Case Studies in Construction Materials 8 (2018) 97–115

Contents lists available at ScienceDirect

Case Studies in Construction Materials

journal homepage: www.elsevier.com/locate/cscm

Case study

Low carbonation of concrete found on 100-year-old bridges

T
a,⁎ a b
Ivan Janotka , Michal Bačuvčík , Peter Paulík
a
Building Testing and Research Institute, Bratislava, Slovakia
b
Faculty of Civil Engineering, Slovak University of Technology, Bratislava, Slovakia

AR TI CLE I NF O AB S T R A CT

Keywords: This work investigates an unexpectedly low carbonation depth found in two bridges. A protection
Old bridge against carbonation is ensured by approximately 2–3 mm thick dense plaster coat covering the
Concrete outer surface of concrete. The plaster coat consists prevailingly of compact carbonate micro
Carbonation particles, showing no open pores and such a density, which gives this extremely thin layer non-
Plaster coat
permeability property for carbon oxide penetration over time.
Permeability
Durability

1. Introduction

A carbonation leads to a reduction in a total porosity whereas a molar volume of a reaction product CaCO3 is higher (36.93 cm3/
mol) than a primary reactant Ca(OH)2 (32.29 cm3/mol) [1]. Ca(OH)2 is gradually converted into CaCO3, which reduces the alkalinity
of the concrete [2]. Due to the reduction in a total porosity, carbonation reaction firstly reduces the permeability [3], increases the
micro hardness of the cementitious matrix [4], changes a pore size distribution at the same time by increasing the proportion of
capillary pore sizes greater than 30 nm [5]. Due to an increase in capillary porosity, further progressing carbonation increases the
diffusivity of ions through the cementitious matrix and finally the permeability [6]. This results in more extensive carbonation
shrinkage connected with crack propagation. Four carbonation stages were determined by XRD and TG-DTA techniques combined
with the set of 4 indicators and steel corrosion measurements by potentiodynamic method as shown in Table 1 [7,8]. Various colored
changes respond to different pH values [9]. The rate of carbonation was determined by a degree of carbonation (DC) and degree of
modification changes [DMC was given by CaOCARB/CaOTOTAL ratio (% wt)] [10,11]. Three distinct regions of fully carbonated, partly
carbonated and non-carbonated zones in carbonated concrete were distinguished by Cheng-Feng Chang [12]. The fully carbonated
zone was identified by the degree of carbonation greater than 50% wt. (pH < 9.0). The degree of carbonation in the partly car-
bonated zone lied between 0% wt. and 50% wt. (9.0 < pH < 11.5).
The non-carbonated zone occurs where the test specimen shows no signs of carbonation. Rates of carbonation for reinforced
concrete bridges were investigated for three localities in South Africa. Carbonation data from approximately 90 in-service bridges
aged between 11 and 76 years showed average carbonation rates approximately 0.3–0.7 mm/year. The carbonation coefficients
correlated well with the permeability for each of the binders [13]. Assessment of carbonation resistance through the use of Torrent air
permeability test was performed [14]. Carbonation depth and carbonation-induced steel corrosion of the concrete bridge depended
on the concrete quality: sound, cracked and joint parts [15]. Organic coatings on the concrete surface substantially reduce the risk of
carbonation. Under real circumstances, the coatings degrade naturally as a result of heat, moisture and ultraviolet radiation [16].
Degradation of concrete was studied on 100-year-old bridges. The objective of this work was to investigate the unexpectedly low
carbonation depth found in two bridges.

⁎
Corresponding author.
E-mail address: janotka@tsus.sk (I. Janotka).

https://doi.org/10.1016/j.cscm.2017.12.006
Received 26 May 2017; Received in revised form 4 December 2017; Accepted 12 December 2017
2214-5095/ © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Table 1
Assessment of carbonation attack of concrete.

TG-DTA, XRD (SEM, MIP) 4 indicator´s method Carbonation stage Potentiodynamic method

Degree of carbonation (% wt.) Carbonation stage pH in profile pH of concrete extract Electrochemical state of steel

< 55 I pH > 11.5 0 0–I pH > 11.7 passive

10 < pH < 11.5 I 11 < pH < 11.7 non-stable
55–65 II 9 < pH < 10 II
65–80 III 8 < pH < 9 III pH < 11 active-corroded
> 80 IV pH < 8 IV

Fig. 1. Reinforced concrete bridge Krásno nad Kysucou before reconstruction in 2014.

2. Experimental investigations

2.1. Materials

The first investigated road bridge is situated above the river Bystrica in the city Krásno nad Kysucou. It is believed to be the oldest
preserved reinforced concrete bridge in Slovakia and one of the oldest Monier Arches in Central Europe, which are still in service. The
bridge was completed in 1892 as one of the series of Monier Arches built in the former Austrian-Hungarian monarchy. It has survived
two World Wars without large damage and served till 2014 without any major repair. The bridge consists of two reinforced concrete
arches built on stone abutments and stone pier, which were a part of the previous stone arch bridge. Each of arch has a span of 16.8 m
and a span to raise ratio of 0.138 m. The thickness of primary arch varies from 400 mm in the springing to 150 mm in the crown of the
first arch and only 130 mm at the crown of the second arch. Above the primary arch, which is reinforced at both surfaces, there is also
an unreinforced overfill that reaches the thickness of up to 600 mm near the springing and gradually diminish towards the crown.
Complete reconstruction of the bridge started in 2014 (Fig. 1) and finished in 2016 (Fig. 2). The carbonation depth was measured
only by phenolphthalein test before and after reconstruction, when the origin concrete surface was covered by a modern covering
plaster.
The second investigated road bridge with evidently reduced carbonation is located on the III. class road nearby the city
Sládkovičovo over a small creek (Fig. 3). The bridge has a single span of 4 m; its superstructure was reconstructed in 1965. Thus from
the contemporary structure only the concrete abutments remained original of the age of approximately 100 years.
The measured carbonation depths at the bridges are reported in Table 2.

2.2. Research methodology

In-situ tests applied in the current project concerned the following procedures: 1) testing concrete by hardness method using

98
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 2. Bridge Krásno nad Kysucou, still in service, after reconstruction in 2016.

Fig. 3. Reinforced concrete bridge near to Sládkovičovo.

Table 2
Carbonation depths.

Bridge Location of measurement Average carbonation depth (mm)

Krásno nad Kysucou Superstructure <4

Superstructure <4
Sládkovičovo Abutment, sheltered from rain <2
Abutment, sheltered from rain <2

Schmidt hammer, model N according to STN 73 1373 Standard [17], 2) ultrasonic examination of concrete by STN 73 1371 [18]
using PUNDIT apparatus searching for the homogeneity of concrete material and hence the mechanical properties, 3) measurements
of surface tensile strength of concrete using square bounded metal plates according to STN EN 12 636 [19], 4) measurements of

99
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Table 3
Mechanical properties of concrete at the bridge in Krásno nad Kysucou (abbreviated K).

Specimen Position of the specimen Volume density Dynamic modulus of Young’s modulus of Compressive strength on
notification on the bridge (kg/m3) elasticity (GPa) elasticity (GPa) cylinders (MPa)

K-1 Arch 2080 No-measured 21,0 23,8

K-2 2080 No-measured 22,0 25,7
K-3 Arch Overfill 2090 29,3 20,9 22,7
K-4 2160 37,1 23,7 30,5
K-5 2050 No-measured 20,7 21,4
K-6 Parapet 2150 35,2 24,1 20,2
K-7 2140 31,9 16,5 22,4
K-8 2130 29,5 15,1 19,2

Notice: bridge Krásno nad Kysucou is abbreviated as K.

surface permeability of concrete by the Torrent air permeability test [20] for checking the surface permeability of uncovered and
covered concrete by plaster coat and 5) carbonation depth by phenolphthalein test. The used cylindrical cores with diameter of
100 mm and lengths about 200–250 mm were tested in a laboratory for dynamic modulus of elasticity [18], Young’s (static) modulus
of elasticity [22] and compressive strength [21]. Cylindrical specimens were broken in half in longitudinal direction. Phenolphthalein
test was repeated on the first part and the second part was used to study the microstructure and pore structure. Consequently two sub-
samples were chosen for in-deep detail verifications — surface: directly exposed to prolonged exposure to CO2 and — internal:
covered with overlay of concrete kept out of direct contact with open air. Subsequently the sub-samples were crushed then heated at
60 °C to constant weight and ground for fineness passing 0.063 mm for next analytical procedures. Chemical analysis of the ground
concrete matrix (without large aggregate grains) was performed in concordance with STN 72 010 [23]. The X-ray diffraction tests
were made on the automatic theta/theta X-ray diffractometer STOE & Cie GmbH (Germany) using the program Bede ZDS W95/98/
NT for assessing the records and run in a 2Θ range of 5–65°. CuKα radiation and Ni-filter was used. Thermal studies were conducted
simultaneously from 20 to 1100 °C at the NETZSCH apparatus STA 449 F3 Jupiter (Germany). Samples weighting 200 mg were
heated in flowing air within the heating range 20–1100 °C and at heating rate 10 °C/min. The pore structure was studied by mercury
intrusion porosimetry (MIP) on the porosimeter Poremaster GT-60, Quantachrome (USA) using 5–10 g bulk samples of concrete
matrix without large aggregate grains. The microstructure of bulk concrete samples (without large aggregate grains) was studied by
scanning electron microscopy on JEOL 7500F apparatus (Japan). The samples were fixed using a two component epoxy paste in
aluminum discs with a diameter of 25 mm, which are used to transfer the sample to the interior of the microscope. This way prepared
samples were plated with a mixture of gold-palladium under argon. For plating was used spreader device Balzers. The image was
captured electronically using the program “PC SEM” (ver. 3.0.1.8). Optical microscopy was carried out on a digital device VHX-5000
with CMOS image sensor 1/1 8 virtual pixel 1600 (H) × 1200 (V) from Keyence International (Belgium).

3. Bridge at Krásno nad Kysucou

3.1. Mechanical properties of concrete and stage of carbonation

Eight drill cores were taken to a laboratory from the bridge Krásno nad Kysucou. Basic mechanical properties of the concrete are
reported in Table 3.
Onsite measurements of the carbonation depth were performed by phenolphthalein test. Thin plaster coat was found on the
surface of the original concrete. Visual inspection confirmed that the plaster coat occurs throughout the whole bridge, except for
some small areas. The following defensive effects prior to penetration of CO2 into concrete were observed: from highly effective to
less protective (Table 4). During reconstruction in 2014 the original bridge structure was plastered from the outside by current
industrially produced plaster (CIPP). The concrete was characterized by varying carbonation depths depending on the long-term
indirect CO2 penetration through the old origin plaster coat (OOC) (Figs. 4 and 5). Undoubtedly the interesting fact is that the depth
of carbonation of 10 mm thick CIPP after one-year exposure on the bridge (5 mm) is higher than the one at 125-year-old concrete

Table 4
Depth of carbonation of old concrete placed 125 years under OOC and the alone CIPP after one-year exposure at the reconstructed bridge.

Specimen notification Age of materials (years) Position of the specimen on the bridge Depth of carbonation (mm)

K6-surface 125 Parapet 20

K1-surface 125 Arch 4
CIPP 1 Arch 5

100
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 4. Depth of carbonation of concrete: K6-surface specimen taken from the parapet.

Fig. 5. Depth of carbonation of concrete: K1-surface specimen taken from the arch.

under the cover of OOC (4 mm) (specimen K1-surface). Differences in the measured depths of carbonation show variability of the
OOC protection. However, 20 mm carbonation depth found in 125 year-old-concrete (specimen K6-surface) opposite to that of CIPP
(5 mm) is still a good result. The CIPP and OOC are shown in Figs. 6 and 7.
Chemical compositions of the tested materials are reported in Table 5. Chemical analysis indicates that the K1-surface concrete
with higher contents of SiO2 and Na2Oeqv. and lower CaO is characterized by the reduced carbonation depth opposite to the K6-
surface concrete with inverse oxide contents. It is currently unknown whether concrete composition of the arch was the same as that
of the parapet. The results indicate that our assumptions of the same concrete composition could be correct. The similar difference in
CaO and SiO2 contents was observed also for the CIPP and OOC. The OOC of thickness approximately 2–3 mm can be specified as
inorganic cement or CaO-based plaster coat, while CIPP of thickness of 10 mm is found richer in CaO and therefore more vulnerable
to carbonation. CIPP differs from OOC in the amounts of MgO and Cl− when containing more chlorides. Total content of CaCO3
occurring in the specimens is reported in Table 6. One would conclude that the higher content of CaCO3 proves the higher carbo-
nation attack, which is also proportional to the CaO contents. As the reason of lower CaO content, OOC would possess lower
vulnerability to carbonation opposite to CaO-rich CIPP; the same assumption may be extended to assess the carbonation of concrete.

101
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 6. New plaster coat (one-year old) separated from the origin (125-years old).

Fig. 7. View on the origin plaster coat (partially removed after reconstruction works).

Table 5
Chemical analyses of the concrete, old plaster coat and current industrially-produced plaster.

Specimen notification Concrete analyzed in depth from the surface Constituent (% wt.)
(mm)
Ign. loss SiO2 CaO MgO Fe2O3 Al2O3 SO3 eqv. Na2O Cl− pH

K1-surface 0–10 12.48 62.47 14.24 0.66 1.97 4.85 0.15 2.28 0.015 9.07
K6-surface 0–10 17.16 50.56 23.11 1.32 1.44 4.24 0.65 1.00 0.014 11.77
K6-internal 120–150 15.60 54.34 21.98 0.76 1.71 3.84 0.17 0.78 0.017 8.93
OOC – 19.00 50.62 19.37 1.05 2.03 4.84 0.50 2.48 0.080 12.22
CIPP – 19.59 40.68 26.07 3.29 2.68 5.37 0.56 1.55 0.110 12.55

102
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Table 6
Total content of CaCO3 estimated by TG calculation found in concretes, CIPP and OOC.

Specimen notification Age of materials (years) Concrete analyzed in depth from the Content of CaCO3 (% Increase of CaCO3 on the surface (%
surface (mm) wt.) wt.)

K1-surface 125 0–10 17.47 –

K6-surface 125 0–10 24.47 5.00
K6-internal 125 120–150 19.47
OOC 125 – 25.04 –
CIPP 1 – 31.38 –

Table 7
Final determination of the carbonation attack.

Specimen Total content of CaO CaO content calculated from thermal analysis Assessment of carbonation
from chemical analysis (% wt.) occurring in
(% wt.)
fine-grained coarse-grained total Degree of Degree of modification Stage of
CaCO3 CaCO3 carbonation (%) changes (−) carbonation

K1-surface 14.24 45.03 4.96 89.99 63.1 1.0 II

K6-surface 23.11 2.09 11.56 13.65 59.1 5.5 II
K6-internal 21.98 1.48 9.43 10.91 49.6 6.4 I
OOC 19.37 7.15 7.03 14.18 73.2 1.0 III
CIPP 26.07 3.95 13.85 17.76 68.1 3.5 III

3.2. Degree of carbonation, modification changes and stage of carbonation

Indicators of critical importance for the assessment of carbonation (degree of carbonation and modification changes as well as
stage of carbonation) are obtained from chemical and thermal analysis (Tables 5 and 6). Final evaluation is reported in Table 7.
One-year old CIPP is specified by markedly higher degree of modification changes (3.5) due to 13.85% wt. of coarse-grained
CaCO3 to 3.95% wt. of fine-grained CaCO3 (Table 7). Large amount of coarse-grained CaCO3 highlights the advanced carbonation
opposite to that of OOC specified with almost equivalent contents of coarse- and fined-grained CaCO3. Besides different chemical
composition of OOC and CIPP, the reduced content of coarse-grained CaCO3 in OOC can be attributed to a narrow space capable for
its formation. When yet formed, one can assume that the spatial deficiency does not allow its further growth or enables its formation
to the extent still sealing the system against further CO2 penetration into the underlying concrete. A narrow free space filled with the
carbonate can cause increased non-permeability of the plaster coat. Torrent air permeability tests were not performed at Krásno nad
Kysucou because the reconstruction took place between the evaluation of the results and next visit to the bridge. The OOC and CIPP
indicate the degree of carbonation III. Both surface specimens K1 and K6 indicate degree of carbonation II. The specimen K1
compared to K6 is characterized by markedly higher degree of modification changes. This is the evidence that surface K6 specimen is
more progressive carbonated than K1 as proved by high content of coarse-grained CaCO3. Internal sample K6 protected by overlay of

Fig. 8. Comparison of TG-DTA plots of OOC and CIPP.

103
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 9. The thickness of OOC observed at the bridge Krásno nad Kysucou.

concrete shows carbonation stage I. Though internal K6 contains less carbonation products, high degree of modification changes (6.4)
demonstrates prevailing share of coarse-grained CaCO3 present in the concrete. This phenomenon can be attributed to the sufficiency
of space in the developed pore system, which enables creating coarse-grained deposits of CaCO3 over sufficiently long time.
Comparison of TG-DTA plots of OOC and CIPP gives clear proof of a different resistance to aggressive effect of CO2 (Fig. 8). In
assessing the extent of carbonation, the age difference between both tested specimens is needed to bearing in mind. Both materials
are specified by the carbonation stage III but OOC after 125 years and CIPP only after one-year exposure to outer atmospheric
conditions. CIPP records the degree of modification changes 3.5, while OOC only 1.00. This difference is caused by a greater amount
of coarse-grained CaCO3 accumulated in CIPP compared to that found in OOC. The detected finding underlines the same stage of
carbonation in one-year old CIPP compared to 125-years old OOC as reports Table 7. The result confirms that the aggressive effect of
CO2 is markedly considered not only by the degree of carbonation and the stage of carbonation but with the same meaning of
importance by the degree of modification changes. Larger weight portion of coarse-grained CaCO3 relative to the fine-grained gives
the evidence of more carbonated concrete over time.
Fig. 9 depicts the OOC thickness by the opposing arrows at the location on the bridge structure, in which it was encountered. The
meter is located at the right angle, perpendicularly to the surface of concrete. The picture leads to the misleading judgment that it is

Fig. 10. Mineral composition of the concrete taken from the bridge Krásno nad Kysucou.
Abbreviations; of identified minerals: Q — quartz SiO2; AL — albite NaAlSi3O8; OS — osumillite KFe22+(Al5Si10)O30; Cc — calcite CaCO3.

104
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 11. Mineral compositions of OOC and CIPP applied at the bridge Krásno nad Kysucou.
Abbreviations of identified minerals: the same as in Fig. 10 + muscovite — Mu
KAl2(Si3Al)O10(OH)2 and kaolinite — Ka Al2Si2O5(OH)4.

situated concurrently.
XRD patterns of the studied concrete specimens, OOC and CIPP are illustrated in Figs. 10 and 11. The concretes are characterized
by a similar mineral composition with the majority of quartz — Q and minor portions of albite — AL and osumillite — OS. Calcite is
regarded as the carbonation product. Differences were observed in the intensities of diffraction lines of calcite, in particular that
occurring near to 35° 2Θ. The largest presence of calcite — Cc indicates K6-surface sample. This result is fully in line with the findings
reported in Table 6.
The records of OOC and CIPP show the related mineral composition. Both materials are composed mainly of quartz — Q that is
slightly impured by albite — AL and osumillite — OS. CIPP compared to OOC contains the traces of muscovite — Mu and kaolinite —
Ka; The OOC specimen points higher silica content. This result is fully in compliance with the finding reported in Table 5.
The exact identification of the aged OOC composition has not been the subject of this research, so it is currently unknown.
Clarification of the material composition of OOC; proposing compositions using today's materials, and verifying laboratory-made
samples of modern coatings are a key intention of a future research project.
The formed carbonation products are illustrated in Figs. 12 and 13. Large crystallites of needle-shaped or plate-shaped form are
observed in K6-surface and K6-internal samples.

Fig. 12. SEM image of the surface (0–10 mm) K6-surface concrete (left) and the internal (120–150) mm K6 concrete (right) taken from the parapet (magnifica-
tion × 20,000).

105
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 13. SEM image of the surface (0–10 mm) concrete from the arch of bridge K1-surface protected by OOC (magnification × 20,000).

Both specimens show well-developed CaCO3 crystals (Fig. 12). Nevertheless, K6-internal sample indicates less profiled coarse-
grained nature of the crystallites compared to K6-surface. Carbonates in K1-surface concrete protected by OOC (Fig. 13) for a long
time exhibit space-saving crystallization character and denser microstructure without evident proof of crystal-developed coarse

Fig. 14. Mainly fine-grained carbonate deposits accumulated in the 2–3 mm OOC.

106
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Table 8
Basic mechanical properties of concrete at the bridge near to Sládkovičovo.

Specimen Position of the specimen Volume density Dynamic modulus of Young’s modulus of Compressive strength on
notification on the bridge (kg/m3) elasticity (GPa) elasticity (GPa) cylinders (MPa)

SL-2 Abutment 2110 30.8 16.1 20.3

SL-3 2157 35.9 20.3 25.5

Notice: bridge Sládkovičovo is abbreviated as SL.

CaCO3.
K1-surface concrete protected by OOC is characterized by a palette-like arrangement of fine-grained carbonates that produces self-
protection of concrete against the further penetration of aggressive CO2 into deeper layers of subsurface concrete.
Images from the optical microscope at various magnifications show at first invisible but at higher magnification demonstrable
accumulated layer of predominantly fine-grained CaCO3 within the OOC as seen in Fig. 14. Shining needles show also the presence of
crystal-developed coarse CaCO3 compressed within the OOC thickness. The OOC is characterized by almost the same, and relative
low contents of fine-grained and coarse-grained CaCO3 7.15% wt. and 7.03% wt. according to Table 7. This is main reason for low
degree of modification changes, although OOC shows carbonation stage III. The fine-grained character of the formed dense, as if
sintered, carbonates in confined spatial conditions inside the 2–3 mm thick OOC creates the best condition for the built limestone-
based barrier with ability to dramatically reduce CO2 penetration into the underlying concrete over time.

4. Bridge near to Sládkovičovo

4.1. Mechanical properties of the concrete

Two drilled cores were taken from the bridge Sládkovičovo. Properties of this concrete are listed in Table 8. The attention was
paid to the protective effect of the OOC with a focus on the relationship between the permeability and the efficiency of the cover to
prevent CO2 ingress into the concrete beneath.

4.2. Onsite measurements of permeability of the plaster coat

The OOC of thickness of 2–3 mm was also found at the bridge near to Sládkovičovo. Carbonation of low depths was observed
bellow OOC. The plaster coats were characterized by either high or low permeability confirmed by Torrent air permeability tests.
After drilling the cores with diameter of 75 mm and length approximately of 200 mm, onsite carbonation depths were measured. The
test was re-performed and carbonation depth of concrete under the OOC was repeatedly verified by phenolphthalein on the site with
low permeability (announced as good site G1 and G2) and with high permeability of the plaster coat (announced as bad site B1 and
B2). The dependence between permeability coefficients and carbonation depths is reported in Table 9. Figs. 15 and 16 show car-
bonation depth of concrete situated below low-permeability and high-permeability OOC.
Carbonation depth of the concrete below high-permeability OOC is 30 mm. This value refers to an annual carbonation rate
0.3 mm. One would conclude that the depth of carbonation 30 mm within period of 100 years is still relative low. This observation is
consistent with the finding reported in [13] on lower carbonation depths of concrete with increased age, mainly due to different
character of cement production for a time longer maturing concrete and acquisition of early concrete strengths. The surprising
discovery has provided the low-permeability OOC because the concrete beneath it did not show carbonation. The results indicate that
the presence of a high-impermeable plaster coat ensures reliable protection against carbonation of concrete. The studied OOCs are
characterized by diverse quality in terms of the observed permeability; only a low-permeability plaster coat creates a highly effective
barrier preventing concrete carbonation.
The sites covered and uncovered by the OOC are shown in Fig. 17. The cylindrical specimens from the abutment provided and

Table 9
Permeability of the OOC and carbonation depth of the concrete underneath.

Marking the measuring point Coefficient of permeability K (×10−16 m2) Carbonation depth (mm)

B1 80.17 30
B2 60.40 30
G1 0.073 0
G2 0.002 0

107
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 15. Carbonation depth of concrete below the plaster coat characterized by permeability of 0.073 × 10−16 m2.

Fig. 16. Carbonation depth of concrete below the plaster coat with the measured permeability of 80.17 × 10−16 m2.

non-provided by the OOC are seen in Fig. 18. The carbonation depth of the specimen with covered (top) and uncovered surface
(below) is illustrated in Fig. 19. The fragments of the OOC from the outer (left) and the inner (right) side contacting concrete are
depicted in Fig. 20. Depths of carbonation of covered and uncovered concrete by plaster coat demonstrates Table 10.

4.3. Degree of carbonation, modification changes and stage of carbonation

Both cylindrical concrete specimens were cut into two segments at a distance of (0–20) mm and (150–180) mm from the outer
surface. The obtained sample segments as well as the sole OOC marked as G1 (Table 9) were crushed, dried at 60 °C to constant
weight and consequently ground to particle sizes below 0.063 mm. Individual powders were subjected to chemical [50], thermal (TG-
DTA) and X- ray diffraction analysis. Smaller fragmented segments of approximately 5 mm in size were analyzed by Hg-porosimetry
and SEM techniques. The results of chemical analysis are given in Table 11. TG evaluations are reported in Table 12. Determination of
degree of carbonation, degree of modification changes and carbonation stage are summarized in Table 13.
The oxide composition of covered and uncovered concrete by the OOC in the depths of (0–20) mm and (150–180) mm are not
diametrically different. The surface (0–20) mm concrete layer is characterized by the finely raised ignition loss and the slightly lower
pH extract value opposite to the internal (150–180) mm layer. Comparison of the protected (SL-P) and the exposed concrete (SL-N),
in particular both surface (0–20 mm) layers show that the pH value of the extract of the exposed concrete without the OOC is lower
compared to the protected one. This fact indicates a greater extent of carbonation of SL-N concrete opposite to that of SL-P. The OOC
records diametrically distinct chemical composition in comparison with those of the concrete. Differences are registered mainly in the
ignition loss, SiO2, CaO, and partially MgO values, and also in pH of water extracts. From pH 12.3 that the OOC has retained for
approximately 100 years of the service-life, still the alkaline nature of material matrix is deduced.
The presence of gel-like hydration products and Ca(OH)2 are not clearly confirmed, though DTA plot shows very weak or

108
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 17. Sites on the bridge abutment with and without OOC.

Fig. 18. Concrete specimens taken from the abutment of the bridge: covered (left) and uncovered (right) by OOC.

negligible signs of endothermic peak between 100 and 300 °C and at about 520 °C. By contrast TG curve is specified by a couple of
partial mass losses, from which the only one clear is that attributed to CaCO3 decomposition at peak maximum 796.3 °C. The mass
loss related to the released CO2 within 600 °C–800 °C takes 11.53% wt. and is proportional to 26,22% of CaCO3 by weight. The
occurrence of CaCO3 in the OOC is thus clearly confirmed (Fig. 21).
The results of TG percentage analysis are presented in Table 12. Unprotected concrete SL-N demonstrates evident increase in
CaCO3 content (10.32% wt.) in the (0–20) mm surface sample (N-1) compared to the (150–180) mm internal sample (N-2) whereas
this difference is negligible in the protected concrete (P-1 vs. P-2, only 0.72% wt.). The low-permeable (G1 in Table 9) OOC placed at
the concrete surface acts as a protective barrier preventing CO2 diffusion over time. The results show that there is a real amount of
carbonates in the plaster coats.
The high-permeable OOC (Table 9) has approximately half of the protective effect against carbonation compared to the uncovered
concrete (Table 10). The high content of CaCO3 in the OOC implies that actual non-permeability is due to the sequential deposition of
carbonates over time.
This hypothesis would be examined by the future detailed research. However, one would assume high-material quality and

109
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 19. The carbonation depth of concrete with OOC (top) and without it (below).

Fig. 20. Fragments of OOC from outside (left) and inside view of concrete base (right).

Table 10
Carbonation depth of concrete covered and uncovered by the OOC.

Specimen notification Surface of the concrete tested Position of the specimen on the bridge Depth of carbonation (mm)

SL-N Without the plaster coat Uncovered abutment with the plaster coat 60
SL-P With the plaster coat Covered abutment 0

technological discipline at placing of the plaster coat as surface protection on the concrete surface as well as no crack occurrence or
any type of other mechanical damage of the plaster coat during the service-life of concrete (bridge) structure. The reason of applying
a surface plaster coat (OOC) over 100 years ago is still not clear. It might be either as a finishing procedure or even an attempt to
mitigate the degradation of the concrete by effects of external factors.
The XRD patterns (Figs. 22 and 23) show quartz (Q) appearing as the predominant phase of the used aggregates. Albite (AL) and
dolomite (D), osumillite (Os) and muscovite (Mu) contents are of marginal importance. The present calcite (Cc) is supposed as the

110
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Table 11
Chemical analysis of concrete specimens and the inorganic plaster coat.

Sample Sampling point Concrete analyzed in depth from the surface Constituent (% wt.)
(mm)
Ign. loss SiO2 CaO MgO Fe2O3 Al2O3 SO3 Na2O eqv. pH

SL-N-1 Concrete without OOC 0–20 10.4 67.93 9.09 1.38 2.08 6.08 0.03 2.19 10.7
SL-N-2 150–180 9.6 49.78 8.51 1.52 1.73 6.07 0.23 1.93 11.8
SL-P-1 Concrete covered by 0–20 9.4 68.76 9.86 1.40 1.91 5.91 0.34 1.87 11.7
SL-P-2 OOC 150–180 8.6 72.90 7.06 0.92 1.84 5.87 0.23 1.97 11.8
SL-OOC – – 19.9 35.64 29.11 2.90 1.85 5.69 0.68 1.65 12.3

The chloride contents of the samples varied between 0.01 and 0.02% wt. that of the OOC was 0.11% wt.

Table 12
The evaluation of thermal analysis results by TG calculations.

Sample Sampling point Depth of concrete from the surface of the CaCO3 content (% CaCO3 increase at the surface opposite to the
taken core drill (mm) hmot.) internal sample (% wt.)

SL-N-1 Without plaster coat 0–20 18.33 10.32

SL-N-2 protection 150–180 8.01
SL-P-1 Protected by plaster coat 0–20 10.07 0.72
SL-P-2 150–180 9.35
SL-OOC Plaster coat alone – 26.22 –

Table 13
Final determination of the carbonation attack.

Sample Total content of CaO from CaO content calculated from thermal analysis (% Assessment of carbonation
chemical analysis (% wt.) wt.) occurring in

fine − grained coarse- grained total Degree of Degree of modification Stage of

CaCO3 CaCO3 carbonation (%) changes (−) carbonation

SL-N-1 9.09 2.40 7.87 10.27 113.0 3.3 IV

SL-N-2 8.51 1.40 3.08 4.49 52.7 2.2 I
SL-P-1 9.86 1.77 3.87 5.64 57.3 2.2 II
SL-P-2 7.06 1.41 3.82 45.24 60,0 2,0 II
SL-OOC 29.11 5.05 10.19 15.24 52.4 2.0 I

Fig. 21. TG-DTA plots of the inorganic plaster coat (OOC).

carbonation product of portlandite Ca(OH)2 prior to coming from the aggregates because of very negligible presence of dolomite,
occurring usually in aggregates rich on calcite. The patterns showing main Ca(OH)2 ray (4.92 Å, 3.11 Å, 2.63 Å, 1.93 Å and 1.80 Å)
are missing. When comparing intensities of diffraction lines, the OOC and uncovered concrete are characterized by the highest calcite

111
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 22. X-ray diffraction patterns of the concrete unprotected (N1-surface and N2-internal sample) and protected (CH1-surface sample) by the OOC taken from the
bridge Sládkovičovo (SL) — the individual minerals are abbreviated in Figs. 10 and 11.

Fig. 23. Comparison of XRD records of the surface concrete without protective plaster coat (N1) and the lone inorganic plaster coat (OOC) — the individual minerals
are abbreviated in Figs. 10 and 11.

portion. Complete evaluation of the samples from the bridge Sládkovičovo is summarized in Table 13. The exposed concrete surface
layer (SL-N-1) is characterized by an extremely high degree of carbonation, the highest degree of modification changes and therefore
the carbonation stage IV that is specified by an extensive carbonation attack. Conversely, the internal sample (SL-N-2) indicates the

Table 14
Pore structure characteristics of concrete and inorganic plaster coat.

Sample Specific surface of open pores
with radii between 1.82 nm
and 0.534 mm (m2/g)
Volume of pores with Volume of total pores Pore median radius Total
radii between 1.85 and between 1.82 nm and porosity (%)
5 250 nm (cm3/g) 0.534 mm (cm3/g) of total pores with pores with radii
radii between 1.85 between 1.85 and 5
and 5 250 nm (nm) 250 nm (nm)

SL-N-1 6.36 0.065 0.090 231.0 64.38 19.65

SL-N-2 9.68 0.069 0.130 125.0 29.96 25.40
SL-P-1 11.39 0.070 0.110 166.0 30.35 21.88
SL-OOC 6.06 0.041 0.075 100,3 32.64 15.77

112
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

Fig. 24. Microstructure of surface concrete SL-N-1 unprotected by the OOC.

stage I of carbonation, characterized by non-detrimental effect on the concrete properties and durability. It is thought that the
differences in mineral composition between uncovered surface and internal sample (N-1 vs. N-2) are due to the shading effect of the
surface (0–20) mm layer, in which carbonates have accumulated. The surface and internal layer of the concrete protected by the OOC
show similar values of the degree of carbonation and modification changes and the same carbonation stages II.
Similarities in the extent of carbonation between SL-P-1 and SL-P-2 samples are the result of the OOC's effect on reducing the CO2
penetration into the concrete beneath. For this reason the carbonation of the surface (0–20) mm (SL-P-1) and internal (150–180) mm
(SL-P-2) concrete layer covered by the OOC (G1 in Table 9) is the same. The plaster coat records the degree of carbonation pro-
portional to the stage of carbonation I despite the confirmed largest content of CaCO3. This finding implies two facts: 1) the OOC is
after 100 years of service-life attacked by carbonation only to the harmless level and 2) the largest content of CaCO3 is not regarded as

Fig. 25. Microstructure of surface concrete SL-P-1 protected by the OOC.

113
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

the only one important critical factor to determining the extent of carbonation. Next factor of the same importance is considered the
degree of modification changes.
The results of the developed pore structure are reported in Table 14. The OOC is characterized by the lowest values of total
specific surface of all open pores, volume of all pores and pore radius between 1.85 and 5 250 nm, pore radius of total pores and total
porosity. This result demonstrates the densest pore structure formed in the OOC compared to those of the concretes. The decisive pore
structure parameters of the surface (0–20) mm SL-P-1are similar to the internal (150–180) mm SL-N-2. Pore structure of the un-
protected (0–20) mm SL-N-1 sample is considered in terms of densification more permeable relative to those developed in SL-P-1 and
SL-OOC.
Figs. 24 and 25 illustrate differences between microstructure of the surface concrete, unprotected and protected by the OOC.
Concrete with direct contact with air proves the apparent amount of coarse crystalline carbonate products, apparently pointing to
more open macropores and thus higher permeability while the protected is thickened with fine-grained substitutes forming a dense
and very low-permeability barrier, in which micropores are predominantly present. Such a highly thickened deposit of finely crys-
talline carbonates in a relatively small space enclosed by 2–3 mm thickness of the coating does not allow continual crystallization into
coarse-grained carbonates. In the end, this results in increasing porosity and permeability over time. An old surface treatment of
concrete used at that time mitigates carbonation reaction by evident reducing further diffusion of CO2 into the subsurface concrete.
Carbonation reaction with its own product form in the thin layer of the coating thus prevents further carbonation in the concrete
itself.

5. Conclusions

The following conclusions have been derived from the results reported in this study:

1) The assessment of aggressive action of CO2 on concrete based on the degree of carbonation, degree of modification changes and
four stages of carbonation is an appropriate method for diagnosing the concrete attacked by carbonation for a long time.
2) The layer of 2–3 mm thin plaster coat has been proven as effective anti-carbonation barrier. The depth of carbonation of over 100-
year-old concrete protected by high-quality, low-permeable plaster coat was ascertained in maximum depth of 4 mm.
3) The presented results show that the carbonation depth of the underlying concrete depends decisively on non-permeability of the
plaster coat placed on its surface.
4) The dense plaster coat can become an effective material for the extended service-life of modern constructions, especially from
carbonation prevention.
5) The exact identification of the aged plaster coat composition has not been the subject of this research, so it is currently unknown.
Clarification of the material composition of old plaster coat; proposing compositions using today's materials, and verifying la-
boratory-made samples of modern coatings are a key intention of a future research project.

Acknowledgements

The authors wish to express their gratitude to Slovak Agency for Research and Development (project APVV No. 0442-12) for
financing this research work.

References

[1] R. Liu, L. Jiang, J. Xu, C. Xiong, Z. Song, Influence of carbonation on chloride induced reinforcement corrosion in simulated concrete pore solutions, Constr.
Build. Mater. 56 (2014) 16–20.
[2] A.M. Aguirre-Guerrero, R. Mejla-deGutiérret, M.J.R. Montees-Correia, Corrosion performance of blended concretes exposed to differenent aggressive environ-
ments, Constr. Build. Mater. 121 (2016) 704–716.
[3] P.A. Claisse, H.I. Elsayad, I.G. Shaaban, Permeability and pore volume of carbonated concrete, ACI Mater. J. 96 (1999) 378–381.
[4] L. Mo, D.K. Panesar, Accelerated carbonation −a potential approach to sequester CO2 in cement paste containing slag and reactive MgO, Cem. Concr. Compos.
43 (2013) 69–77.
[5] S.E. Pihlajavaara, Some results of the effect of carbonation on the porosity and pore size distribution of cement paste, Mater. Constr. 52 (1968) 1–52 (6).
[6] V.T. Ngala, C.L. Page, Effects of carbonation on pore structure and diffusional properties of hydrated cement pastes, Cem. Concr. Res. 27 (1997) 995–1007.
[7] I. Janotka, L. Krajci, Degradation of concrete in panel housing due to carbonation, Proceedings of the 1st Slovak Conference on Concrete Structures, Bratislava,
(Slovakia), September 10–12, 1994, pp. 362–371.
[8] I. Janotka, I.L. Krajci, Assessment of carbonation in concrete by pH and reinforcement corrosion, Proceedings of the 4th CANMET/ACI International Conference
on Durability of Concrete, Supplementary Papers, Sydney, (Australia), August 17–22, 1997, pp. 147–158.
[9] I. Janotka, L. Krajci, Concrete deterioration determined by four stages of carbonation, Proceedings of the 13th FIP Congress on Challenges for Concrete in the
Next Millennium, Amsterdam, (The Netherlands), May 23–29, 1998, pp. 1051–1054.
[10] M. Matoušek, R. Drochytka, Atmosférická koroze betonů − in Czech (Atmospheric corrosion of concrete − in English), first ed., IKAS −ČKAIT, Praha (Prague),
1998.
[11] L. Krajci, I. Janotka, Measurement techniques for rapid assessment of carbonation in concrete, ACI Mat. J. 97 (2000) 168–171.
[12] Cheng-Feng Chang, Jing-Wen Chen, The experimental investigation of concrete carbonation depth, Cem. Concr. Res. 36 (2006) 1760–1767.
[13] M.G. Alexander, J.R. Mackechnie, W. Yam, Carbonation of concrete bridge structures in three South African localities, Cem. Concr. Compos. 29 (2007) 750–759.
[14] R. Neves, B. Sena da Fonseca, F. Branco, J. de Brito, A. Castela, M.F. Montemor, Assessing concrete carbonation resistance through air permeability mea-
surements, Constr. Build. Mater. 82 (2015) 304–309.
[15] K.Y. Ann, S.-W. Pack, J.-P. Hwang, H.-W. Song, S.-H. Kim, Service life prediction of a concrete bridge structure subjected to carbonation, Constr. Build. Mater. 24
(2010) 1494–1501.
[16] D.C. Park Carbonation of concrete in realtion to CO2 permeability and degradation of plaster coats, Constr. Build. Mater. 22 (2008) 2260–2268.

114
I. Janotka et al. Case Studies in Construction Materials 8 (2018) 97–115

[17] STN 73 1373, Testing of Concrete by Hardness Testing Methods, (1981).

[18] STN 73 1371, Method of Ultrasonic Pulse Testing of Concrete, (1981).
[19] STN EN 12 636, Products and Systems for the Protection and Repair of Concrete Structures. Test Methods. Determination of Adhesion Concrete to Concrete,
(2001).
[20] R. Torrent, G. Frenzer, Study on methods to determine and judge characteristic values of the cover concrete on site (in German), Bundesamt für Strassenbau,
Bern 516 (1995) 105.
[21] STN ISO 6784, Concrete. Determination of Static Modulus of Elasticity in Compression, (1993).
[22] STN EN 12390-3, Testing Hardened Concrete. Part 3: Compressive Strength of Test Specimens, (2010).
[23] STN 72 0101, Basic Analysis of Silicates. Common Regulations, (1973).

115