydrogen evolution of zinc electrodes in alkaline solutions,
Electrochem. Techno!. 3(5-6), 161-166, 1965.
1145. Bell, G. S.: The stability of zinc in zinc-chloride/ammonium-chloride electrolytes, Electrochim. Acta
13,2197-2202,1968.
1146. Riietschi, P.: Solubility and diffusion of hydrogen in strong electrolytes and the generation and
consumption of hydrogen in sealed primary batteries, J. Electrochem. Soc. 114,301-305,1967.
1147. Sato, Y., Takahashi, M., Asakura, H., Yoshida, T., Tada, K., Kobayakawa, K., Chiba, N., and Yoshida,
K.: Gas evolution behaviour ofZn alloy powder in KOH solution, 1. Power Sources 38,317-325, 1992.
1148. Dube, G., Renaud, R, and Huot, J. Y.: Amalgamation of battery grade indium-zinc powders, Prog.
Batteries Battery Mater., 10, 151-155, 1991.
1149. Huber, R: Leclanche batteries, in Batteries, K. V. Kordesch (ed.), Vol. I, pp. 1-239, Marcel Dekker,
New York, 1974.
1150. Baugh, L. M., Tye, F. L., and White, N. c.: Corrosion and polarization characteristics of zinc in
concentrated alkaline solution and the effect of amalgamation, Power Sources 9, 303-323, 1983.
1151. Miura, A, Takata, K., Okazaki, R, Ogawa, H., Uemura, T., Nakamura, Y., and Kasahara, N.: Development of corrosion resistant zinc alloys for alkaline-manganese batteries, Denki Kagaku 1989(6), 57,
459.
1152. Meesus, M., Strauven, Y., and Groothaert, L.: New development in reduction of mercury content in zinc
powder for alkaline dry batteries, Power Sources 11, 281-299, 1987.
1153. Glaeser, W.: Gas evolution data on very-Iow-mercury-content zinc powders for alkaline batteries, Power
Sources 12, 265-296, 1989.
1154. Dirkse, T. P., De Wit, D., and Shoemaker, R: The anodic behaviour of zinc in KOH solutions, J.
Electrochem. Soc. 115,442-444, 1968.
1155. Dirkse, T. P.: The behaviour of the zinc electrode in alkaline solutions, J. Electrochem. Soc. 126,
541-543,1979.
1156. Meeus, M. L., Crocq, G. G., and Lemaitre, C. A.: Metallurgical, structural and electrochemical factors

influencing the production of optimal zinc cans for leclanche dry cells by impact extrusion, Power
Sources 7, 463-479, 1979.
1157. Mayer, S. T.: Anodic film formation on Cu, Zn, Na, Ag in alkaline media, Thesis, University of
California, Berkeley, 1989.
1158. Lindsay, 1. H., and O'Keefe, T. J.: Electrogalvanizing, in Modem Aspects of Electrochemistry, No. 26,
B. E. Conway, 1. O'M. Bockris, and R. E. White (eds.), Plenum Press, New York, pp. 165-228.
1159. Hafford, B., Pepper, W. E., and Lioyd, T. B.: Zinc Dust and Zinc Powder, International Lead Zinc
Research Organization, New York, 1982.

REFERENCES

457

1160. Feliu, S., Jr., Morcillo, M., Bastidas, J. M., and Feliu, S.: Evolution of the protective mechanisms of
zinc-rich paints during atmospheric exposure, 1. Coating Techno!. 65, 43-48, 1993.
1161. Szauer, T., and Miszczyk, A.: Zinc pigmented coatings: Ways of improving performance, Polymeric
Materials for Corrosion Control. Proceedings of the American Chemical Society Conference, Chicago,
Sept. 8-13,1985, pp. 849-854.
1162. Montie, J. E, and Hasser, M. D.: Mater. Peiform. 1976(August), 15-18.
1163. Odnevall, I.: Atmospheric corrosion of field exposed zinc, Ph.D. Thesis, Royal Institute of Technology,
Stockholm, 1994.
1164. Odnevall, l., and Leygraf, c.: A comparison between analytical methods for zinc specimens exposed in
a rual atmosphere, J. Electrochem. Soc. 138, 1923-1928, 1991.
1165. Odnevall, I., and Leygraf, c.: The formation of Zn4CI2(OH)4S045H20 in an urban and an industrial
atmosphere, Corros. Sci. 36, 1551-1567, 1994.
1166. Odnevall, 1., and Leygraf, c.: Reaction sequence in atmosphere corrosion of zinc ASTM STP 1239.
American Society for Testing and Materials, Philadelphia, 1994.
1167. International Lead Zinc Research Organization: Zinc Chemicals, ILZRO, New York, 1973.
1168. Bornholdt, R. A, and Perkins, 1.: SEM examination of corrosion product morphology for anodically
polarized zinc, Metallographv 8, 401-409, 1975.
1169. Perkins, 1., Luebke, W. H., Graham, K. 1., and Todd, J. M.: Anodic corrosion of zinc alloysin seawater,
1. Electrochem. Soc. 124, 819-826, 1977.
1170. Tran, Q. T., and Dabosi, E: International Lead Zinc Research Organization Report ZE416, 1994.
1171. Flinn, D. R., Cramer, S. D., Carter, 1. P, Hurwitz, D. M., and Linstrom, P. J.: Materials degradation
caused by acid rain, R. Bahoian (ed.), pp. 119-151, ACS Symposium Series 318, American Chemical
Society, Washington, D.C., 1986.
1172. Romagnoli, R., Vetere, V. E, and Armas, R. A.: Influence of the composition of zinc-ethyl silicate paints
on electrochemical properties. J. Appl. Electrochem. 24, 1013-10 18, 1994.
1173. Hime. M., and Machin, W. G.: Performance variances of galvanized steel in mortar and concrete,
Corrosion 49, 858-860, 1993.
1174. Mattsson, E.: Corrosion: An electrochemical problem, Chemtech 15, 234-243, 1985.
1175. Rausch,
The Phosphating of Metals, Finishing Publications Ltd., Middlesex, England, 1990.
1176. Bard, A. J., and Wrighton, M. S.: Thermodynamic potential for the anodic dissolution of IHype
semiconductors, J. Electrochem. Soc. 124, 1706-1710, 1977.
1177. Mon'ison, S. R.: Electrochemistrv at Semicolldutor and Oxidized Metal Electmdes, Plenum Press, New
York, 1980.
1178. Tench, D. M., and Gerischer, H.: The phototransition in ZnO at 380 nm studied hy anodic photocurrents,
J. Elcctrochem. Soc. 124,1612-1618,1977.
1179. Brown, H.: Zinc Oxide-Properties and Applications, International Lead Zinc Research Organization,
New York, 1976.
1180. Butler, M. A., and Ginley, D. S.: Prediction of f1atband potentials at semiconductor-electrolyte
interfaces from atomic electronegativities, 1. Electrochem. Soc. 125, 228-232, 1978.
1181. Blok, L., and De Bruyn, P. L.: The ionic double layer at the ZnO/solution interface. 1. Colloid Interface
Sci. 32, 518-526,1970.
1182. Memming, R.: The role of energy levels in semiconductor-electrolyte solar cells, 1. Electrochem. Soc.
125,117-123,1978.
1183. Pleskov, Yu. V., and Gurevch. Yu. Ya.: Semiconductor Photoelectrochemistry, P N. Bartlett (translation
ed.), Consultants Bureau, New York, 1986.
1184. Ismail, K. M., Sikora, E., and Macdonald, D. D.: Electrochemical Society Extended Abstracts, Vol. 94-2,
Abstract No. 203, 1994.
1185. Parks, G. A: The isoelectric points of solid oxides, solid hydroxides, and aqueous hydroxo complex
systems, Chon. Rev. 65,177-198,1965.
1186. Gerischer, H., and Sorg, N.: Chemical dissolution of oxides: Experiments with si ntered ZnO pellets and
ZnO single crystals, Werhl. Korros. 42, 149--157, 1991.
1187. Fruhwirth, 0., Herzog, G. W., and Poulios, 1.: Dark dissolution and photodissolution of ZnO, SlII/
Techno!. 24, 293-300,198:;.
1188. Bernas, A: A few cases of electron emission and transfer in heterogeneous media, J. Ph"s. Chem. 68,
2047-2052,1964.

w.:

458

REFERENCES

1189. Miles, M. H., and Gerischer, H.: The ZnJZn(Hg) electrode reaction in binary mixture of water and
n-propand, 1. Electrochem. Soc. 118, 837-841, 1971.
1190. Porter, E C: Corrosion Resistance of Zinc and Zinc Alloys, Corrosion Technology Series, Vol. 6, P. A.
Schweitzer (ed.), Marcel Dekker, New York, 1994.
1191. Romagnoli, R, and Vetere, V. E: The mechanism of the anti-corrosive action of zinc ethyl silicate paints,
Surf. Coat. Int. 76, 208-210, 212-213,1993.
1192. Mackay, W B.: Chapter 11.2 in Corrosion, Vol. 2, L. L. Sheir (ed.), Newnes-Butterworths, London,
1976.
1193. Morgan, J. H.: Zinc anodes for cathodic protection, Carras. Techno!. 272-277, 1957.
1194. Crennel, J. T., and Wheeler, W C G.: Zinc anodes for use in sea water, 1. App!. Chem. 6,415-421,
1956.
1195. Cachet, C, Saidani, B., and Wi art, R: The behaviour of the zinc electrode in alkaline electrolytes. 1.
Eletrochem. Soc. 139,644-654, 1992.
1196. Lohmann, E: Der einfluB des pH auf die elektrischen und chemischen eigenschaften von zinkoxidelektroden, Ber. Bunsenges. Phys. Chem. 70, 428-434. 1966.
1197. Sze, S.: Physics of Semiconductor Devices, John Wiley & Sons, New York, 1969.
1198. Gehmecker, H.: National Association of Corrosion Engineers Corrosion '91 Conference, Cincinnati,
Ohio, March II-IS, 1991, Paper No. 380.
1199. Vincent, J. 1., and Coon, C L.: The relative effect of paint film thickness on bimetallic and crevice
corrosion, Society of Automotive Engineers, Paper No. 860109,1986.
1200. Cormier, G. 1.: Effective cleaning of zinc and aluminum surface, Society of Automotive Engineers
Automotive Corrosion and Prevention Conference and Proceedings, Dearborn. Michigan, 1989, Paper
No. 892555, pp. 7-16.
1201. Yoshikawa, y., Itoh, M., Wakano, S., and Shiota, T.: The properties of organic composite coated Zn-Ni
plated steel sheets, Society of Automotive Engineers Automotive Corrosion and Prevention Conference
Proceedings, Dearborn, Michigan, 1989, Paper No. 892565, pp. 115-120.
1202. Van Eijnsbergen, 1. E H.: Duplex Systems, Elsevier, Amsterdam, 1994.
1203. Nartey, V. K, Binder, L.. and Kordesh. K: Identification of organic corrosion inhIbitors suitable for use
in rechargeable alkaline zinc batteries, 1. Power Sources 52, 217-222. 1994.
1204. Glaeser, W.: Gas evolution data on very-Iow-mercury-content zinc powders for alkaline batteries, Power
Sources 12, 265-296,1989.
1205. Yoshizawa, H., and Miura, A.: Mercury free alkaline manganese batteries, Prog. Batteries Battery Mater.
12,132-139,1993.
1206. Bhatt, D. P., and Udhayan, R: Electrochemical studies on a zinc-lead-cadmium alloy in aqueous
ammonium chloride solution, 1. Power Sources 47, 177-184, 1994.
1207. McBreen, J.: Nickel/zinc batteries, 1. Power Sources 51, 37-44, 1994.
1208. Strauven, Y. A. 1., and Skenazi, A.: European Patent 0441420 AI, 1991.
1209. Gonzalez,1. A., Lopez, W, and Rodriquez, P.: Effects of moisture availability on corrosion kinetics of
steel embedded in concrete, Corrosion 49, 1004-1010,1993.
1210. Thompson, N. G., Islam, M., Lankard, D. R, and Virmani, Y. P.: Effect of environmental factors on
corrosion induced deterioration of reinforced concrete bridge structures, Corrosion '95, National
Association of Corrosion Engineers International, Houston, Paper #4, 1995.
1211. Illston, J. M., Dinwoodie, J. M., and Smith, A. A.: Concrete, Timber and Metals, Van Nostrand Reinhold,
New York, 1979.
1212. Suda, K, Misa, S., and Motohashi, K: Corrosion products of reinforcing bars embedded in concrete,
Corros. Sci. 25,1543-1549,1993.
1213. Yamashita, T.: Corrosion of iron and zinc in concrete, Corros. Eng. 40, 345-357, 1991.
1214. Andrade, M. C, and Macias, A.: Galvanized reinforcements in concrete, Suiface Coatings-2, Chapter
5, pp. 137-182, Elsevier Applied Science, Amsterdam, 1988.
1215. Baker, E. A., Money, K L., and Sanborn, C 8.: Marine corrosion behavior of bare and metallic-coated
steel reinforcing rods in concrete, ASTM STP 629, pp. 30-50, American Society for Testing and
Materials, Philadelphia, 1977.
1216. Cornet, I., and Bresler, B.: Galvanized Reinforcement for Concrete-II, International Lead Zinc
Research Orgainization, 1981, p. 1.

REFERENCES

459

1217. Tonini, D. E., and Cook, A. R.: The performance of galvanized reinforcement in high chloride
environments-Field study results, International Lead Zinc Research Organization, 1981, pp. 57-88.
1218. Clear, K. c.: Federal Highway Administration, Washington, D.C., FHWNRD-82/028, December 1981.
1219. Slater, J. E.: Corrosion of Metals in Association with Concrete, STP 818, American Society for Testing
and Materials, Philadelphia, 1983.
1220. Federal Highway Administration, Publication No. SA-91-023, HTA-13/6-91 (2.5M) QE, June 1991.
1221. Hansson, C. M.: Oxygen diffusion through portland cement mortars, Corms. Sci. 35, 1551-1556, 1993.
1222. Baldwin, K. R., Robinson, M. 1., and Smith, C. J. E.: Galvanic corrosion behavior at electrodeposited
zinc and Zn-Ni coatings coupled with aluminum alloys, Br. Corros. 1. 29,293-298, 1994.
1223. Baldwin, K. R., Robinson, M. J., and Smith, C. J. E.: Galvanic corrosion behavior of electrodeposited
Zn-Ni coatings coupled with steel, Br. Corros. 1.29,299-304, 1994.
1224. Weast, R. C. (ed.): Handbook of Chemist(\' and Physics, 68th ed., CRC Press, Boca Raton, Florida.
1988.
1225. Zhang, X. G . Corrosion protection of rebar using a zinc wire, Corrosion 51. 679. 1995.
1226. Dugan, B. P., Stejskal. B . and Stoneman. A. M.: Survey of highway bridges containing galvanized rebar.
in Proceedings ofhllernational Conference on Corrosion and Corrosion Protection of Steel in Concrete.
Vol. I, R. N., Swamy (ed.). p. 157. Sheffield Academic Press, Sheffield, United Kingdom, 1994.
1227. Lee, H. H.: Galvanized steel with improved resistance to intergranular corrosion, Proceedings of the
Galvanizing Committee of the Zinc Institute, Chicago, December 1977.
1228. Baker. H. (ed.), Alloy Phase Diagrams, Vol. 3, ASM Metal Handbook. pp. 2.56. 2.206, ASM International. 1992.
1229. Bailey. R. W, and Ridge, H. G.: Corrosion of metals in buildings, Chon. Ind. 1957. 1222-1227.
1230. Edgar, T. V: In-service corrosion of galvanized culvert pipe, in EfieCls of Soil Characteristics 0/1
Corrosion, ASTM STP 1013, pp. 133-143, American Society for Testing and Materials. Philadelphia.
1989.
1231. Cabrillac, c., and Exertier, A.: The effect of coating composition on the properties of a galvanized
coating, in Proceedings of the 7th International Conference on Hot Dip Galvanizing, Paris, June 1964.
pp. 289-313, Pergamon Press.
1232. Whiting, D., Stark, D., and Schutt, W.: Galvanic anode cathodic protection system for bridge decksupdated results, Corrosion/Xl. International Corrosion Forum, Sponsored By the National Association
of Corrosion Engineers. Toronto. Ontario, April 6-10, 1981, Paper #41.
1233. Doig. P, and Flewitt. P. E. 1.: A finite difference numerical analysis of galvanic corrosion for
semi-infinite linear coplanar electrodes. 1. Electrochem. Soc. 126. pp. 2057-2063, 1979.
1234. Asphahani. A .. and Uhlig. H. H.: Stress corrosion cracking of 4140 high strength steel in aqueous
solutions. 1. Electrochem. Soc. 122, pp. 174-179, 1975.
1235. McCafferty, E.: Distribution of potential and current in circular corrosion cells having unequal
polarization parameters, 1. Electrochem. Soc. 124, 1869-1878, 1977.
1236. Morris, R., and Smyrl, W: Current and potential distribution in thin electrolyte layer galvanic cells, 1.
Electrochem. Soc. 136,3229-3236, 1989.
1237. Tierra, P, Bernal, M., Molera. P., and Fernandez, L.: Comparative study on the cathodic protection
supplied to steel by hot dip coatings, INTERGALVA '91, 3rd International Zinc Coated Sheet Conference.
Barcelona, June 1991.
1238. Wetzel, D.: Batch hot dip galvanized coatings, inASM Metals Handbook, Vol. 5, pp. 360-371, Amcrican
Society for Metals, Materials Park, Ohio, 1994.
1239. Compton, K. G.: Sources of underground corrosion potential differences, Corrosion 16, 87-90, 1960.
1240. Wilhelm, S. M.: Galvanic corrosion caused by corrosion products, in Galvanic Corrosion, H. P. Hack
(ed.), ASTM STP 978, pp. 23-34, American Society for Testing and Materials, Philadelphia, 1988.
1241. Suzuki, S., Kanamaru. T.. Arai. K., and Morita, J.: Edge corrosion resistance of various coated stecl
sheets, Corrosion/91. National Association of Corrosion Engineers Annual Conference and Corrosion
Show, Cincinnati. Ohio. March I I - I 5. 1991, Paper No. 415.
1242. Pryor. M. J., and Keir, D. S.: Galvanic cOITosion: current flow and polarization characteristics of the
aluminum-steel and zinc-steel couples in sodium chloride solution, 1. Electrochell1. Sue. 104,269-275,
1957.
1243. Pryor, M. J.. and Keir, D. S.: Galvanic corrosion: Effect of pH and dissolved concentration on the
aluminum-steel couple, 1. Electrochem. Soc. 105,629-635, 1958.

460

REFERENCES

1244. Munn, R. S., and Devereux, O. E: Numerical modeling and solution of galvanic corrosion systems,
Corrosion 47, 612-634, 1991.
1245. Huang, S., and Oriani, R. A.: The corrosion potential of galvanically coupled copper and zinc under
humid gases, Fall Meeting of the Electrochemical Society, Minneapolis, Minnesota, 1993, Abstract No.
91.
1246. LaQue, E L.: Corrosion by sea water, in Corrosion Handbook, H. H. Uhlg, (ed.), p. 383, John Wiley &
Sons, New York, 1948.
1247. Wellington, J. R.: The low potential zinc anode in theory and application, Corrosion 17, 122-128, 1961.
1248. Waber, J. T.: Mathematical studies on galvanic corrosion, J. Electrochem. Soc. 101, 271-276, 1954;
102,344-353,420-429,1955;103,64-72, 138-147,567-570, 1956.
1249. Gukild, 1., Husevag, R., and Markestad, A.: Reactivity of zinc and aluminum in fresh concrete, 5th
Scandinavian Corrosion Congress, 28/1-12, 1968.
1250. Townsend, H. E.: Continuous hot dip coatings, inASM Metals Handbook, Vol. 5, pp. 339-348, American
Society for Metals, Materials Park, Ohio, 1994.
1251. Wi art, R., Cachet, C., Bozhkov, c., and Rashkov, S.: On the nature of the "induction period" during the
electrowinning of zinc from nickel containing sulphate electrolytes, J. Appl. Electrochem. 20, 381-389,
1990.
1252. Wang, Y., O'Keefe, T. J., and James, W. J.: Voltammetric evaluation of zinc electrowinning solution
containing nickel, J. Electrochem. Soc. 127,2589-2593, 1980.
1253. Maja, M., Penazzi, N., Fratesi, R., and Roventi, G.: Zinc electrocrystallization from impurity-containing
sulfate baths, J. Electrochem. Soc. 129, 2695-2700, 1982.
1254. Rashkov, S., Petrova, M., and Bozhkov, c.: The effect of nickel on the mechanism of the initial stages
of zinc electrowinning from sulphate electrolytes, Part I, J. Appl. Electrochem. 20, 11-16, 1990.
1255. Rashkov, S., Petrova, M., and Bozhkov, c.: The effect of nickel on the mechanism of the initial stages
of zinc electrowinning from sulphate electrolytes, Part II, 1. Appl. Electrochem. 20, 17-21, 1990.
1256. MacKinnon, D. J., and Brannen, J. M.: Aluminum cathode effects in zinc electrowinning from industrial
acid sulphate electrolyte, J. Appl. Electrochem. 16, 127-133, 1986.
1257. O'Keefe, T. J., Chen, S. E, Cole, E. R., Jr., and Dattilo, M.: Electrochemical monitoring of electrogalvanizing solutions, J. Appl. Electrochem. 16,913-919,1986.
1258. Hagans, P. L., and Haas, C. M.: Chromate conversion coatings, in ASM Metals Handbook, Vol. 5, pp.
405-411, American Society for Metals, Materials Park, Ohio, 1994.
1259. Johansson, E., and Gullman, J.: Corrosion study of carbon steel and zinc, in Atmospheric Corrosion, W
W. Kirk and H. H. Lawson (eds.), ASTM STP 1239, pp. 240-256, American Society for Testing and
Materials, Philadelphia, 1995.
1260. Arnold, C. J.: Galvanized Steel Reinforced Concrete Bridge Decks, Michigan State Highway Commission, Research Report No. R-1033, Lansing, December 1976.
1261. Sun, X., and Tsujikawa, S.: Initiation of under film corrosion for zinc in galvanic contact with steel
under cyclic corrosion tests, Corros. Eng. 41, 741-750,1992 [in Japanese].
1262. Garcia-Anton J., Perz-Herranz, V., Guinon, J. L., and Lacoste, G.: Use of diiferential pulse polarography
to study corrosion of galvanized steel in aqueous lithium bromide solution, Corrosion 50, 91,1994.
1263. Hwang, J., and Zhang, X. G.: Cominco PTC Internal Report, 1995.
1264. Schikorr G.: Relation between the atmospheric corrosion of zinc and nickel and the sulphur dioxide
content of the air, Metal 15, 981,1961 [in German].
1265. Strekalov, P. v., Agafonov, V. v., and Mikailovskii, Y. N.: Prot. Met. 8, 521, 1972.
1266. Greenburg, L., and Jacobs, M.: Ind. Eng. Chern. 48,1517,1956.
1267. Sydberger, T., and Vannerberg: The influence of the relative humidity and corrosion products on the
adsorption of sulfur dioxide on metal surface, Corros. Sci. 12,775, 1972.
1268. Anderson, E. A., and Reinhard: Chemical removal of corrosion products in the determination of the
corrosion rate of in zinc, ASTM 139, 691, 1939.
1269. Hudson, J. c.: Sixth Report of the Corrosion Committee, Special Report No. 66, Iron and Steel Institute,
1959, p. 25.
1270. Rajagopalan, K. S., and Ramaseshan, G.: Relative corrodibility of zinc and steel in unpolluted
atmospheres, J. Sci. Ind. Res. (India) 18B, 87,1959.
1271. Barton, K., and Beranek, E.: Reaction mechanism of the atmospheric corrosion of metals in moist air
contaminated with sufur dioxide, Werkst. Korros. 10, 377, 1959.

REFERENCES

461

1272. Newman, K.: Composite Materials, L. Holliday, ed., Elsevier, 1966.


1273. Power, T. C. et al: 1. Am. C(lIlcr. Inst. 1954.
1274. Tuutti, K.: Corrosion of steel in concrete, presented at the 6th European Congress on Metallic Corrosion,
London, 1977.
1275. Jaenicke, w., and Scheitzer, H.: Z. Phvs. Chelll. (Frankfurt) 52,104, 1967.
1276. Aufenast, E: 1st International Symposium on Batteries, Christchurch. New Zealand, 1958.
1277. Stutz, G.: Ind. Eng Chem. 19(1),52,1927.
1278. Erbse, H., Hauffe, K., and Range, J.: Z. Phys. Chem. N.F 74, 248, 1971.
1279. Stark, D., and Perenchio, W.: The Performance of Galvanized Steel Reinforcement in Concrete Bridge
Decks, Portand Cement Association, Project No. 2e-206, 1975.
1280. Harris, 1. 0.: Soil in the con'osion process, in Corrosion, Vol. 1, L. L. Shreir, cd., p. 2.37, George Newnes
Ltd., London, 1963.
1281. Tang, N. Y: Comment on Fe-AI-Zn.1. Phase Equilihria 15, 237, 1994.
1282. Tang, N. Y, Adams, G. A., and Kolisnyk, P. S.: On determining effective aluminum in continuous
galvanizing baths, GALVATECH '95, Chicago, Iron and Steel Society, 1995, pp. 777-782.
1283. Schikorr, G.: Relation between the atmospheric corrosion of zinc and nickel and the sulfur dioxide
content in air, Metal 15, 981, 1961 lin Germani.
1284. Whiting, D. A., Nagi, M. A., and Broomfield, 1. P.: Laboratory evaluation of sacrificial anode materials
for cathodic protection of reinforced concrete bridge decks, Corrosionl95, Paper No. 513, National
Association of Corrosion Engineers, Houston, 1995.
1285. Saques, A., and Powers, R.: Sprayed-zinc sacrificial anodes for reinforced concrete in marine service,
Corrosionl95, Paper No. 515, National Association of Corrosion Engineers, Houston, 1995.
1286. Brousseau, R., Arnott, M .. and Baldock, B.: Laboratory performance of zinc anodes for impressed
current cathodic protection of reinforced concrete, Corrosion 51, 639-644, 1995.
1287. Brousseau, R., Arnott, M .. and Baldock, B.: Improving the adhesion of zinc coatings used to metallize
concrete, Mater. Perform. 33,40-42, 1994.
1288. Kessler, R. J., Powers, R. G., and Lasa, 1. R.: Update on sacrificial anode cathodic protection on steel
reinforced concrete structures in seawater, Corrosiol1195, Paper No. 516, National Association of
Conosion Engineers, HOllston. 1995.
1289. Wills, D. J.. De Lisco, A., and Gleeson, B.: Intergranular corrosion in continuously galvanized steel,
GALVATECH '95, Chicago, Iron and Steel Society, 1995, p. 139.
1290. Kehrer, H. P.: The effect of small impurities on grain boundary corrosion of zinc die castings, Metal!.
28,883, 1974 [in Germani.
1291. Keifets, V L., and Krasikov, B. S.: Dokl. Akad. Nauk SSSR 109, 586,1956.
1292. John, V: Durability of Galvanized Steel Building Components in Domestic Housing, Report No.
WL/SMPIl106EIl09IfD, British Steel, 1991.
1293. Zhang, X. G.: Unpublishcd results, Cominco Product Technology Centre, 1995.
1294. Zhang, X. G.: Galvanic corrosion of zinc and its alloys, 1. Electrochem. Soc. 143, 1472-1484, 1996.
1295. Zinc Chemicals, International Lead Zinc Research Organization, New York, 1983.
1296. Fountoulakis, S. G.: Continuous electrodeposited coatings for steel strip, in ASM Metals Handhook,
Vol. 5, pp. 349-359, Matcrials Park, Ohio, 1994.
1297. Sato, A.: Zinc plating, in ASM Metals Handbook, Vol. 5, pp. 227-235, American Society for Metals,
Materials Park, Ohio, 1994.
1298. Zhang, X. G., and Hwang, J.: Zinc wired rebar, National Association of Corrosion Engineers, Corrosion
'96, Paper No. :lOO, 1996.
1299. Townsend, H.: Bethlehem Steel Co., private communication.

Index
Absorption
atomic, 155
gas in ZnO, 97
light, 86, 94, 95. 98. 109-117. 350
water. 279, 236
Accelerate tests, 266, 317, 318, 325, 360, 368
Acceleration factor, 266
Accumulation layer, 107
Acetate, 30, 57
Acetic acid, 122, 40 I
Acetone, 178, 395, 396
Acidification, 91, 225, 226, 280
Acids, 249, 266, 276, 395, 401
Acoustic effect, III
Activation process, 16,45,126, 143, 188, 192,381
Activation energy, 72, 80, 135, 390, 399
Adhesion
coatings, 12,407
cOlTosion products. 163, 296, 169, 175
paint, 15.82.328-330
Adsorption
capacitance, 27
corrosion products, 135, 137, 149
current doubling, 113
dissolution and deposition, 34, 38
dissolution of ZnO, 119, 122
double layer, 97,101, 105
f1atband potential, 115
hydrogen evolution, 45
inhibitors, 384
intergranular corrosion, 59
passivation. 85. 91
pitting. 226, 227
in sol vents. 399
of sulfur dioxide, 244
of water layer, 242
Aeration, 141, 305
capacitance, 28
corrosion potential and current, 130, 133

corrosion rates, 30 I
dissolution, 34, 35, 36
impedance, 57, 62,
surface film, 85
Aerosol, 243
Aging, 231
Agitation: see Convection
Air, 243
aeration, 174
in concrete, 353
cooling, 12
corrosion products, 162, 174, 176- 178
pollutants, 243, 244
soil, 307
wet storage stain, 237
Alcohols, 112, 178, 293, 393
Alloying, 3
anode, 214, 215
corrosion potential. 144-149
corrosion rates, 258, 260, 385, 387
galvanic corrosion, 184, 198,216
intergranular corrosion, 227-229
Aluminum
adsorption of SO" 244
affinity to oxygen, 10
alloying element, 145,363,387,404
anode, 214,403
atmospheric corrosion, 258
corrosion potential, 145,263
galvanic corrosion, 184, 188, 197,202,210
intergranular corrosion, 181, 227 -231
Zn-AI phase diagram, 4
Aluminum oxide, 323
Amalgamation, 385, 391, 404
Ammonia, 285, 375
Annealing, 232, 257, 359
Antimony, 260
Anodes, 184, 186,212,213,305,403
alloying, 214, 215, 404

463

464
Anodes (cant.)
cathodic protection, 407
efficiency, 213, 215,403
passivation, 403, 407
Anodization, 85-87,115
Area
adsorption, 135
effective surface, 143,202, 343
gal vanic corrosion, 184, 196, 199-121, 214
polarity reversal, 206, 217
surface roughness, 143
Atmosphere, 241
corrosion forms, 184, 261
corrosion product, 158-168, 261
corrosion rates, 245-248
galvanic corrosion, 208
intergranular corrosion, 227
pitting, 220
Auger, spectroscopy, 157, 162
Automobile, 6, 215, 263, 315
Axis, 174
Bacteria, 244, 285, 295, 305
Band bending, 97-100, 104, 108
Band edge, 98,101-103,115,117,120
Band gap, 96, 98, 103, 108, III, 115, 118
Band structure, 96, 103-105, 114, 120
Barrier height, 106, 117
Batteries, 68, 373
cathode, 374
electrolyte, 374
pitting, 220, 225
Benzene, 78, 395,396
Benzoate, 299
Binding medium, 337, 344
Black film, 86
Blistering, 233, 315, 321
Bond strength, 359
Bonding, 159; see also Adhesion
Brass, 3, 184, 375
Breakdown
of passivation, 66, 89, 221, 225
pitting. 221
of semiconductor, 107-109, 115, 119
Bridge, 51, 351, 363
Buffering, 84, 91, 173
Bundle, 262
Cadmium, 184, 198,231,388,404
Capac itance
adsorption, 35
corrosion product film, 62, 63
double layer, 27, 55, 62, 63
space charge layer, 100, 103, 106
surface coatings, 56

INDEX
Capillary effect, 88, 237, 242, 245, 307, 353
Car: see Automobile
Carbon, 245,338, 346.374
Carbon dioxide
anodiztion, 86
corrosion products, 168, 170, 176
corrosion rates, 270, 287, 289
galvanic corrosion. 203
pitting, 219
polarity reversal, 203
solubility in water, 284
Carbonates
corrosion products, 158-181, 237, 278
dissolution. 31
formation condition, 23, 24 278
intergranular corrosion, 234
passivation, 77-80, 85-87
polarity reversal, 203
Carbonation, 358, 366
Casting, 7
Catalytic reaction, 34, 38, 45,108,113,269,338
Catalyst, 75
Cathodic protection, 213, 359, 407
Cells, 195,201,301,334,373
Cement, 352
Charge carriers, 96, 105, 109, 110
Charge transfer. 32, 59, 62, 10 I, 104, 115, 122, 333
Charge transfer coefficient, 35, 40, 44, 48
Chemicals: see Ions
Chipping, of paint, 328, 331
Chlorides; see also Ions
adsorption, 27, 28
battery electrolytes, 374, 384
complexing with zinc, 24
in concrete, 351, 357
conductivity, 13, 127
corrosion current, 128, 135
corrosion products, 171, 172, 174, 177
corrosive agents, 357
galvanic corrosion, 198-202
intergranular corrosion, 229
marine, 252, 255, 266, 291
pitting, 221-223
in solutions, 296
current doubling, 113
dissolution. 31. 124
flux, II
impedance, 60
oxygen reduction, 51
passivation, 78
passivation breakdown, 89
plating bath, 13
polarity reversal, 204
road salt, 263
salt spray test, 270

INDEX
Chlorides (cont.)
in solvents, 395, 399
vapor, 400
Chromium, 184. 260
Chromate, 85; see also Chemicals
Chromating,16
Chromate film
characterization, 16, 17, 171. 175
effect on corrosion, 144, 238, 260, 329, 362
properties, 17, 180
Clay, 305, 311
Cleaning, 11, 12
Cleavage, 165, 170
Climate, 254
Coating life, 256
Coating weight, 323
Coating thickness, 9
Coatings, 12, 14
Cobalt, 188
Colloidal solution, 16,305
Composite coatings, 323
Complexes, 25, 33-39
Condensation, 233, 237, 341, 252
Concrete, 184,227,351,403,407
Conductivity
concrete, 354
corrosion products. 187
galvanic corrosion, 188,213
metal. 2
pitti ng, 225
soil, 305, 307, 310
solution, 26, 27, 301
surface films, 74, 79, 115
waters, 285
zinc oxide, 95, 96, lIS
Contact line, 195,218,277
Contaminants, 161, 163, 176,241,265
Convection, 72, 75. 88, 139, 141,275
Conversion coatings, 16,260
Conversion factors, 129, 130
Cooling, 12, 162
Coordination number, 25
Copper, 145,245,391
alloying elements, 145, 146,307
galvanic corrosion, 184, 194, 196,210
Corrosion, 54
Corrosion current, 125,314,365
calculation, 125-127
correlation to weight loss, 153
definition, 125
measurement techniques, 127, 128
relation to corrosion potential, 130, 368
Corrosion environments
atmosphere, 241
batteries, 373

465
concrete, 351
Corrosion environments (cont.)
gaseous, 399
lab testing chambers, 267, 270, 272
organic sol vents, 393
paints, 315
soils, 305
solutions, 296
waters, 283
Corrosion forms, 183,293
Corrosion potential, 125-156
in concrete, 363, 367
definition, 125
painted, 332
pitting. 224
relation to corrosion current, 153
respect to passivation. 68. 85, 207
in soil, 211, 212. 312
in solution, 85.133-153
zinc-rich paints, 347, 348
Corrosion products, 157,261,399
characterization techniques, 157
classification, 157
corrosion current, 141, 149. 152
corrosion potential, 149, 152
effect on corrosion, 54, 178
effect of climate, 254
galvanic corrosion, 184,202,210.224,340
impedance, 62, 63
oxygen reduction, 51, 53
relative to weight loss. 272
stability, 31, 165, 170, 176
transformation, 168-176
Corrosion rates, 130
atmospheric, 245
in concrete, 360-363, 366
galvanic corrosion, 208-212
gaseous, 400-402
intergranular corrosion, 228-234
metals other than zinc and steel, 248, 292
in organic sol vents, 395-398
pitting, 219
simulated tests, 266-277
in soil, 308-311
in solutions, 296-304
under-paint, 319
units, 129, 130
in waters, 286-296
Corrosion tests, 211, 237, 317, 325
Cosmetic corrosion, 319, 320, 360. 378
Countries, 248
Cracking, 183,238, 358, 364
Creep, 2
Creeping, 319-323
Crevice, 218, 236,357

466
Crevice corrosion, 183
Cross section, 107
Crystal
polycrystal, 115-117, 144
powder, 94
size, 94, 171, 175,320
zinc single crystal, 27, 32, 47 165, 174
Crystal structure
corrosion products, 160, 161, 165, 175
zinc, I
zinc oxide, 94
Crystal orientation
breakdown, 108
capacitance, 28
cast and rolled products, 2
corrosion potential, 143
corrosion rates, 143, 302
decomposition, 122
dissolution and deposition, 35
photocurrent, III
pitting, 225
standard potential, 20
Curing, 329, 337, 352
Current
dark current, 105
distribution, 184, 192, 194
limiting current, 31
peak current, 68, 69, 82, 83
saturation current, 110
versus potential curves: see I-V curves
Current collector, 375, 390
Current distribution, 192
Current doubling, 111-113
Current efficiency, 32, 79, 80
Current output, 404
Curvature, 39
Cut edge, 216, 262
Crystallization, 16, 323
Cyclic test, 216, 272, 318, 325
Cyclic wet/dry pattern, 254, 272
Deaeration: see Aeration
Decomposition
corrosion products, 173, 176
HP2,51
zinc oxide crystal, 110, 119-124
Deformation, 303, 331
Deicing, 263
Delamination, 315, 333
Dendrite, 39, 377
Depassivation,91
Depletion layer, III
Depolarization, 207
Deposition, 36-39, liS, 117

INDEX
Depth
intergranular corrosion, 227
pitting, 217, 220, 225
sea water, 285, 294
Desorption, 34, 101; see also Adsorption
Dew, 242
Dezincification, 149
Die casting alloy, 3, 7, 227
Dielectric properties, 74, 89, 395
Diffusion
in concrete, 354
deposition, 39
dissolution, 31, 35, 162
of electron and hole, 110
of hydrogen, 40, 139
of mercury in zinc, 386
in organic solvents, 396
in paint, 316
passivation, 62, 69-73, 77, 80, 84, 88
of salts, 316, 357
of various ions, 26
of zinc in zinc coatings, 8
in zinc oxide, 97
Diffusion coefficients, 25, 40, 71, 190, 316, 354
Diffusion length, 110
Discharging, 377, 390
Disintegration, 32
Distance
from city center, 244
gal vanic action, 20 I, 209, 213
from sea coast, 244, 252
from waterline, 301
Dislocation, 39, 225
Dissolution, 29
at corrosion potential, 125-129
corrosion products, 30, 36, 168, 171
efficiency, 32
exchange current, 30, 31
inhibition, 62
paint delamination, 333
at passivation, 66, 80, 87-89
pitting, 226
process, 62
of zinc oxide, 119-124
Distilled water, 140, 169, 170, 286
intergranular corrosion, 227
pitting, 217, 224
polarity reversal, 203, 204
Donor,96, liS, 118, 122
Doping, 117
Double layer, 27-29, 55, 62,115, 173,396
Drying
accelerated test, 267, 325
battery, 221

INDEX
Drying (coni.)
concrete, 356
corrosion current, 144
corrosion products, 160, 177, 242, 254
wet storage stain, 257
Ductility, 227, 238, 359
Duplex coating, 323
Dust, 215, 245, 316
Dye, III
Edge, 196,217,277
Electrochemical techniques, 29, 54, 153
Electrogalvanizing, 13, 14
Electrolumineseces, 114
Electron acceptor, 96
Electrolyte
battery, 374, 379
conductivity, 26, 27. 191, 192
near surface, 32
pitting, 225, 227
resistance, 55, 62, 144, 211
thickness, 190, 192-197,213,276
Electroplating, 13, 14,260
Electrowinning, 19, 39
Elevation, 252
EMF series, 144, 185, 186,211
Enthalpy, 390
Epoxy, 339
Equilibrium, 19,20-25, 120,127,133
Equivalent electrical circuit: see Impedance
Etch, 74, 102,221
Eutectic phase, 3, 13, 323
Exchange current, 30, 36, 42,125,127,130,151,
396
Extracts, 356
Extrusion, 232
Faradaic process, 32
Fatigue, 369
Fermi level, 96, 98, 100-103
Field strength, 90, 104
Film: see Surface films
Flake, 7
Flame: see MetaJlizing
Flat band, 97, 98
Flatband potential, 100
definition, 98
determination, 100
oxide film, lIS, 116, 118,202
single crystal, 100, 10 I
Flow, 144, 178, 184,266,289,294,348
Flux, II, 110,237,354
Fluxing, II
Fly-ash,269
Fog, 243

467
Formability, 321
Formic acid, 112, 113
Galfan, 12, 162,259,263
Galvalume, 12, 162,263,324
Galvanneal, 12, 145, 147, 148,263
Galvanic corrosion, 183, 261, 312, 332, 339
corrosion rate, 198,201,210-213
of coupled metals, 184, 196-198, 210
definition, 183
factors, 188, 209
testing, 184,209,211
theory, 185-196
Galvanic current, 185, 191, 194-197,213
Galvanic protection, 213
anodes, 215
coatings, 215
definition, 185
field data, 212
throwing power, 213
zinc-rich paint, 339
Galvanic series, 187,211
Galvanizing: see also Electroplating
batch process, 10
classification, 7, 8
continuous process, II
Galvanized steel, 256
coating methods, 7, 256
galvanic protection, 213
life, 256
pitting, 203, 219, 225, 308
polarity reversal, 203
premature darkening, 262
surface treatment, 12
wet storage stain, 183, 236
Galvanized rebar, 359
Gases, 184,339
CO,: see Carbon dioxide
H,: see Hydrogen
H,S,243,268,285,401
N,,242
NH 3,243
0,: see Oxygen
SO,: see Sulfur dioxide
Gassing, 224, 338, 378
Gel, 353, 382
General corrosion, 183, 225
Geometrical effect, 185, 186, 189-192, 213
Gravitation, 217
Grain, crystal
orientation, 2
potential, 144
shape, 74
size, 3, 230, 235, 388
Grain boundary, 3, 32, 225, 352

468
Grid,390
Grinding, 143
Hardness, 284
Hardware, 213
Heat treatment, 231
Highway, 262
Hole, 96, 217
Hot dip: see Galvanizing
Hot water
corrosion products, 168-170, 179
corrosion rates, 288, 289
pitting, 219, 291
polarity reversal, 203, 208
Humidity
atmosphere, 241
intergranular corrosion, 226
test, 267
time of wetness, 248, 252, 280
wet storage stain, 183, 236
Hydration number, 25
Hydration process, 36, 45, 67, 352
Hydration shell, 39
Hydrogen, gas
anode,214
corrosion rate in water, 299
diffusion, 40, 139
intergranular corrosion, 235
measurement device, 378
rebar/concrete surface, 359
solubility, 40
surface preparation, 78
units, 129
Hydrogen embrittlement, 183, 238
Hydrogen peroxide, 51, 107, 112
Hydrogen reaction, 39
corrosion potential, 134
exchange current, 40, 42
inhibition, 47
over potential, 41, 45, 230
oxide film, 45
thermodynamics, 21
Ice, 237
Illumination, 109
Impedance, 54
corrosion current, 115
dissolution, 35
equivalent circuit, 56, 62
paints, 318, 340
polarization resistance, 127
semiconducting film, 118
spectrum, 56-61, 74
thin-layer electrolyte, 277

INDEX
Impurities, in
electrodes, 47
electrolyte, 39, 51, 227
grain boundaries. 235
water, 207
zinc, 47, 183,258,377,404
zinc oxide, 93, 96, 100
Incubation, 319
Indium, 387
Indoor atmosphere, 16 I. 162, 264
Inductance, 55
Inhibitors, 359, 384
Interfaces, 55, 67, 90, 103, 137,235,316,339,359
Intergranular corrosion. 129, 183.225.239,407
Intermetallic compounds, 3. 8-12, 145, 227-238
Ionic properties, 25-27
Ionic strength, 72
Ionization, 135
Ions. in solutions
Ag+, 114, 175
As'+, 46, 112, 382
AsO~-, 85, 89
BO;-, 89, 116, 117,221
Br-, 51, 60, 89,113, 124, 135,383
BrO,,32
Ca2+~ 85. 137. 140, 174, 179,205,357
Cd'+, 220
Ce ' +, 137, 180
CH,CO;,89
Ct: see Chlorides
CW, 13, 112
C10:;, 32, 137
CIO.;, 28, 51, 89, 113, 128. 171,384
Co2+, 298
col-: see Carbonates
complexing agents, 25
concentration, 298
corrosion current, 131, 135, 145, 150
corrosion potential, 131, 135-137, 145, 149
corrosion products, 172, 177
corrosion rates, 297, 298
current doubling, 113
crO;-, 16,31,89,223
Cu'+' 46,175,201,291,382
diffusion coefficients, 26
exchange current, 30
F-,89
Fe'+, 46, 382
Fe(CN)t, 103
f1atband potential, 101
galvanic corrosion, 184,201,204
HCOO-112
HP-, 51, 53
hydrogen evolution, 41

469

INDEX
Ions, in solutions (collt.)
1-,89, 107. 113, 123
In'+.383
I-V curve ofZnO. 106
K+, 45
Mg2+, 201
MnO:;, 31, 113
MoOt, 89
Ni 2+, 39, 46. 382
Na+,201
NH;, II. 13,28,34,46,60,128, 173,244,374,
384
NO~, 30, 51, 89,107,112,135,137,140,205
passivation, 66, 89, 90
Pb2+, 36, 47. 73, III. 220. 383
pitting, 222
PO~-, 51. 115, 139. 171,223
S2-,296
Sb'+, 46, 382
SiOl-,205
Sn 2+, 31, 46, 73, 383
SOl-, 137
S01-, 14,40,51,57, 113, 135, 140, 171,201,
244,384
WO~-, 31. 135, 136, 223
Zn 2+, 133
battery, 373
current doubling, 113
corrosion current, 134
corrosion potential, 130, 133. 138
dissolution, 29, 63, 396
hydrogen reaction, 46
photo decomposition, 122
standard potential, 19
Zn(OH)i-. 113
battery, 381
corrosion potential, 134
dissolution, 29, 63
hydrogen reaction, 46, 47
passivation, 75, 77
solubility, 22
Iron, steel
adsorption of S02' 244
alloying element, 198,216,293,404
concrete reinforcement, 358
corrosion potential, 145,312
corrosion products, 202, 332
corrosion protection, 208, 210-217, 259
corrosion rates, 208, 210, 212
atmospheric, 246, 252
gaseous, 402
soil,310
under-paint, 263
Fe-Zn phase diagram, 5

galvanic action, 184, 188.203,208. 345


Iron. steel (cant.)
intermetallic compounds, 3, 219, 279
phosphatising, 330
solubility in zinc, 3
IR drop, 74, 403
Isotherm, 105
I-V curves
breakdown of ZnO, 108
cathodic reaction on ZnO, 106
corrosion current, 128
decomposition ofZnO, 121
galvanic corrosion, 184,200
intergranular corrosion, 233
passivation, 67, 68, 73. 74, 79-84
passivation breakdown, 89
photocurrent, 110
pitting, 221
in soil, 313
in solvents, 399
zinc rich paints, 345
Kelvin probe, 184
Kinetics
corrosion processes, 125-127
dissolution, 29-39
electrochemical techniques, 29
galvanic corrosion, 188-191
hydrogen reaction, 39-46
oxygen reduction, 48-53
passivation, 68-85
semiconductor electrochemistry, 105-115
Kink sites, 35, 122
Lattice constant, I, 94
Leaching, 356, 358
Lead, 258
galvanic corrosion, 184,210
intergranular corrosion, 227, 229-231
Leclanche cell, 38, 220, 225, 373
Life, 256, 333, 360, 375
Lifetime, 113. 115, 346
Light
absorption, 86, 94,109-115,350
deflection, 94, 95
emission, 114, 115
penetration depth, 110
transmission, 94, 95
Linear polarization, 54
Loam, 305
Local corrosion, 185, 200
Localized corrosion, 66, 91,129,225
Luminescence, 114

470
Magnesium, 145,260,388
anode,214,403
gal vanic corrosion, 184, 210
intergranular corrosion, 229
Manganese, 145, 146
Mass transport, 71, 77, 88
Mechanisms
atmospheric corrosion, 278
carrier recombination, III
current doubling, III
deposition, 38
dissolution, 32
formation of corrosion products, 181
hydrogen reaction, 43
intergranular corrosion, 235
metallic surface film, 86
oxygen reaction, 49
passivation, 75, 80
passivation breakdown, 89, 225
pitting, 225
polarity reversal, 207
premature darkeni ng, 262
protection by zinc rich paint, 339
under-paint corrosion, 333
wet storage stain, 236
Mechanical properties, 2. 90. 231. 234
Melting point. I
Meniscus, 30 I
Mercury. 385
Metallizing. 6, 407
Metallography. 8, 9. II
Metallurgical effect, 229. 302
Metals (other than zinc and steel)
adsorption, 244
atmospheric corrosion, 248
gal vanic corrosion, 184. 198. 210
hydrogen exchange current, 42
sea water corrosion. 292
Microstructure. 232
Mill scale, 347
Minerals, 283
Mixing, 53
Mobility. 25-27. 36
Moisture. 175,237,242.355,400
Monolayer, 29, 78. 79, 82
Morphology, 39, 74. 75,163-165,172.175,244
Mortar. 358, 365
Mott-Schottkyequation. 100
Nernst equation. 133, 138
Neutralization, 16
Nickel, 184, 198,210,260.387
Normal corrosion, 185.210,211
Nucleation, 16,77,79.171

INDEX
Ohmic effect
CUlTent measurement, 156
gal vanic corrosion, 188. 190. 196
sacrificial protection. 403
Open circuit condition, 68. 82. 87. 125, 188,225.
345
Orbital. 103
Organic substances. 384. 393. 402
Oscillation. 39, 74
Osmatic pressure, 316
Overpotential, 127. 188.386
anode. 215.403
dissolution. 33
hydrogen evolution, 41. 43
oxygen reduction. 50
passivation. 66
Oxidation
of zinc, liS. 162
of water. 120. 122
Oxide: see Zinc oxide
Oxygen
in concrete. 354
diffusion, 51. 53. 274,195.334
diffusion coefficient, 48
effect on corrosion, 141-143, 181
evolution, 78, 88
galvanic corrosion, 190
intergranular corrosion, 287
kinetics, 48-54, 202
paints, 334, 339
pitting, 218
polarity reversal, 203, 206-208
stability, 21
standard potential, 48
solubility, 48, 284, 298
Paint, 328. 337
adhesion, 15,82.328-330
galvanic corrosion, 184
paintability, 12, 14
permittivity, 316
properties, 316. 337
thicknesi>, 324
Particles, 339, 341, 343, 352, 389
Particulate film, liS, 117
Passivation, 65-92
in batteries, 375, 376
breakdown, 66. 81, 87. 89-91. 221, 225
condition, 21, 23, 67, 78
definition. 65, 87
effect on corrosion, 130-133, 138, 191,233
overpotential, 66
passivation current, 67, 75, 87-89
passivation potential, 67,81, 84, 85

INDEX
Passivation (COlli.)
polarity reversal 207-208
process. 67. 74-77
time to passivation. 65. 70-73. 77. 78
Passive films: see a/so Surface films
characterization. 75. 78
properties. 75. 93.115-118.149.179.207
stability. 67. 86-92. 299
Passive state. 65. 67. 87. 88, 208, 363
Paste, 352
Permeation, 176. 178. 215, 316, 339, 353
Perforation, 221, 319
pH, 137.206.298
buffer, 82. 84, 173
capacitance. 27
concrete. 357
corrosion current. 131. 137-140, 148, 152
corrosion potential, 131, 137-140, 148, 152
corrosion products. 171-173. 273
corrosion rates, 273. 298, 311
dark potential, 102
decomposition ofZnO, 119, 122
flatband potential, 102-105, 115, 116
intergranular corrosion, 233
passivation, 66, 80, 83, 299
passivation breakdown, 89
pitting. 222
polarity reversal, 206
soil, 309-311
solubility, 22. 23. 299. 357
waters. 285
pH-potential diagram, 21,23,374
Phase diagram, 4, 5
Phosphates, as chemicals
corrosion potential. 136
dissolution, 31. 62
oxygen reduction, 5 I
passivation, 80-84, 86, 89
zinc-rich paint, 346
Phosphate coating, IS, 171, 180,329,334
Phosphating. 15,80.260.329
Photo: see Light
Photo catalysis, 113. 269
PholOcurrent. 95, 98. 100. 109
current doubling, III
dissolution of ZnO. 121-123
luminesces, 115
passive film, 116, 117
Photoeleetrochemistry, 109-118
Photoluminensece, 114
Photopotential, 109, 110, 117, 207
Physical properties, I. 93, 94
Pickling. I I
Pigment, 316, 338. 341, 345
Pipe, 217. 262, 291, 312

471
Pitting, 217, 290
characterization, 66, 89. 217
corrosion current, 129
corrosion rate 219-220
galvanic, 184
in hot water, 290
occurrence, 202. 217, 261, 263. 293. 308
measurement, 217
mechanism, 89-91
Pitting potentia!, 66, 89, 217, 221-224, 226
Plastics, 184,213,218,401
Plate, 209
Polarity reversal, 179, 184, 202-208. 219, 407
Polarization resistance, 127
conductivity, 367
corrosion rate, IS I, 154
impedance, 51
zinc-rich paint, 342, 349
Polarization parameter, 192
Polishing, 32, 143.225
Pollutants, 242, 249, 269
Polymers. 316, 338
Polymerization, 82
Porosity, 74, 353
Potential, electrode
breakdown potential. 66
dark potential, 102, 104
half-wave potential, 51
Helmholtz potential, 101-105
mixed potential, 125
open circuit potential. 68,82,87. 125, 188
photopotential, 109, 117,207
pitting potential, 221
rest potential, 102, 104, 125,221, 222
reversible potential, 39, 65, 138, 185
standard potential, 19. 39
Potential barrier: see Surface barrier
Potential bias, 105, 114
Potential distribution, 184. 192, 195
Potential of zero charge, 28, 104
Pourbaix diagram, 21
Powder, 170, 177
Premature darkening, 262
Pressure, 170, 291
Primer, 316, 323
Quenching, 232
Quantum yield, efficiency, 110, 113-117
Radiation, 241
Radicals, 112, 114,285
Rain, 158, 163, 164,242
galvanic corrosion, 211
synthetic, 271
Rebar, 359

472

Recombination, 98,110, III, 113-115


Recrystallization, 2, 12
Red rust, 256, 320
Redox couple, 98,105-108
corrosion potential, 127
dissolution of ZnO, 122, 123
Redox potential, 120
Repassivation,91
Resin, 316
Resistivity: see Conductivity
Resource, zinc, I
Rock,305
Rolling, 303
Rolled zinc, 7
Roof,249
Rotating electrode, 40, 51, 68-70, 82, 123, 143
Roughness: see Surface
Rubber, 337,401
Runoff, 164,250
Sacrificial protection: see Galvanic protection
Salinity, 252
Salt, 358
Salt layers, 67, 88, 223, 266, 270, 318, 325
Salt spray test, 266, 270, 318, 325
Salvation, 36
Sample, for testing, 188, 195, 196
orientation, 72, 191, 254, 262
position, 201, 301
preparation, 143
shape, 191, 254
size, 191,201
Sand,305,311,352
Scanning rate, 156
Scribe, 323
Sea water, 169-171,284,291
corrosion rates, 291
galvanic corrosion, 184, 187,211
intergranular corrosion, 227
Sealing, 313, 377
Season, 158,247,249
Self corrosion, 214, 403
Semiconducting behavior, 75, 89, 93-124
polarity reversal, 202, 208
Sensitization, 110
Sheet, 237,255,258,262
Sheltering
corrosion products, 160-166
corrosion rates, 255
galvanic corrosion, 211
Sherardising, 7
Ship, 215
Silt, 305
Silicate, 313, 338
Silicon, 9, 257

INDEX
Silver, 260, 387, 391
Slip, 3
Snow, 242
Soil,305
galvanic corrosion, 184,211
pitting, 220
Solidification, II. 353
Solubility
air, 49
alloying elements, 3, 229, 230
hydrogen, 39
minerals, 284
oxygen,48,49, 298
passive films. 87. 91, 299
salts in organic solvents, 393
zinc compounds, 22, 66, 91, 158, 171. 299
Solutions, 171-176,201,204,219.296; see also
Electrolytes and Ions
Solvation, 274, 280, 393, 396
Solvent, 170.393
Space charge layer, 98, 100. 104-106, 108
Spalling, 358, 364
Spangle, 10
Spectra, 116, 117
Splash zone, 294, 365
Spray, 6, 7
Stability
carbonates, 23
concrete, 358
corrosion products, 31, 165, 170, 176
hydroxides, 21
oxide, 21,119-124
passivation, 87-92, 299
Stain. 257, 268, 349
Stainless steel. 184, 196, 198, 210
Steam, 235
Steel: see Iron
Steady state, 35, 135, 153. 155
Stern-Geary constant. 127
Stirring, 139. 168, 173
Strength, 227, 234, 317, 352. 359
Stress, 90, 231, 239, 303
Strain, 332
Stress corrosion, 183. 184, 238
Superplasticity, 2
Surface area, 143, 149,341.343,375,390
Surface barrier, 105-110
Surface condition, 143, 144.202,347,363
Surface films; see also Corrosion products and
Passive films
catalytic effect, 45, 75
classification, 67
color, 74, 75, 80, 85-87, 204
corrosion potential, 207
formation, 75-78, 80, 84, 399

473

INDEX
Surface films (conI.)
impedance. 62. 63. 79
inhibitive effect. 3 I. 5 I
metallic, 73, 80, 86
nature, 75, 77-80, 84-86, 89, 178-181
phase transformation, 168-17 I. 208
semiconducting, 102, 115-118,202
stability, 78,87-92
Surface roughness, 149, 150
Surface states, 101.103, 105, 107, 117, 140
Surface structure, 242
Sulfur dioxide
adsorption, 245
air pollutants, 243
corrosion mechanism, 280
corrosion rates, 249, 280, 400
intergranular corrosion, 233
solubility, 249
Surface treatment. 260, 302, 388
chromating, 16, 180
gal vanizing, 12
phosphating, 15
plming, 13
ZnO crystal, 102, 103, I I 1
Sweep rate, potential, 69, 80, 82, 88
Tafel slope
corrosion current, 125-127, 153, 156
determination, 127, 128, 156
dissolution, 30, 33, 34, 36-38
hydrogen reduction, 40, 41, 48
oxygen reduction, 49
Tank,208, 214,219,237,268,403
Temperature, 140,203
anode efficiency, 406
corrosion products, 168-170
corrosion rates, 14 I. 390
passivation, 78, 80, 85
pi tli ng, 217, 220
polarity reversal, 203
potential,20, 140, 141
semiconductor, 97
solubility, 48
Texture, 10, 15, 232, 303, 306, 390
Thermodynamics, 19-25
Thin-layer electrolyte, 274
capacitance, 28
concentration, 276
corrosion potential, 142
gal vanic action, 176, 193, 333
hydrogen reduction, 276
oxygen reduction, 53, 279
Throwing power, 213, 217, 333
Tidal zone, 293, 294, 364
Time, effect, 149, 144,206

concrete curing, 353


Time, effect (conI.)
corrosion current. 144, 150, 152
corrosion potential, 149. 151. 152
corrosion products, 158, 163-165, 170
corrosion rates, 253, 293
galvanic corrosion, 193
gassing, 379
paint curing, 339
polarity reversal, 206
Time of the day, 253
Time of the year, 163
Time of wetness, 248, 252, 280
Tin, 184, 198,230,260,387
Titanium, 184, 185, 196, 198
Transient effect, 35, 154, 184
Transpassive state, 65
Transport number. 27
Tunnelling, 108. 109. 115
Twinning, 3
Ultraviolet light, 94, 317
Under-paint corrosion. 315
Under-vehicle corrosion. 263
Underground. 213. 217, 305
Uniform corrosion, 185
Valence, 32
Vanadium, 260
Vapor, 178.228, 269. 387
Viscosity, 73, 393
Wall. 249
Water, 9
in concrete, 355
content in air, 242
corrosion products, 168-171. 177
intergranular corrosion. 227
pitting, 219
polarity reversal, 203
in solvents, 393
Water drop, 274
Water line, 301, 31 I
Water spray, 270
Weather, 252, 339
Weathering, 305, 347
Weight loss
relation to corrosion current, 129. 151-156
relative to corrosion products, 163
units, 141-143
Wetting, 177,241,271,325,344
Wet storage stain. 183.236.262
White rust. 237
Wire, 176,209. 254. 359
Wind. 242. 245

474
Woods, 401
X-ray, 94, 157,207,224
Yellow rust, 257
Zinc alloys; see also Alloying and Zinc coatings
battery anodes, 387
corrosion current and potential, 144-149
corrosion products, 176
die-casting. 3, 7
sacrificial anodes, 403
Zinc coatings, 12, 14,321. 407
applications, 6. 322
classification, 3
composite coatings, 323
corrosion products, 180
duplex coatings, 337
gal vanic corrosion, 216
production methods, 7-14
under-paint corrosion, 321
under-vehicle corrosion, 263
Zn-AI, 12, 145,324

INDEX
Zn-Co, 118, 145-147, 181
Zinc coatings (cant.)
Zn-Cr, 145,216
Zn-Cu, 145. 146,216
Zn-Fe, 12, 145, 147, 148,263,323
Zn-Mg. 145,216
Zn-Ni, 118, 145-147, 184,216
Zn-Ti, 145,216
Zinc dust, 7, 316, 338, 341
Zinc oxide, 93-124
applications, I, 93, 338
corrosion products. 157.263,399
oxide films: see Surface films
photo electrochemistry. 107
physical properties, 93-97
semiconducting property, 97
stability, 107-109
Zinc powder, 7, 375, 381,389
Zinc products, 3, 6
Zinc-rich paint, 183, 184,215,337
Zinc tape, 7
Zincate, 382
Zincrometal, 338

Vous aimerez peut-être aussi