Vous êtes sur la page 1sur 224

Springer Series in Materials Science

Volume 176

Series Editors

Robert Hull, Charlottesville, VA, USA


Chennupati Jagadish, Canberra, ACT, Australia
Richard M. Osgood, New York, NY, USA
Jrgen Parisi, Oldenburg, Germany
Zhiming M. Wang, Chengdu, P.R. China

For further volumes:


http://www.springer.com/series/856
The Springer Series in Materials Science covers the complete spectrum of mate-
rials physics, including fundamental principles, physical properties, materials
theory and design. Recognizing the increasing importance of materials science in
future device technologies, the book titles in this series reflect the state-of-the-art
in understanding and controlling the structure and properties of all important
classes of materials.
Dmitrij Lyubimov Kirill Dolgopolov

Leonid Pinchuk

Micromechanisms
of Friction and Wear
Introduction to Relativistic Tribology

123
Dmitrij Lyubimov Leonid Pinchuk
Kirill Dolgopolov National Academy of Sciences of Belarus
Department of Antifrictional Materials Metal-Polymer Research Institute
South-Russian State University Gomel
Shahty Belarus
Russia

ISSN 0933-033X
ISBN 978-3-642-35147-1 ISBN 978-3-642-35148-8 (eBook)
DOI 10.1007/978-3-642-35148-8
Springer Heidelberg New York Dordrecht London

Library of Congress Control Number: 2012954058

 Springer-Verlag Berlin Heidelberg 2013


This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief
excerpts in connection with reviews or scholarly analysis or material supplied specifically for the
purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the
work. Duplication of this publication or parts thereof is permitted only under the provisions of
the Copyright Law of the Publishers location, in its current version, and permission for use must always
be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright
Clearance Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science?Business Media (www.springer.com)


Preface

all sciences become mathematical by nature as long as they improve


N. Whitehead, English Mathematician and Philosopher
during the last 5 years, the relations between individual sciences and
between science and technology have changed. The growing awareness
and effectiveness of scientific methods has enabled fruitful application of
one science to another science Meanwhile, the engineering sciences have
become so complicated that the empirical methods alone have become
helpless. In short, the science has become both feasible and mandatory to
use in the engineering. The difference between aspirations still remains and
frequently happens to be decisive, while methods and challenges are
growing more and more similar
G. Smith, American Mathematician
The friction phenomena attract both the physicist and the engineer alike;
their studies relate to the hardest sphere of the physical boundary-value
problems
W. Hardy, British Tribologist

The progress of engineering and technology stimulated the ideas of friction from
the simplest mechanistic views of Leonardo da Vinci and Charles-Augustin de
Coulomb to the modern tribophysical viewpoint considering the friction as the
superposition of adhesive and cohesive forces. It presumes that friction belongs to
the category of nanotheories, though now we are more accustomed to using the
terms microlevel, micromechanisms, etc.
Unfortunately, the development of models describing the process of friction and
wear using the ideas of the matter electron molecular structure confronts if not
principal then considerable challenges. The fact can be related to them that the
matter structure changes during friction at the microlevel which is hard to trace
and, therefore, hard to study. The structure of the substance exposed to friction is
in the dynamically mobile variable state having the properties hard to assess or
predict.

v
vi Preface

The empirical descriptive approach prevails in the most theoretical tribological


works without using comprehensively the modern physical apparatus. A. S.
Akhmatov authored the outstanding book Molecular Physics of Boundary
Friction; it is exceptional book confirming the rule. The author manifests that the
achievements of the modern physics of solids permit to approach close to at least
the correct physical description of the microprocesses governing the tribounit
macrocharacteristcs, like the wear rate and the friction coefficient. A. I. Sviride-
nok, the Belarusian tribologist, believes that all triboeffects display themselves at
the macroscale, while the phenomena causing and accompanying them evolve at
the microlevel.
The authors endeavor to describe in the present paper theoretically correctly the
transformations the friction forces induce in the substance structure using the ideas
and the apparatus of crystal physics, the solid body quantum and the chemical
bonding theories. When writing, the authors were surprised to discover that the
book was structured so that it resembled the above-mentioned book of A. S.
Akhmatov. The present book continues, in some aspects, the monograph Modern
Tribology published earlier. We describe it in much more detail the micropro-
cesses in which Professor D. N. Garkunov shared when reviewing the preceding
works. The detailed elaboration and more profound revision of the material nar-
rowed the framework of the problems elucidated omitting a variety of the
appealing problems from the study while they are worthwhile to be described
separately.
The book highlights, in the first place, the transformations evolving in the
crystalline lattice in friction, how these transformations influence the appearance
of overactive state capable to become reactive states and some little-known phe-
nomena affecting the friction characteristics. The findings in recent decades
expanded substantially the tribological knowledge which relates in this or another
way to the application of relativistic ideas based on the theory of relativity
approaches to the friction physics study. The discovery by B. V. Deryagin and
colleagues of the mechanic emission electrons and X-ray photons from the friction
surface served the experimental basis; A. L. Zharin, Belarusian tribologist, dis-
covered the dependence of the electron work from the surface of metallic parts on
the extent of friction deformation of surface layers; the discovery by the physicists
of the school of B. V. Deryagin of the ability of free surfaces of solids to emit
heavy particles in friction, etc. The authors of the present book have attempted to
adapt the ideas of the relativistic mechanics to the description of tribophysical
phenomena.
The paper provides physical grounds to validate new models and mechanisms
of triboprocesses. They served as the physical grounds for the tribological concept
which the authors adhere. The concept implies development of the model of
mechanisms of appearance of the third body as the most essential factor of evo-
lution which governs the main tribosystem behavior features. The reader of this
book will need certain knowledge of the solid body physics, electrodynamics,
continuum physics, and quantum mechanics. It is intended for the prepared reader
for whom the contents may be appealing.
Preface vii

The contents of the present monograph are at the borderline between different
trends of natural and engineering sciences, like electrodynamics, the theory of
relativity, the quantum mechanics, the solid body and the surface physics, the
physical and chemical kinetics, and others. Each of the above-listed sections
contains a long established system of common designations, constants, and values.
These designations coincide sometimes. For instance, one and the same symbol
S designates the fundamental values, such as action in the mechanics,
entropy in the thermodynamics, spin in the quantum theory, Umov-Pointing
vector in the electrodynamics, and others. This symbolic designation is so much
common that any effort to change it can cause confusion in the perception of
formulas like the use of similar symbols can be misleading. Therefore, the for-
mulas are deciphered the way they are written in the text; in addition, the book
contains at the end the list of main designations included into the formulas in each
chapter.
With sincere respect to our readers who open this book for the first time,

Shahty, Russia Dmitrij Lyubimov


Kirill Dolgopolov
Gomel, Belarus Leonid Pinchuk
Contents

1 Interaction Between Atoms in the Substance . . . . . . . . . . . . . . . . . 1


1.1 Van der Waals Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Quantum Mechanical Description of Atom Interactions . . . . . . . 5
1.3 Strong Chemical Bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.1 Ion Bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.2 Covalent Bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.3.3 Metallic Bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.3.4 Hydrogen and Resonance Bonds . . . . . . . . . . . . . . . . . . 14
1.4 Collective Interaction Between Atoms . . . . . . . . . . . . . . . . . . . 15
1.5 Crystallization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.6 Crystalline Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.7 Electronic Structure of Materials . . . . . . . . . . . . . . . . . . . . . . . 29
1.8 Interaction Between Molecular Chains . . . . . . . . . . . . . . . . . . . 33
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

2 Crystalline Materials: Surfaces of Solid Bodies . . . . . . . . . . . . . . . 37


2.1 Defects of Crystalline Structure. . . . . . . . . . . . . . . . . . . . . . . . 38
2.2 Polycrystals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.3 Solid Body Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.4 Phase Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

3 Dynamic Microprocesses in Solid Bodies. . . . . . . . . . . . . . . . . . . . 67


3.1 Elastic Waves in Crystals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.2 Quantum Crystalline Lattice Oscillations . . . . . . . . . . . . . . . . . 72
3.3 Diffusion in Crystals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.4 Motions of Dislocations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.5 Dependence of the Crystalline Lattice Parameters
on Dynamic Factors. . . . . . . . . . . . . . . . . . . . . . .......... 90
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......... 93

ix
x Contents

4 Friction Microdynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.1 Friction Effect on Solid Body Surface . . . . . . . . . . . . . . . . . . . 95
4.2 Diffusive Triboprocesses in Friction . . . . . . . . . . . . . . . . . . . . 101
4.3 Motion of Dislocations in Friction . . . . . . . . . . . . . . . . . . . . . . 111
4.4 Relativistic Effects in Friction. . . . . . . . . . . . . . . . . . . . . . . . . 116
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

5 Plasma Triboprocesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127


5.1 Mechanoactivation of Surfaces of Tribocouples. . . . . . . . . . . . . 127
5.2 Triboplasma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
5.3 Wave Processes in Triboplasma . . . . . . . . . . . . . . . . . . . . . . . 146
5.4 Effect of External Electromagnetic Fields on Triboplasma . . . . . 150
5.5 The Experimental Methods of Triboplasma Study . . . . . . . . . . . 155
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

6 Wave Tribochemistry of Post-Plasma States . . . . . . . . ......... 165


6.1 Interaction Between Triboplasma with Tribosurface. ......... 166
6.2 Post-Plasmatic States and Wave Processes Due
Tribochemical Synthesis. . . . . . . . . . . . . . . . . . . . ......... 170
6.3 Apparatus of Wave Mechanics for Description
of Tribocouple Micromechanics . . . . . . . . . . . . . . ......... 176
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......... 183

7 Antifriction Micromechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . 185


7.1 Lubricating Layer Formation . . . . . . . . . . . . . . . . . . . . . . . . . 186
7.2 Chemical Interaction in Friction . . . . . . . . . . . . . . . . . . . . . . . 190
7.3 Complex Compound of Tribomaterials. . . . . . . . . . . . . . . . . . . 192
7.4 Tribocoordination or Surfing Effect . . . . . . . . . . . . . . . . . . . . . 201
7.5 Additives Based on Complex Compounds . . . . . . . . . . . . . . . . 209
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211

Afterword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Chapter 1
Interaction Between Atoms in the Substance

Each science discovers the things which are far beyond the
immediate observation,
Ch. Kittel, American philosopher
In fact, the physics of solids reduces to the establishment of
relations between the behavior of individual atoms and
molecules they manifest when the atoms and molecules combine
into huge associations like regularly ordered crystals
Ch. Kittel, American philosopher
From the viewpoint of the physical principles, the most
interesting phenomena occur in the new spots where the rules
are not suitable or where they are ineffective,
Richard Phillips Feynman, American physicist

Abstract The basic models of atoms and molecular electromagnetic interaction


in solids are presented. Generally accepted systematization of chemical bonds as
well as fundamentals of nucleation centre model is exhibited. The basic notions
about structure of ideal crystals and polymers are adduced. Computation models of
molecular chains interaction are examined.
It can be asserted with certainty that the transition from the purely mechanistic
ideas about the physical processes of friction interactions between solids to the atom
and molecular interactions has been completed so far. The forces acting theoretically
between the contacting solids include all types of interactions; they include the cohe-
sion between solids: metallic, covalent, ion bonds which are short-term forces, the
Van der Waals bonds relating to the work of acting forces.
Let us recall the main notions come across in the theory of the above-mentioned
interactions which we will address when describing the processes accompanying
friction. It will help understand better the subsequent chapters.

D. Lyubimov et al., Micromechanisms of Friction and Wear, 1


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8_1,
Springer-Verlag Berlin Heidelberg 2013
2 1 Interaction Between Atoms in the Substance

1.1 Van der Waals Forces

When two solid bodies approach mutually, the attraction forces begin to act due to
the Van der Waals interactions or the Van der Waals forces (J. van der Waals is the
Dutch physicist, the Nobel Prize winner). These forces ate intrinsic to the neutral
electric systems and they appear due to the fluctuations in the distribution of charges
in atoms or molecules of the converging bodies. These charges produce momentary
continually displacing dipoles in the adjacent atoms. The Van der Waals forces are
divided into three types [1].
A. Dutch physicist A. W. Keesom, was the first to treat the interaction between
constant dipoles within the framework of the classic electromagnetic theory and
which he termed the orientation interaction. Keesom studied the orientation of
two dipoles corresponding to the least potential energy in this system. The energy
of their interactions is

2 p12 p22 6
Uor = r , (1.1)
3 kT

where p12 , p22 dipole moments; kBoltzmann constant; T temperature;


r radius of action of forces.
Following the classic mechanics rules, the Van der Waals forces are determined as
 
dUor
For = r 7 . (1.2)
dr

Thus, the orientations forces is inversely proportional to the spacing to the order
of the seventh power and they can be both be the forces of attraction and the forces
of repulsion in response to the orientation of dipoles. The so-called Keesom effect
means that the attraction of dipoles comes to prevail when the temperature lowers.
B. The experiment did not confirm the temperature dependence (1.1). In real-
ity, the energy Uor reduces much slower as the temperature drops. P. Debye,
German physicist, the Nobel Prize winner, assumed that the molecule with the
constant dipole moment produces the induced moment in the adjacent molecule.
Therefore, the expression to calculate the energy of dipole-to-dipole interactions
should contain the second term independent of the temperature. This interaction
was called induction or polarization. The term originates from the displacement
of the centers of gravity of electric charges of a particle under the external field
effect from the adjacent atoms and molecules. The values of the induced electric
models in this case are relatively insignificant bearing in mind the weakness of
the molecular fields. The energy of interactions depends on the magnitude of the
moment in the system, the polarization coefficient and inversely proportional
to the spacing to the power of six:
1.1 Van der Waals Forces 3

p2
Uind = . (1.3)
r6
C. The model of P. Debye was unfit to the molecules without any constant dipole
moment. The problem how the forces of attraction appear between two neutral
particles was solved by German physicists W. Hatler and F. London using the
quantum mechanics apparatus. Afterwards, this type of the Van der Waals forces
was called the dispersion or London forces.
The result to which London arrived is, in general, that the energy of attraction
between two particles is inversely proportional to the spacing between them to the
power of six.

B
Udisp = , (1.4)
r6
where Bthe constant for this type of atoms.
London proposed the approximate formula for simple molecules:

3 n 0 2
Udisp = , (1.5)
4 r6
where  = h/2 the Plank constant; n 0 the so-called characteristic quantum
or the term which is found experimentally using the light dispersion equations.
The measurements have shown that the term values are close to the value of the
ionization potential i that is why formula (1.5) is often rewritten in the following
way:

3 i 2
Udisp = . (1.6)
4 r6
J. Slater and E. Kirkwood develop the ideas of London further and obtained the
expressions well correlating with the experiment:

3e N 3
Udisp = , (1.7)
4 r6 me

where e, m e electron charge and mass; N number of electrons in the atom external
shell.
The improvement of the quantum models of the Van der Waals forces has enabled
to explain why atoms repulse one another. As the spacing between two atoms gets
shorter, the electron clouds begin to overlap so the systems electrostatic energy
changes. If the spacing is rather intimate, the energy of interactions transforms
into the energy of repulsion due to the overlapping electron clouds. The energy
of interactions is the energy of repulsion at any spacing whatsoever between the
atoms with filled electron shells. It results mainly from the Pauli Exclusion Principle
4 1 Interaction Between Atoms in the Substance

(W. Paulithe Swiss physicist-theoretician, the Nobel Prize winner) [2, 3] which
states that two electrons cannot stay in the same quantum state. The Pauli Exclusion
Principle precludes multiple occupation of this state; the electron clouds of two inti-
mate atoms can overlap when the electrons leap into a higher energy state. Thus, the
process of overlapping of electron clouds boosts the full energy of the system and
corresponds to the appearance of repulsion forces. The potential repulsion forces 0
in inert gases can be represented in the following way:

B1
0 = , (1.8)
r 12
where B1 the empirical parameter determined by independent measurements.
The expression of full potential energy of interactions between two atoms at the
spacing r is usually recorded as [3]:
 12  6 
B3 B4
U (r ) = 4 B2 , (1.9)
r r

where B2 , B3 , B4 new constants determined by independent measurements.


Expression (1.9) came to be called as the Lennard-Jones potential (J. Lennard-
JonesEnglish chemist); Fig. 1.1 shows its graph.
To describe changes in the potential repulsion forces in response to the spacing
changes, the following empirical formula is used:
 r
exp , (1.10)
l
where lthe dimension of the interaction area.
The quantum substance models permit to calculate the interaction forces in addi-
tivity approximation by summing up the elementary forces of each pair of atoms
in two bodies. Using this assumption, Kasimir (H. Kasimira German physicist)

Fig. 1.1 Lennard-Jones


potential graph for atoms
of inert gases [3]
1.1 Van der Waals Forces 5

derived the equation to estimate the adhesive interactions between mirror planes
under the effect of the Van der Waals forces [4]:

hc A
FA = , (1.11)
480 r 4
where Athe surface area.
The attraction induced by the Van der Waals forces is compensated by the repulsion
forces which inhibit the mutual penetration of electron clouds of atoms or molecules.
Because the emerging Van der Waals forces do not cause any electron exchange, the
molecular structure of crystals appearing when particles converge is determined
mainly by the geometrical factors. Since Van der Waals forces are not directional,
the atoms tend to pack densely in the lattice. The resulting crystals are dielectric
because the electrons are tightly bonded.

1.2 Quantum Mechanical Description of Atom


Interactions

The Van der Waals forces are described using the classic electrodynamic appa-
ratus containing some elements of the quantum mechanics: the Planck constant
(M. PlanckGerman physicist, The Nobel Prize winner) and the Pauli principle
of exclusion. The latter opens a new class of interactions between the particles hav-
ing the non-Coulomb nature, or the exchange forces. The particles proper, irre-
spective of their nature, possess simultaneously the wave and corpuscular properties
according to the de Broglie hypothesis (L. de BroglieFrench physicist-theoretician,
the Nobel Prize winner). The principle of mechanical determinism becomes upset
instead of being customary from the classic thermodynamic viewpoint. It com-
pelled W. Heisenberg (W. HeisenbergGerman physicist-theoretician, the Nobel
Prize winner) to develop the uncertainty principle. A new class of physical values
was added to the physical theory; these values are impossible to measure concur-
rently: the energy and the time; the pulse and the coordinate. It is impossible to
know simultaneously the object parameters like the coordinate, the velocity and the
time following the uncertainty principle. Hence, the body location in the space-time
coordinates is unknown validly, it is known with some probability. That is why
A. Einstein (A. Einsteinthe founder of the modern physics) called the quantum
mechanics the Gods game of dice. This game demanded to invent the vector of
state as the new physical value in order to describe physical systems. This value
has several names: the wave function, the probability density, the wave vector, and
others; whence it is assumed that the lacks any strict physical sense. In fact, the
vector of state represents some information field or a quantum object which is the
complex function of the coordinate and the time acquiring a single-valued physical
sense only for a particular physical problem. For instance, when models of solid
6 1 Interaction Between Atoms in the Substance

bodies are considered, the value 2 is the charge density, while it is the probability
density in the quantum statistical physics.
The main equation of the quantum mechanics permitting to combine the vector
with any system using the analytic expression = (x, y, z, t) is the equation
of Schrdinger (E. Schrdingeran Austrian physicist-theoretician, the Nobel Prize
winner). The Schrdinger equation derived in 1926 for motion of electrons has the
following general look:

(x, y, z, t)
i = H , (1.12)
t

where ithe imaginary unity; H the Hamilton (Hamiltonian) operator named after
the Irish mathematician W. Hamilton.
The precise solution of the Schrdinger equation as it looks (1.12) was found only
for the limited class of quantum objects, like the hydrogen atom. The electron in these
quantum systems moves in respect to the atom nucleus in the central coulomb field
within which the potential electron energy U depends only on the distance r to the
atom nucleus. The relevant wave functions characterizing the electron state and its
energy are obtained by solving the Schrdinger equation (1.12):

h c R p Z 2
Un = ,
n 2 (1.13)
(r ) = R (r ) Y (, ) ,

where c the light speed; R p the Rydberg constant equal to 3.29 1015 Hz
(J. Rydberga Swedish physicist); Z atom nucleus charge; n  = 1, 2, 3 . . . , main
quantum number; (r )wave function; R(r )wave function radial portion; ,
angles, spherical coordinates of the electron radius-vector.
The value h c R p is the energy of ionization of the hydrogen atom in the
main state taken with the opposite sign. In addition to the main quantum number n  ,
the atom state is set by the azimuth or orbital quantum number l  = n  1 which
determines the moment of electron pulse in respect to the atom nucleus. Because the
moment value has different projections in the randomly selected direction, still one
more: the magnetic quantum number m e = l  , l  1, . . . , l  is introduced. The
numbers n  , l  , m e characterize fully the electron state in the hydrogen atom. The
electron state in the atom has the following alphabetic designations in the quantum
mechanics, chemistry and spectroscopy:

n  = 1, 2, 3. . ., s, p, d. . .
l  = 0, 1, 2. . ., K, L, M. . .

Because it is impossible to determine exactly the electron position due to the


uncertainty principle, the squared wave function characterizes the probability density
of the electron location at a given space spot. In other words, the electron seems to be
1.2 Quantum Mechanical Description of Atom Interactions 7

smeared in the space forming the electronic cloud with the dimensions proportional
to n 2 . Then the value R 2 (r ) determines the radial and the Y 2 (, ) determines the
angular distribution of the electron density.
The atoms containing two or more electrons have the wave functions to describe
which of the interelectron interactions should be taken into account including the
interelectron repulsion, the magnetic interactions, and the exchange forces. It is
because the electron has the spin which is the quantum mechanic analogue of the
proper moment of rotation and the Pauli moment of exclusion. This spin produces an
additional quantum number m s = 1/2, 1/2 complicating significantly the electron
state description in the multielectron atom. Equation (1.12) has no accurate analytic
solution for these objects. That is why the solutions like (1.13) serve for the similar
quantum systems.
The electron energy in the complex atom can be represented similarly to (1.13):

h c R p (Z )2
Un = , (1.14)
n 2
where the screening constant taking into account approximately the interaction
between electrons.
The perturbation theory apparatus yields more accurate results; this apparatus
is one of the most common methods of approximate calculation in the quantum
mechanics. The perturbation theory is applied to the physical systems which with a
number of simplifications can be represented as the object which has a precise ana-
lytic solution of the Schrdinger equation. A certain value is added to the obtained
expression to take into account the additional interaction (perturbation) discriminat-
ing the atom in question from the hydrogen-like analog. It looks mathematically in
the following way. Equation (1.12) is represented as

(t)
i = H0 + U (t) (t), (1.15)
t


where H0 the hydrogen-like atom operator; U (t)the perturbation operator.
Initially the moment of the quantum system is in one of the steady states deter-
mined by the vector m0 as the plane wave with the energy Um0 . Then the solution
(1.15) looks like a series:
 
i
(t) = Cmn (t)m0 exp Um0 t , (1.16)
m


where Cmn the coefficients of expansion of the wave vector (t).


8 1 Interaction Between Atoms in the Substance

The coefficients of expansion are determined by the differential equation:

Cmn
i = Umk (t)Ckn (t),
t
k
 (1.17)
 
i 

Umk (t) = m U (t) k exp t Uk Um dCk ,


0 0 0 0


where (m0 ) the wave vector of the unperturbed state conjugated in complex with
the unperturbed wave vector m0 ; k, mthe indices of the steady state of the
quantum system.
The function |Cmn (t)|2 determines the probability of transition of the quantum

system characterized by the Hamiltonian H0 , from the initial state m to another state

n under the effect of perturbation U (t).
A particular case of solving the quantum mechanical problems of the theory of
perturbation is stationary problems independent of the time of perturbation. In this
case, Eq. (1.15) is rewritten as follows:
 

H0 + U m = Um m . (1.18)

The solution of this equation corresponds to type (1.16) which is much simplified
for the stationary case:

n = Cm m0 . (1.19)
m

The stationary theory of perturbations illustrates well the shifts and splits of the
atomic levels. Equations (1.151.17) is applied to the study of transitions from one
quantum system state to another.
Atoms with Z 2 have the electrons each exposed to the action of fields of
other electrons of this atom. Their energies can be determined only on the basis
of approximate quantum mechanical calculations. The Hartree-Fok method is the
most popular (D. Hartree, V. Fok are the British and the soviet physicists) [5], who
proposed the approximate quantum mechanical solution of the multielectron atom
states. The solution can be obtained by reducing the original equations to the already
known and solved problem for the single electron atom. A system of equations is
solved for each electron moving in the averaged field of the remaining electrons.
The obtained electron density distribution reflects the structure of electron shells.
Because the solution algorithms are well tried out, it can be assumed that the wave
functions and the energies of the stationary atom states are known. More complicated
quantum systems, molecules among them, are considered by making the following
assumptions [6]:
1.2 Quantum Mechanical Description of Atom Interactions 9

(1) the solution of Schrdinger equations for molecules yields the wave functions
describing the energy levels of molecules (the molecular orbitals);
(2) the molecular orbitals embrace all nuclei of molecules and have the same physical
sense as the atomic wave functions;
(3) the electronic structure of the molecular orbital follows the same law of the
atomic quantum mechanics as the atomic orbital.
It follows from these assumptions that the molecular orbital is a single electron
multicenter wave function describing the state of individual electrons in the molecule
moving in the averaged field of other electrons and in the field of the molecules
nuclear carcass. The molecular orbital includes the spatial 1 (r ) and the spin 2 (S)
components, hence, they are often called the spin-orbitals (r, S) [7]:

(r, S) = 1 (r )2 (S). (1.20)

The designations of the molecular orbitals differ from those of the atomic orbitals.
For instance, the orbitals with the quantum number l  = 0, 1, 2, 3 are designated as
-, -, -orbitals (the Greek characters replace the Latin ones). Since the molecules
can have several orbitals of one type, the additional indexes 1, 2, 3, 4, are introduced
to show how the orbitals rank by energy. Special calculation methods are used to
find the molecular orbitals. Most often they are the method of a linear combination
of atomic orbitals (LCAO) and the method of valent schemes.
Following the LCAO method, the molecular orbital m of the molecule consisting
of N atoms is represented as the linear combination of the atomic orbitals n :


N
m = Cn n , (1.21)
n=1

where the coefficients Cn should satisfy the following relation:

Un
= 0. (1.22)
Cn

If the combination of orbitals boosts the electron density among the nuclei,
the combined orbital is called the connective and its designation is + . If the super-
position of orbitals reduces the electron density, it is the loosening orbital . The
higher energy states with the electron density offset from the internuclear region
correspond to the loosening orbitals. Following the LCAO method, interelectron
repulsion is ignored leading to the equal contribution to the energy of ion and cova-
lent summands, though it is not always true.
The method of valent bonds expands the LCAO method implying that each wave
function incorporated into the linear combination describes the state of a specific
configuration of protons and electrons. For instance, the LCAO method dictates that
the orbital of hydrogen molecules should be (1.21): = C1 1 + C2 2 . The orbital
according to the method o valent bonds looks more complicated:
10 1 Interaction Between Atoms in the Substance

= 1 1 + 2 2 + 1 2 + 1 2 , (1.23)

where 1 , 2 the system in which the first electron is bonded to the nuclei A and
B, respectively; 1 , 2 the system in which the second electron is bonded to the
nuclei A and B.
The exhaustive consideration of all the interactions augments the number of terms
in relation (1.23). Paper [6] provides the information about the equation and the
molecular orbitals of the hydrogen atom containing 50 terms.

1.3 Strong Chemical Bonds

The Van der Waals forces are too weak to form the structures of solid and fluid
bodies. Their structure and behavior are governed by stronger cohesive interactions
which are described with the quantum mechanics apparatus. Let us briefly consider
their main regularities.

1.3.1 Ion Bonds

The ion bonds result from the electrostatic attraction of oppositely charged ions.
When they approach intimately enough, the electron shells touch, but without mutual
penetration because the Pauli Exclusion Principle imposes limitations [7]. That is
why it is believed that the distribution of charges in the substances with ion bonds has
a spherical symmetry which is somewhat upset in the areas of contact with adjacent
atoms (Fig. 1.2).
Ch. Kittel (Ch. Kittelthe American physicist) comments that we are likely
to make mistake believing that the energy and the bonds in ion crystals are mostly
due to the Coulomb interaction The spacing between the positive ion and the
nearest negative ion in the sodium chloride crystal equals to 2.81 108 cm, that is
why the potential attraction energy is equal to 5.1 eV. This energy can be compared
with the energy of bonds of the crystal NaCl7.9 eV [3]. It can be concluded
from the above-said, that it is advisable to use the electrostatic approaching when
calculating the energy of ion bonds. This energy is termed the Madelung energy
(E. Madelung) [8].
If we designate the energy of interactions between ions i and j with Ui j , then the
full ion energy Ui with the account of all interactions is equal to

Ui = Ui j . (1.24)
j
j=i
1.3 Strong Chemical Bonds 11

Fig. 1.2 Distribution of electron density in common salt crystal basic plane

Let us assume that the Ui j can be represented as the sum of two Coulomb potential
and some potential of the repulsion forces variable following the exponential law
 
ri j q2
Ui j = B5 exp , (1.25)
B6 ri j

where B5 , B6 are the constants determined experimentally; q is the ion charge (the
plus sign corresponds to the same ad the minus sign corresponds to the opposite
charges).
The full energy of the lattice comprising the N ions can be written in the following
way:

Ufull = N Ui , (1.26)

or with the account of (1.24) and (1.25) as


 
r M q 2
Ufull = N z B5 e B6 , (1.27)
rij

where z is the number of the neighbors nearest to the ion selected as the coordinate
start; M is the Madelung constant (E. MadelungGerman physicist-theoretician).
12 1 Interaction Between Atoms in the Substance

Formula (1.27) is used more often as


 
N M q 2 B6
Ufull = 1 , (1.28)
r ri j
 
where ri j is the spacing between adjacent ions; the value N rMi j q is called the
2

Madelung energy; B6 0, 1 ri j ; the Madelung constant M approximately assumed


equal to 2ln2.
As a rule, the crystals with predominantly ion bonds feature a higher melting
point, strength, stronger hardness and a lower thermal expansion coefficient. It proves
stronger ion bonds. Since all electrons concentrate in the shells of individual atoms,
these materials possess better dielectric properties.

1.3.2 Covalent Bonds

The classic definition of the covalent bonds is the bonding of electron couples with
covalent bonds [911]. These bonds are called in the chemistry and particularly in
the organic chemistry homopolar (homeopolar) bonds. The covalent bonds are strong
bonds because the energy of bonds between the carbon atoms in the diamond is equal
to 7.3 eV. This value is comparable with the energy of ion bonds, notwithstanding
the fact that the covalent bonds appear between neutral atoms.
When a common electron pair bonds following the Pauli Exclusion Principle,
the bonded electrons should have oppositely directed spins. Hence, these bonds can
be represented as the interaction between atoms and electrons when both electrons
cease virtually to be bonded to particular atoms. Double or triple bonds can appear
in this way when two or three electrons come into common use.
All covalent bonds originating from each atom bond it rigidly to other atoms
and possess vividly pronounced directivity. Each atom in the carbon, silicon and
germanium crystals locates in the tetrahedron center which is formed by four nearest
atoms (Fig. 1.3).
The precise angular arrangement of the bonds depends on their number and the
type of electrons involved in the formation of covalent bonds. For instance, the
electron clouds of the s-orbitals are spherically symmetric; the electron clouds of
three p-orbitals in three mutually perpendicularly directions, those of the d- and
f-orbitals have a more intricately structured cloud (Fig. 1.4).
The electrons located over the hybrid orbitals are involved in the bond formation
of the majority of crystals. The bond direction depends on which orbitals dominate.
For instance, one of the diamond electrons can pass to the p-orbital so that all four
outer electrons happen to get unpaired. When three p-orbitals arrange at an angle
90 one to another and the s-orbital is unidirectional, four hybrid sp-orbitals appear
directed towards the tetrahedron apexes (Fig. 1.3) typically for the diamond.
1.3 Strong Chemical Bonds 13

Fig. 1.3 Tetrahedral structure of covalent bonds

Fig. 1.4 Schematic plot of configuration of electron clouds

The crystals appearing similarly to the diamond are mainly the covalent bonds so
tat they feature high strength, melting point and hardness. They are dielectric like
the ion crystals because all outer electrons bond strongly into the covalent bonds.
The dielectric characteristics of the covalent crystals withstand the melting and
dissolution unlike the substances with the ion bonds.

1.3.3 Metallic Bonds

The metallic bonds appear due to the forces of attraction between the lattice of pos-
itively charged ions and the surrounding gas of free electrons. The free electrons
(the electrons of conductance) move under the external field effect through the metal
like a gas molecule in some volume. This idea agrees poorly with the habitual chem-
ical notion of bonds; the latter states that the bonded valent electrons localize in the
ion islands. It leads to the ion bonds which distribute between atoms forming lig-
ands or behave like the covalent bonds. The valent bonds are saturated in both cases
because the valent electron has to remain all the time in the localized bonded state,
or electrons form pairs following the Pauli Exclusion Principle.
The free electrons possess high motility not bonded to any particular atom; they
render the bodies with metallic bonds highly heat and current conductive. The ideas
14 1 Interaction Between Atoms in the Substance

were that the metallic bonds evolution would explain the magnitude of forces of
adhesion between the atoms in the metallic crystalline lattice rather than the physical
properties of metals. The energy of interactions between the atoms in the system
with metallic bonds is determined by the following relation [2, 8]:

A1 e2 A2 A3 e2
Umet = + + , (1.29)

3
1 2

3

where A1 , A2 , A3 are empirical constants independent of crystalline lattice features;


eelectron charge; atomic volume.
The first term of Eq. (1.29) represents the potential energy of free electrons, the
second term is their kinetic energy, and the third term is the kinetic energy of electrons
occupying lower energy states. The sum of the three states yields the function of
energy distribution (Fig. 1.5) which enables to estimate the heat constants (from
U ) and the compressibility of metals. This model agrees well with the experiment
for the alkaline metals and yields the underrated values for noble metals [10].

1.3.4 Hydrogen and Resonance Bonds

Since the neutral hydrogen atom has just one electron, it should possess one bond
which permits it to interact just with one alien atom. But the hydrogen atom combines
with two atoms under some circumstances producing the so-called hydrogen bond
with the energy about 0.1 eV. It is believed that the oxygen bond has mainly the
ion nature because it appears only between the most electrically negative atoms
like those of fluorine or oxygen. In the extreme case, when the oxygen bonds are
purely of the ion nature, the hydrogen atom gives away its electron transforming into
a proton which bonds with two electrically negative atoms (Fig. 1.6a). The proton

Fig. 1.5 Dependence of full energy on spacing between atoms


1.3 Strong Chemical Bonds 15

Fig. 1.6 Examples of hydrogen bonds: a molecule of hydrogen fluoride; b protein molecules

small dimensions permit to the proton to have only two adjacent atoms; therefore,
the oxygen bonds are achievable only between two atoms.
The oxygen bonds are the most essential type of interactions between water mole-
cules governing to large extents its abnormal behavior when heated. The oxygen
bonds limit the size of protein molecules and govern their geometrical structure
(Fig. 1.6b).
L. Pauling discovered the resonance bonds (L. PaulingAmerican physicist
and chemist) in transitional metals [11]. He assumed that the electrons of the
(n1) dsubshell distribute between the atomic and bonding orbitals. The electrons
within the atomic orbitals bond with individual atoms and influence the magnetic
properties of the substance. The electrons at the bonding orbitals form the covalent
bonds which resonate between a great numbers of directions of the bonds.
The resonance idea presumes existence of single- and triple-electron covalent
bonds with the energy equal to half of that of the relevant single and triple bonds.
The latter takes place in the benzene cycle structure. Whence the conclusion is that,
if several configurations of the covalent bonds, the resonance are able to produce the
structure corresponding to a smaller energy, it is thus more stable compared with the
original structures.

1.4 Collective Interaction Between Atoms

The cohesive forces bonding the atoms and the molecules in the solid body are quite
hard to measure experimentally. The chemical bonds in crystalline bodies have a
mixed nature with bonds of one type dominating. Therefore, the solids can be iden-
tified by the dominating type of cohesive interactions [12] (Fig. 1.7). It is intrinsic to
metals to have the metallic type of bonds little ion or covalent contribution. Chemical
compounds of various metals (intermetallides) are often characterized by a consider-
able covalent component. Ceramic and polymeric materials contain mixed types of
bonds. For instance, covalent bonds are effective in polymeric chains; Van der Waals
forces are effective between chains producing a supramolecular structure. The latter
forces can be strong enough to induce crystallization of polymers.
16 1 Interaction Between Atoms in the Substance

Fig. 1.7 Schematic of contribution of different types of bonds into structures of most common
materials [12]

Hence, the theoretical models permitting to estimate the forces which act between
particles in the solid bodies become ever essential. A.S. Akhmatov emphasizes that
the essential theoretically and practically problem emerges how the law of inter-
actions between the condensed system and the isolated particle close to its surface
produces its effect and how two systems of particles interact. These laws are obvi-
ously the foundation of all phenomena of interactions between the surface of this
phase and the environment the theory of boundary friction and lubrication has
particularly fundamental significance [1].
The energy and interactions between condensed bodies (the term of
A.S. Akhmatov) are determined using the models representing the energy as a sum
of paired interactions between isolated atoms and molecules. The drawbacks of this
assumption are apparent. The external forces field on the solid body surface cannot be
represented as a simple sum of elementary fields of individual particles. It is hardly
doubtful that the phase interactions of the substance in the gaseous, fluid or sold state
are different categories of phenomena. At the same time, the interaction between con-
densed media should be determined by the common properties of the substance. The
best fitting models of calculation are those of Kazimir and Soviet physicist E.M.
Livshits which permit to calculate the energy of interactions between condensed
bodies independently of paired interactions. For instance, Kazimir obtained the fol-
lowing relations for the metallic placed spaced at r one from the other (SGSE) [4]:

0.013
F= . (1.30)
r4
E.M. Livshits treats the interphase field assuming that both solid phases are iden-
tical semispaces separated by a narrow slot r wide [13]. He considered two extreme
cases: r is small and r is larger compared with the main wavelength of the spectrum
of absorption (emission) of the substance. For the r < , the forces of interactions
between phases can be represented in the following way:
1.4 Collective Interaction Between Atoms 17

 2
 1
F= d,
8 2 r 3 +1
0

for r >
 2
 c 2 1
F= (), (1.31)
r 4 240 +1

where  = h/2 Planck constant; substance dielectric permeability; ()


 1 2 
the function obtained by integrating numerically the value +1 d ; c speed
0
of light.
For metals at r > and = , (1.31) transforms into a simple expression:

 c 2 0.013
F= = , (1.32)
240 r 4 r4
which is identical to Kazimir equation (1.30).
E.M. Livshits investigated also the dependence of force F on temperature T . For
metals under the condition r < the following relation was obtained:
   
 c 2 r kT 4
F= 4 1 5.33 . (1.33)
r 240 c

The energy of interactions U (r ) between surfaces of two solids spaced sufficiently


can be calculated by integrating expression (1.31):
 2
 c 2 1
U (r ) = 3 (),
3r 240 +1 (1.34)
k
U (r ) = 3 .
r
The main advantage of Livshits theoretical models is their good correlation with
the experiment and full generality, or what Ahmatov calls independence from any
ideas about the nature and laws of interactions between molecules.
The analysis of the crystal physical parameters (temperature of melting, ther-
mal expansion coefficient, sublimation, compressibility, lattice constant) permits to
obtain a general characteristic of the force of cohesion in the crystal. Large forces
of cohesion are manifested, as a rule, by the crystals with high melting temperature
and high heat of sublimation.
Figure 1.8 that the melting temperature variations have two sets of maxima, one
corresponding to the middle of rows of transitional metals, the other belonging to
18 1 Interaction Between Atoms in the Substance

Fig. 1.8 Melting temperature variations in response to number of subgroup in periodic system [8]

the elements of group IV of the periodic system. When these forces are used as a
measure of forces of cohesion in crystals, it is essential to remember the solid and
fluid phases are at equilibrium at the melting point, while the heat of sublimation is
determined by the amount of work needed to transform one gram-atom of the solid
substance into stem consisting of neutral atoms. Therefore, both physical constants
of the elements are the properties of the substance in all solid, fluid and gaseous
states.
The compressibility of the solid body cm is a relative reduction of its volume as
the hydrostatic pressure augment, i.e. it is a measure of the possibility that the atoms
approach. Hence, the compressibility is the inverse value to the forces of cohesion
in the crystal (Fig. 1.9).
The thermal expansion coefficient of the solid body characterizes the variations of
dimensions (length, volume) of the specimen when the temperature changes by one
degree; hence, it is a measure of the amplitude of thermal oscillations of atoms. The
low thermal expansion coefficient indicates considerable forces of adhesion between
atoms. The dependence of the thermal expansion coefficient on the element atomic
number resembles the similar dependence for compressibility.
1.4 Collective Interaction Between Atoms 19

Fig. 1.9 Variations of compressibility of elements in response to subgroup number in periodic


system [8]

The electronic cloud enveloping the atom nucleus is not a rigid sphere and can
change its size and shape in response to the nature of forces of interatomic inter-
action involving the external electrons of atoms in the solid body. The interatomic
distance d can be assumed a measure of strength of interatomic bonding. However,
direct comparison of the value d among the elements with different crystalline struc-
tures is complicated, the general tendency of this parameter is to reduce when the
coordination number (CN) diminishes.
Norwegian geochemist V. Goldschmidt demonstrated, in particular, that reduction
of the CN from 12 to 8, 6 or 4 shortens the interatomic distance by 12 %. So, he
proposed to reduce the interatomic distances in crystals to the CN equal to 12 and
hold half of this value of the atomic radius. The value of the atomic radii obtained
in this way suits well to the metals with comparatively simple structures, but it
is unsuitable for calculation of the group of transitional metals. Another method to
estimate the atomic radius is to exclude the CN effect o the obtained result. It is based
on the definition of the volume (
) of the substance per atom; if the linear parameter
is needed, then the atomic radius is calculated with the expression
= 43 r03 .
The idea about the atomic volume is convenient because this value can be easily
determined for any structure dividing the elementary cell volume by the number of
atoms the cell contains. Figure 1.10 shows the values of atomic radii.
20 1 Interaction Between Atoms in the Substance

Fig. 1.10 Variations of atomic radius r0 in response to subgroup number in periodic systems [8]

1.5 Crystallization

The collective effect of atoms on the cohesive forces produces ordered structure or
crystals. The process of formation of crystals from the melt, gas, solution or another
crystal is called crystallization [5]. The crystallization as a physical phenomenon
means that original atoms, molecules or ions are packed into the crystalline lattice.
The ordered structure lattice implies that is entropy and chemical potential diminishes
versus the original medium. In other words, the formation of the crystalline structure
is determined by the principle of energy expediency: the full energy of a single
crystal is less than the combined energy of its individual components so the process
of crystallization is accompanied by liberation of latent heat.
The growth of crystals is a multistage process: gas or fluid atoms and molecules
interact and produce specific structures or nucleus at the initial stage which can be
spontaneously disintegrate, but, after a certain of number of Ncr atoms (it is called
the nucleus of critical size) has been collected, they will only grow. The rate of
nucleation is determined by the following relation:
 
Unu
Jnu = Bnu exp , (1.35)
kT

where Bnu the coefficient of proportionality between the density of particles in


the mother medium and the rate with which they join the critical nucleus, it is in
solutions Bsol 1025 1030 cm13 s , in melt Bmel 1037 1042 cm13 s ; Unu the work
of appearance of the critical nucleus; T substrate temperature.
1.5 Crystallization 21

2
Umb = N chm + p (c Ncr ) 3 , (1.36)

where chm the difference between chemical potentials; p the specific fee
energy over the nucleusmedium interface; the averaged surface energy; c
the specific substance in the crystalline state.
 
P
chm = kT ln 1 + , (1.37)
P

where P/P is the relative variation of the condensed steam pressure.


The number of atoms Ncr in the nucleus of critical size is calculated with the
following formula:

2 p
Ncr = , (1.38)
3 chm

According to the classic theory of crystallization, the nuclei appear due to tem-
perature fluctuations initiating aggregation of a new daughter phase. Einstein used
the principles of statistic physics to demonstrate the probability that the fluctuations
alter the equilibrium system entropy proportionally to exp( S/k); soviet physicist-
theoretician, Nobel Prize winner, L.D. Landau verified this provision demonstrating
that entropy changes S are equal to S = W/T , where W is the lest work needed
for fluctuations, T is the medium temperature [14]. The energy W can be replaced
with the energy of Gibbs G or free energy in a number of cases.
The formal theory of heterogeneous nucleation assumes that the nucleus on the
alien substrate has a dome shape with the spherical symmetry and it is character-
ized by the equilibrium contact angle . This angle is determined by the following
relation:
m = nu + s cos , (1.39)

where m , nu , s is the averaged surface energy relating to the melt, nucleus and
substrate.
The nucleation on the easily wettable substrate requires the overcoming of a
weaker potential barrier than on the non-wettable one, so the overchilling is less.
The temperature reduction slows down nucleation because the melt viscosity grows
exponentially so that the nuclei adhere to new particles less frequently. At first the
rate of nucleation is the maximum; it slows down as the temperature reduces and may
become so slow that the melt hardens remaining amorphous. Thus, the packing of
particles in smaller nuclei may differ from that in heavier nuclei. That is spontaneous
nucleation is avoided when growing larger crystals [15].
Crystals can grow layer by layer or normal to the crystal surface depending if it
smooth or rough in the atomic scale (Fig. 1.11). The atomic planes form a smooth
surface of almost fully packed atoms; the planes contain rather few vacancies and
atoms adsorbed in the spots of the relevant crystalline lattice nodes in the next layer.
22 1 Interaction Between Atoms in the Substance

Fig. 1.11 Atom smooth (a) and atom rough (b) surfaces

The edges of incomplete atomic planes form steps containing some three dimensional
angles. When new particles adhere to them, the energy of the atom smooth surface
remains unchanged.
The surface roughness (in the atomic scale) consists of multiple steps. When
a new atom attaches to any its point, but the three-dimensional angle of the step
changing its energy, this growth of crystals is termed directional. The filling of few
vacancies reducing the rough surface energy cannot start any new atomic layer. The
concentration of in the relevant crystalline lattice nodes augments the surface energy
and that is why it is small. The crystal can grow only when the three-dimensional
angles arrange dense on the surface.
The density of steps depends on whether they are generated by two-dimensional
nuclei or dislocations. The nucleation requires the overcoming of a barrier determined
by the step energy. If the spiral dislocation grows, its step acquires a spiral shape
(Fig. 1.12).
The layerwise crystallization from the gaseous phase does not evolve through
direct targeting of atoms at the steps due to a low density it evolves through diffusion
of particles adsorbed on the atom smooth surfaces called terraces.
The shape of growth of crystals (habitus) is determined by the anisotropy of the
crystallization rate and the conditions of heat and mass transfer. The crystals with
rough surfaces have usually a circular shape with rough surfaces. The atom smooth
surfaces appear like facets. The stationary shape of the crystalline polyhedron is such
that the distance from its geometrical center to each facet is proportional to its growth
rate. The crystal results formed by the facets with the least growth rates.

Fig. 1.12 Spiral crystal growth


1.6 Crystalline Structure 23

1.6 Crystalline Structure

The perfect crystal can be constructed by endless regular repetition in space of similar
structural unities. The structural unity of simple crystals consists of one atom, for
instance, copper, gold and others. The structural unity of crystals in more complicated
substances may contain several atoms or molecules. For instance, their number may
reach 104 atoms in crystal. The crystal can consist of several chemical elements
(the common salt NaCl) or contain combined groups of atoms (hydrogen H2 ). The
crystalline structure is described using the elementary portion of the crystalline lattice
periodically repeating in space called the elementary or primitive cell.
The perfect crystal is a body consisting of atoms arranged in the spatial lattice
c follows the rule
so that the vector of main translations a , b,

r  = r + n 1 a + n 2 b + n 3 c , (1.40)

where n 1 , n 2 , n 3 are arbitrary integers.


Ratio (1.40) proves that the crystalline lattice looks the same when viewed from
arbitrary points r or r  . The crystalline lattice is a mathematical abstraction. The
operation of moving the crystal as the integrity parallel to itself is described by the
following vector T
T = n 1 a + n 2 b + n 3 c (1.41)

called translation.
If the individual perfect crystal is assumed as the integrity of atoms, then it fol-
lows from relations (1.40) and (1.41) that their arrangement should obey definite
rules of symmetry; French crystallographer O. Bravais reduced them to 14 types of
elementary cells (Fig. 1.13).
The so-called triclinic crystal lacks the symmetry. Both the angles and the stretches
of the crystal axes are unequal. The cubic crystals are the most symmetric because
all the facets are equally long and the angles between them are right.
In addition to the elementary lattices which coincide only in apexes, there centered
lattices which, depending on the symmetry type, contain structural entities in the
center of facets or in the crystal center. The cubic cells (I and F in Fig. 1.13) are
(BCC) body-centered and (FCC) face-centered cubic lattices.
The number of ways of arranging atoms which satisfy the rules of symmetry is
limited to 230 variants termed spatial groups. The spatial groups differ from the
pinpointed classes of symmetry that the operation of symmetry is conducted in them
not in respect to the beginning of coordinates but in respect to any point in the lattice.
All spatial groups are described in the crystallographic table and designated with
alphabetic symbols and numeric index correspond to the set of elements of symme-
try. The letter designates the crystal chemical composition: Asimple substances,
B, C, Dcompounds, Lalloys, Oorganic substances, Ssilicates. The numeric
index designates the structure type. For instance, A1 designates the cubic FCC.
24 1 Interaction Between Atoms in the Substance

Fig. 1.13 Bravais elementary cells [12]

Crystalline structure of materials features a broad variety. Metals crystallize


predominantly into one of three structure type: body-centered cubic (BCC), face-
centered cubic (FCC), hexagonal densely packed (HDP). Many features of metals
depend on the structure type.
The ceramic materials are compounds having the chemical bonds with oxygen,
nitrogen and carbon. The ceramic crystalline structure depends on several factors,
particularly, on the size of atoms which determine the coordination type. The ceramic
materials form the crystalline structure groups the simplest being CsCl, NaCl, CaF2
which correspond to the cubic symmetry.
1.6 Crystalline Structure 25

Fig. 1.14 Structure of crystalline (a) and vitreous (b) SiO2

The strong dependence of covalent bonds between atoms on the crystallographic


directions can hinder the formation of the crystalline structure. In this case, arbitrar-
ily ordered vitreous bodies appear resembling glass silicates are the most essential
example because they contain the structural complex (Si4+ ) (O2 )2 . It is hard to
achieve a fully ordered crystalline structure; instead chain molecules appear most
often having the nearest order (Fig. 1.14).
The structure of polymeric materials is much more complicated than other materi-
als. The basic distinction of polymers is that their structural unity is the molecular or
polymeric chainthe macromolecule. The dimensions of the polymeric chain reach
thousands of Angstrom that it consists of the blocks called monomer links [16].
The monomers form the internal structure of the polymeric chain and make the
polymers similar to simple substances. According to A.S. Akhmatov, the struc-
ture of the methylene chain resembles the diamond structure with the same inter-
atomic distance and the same tetrahedral angles between valent bonds. The structural
closeness determines the similarity of some properties of these substances: the axial
modulus of elasticity of the methylene chain happens to be equal to the diamond
modulus of elasticity [1].
Let us designate the monomer size in the molecular chain with l/2. A.S. Akhmatov
calls l the period of reproducibility of the structure in which the methylene chain
embraces thee atoms of hydrogen, while the period of reproducibility is 254 , the
molecules of the monomer are 1.27 long, respectively. The polymeric chain is
equal to

l
L chain = n mon , (1.42)
2
where n mon is the number of monomers in the polymeric chain (Fig. 1.15).
The molecular chain length is usually assumed as the shortest distance between
the terminals atoms of the macromolecule stretched without deformation. This value
extensively used in publications is the structural characteristic of the carbon skeleton
of the polymer molecule.
Geometrically there are polymeric chains of linear, ramified and reticular struc-
tures (Fig. 1.16). The macromolecules of linear polymers are long nonramified
molecular chains containing similar or dissimilar monomeric chains. The ramified
polymers have the basic (main) molecular and lateral chains. Chemically the main
chain has homogeneous-chain and heterogeneous-chain polymers. The macromole-
26 1 Interaction Between Atoms in the Substance

Fig. 1.15 Schematic view of methylene chain structure

Fig. 1.16 Schematic view of structures of polymeric chains: a linear; b ramified; c reticular

cules of the homogeneous-chain polymers contain identical atoms, for instance,


carbon atoms. The macromolecules of the heterogeneous-chain polymers contain
different atoms. The macromolecules of the reticular polymers combine together by
transverse chemical bonding producing spatial structures.
The polymeric material always has set of macromolecules with different length of
the molecular chain; therefore, their masses are different. That is why the molecular
mass of a material is determined as the mean value of the molecular mass of all
macromolecules the material contains.
Each macromolecule has its intrinsic order of atom arrangement and availability
of certain links between them. The arrangement of atoms in the macromolecule
determines its spatial configuration, in particular, the presence of definite angles
between chains due to the particular features of covalent bonds (Fig. 1.15). The
modern interpretation of the term configuration of macromolecule relates to the
integral characteristic of the polymeric chain which can acquire different spatial
forms or conformations. There are four configuration levels: link configuration, local
configuration order, long-rage configuration order and chain configuration.
The polymeric chains form more intricate structures due to the appearance of
van-der-waals and hydrogen bonding. When these structures appear, the macro-
molecule folds into a ball or various macromolecules may interact. These formations
acquired the name of supramolecular structures (SMS).
Depending on the nature of the SMS polymers, they are subdivided into crystalline
and amorphous. The crystalline polymeric materials produce the SMS which include
the crystallites with a definite crystalline lattice. As a rule, the elementary cell in these
lattices contains a great number of atoms (Fig. 1.17).
The noncrystalline polymers produce the SMS with the structures lacking order.
It is worthwhile to note that the chemical structure of polymeric molecular chains
does not reflect fully the physical properties. It is exactly the SMS which deter-
mines them to a significant degree. Usually a polymeric material has the crystalline
1.6 Crystalline Structure 27

Fig. 1.17 Polyhexamethylene


dipamide elementary cell

(crystallites) and amorphous interim chains in macromolecules contained in two


crystallite portions. Their relationship characterizes the parameter called degree of
crystallinity.
The crystalline structure is designated following definite rules. The atomic
planes in crystallography serve to characterize the atomic planes in the elemen-
tary cell by introducing some system of coordinates. The internal coordinates in
this system are assume as the unify of measurement on the side of the elementary cell
(Fig. 1.18). The atoms inside the cell are characterized by the fractional units. The
position of atoms in other cells can be obtained by adding the vector of translation T
(1.41) to the internal coordinates combining the general beginning of the coordinates
to the apex of this cell.
The Miller indexes are used to describe the crystallographic planes and directions
quantitatively (W. MillerEnglish crystallographer). They are determined in the
c (1.401.41), which
following way: if the lattice is characterized by vectors a , b,
might be differently long, then the points of intersection of the coordinates with the
atomic planes lie along the axes of the internal coordinates ma, nb, qc. The inverse
multiples (1/m, 1/n, 1/q) serve to obtain the Miller indexes taking them  times; its
value is selected such that the products would be simple digits, in other words:
28 1 Interaction Between Atoms in the Substance

Fig. 1.18 Determination of crystallographic coordinates of atoms in elementary cell

 
1 1 1  
 , , = h, k, l  , (1.43)
m n q

where h  , k  , l  simple digits.


Interaction between particles in crystals is estimated in the following way.
Atoms and molecules in condensed substances stay in the state of relative rest due to
the balance of forces of attraction and repulsion. It is mentioned above that the forces
of attraction prevail at larger distances between particles; as the distance shortens,
the particles approach and electron clouds overlap. The proportion of the forces of
repulsion grows in the total balance of forces due to the two types of interactions: the
coulomb repulsion of the same charges and the quantum exchange forces relating to
the effect of the Pauli principle of exclusion. The balance of these forces insures the
substance steady state manifested by the mechanical properties (hardness, elasticity,
viscosity, etc.).
The shape generating force in the ion crystals is the main electrostatic attraction
of oppositely charged ions. Bonds of metal atoms are determined considerably by
the kinetic energy loss of valent electrons versus the free atom energy. The covalent
bonds appear together with the overlapping of electron shells of atoms; following
the Pauli principle of exclusion, the bonding of electrons takes place among those
with antiparallel spins. The electrons of overlapping shells bond the skeleton ions by
electrostatic attraction.
The forces fields in crystalline lattices influence strongly the energy levels of
atoms. The energy of the main state of the metallic crystal is much smaller than that
of the isolated metal atom. The energy reduction of the main atom in the crystal is
similar to the energy gain by bonds. It corresponds to the alteration of the boundary
conditions of the wave functions. The boundary conditions of free atoms are 0
at r , while the condition of continuity is d/dr . The condition of continuity
of the periodic crystal structure remains the same providing the wave vector k is zero
over the crystalline planes passing through the middle between the adjacent atoms.
The derivative d/dr turns to zero in this case (the wave function recording for the
periodic crystal structure is shown below). This provision permits the wave function
of the main crystal state to have a smaller curvature than the similar dependencies
1.6 Crystalline Structure 29

for the free atom. From the quantum mechanics viewpoint, it is exactly this fact
which proves that the kinetic energy of electrons is smaller in the main crystal states.
It is impossible to calculate the mechanical characteristics of solid materials rely-
ing just on the regularities of atomic interactions due to the reasons disclosed in
Sect. 1.6. But these problems are solved rather easily for perfect crystals [3]. For
instance, the volume modulus of elasticity of ion crystals E y is found in the follow-
ing way:

M e2
E y = Aq , (1.44)
18ri4j

where Aq the constant depending on the electron structure of solids; ri j the spac-
ing between nearest ions; M the Madelung constant (see 1.27); ethe electron
charge.
The common salt perfect crystals have the constant Aq equal to 10.2; it corresponds
to the domination of forces of repulsion at distances 0.3 108 cm [3]. To proceed
to the engineering materials, the deviations from the periodicity of their crystalline
structure and the collective action of cohesive forces should be taken into account.

1.7 Electronic Structure of Materials

The electronic structure of materials is described by the equations of quantum


mechanics; some are shown in Sect. 1.11.6. The quantum mechanical equivalent
of the law of substance preservation formulated in [8] states: if N separated quanta
are collected together so that the earlier independent states become mutually bonded,
this system would comprise only N quantum states. The energy and wave functions
describing these states can vary, but the number remains unchanged. Other quantum
regularities of the electronic structure of materials are the following. The energy
zone width W in the substance with the given spacing between atoms does not
depend on their number in the specimen providing that this number is large. In this
case, the number of states is proportional to the number of atoms. If the crystal has
macroscopic dimensions, in other words, it contains the number of atoms compara-
ble with the Avogadro number (1023 atoms), the distance between energy levels is
infinitesimal and the crystal energy spectrum is quasicontinuous.
According to the Pauli principle, the calculation of the crystalline state should
take into account the indistinguishability of electrons. The law of distribution of
particles is effective in the quantum systems by the energy distribution different
from the classic Boltzmann distribution (L. Boltzmannthe Austrian physicist, one
of the founders of the statistical physics and the physical kinetics) which was named
the Fermi distribution (E. Fermithe Italian physicist, the Nobel Prize winner). The
probability that the quantum state with the energy W is occupied is predicted by the
Fermi statistics in the following way:
30 1 Interaction Between Atoms in the Substance

 
1
W Wf
f (E) = exp +1 , (1.45)
kT

where W f the Fermi energy.


The value W f is the energy to which the probability f (W ) corresponds equal to
0.5. Figure 1.19 shows the Fermi distribution function at T = 0 K and at T W f /k.
The W f at zero temperature corresponds to the upper filled energy state among others.
The W f at other temperatures is found from the condition of being equal to the unity
of the sum probability of all occupied states N .
To describe fully the permitted crystal states, the main equation of the quantum
mechanics, that is to say the Schrdinger equation [17] for a large number of interact-
ing electrons moving in the potential field of the crystalline lattice. The Hamiltonian
operator (energy operator) of this equation is

N 

1
N
2 2
H= i + Ui (r ) + Ui j , (1.46)
2m 2
i=1 i, j


kinetic energy operator; i = xi + yi + z i Laplace operator
2
where 2m
(P. Laplacea French physicist); Ui (r )potential energy of the interactions between
the ith electron and the nucleus; Ui j the energy of interactions between electron.
This problem cannot be solved without certain approximation of the problems
already solved. The best elaborated approximation of free electrons is the one which
excellently describes the structure of metals.
The model of the potential of free electrons added by the ions and electrons of
the crystalline lattice is represented as a homogeneous background with its mean
potential energy being a constant value U(r ) = const. It is counted from U(r ) = 0,
then the Hamiltonian value from (1.46) depends only on the kinetic energy of elec-
trons:

Fig. 1.19 Fermi distribution view: 1 at absolute zero; 2 at limit temperature satisfying condition
kT W f
1.7 Electronic Structure of Materials 31

2 2
H = . (1.47)
2m
The kinetic energy corresponding to the Hamiltonian is the following:

2 2
Wk = , (1.48)
2m
where wave number, = 1/; wave length.
The wave function proper describing the electron state in the crystal is the fol-
lowing:
1
= ei r , (1.49)

where
crystal volume.
For the three-dimensional lattice, equation (1.49) can be rewritten as
 3

1 2 2
= exp i (n x x + n y y + n z z) . (1.50)
L cr L cr

Correspondingly, the crystal energy is


 2
2 2
W = (n 2x + n 2y + n 2z ), (1.51)
2m L cr

where L cr crystal length; n x + n y + n z integers capable to assume positive and


negative values, including zero.
It is apparent from comparing expressions (1.48) and (1.51) that

2
= n , (1.52)
L cr

where n = i n x + j n y + k n z the vector with the components n x , n y , n z .


The Fermi level is the last filled level of the crystal at T = 0 , n 2 = n 2x + n 2y +
n 2z = 0. Assuming that the surface energy levels are spherically symmetric, the
energy of the crystal with the linear dimensions L cr containing N electrons can be
calculated:
2 2
Wf = (3 2 N ) 3 . (1.53)
2m

The density of states N (W ) per single energy interval is found from the equality

W f
N= N (W )dW , (1.54)
0
32 1 Interaction Between Atoms in the Substance

It follows from (1.53) that

W f  3
1 2m 2 3
N (W )dW = W f2 . (1.55)
3 2 2
0

The density of states satisfies equality (1.55):


 3
1 2m 2 1 1
N (W ) = W 2 = CW 2 . (1.56)
2 2 2

Now the mean kinetic energy of free electrons or the electrons of conductivity
can be calculated.
W f
1 2
Wmean = WN(W )dW = W f . (1.57)
N 5
0

The following expression was obtained experimentally for the heat capacity of
metals:
C v = a1 T + a2 T 2 + a3 T 3 . (1.58)

The second and third summands in expression (1.58) can be ignored at low tem-
peratures, then:

2 N k2
a1 = ,
2W f (1.59)
W f = kT f ,

where kthe Boltzmann constant; T f = 5 104 the Fermi temperature.


The constant a1 is about 2 104 cal/(mole g). Substitution of the equation to
find a1 into expression (1.57) permits to find the effective mass m of free electrons
in the metallic specimen:
 2  1
a 1 2 3 3 1 3
m = 2 . (1.60)
k N

The existence of the long-range forces in the metals lattice can be explained from
the viewpoint of interactions between electrons. Let us consider the following system:
the electron gas with the density eN against the homogeneous background of
the positive charge +eN. Let us displace the electrons radially in respect to the
charged background so that the U (r ) acts as the function of displacement r . Then
the charge value moving away from the sphere with radius r is approximately equal
to 4 e2 Nr 2 U (r ). Since the coulomb forces follow the law of inverse squares, the
1.7 Electronic Structure of Materials 33

electron at a distance r from the coordinate start is exposed to the force of attraction
..
equal to 4 e2 NU (r ). Application of U (r ) and the second Newtonian law yield the
equation describing electron motion:

.. 4 e2 N
U (r ) + U (r ) = 0. (1.61)
m
Equation (1.61) is the equation of harmonic oscillations with the frequency

4 e2 N
p = . (1.62)
m

The frequency p , calculated with formula (1.62) is called plasma frequency and
similar oscillations induced by far acting interactions between electrons are called
plasma interactions. Their energy amounts to about 4 eV, the heat oscillations change
somewhat the value p .

1.8 Interactions Between Molecular Chains

The interactions between molecular chains of polymers within the SMS are compli-
cated and more varied than similar processes in metals. It is shown in Sect. 1.6; the
SMS structure of polymers resembles the crystalline structure. The crystal energy U
in the calculation per 1g/mole of the substance is measured by the work of is isother-
mal sublimation during transition into the gaseous state [3]:
 
1
U = U + f 1 f 1 + f 2 f 2 + . (1.63)
n t t

where f the specific surface energy on the crystal facet; W f the total value of
the surfaces of the crystal of this class, the indexes f designate the crystal phase
surfaces.
If these ideas are expanded to the crystallizing polymers, the first term in the
formula can be called the volume energy of the molecular lattice because this portion
of the full energy is determined by the potential energy of macromolecules that are
beyond the action of the surface forces. The second term (the polynomial) expresses
the sum of specific surface energies o the facets of the molecular crystallite (nthe
number of grammoles of the substance). Its value in respect to the first term can be
estimated knowing the relation between the number of particles in the surface layers
and the total number of particles making up the crystallite.
The energy of attraction and repulsion between the links of molecular chains
determines the crystalline lattice energy, hence, the physical and chemical polymer
characteristics too, the melting point in the first place, the constants of elasticity
34 1 Interaction Between Atoms in the Substance

and the plastic flow parameters. This energy can be expressed through the magnetic
susceptibility of the atomic groups [18]:

6mc2 1 2
UA = r 6 , (1.64)
11 1 + 21 2

where 1 , 2 magnetic susceptibilities; 1 , 2 coefficients of polarization.


For the SMS, formula (1.64) is the following
 
3 2 21 2
6 6 6
U A = mc 1 1 r1 + 1 r1,2 + 2 2 r2 .
2 1 1 + 21 2
(1.65)

The first and third terms in formula (1.65) express the energy of interactions
between similar atomic groups, he second term expresses the energy of interac-
tion between dissimilar atomic groups. The components in these formulas are the
following:
3
i = Mr
Na
1 np (1.66)
Mr =
2 + n 2p

where Nathe Avogadro number; nr the refraction indicator; molecular mass;


density; Mr molecular refraction.
The calculations of the CHn -groups show that the coefficient of polarization is
equal to i = 2 1024 cm3 , the coefficient of magnetic susceptibility is
2 1029 cm3 .
To calculate the energy of interactions in the molecular chains, the following
approximation is used frequently: each macromolecule is represented divided into a
series of atomic groups which are considered as pinpoint force centers. The position
of these centers should coincide with the center of gravity of electric charges, in
other words, the atomic groups are treated as pinpoint charges; it is true if the chain
is much smaller than the molecular chain. Then the energy of interactions between
two similar atomic groups is the following:

3
U A = mc2 i r 6 = B A r 6 , (1.67)
2
and between dissimilar groups

3 1 2
UAB = mc2 1 r 6 = B AB r 6 .
2 1 1 + 21 2
1.8 Interactions Between Molecular Chains 35

The calculations show that BCH2 0.4 1010 erg cm6 , and BCH2 CH3 0.5
1010 erg cm6 . The energy of attraction of the CHn -groups turned out within 0.6
1012 erg [1].
In addition to the forces of attraction of polymeric molecules, there are forces of
repulsion; to calculate them the repulsion potential r ep is used. According to the
Slater formula (J. SlaterAmerican physicist-theoretician) the following is obtained:

r ep = M er , (1.68)

where M = 7.7 1010 ; = 2.43rB1 ; rB the Bohr radius called so after N. Bohr,
one of the founders of modern physics; r the spacing between atoms (nuclei).
The energy of repulsion within the molecular chains governing the elasticity (or
compressibility) of macromolecules is calculated with the formula:

U = B ri6 + r ep . (1.69)
i

The calculations show that the full potential of the repulsion value for the
macromolecules consisting of the CHn -groups amounts to 0.2 1013 erg and
the full potential of the interactions of these molecules is approximately equal to
1.3 1013 erg.
The above-listed formulas [1] served to calculate the Youngs modules of elas-
ticity (T. Younga British physicist) among three axes of the coordinates x, y, z,
yielding good agreement with the experimental values. The modulus of elasticity of
the molecular chain of paraffins turned out equal to E x = 1.25 1011 din/cm2 ,
E y = 1.25 1011 din/cm2 , E z 33.3 1011 din/cm2 .

References

1. A.S. Akhmatov, Molecular Physics of Boundary Friction (physical and mathematical literature,
Moscow, 1963), p. 472
2. D.N. Lyubimov, V.A. Ryzhikov, Processes in Friction: Manual (South Russia State Engineering
University, Novochherkask, 2006), p. 147 (Phys. Chem.)
3. Ch. Kittel, Introduction into Solid Body Physics (Mir, Moscow, 1978), p. 791
4. H. Casimir, Proc. Ned. Ac. Watensch. 51, 793 (1948)
5. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 2 (Sov. Encyclopedia, Moscow, 1990), p.
703c
6. E. Cartnell, G.V.A. Fowls, Valence and Structure of Molecules (English translation) ed. by
M.V. Bazilevsky (Chemistry, Moscow, 1978), p. 360
7. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 3 (Sov. Encyclopedia, Moscow, 1992), pp.
672672c
8. R. Kana, Physical Science of Metals, vol. 1 (Mir, Moscow, 1967), p. 329
9. N.L. Glinka, (ed.), in General Chemistry: University Manual, 23 edn, ed. by V.A. Rabinovich.
Chemistry (Khimiya, Leningrad, 1983) p. 704
10. M.K. Strougatskii, B.P. Nadeyensky, General Chemistry (Vysshaya Shkola, Moscow, 1962),
p. 407
36 1 Interaction Between Atoms in the Substance

11. L. Pauling, Theory of Alloy Phases ASM (Cleveland, Ohio, 1956)


12. G. Gottstein, Physical and Chemical Principles of Science of Materials (Binom, Moscow,
2009), p. 375
13. E.M. Livshits, Acc. Sci. Trans. 100(5), 879 (1955)
14. L.D. Landau, E.M. Livshits, Theoretical Physics, vol. 5 (Nauka, Moscow, 1964)
15. A.V. Chichinadze, Fundamentals of Tribology. Science and Technology. Moscow 778 (1995)
16. L.S. Pinchouk, V.A. Strouk, N.K. Myshkin, A.I. Sviridyonok, Science of Materials and Struc-
tural Materials (Vysshaya Shkola, Minsk, 1989) p. 459
17. L.D. Landau, E.M. Livshits, Quantum Mechanics. Non-Relativistic Theory, Vol. 3 (Nauka,
Moscow 1974) p. 464
18. E.S. Kirkwood, Phys. Z. 33, 57 (1952)
Chapter 2
Crystalline Materials: Surfaces of Solid
Bodies

It was not the same Jupiter he new on TV. He had expected


Jupiter would be different, but not much.
American writer Clifford D. Simak

The idea that planets follow absolutely circular orbits is simple


and elegant, but Brahe, Kepler and Newton proved that it was
untrue.
American physicist and writer Alan Lightman

We ask ourselves is sand different from stone? Or may be they


are mainly pebbles? Or be Moon is a huge stone too? Then, if
we understand what stones are, shall we understand the nature
of sand and Moon?
American physicist Richard Phillips Feynman

Abstract The crystal structure of technical materials is considered, systematization


of crystal structure defects is given. Special attention is devoted to dislocations,
their interactions with crystal structure defects and influence thereof on mechanical
properties of solids. The basic notions are reviewed about interfaces, surface layer
of solids and principal surface phenomena (adhesion, adsorption, surface diffusion).

Material point, absolutely solid body, perfect gas, heat machine of Nicolas Carno, are
physical models intended to help researchers study natural phenomena by simplifying
the objects of material world. The perfect crystal treated in the first chapter belongs to
the models. Solid materials are different because their crystalline space is distorted.
That is why additional notions were needed in the near and far order to describe
materials; the notion which were unnecessary before for perfect crystals. The near
order is the substance condensed state in which the ordered arrangement of particles
appears at distances comparable with atomic distances. The far order is the strict
reproducibility of one and the same structural element in all directions throughout
hundreds and thousands of periods of the crystalline lattice. The imperfect structure
of engineering materials governs the values of effective cohesive forces.

D. Lyubimov et al., Micromechanisms of Friction and Wear, 37


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8_2,
Springer-Verlag Berlin Heidelberg 2013
38 2 Crystalline Materials: Surfaces of Solid Bodies

2.1 Defects of Crystalline Structure

The structure of the majority of natural solid bodies is far from that of the perfect
crystal because they contain a great number of defects. British physicist D. Griffiths
believes that the solid body is a collection of cracks [1]. Unless the irregularities
of the structure are taken into account, the modern theory of the solid body would
be unable to assess adequately and quantitatively the physical properties of natural
materials.
Cracks, imperfections, deviations from perfect structure called defects
appear when substances crystallize deforming the physical and chemical homogeneity
of the lattice structure. There are three main types of defects of tribosystems: chemical
composition inhomogeneity, dislocations, and voids [13].
The chemical composition inhomogeneity appears due to the foreign atoms and
molecules in the crystallizing substances. They can occupy the nodes in the crys-
talline lattice under formation producing pinpointed defects which modify the crystal
parameters. Such modifications occur because the extrinsic atom has the dimensions
incompatible with those of the main atoms in crystals. These distortions of the crys-
talline lattice reduce the crystal full energy.
The changes in the crystal volume V due to the intrusion of extrinsic atoms are
calculated using formula [4]

V = 4 Pcr , (2.1)

where Pcr the power of crystalline space expansion; = 3(1 )(1 + )1 ;


Poisson coefficient (S. PoissonFrench mathematician, mechanic, physicist). It is
1.5 for most metals.
The atoms are displaced around the defect in the way that the crystal energy
reduces to the minimum. Concurrently the energy of repulsion of ions in the defective
crystal lattice diminishes compared with the perfect lattice. When an extrinsic atom
is introduced, the crystal energy changes by the following value:

Wd = Wr + We , (2.2)

where Wd the full energy for point defect appearance; Wr the energy of repul-
sion of atoms in closed shells; We changes in the energy of electrons.
 
4 V
We = Wf z , (2.3)
15 

where zthe charge number; V the volume change around the defect in the
infinite crystal; atom volume.
The constant is = 1 for the infinite crystal in (2.1), hence, according to (2.1):

V = 4P, (2.4)
2.1 Defects of Crystalline Structure 39

The point defects imply both cases when the extrinsic atom is present and absent
in the crystalline lattice node. This phenomenon acquired vacancy or the Schottky
defect (W. SchottkyGerman physicist) [5]. It is probable that this lattice node is
vacant in the thermal equilibrium state, when the energy is equal to
 
Wv
pe = exp , (2.5)
kT

where Wv the energy needed to displace the atom from the crystalline lattice node
of the surface crystal; kthe Boltzmann constant.
Another type is the Frenkel defect (Ya.I. Frenkelsoviet physicist-theoretician):
it is the atom displaced into the internode position uncommon for the atom. The
number of defects according to Frenkel is the following:
 
 1 Wy
N F = N N  2 exp , (2.6)
2kT

where N the total number of nodes, N  the number in internodal positions; W y


the energy needed to displace the atom from the lattice into the internodal position.
The deformation of the crystalline lattice with the point defect (Fig. 2.1) leads to
the mechanical stresses in the solid body [2]:
 
d
= C E, (2.7)
d

Fig. 2.1 Effect of point defects on crystalline space: a vacancy; b internodal atom; c slight inden-
tation defect; d slight displacement defect; e Frenkel defect; f Schottky defect (pair of vacancies in
cation and anion sublattices)
40 2 Crystalline Materials: Surfaces of Solid Bodies

Fig. 2.2 Arrangement of atoms around boundary dislocation: a schematic image; b electron
microscope image of dislocations in crystal

where Cv the gradient of point vacancies; dthe lattice period of the main sub-
stance (the solvent); dthe difference between lattice periods of the solvent and
the soluble component; Ethe modulus of the solvent elasticity.
The solvent implies the substance in which crystallization takes place; the soluble
component is the extrinsic substance. If the concentration of the extrinsic phase is
high enough, the distribution of the extrinsic substance in the solvent solid phase is
laminar, the isoconcentrates (the lines equal to the concentration of vacancies) being
parallel to the crystallization front in the plane in which the interface surface moves.
The process evolves more intensively of the interface surface is due to absorption of
extrinsic elements.
Dislocations are another type of defects of the crystalline structure [2, 57]. The
dislocations are the crystalline structure defects along some line. Three mechanisms
determine mainly the dislocations appearing when crystals grow: thermal stresses due
to the stresses induced by the fluctuations in the solid phase and the oversaturation of
the lattice with vacancies. The simplest types are boundary and helical dislocations.
The simplest presentation of the boundary dislocation is the line along which the
crystallographic semiplane breaks (Fig. 2.2). The electron microscopes with high
resolution power enable to observe arrangement of atom rows specific for boundary
dislocations in some crystals (Fig. 2.2b).
The helical type dislocations result when crystal portions displace parallel to the
broken crystalline semiplane rather than perpendicularly (Fig. 2.3). These disloca-
tions are called so because their contour on the plane perpendicular to the line of
break (the line of dislocations) is a helix around the line of dislocations.
The dislocation of crystalline structure fragments in the break plane can occur
at any angle to the line of dislocation rather than parallel or perpendicular. These
mixed dislocations can be considered consisting two types of boundary and helical
dislocations.
2.1 Defects of Crystalline Structure 41

Fig. 2.3 Configuration of helical dislocations

The geometrical dislocations are characterized by the linear element S and the
Burgers vector b (see Fig. 2.3) called after British physicist J. Burgers.
The linear element is the vector tangent to the line of dislocations. If the line is
curved, the direction S is variable. The value and direction of the Burgers vector
coincide with vector towards which the crystal portion above and beneath the plane
of dislocations displace mutually. The modulus of the Burgers vector is usually equal
to the least spacing between atoms in the lattice.
Since the linear element vector S can vary, while the vector b remains constant, the
type of dislocations along its line can vary too. The boundary and helical components
of the mixed dislocation characterized by the Burgers vectors b and be , respectively,
depend on the orientation of the mixed dislocation Burgers vector b to S.  If the angle
between them is equal to , then:
    
  
b = S b S = b cos S,

   (2.8)
 
be = S b S = b sin n,

where nthe normal to the line of dislocations.


The dislocation can be configured as a close loop staying under the surface.
(Dislocation loops appear predominantly during material plastic flow it will be
discussed further). It can be explained in the following way: the crystal becomes
sectioned through the plane along some direction ABCD (Fig. 2.4); the lower part
 After two parts of the
displaces in respect to the upper part on the Burgers vector b.
42 2 Crystalline Materials: Surfaces of Solid Bodies

Fig. 2.4 Origination of dislocation loop: a view of rectangular closed dislocation loop (sectioned
through AACD plane); b arrangement of atoms in upper plane; c arrangement of atoms in lower
plane

crystal coincide, the ABCD boundary of this section represents the dislocation loop.
It is characterized by identical value of the Burgers vector. The loop in Fig. 2.4. can
be represented composed of two boundary AB and CD and two helical dislocations
BC and DA. Both the boundary and helical components are antiparallel.
The closed dislocation loop changes direction at least in two points, but the vector
b remains constant along all loop portions that is why it can consist only of heli-
cal dislocations. The closed loop can be plotted with only boundary dislocations if
the Burgers vector is parallel to the loop plane. These dislocations were called pris-
matic or FrankReed dislocations (J. FrankGerman physicist, W. ReedAmerican
physicist). If the loop ends are fixed, the FrankReed dislocations can multiply under
the effect of sufficiently strong stresses.
The expanding dislocation loop reaches the critical state when its radius becomes
equal to the distance l between the points where the lop ends are fixed. The relevant
displacement stress cr can be calculated with the formula:

Gs b
cr = , (2.9)
l
where G s the shear modulus; bthe Burgers vector modulus.
The continuum physics apparatus is usually used to treat the effect of dislocations
on elastic characteristics of solid bodies [8].
If the shear dislocations u i are known in the Cartesian system of coordinates,
the tensor component of the shear stresses jk , appearing when dislocations move
following the Hook law. The approximation of the isotropic crystal can be rerecorded
as follows [3]:

jj =  + 2G s ejj ,
(2.10)
jk = G s ejk ,
2.1 Defects of Crystalline Structure 43

u j u k u j
ejk = + ; ejj = ;
k j j

= ejj ; = 2G s (1 2)1 ,
j

where i, j, kindexes of the tensor matrix components; the Poisson coefficient;


eij , ejk deformation components.
Each solution of this problem in the theory of elasticity should satisfy the condi-
tions of equilibrium:
jj jk ej
jk = kj ; + + + P j = 0, (2.11)
j k e

where P j the volume forces affecting the dislocation in the crystal.


The solution should obey the boundary conditions corresponding to this problem
and the conditions of compatibility which are recorded as follows:

2 ejk 2 ejj 2 ekk


= + ,
jk k 2 j2
  (2.12)
2 ejj eki eji ejk
2 = + + .
ki j j k i

The solutions for stresses and deformations relating to the helical dislocations can
be obtained with the three stage routine:
1. Let us apply the Hook law (2.10) and determine the stresses which result from
the deformation due to the appearing helical dislocations which can bee recorded as
follows:
b y
u x = u y = 0; u z = arctg , (2.13)
2 x
where x, ythe Cartesian coordinates of dislocations.
Hence, the only non-vanishing stress components are the following:
G s by
x z = zx =  ,
2 x 2 + y 2
(2.14)
G s bx
zy = yz =  .
2 x 2 + y 2

Let us introduce the parameter and r = (x2 + y2 )1/2 , then:


Gs b y
z = ; = arctg . (2.15)
2r x
2. Let us verify if solutions (2.142.15) satisfies the equation of equilibrium and
the boundary conditions. The verification of the compatibility of the equation of
44 2 Crystalline Materials: Surfaces of Solid Bodies

equilibrium can be made directly by substitution providing Eqs. (2.11) and (2.14) are
fulfilled if [4]

2uz 2uz
+ = 0. (2.16)
x2 y2

Due to the stresses zx and zy , the dislocations can be twisted into the loop with
the value equal to

1  
M= G s b R0disl 2 r0disl
2
, (2.17)
2
where R0disl , r0disl are internal and external radii of the helical dislocations.
3. Then the component of stresses z after correction for twisting becomes equal to

Gs b 1 1
z = . (2.18)
2r0disl 2
R0disl + r0disl
2

To find the deformations following the method applied to the helical dislocations
is impossible when applied to the boundary dislocations. Instead the so-called Airy
function of stresses , is used which subordinates the following equation:
  
2 2 2 2
+ + = 0. (2.19)
x2 y2 x2 y2

The Airy function corresponding to the boundary dislocations is the following:

G s by  1
2
= ln x 2 + y 2 . (2.20)
2 (1 )

To calculate the stresses and displacements induced in the crystal the following
expressions are used:

G s by 3x 2 + y 2
x x =   ,
2 (1 ) x 2 + y 2 2
G s by x 2 y2
yy =   ,
2 (1 ) x 2 + y 2 2
(2.21)
G s by y
zz = ,
(1 ) x + y 2
2

G s bx x 2 y2
x y =   ,
2 (1 ) x 2 + y 2 2
2.1 Defects of Crystalline Structure 45
 
b 1 bxy
ux = +  ,
2 2 4 (1 ) x 2 + y 2
(2.22)
b (1 2) x 2 + y2 b x2
uy = ln .
8 (1 ) 2
r0 4 (1 ) x + y 2
2

No matter what mechanism causes stresses in the crystal, it results in the deforma-
tion which would be partly elastic, in the general case, and partly plastic. The relative
value of each of these components depends on the material yield strength Y . The
plastic deformation p resulting in generation of dislocations is determined by he
relation:
Y
p = , (2.23)
E
where Ethe elasticity modulus.
The number of dislocations d per unit of the crystal volume is determined with
the following formula:
p
d = ae , (2.24)
b
where ae the experimental coefficient.
Usually d is inversely proportional to the distance between dislocations.
The energy of the field of elastic stresses around dislocations can be found by
integrating the elastic energy density through the crystal volume. Incase of the helical
dislocations located in the hollow cylinder L long with the radii R0disl and r0disl ,
this energy is

G s b2 L R0disl
Uv = ln . (2.25)
4 r0disl

In respect to the boundary dislocations, the elastic energy is:

G s b2 L R0disl
Uc = ln . (2.26)
4 (1 ) r0disl

The elastic energy of the mixed dislocation is the superposition of boundary and
helical dislocations is found in the following way:
 
G s b2 L R0disl  
Uc = ln 1 cos2 , (2.27)
4 (1 ) r0disl

where the angle between the Burgers vector and the axis of dislocations.
Thus, the dislocations possess a certain energy activity. Their interaction with
point defects in the crystal concentrates them close to the axes of dislocations.
46 2 Crystalline Materials: Surfaces of Solid Bodies

concentrating around dislocations the point defects form cloud or the Kottrell
atmosphere (A. KottrellBritish metallurgist) [7, 9]. The concentration of impu-
rities in the Kottrell atmosphere around the mixed dislocation is calculated from the
following relation [3]:
 
4 (1 + )    
1
Cimp = C0 exp G s b sin Rat Rat Rimp sin (r kT )
2
,
3 (1 )
(2.28)

where C0 pre-exponential multiplier; Ram the atom radius of the main substance;
Rimp the atom radius of the impurity; p the angle of disagreement; r the
dislocation nucleus radius.
The concretion of the Kottrell atmosphere in transparent crystals results in
decoration of dislocations making them visually perceptible. The defect of the Kot-
trell atmosphere can coagulate in the oversaturated solid solutions. Hey are likely to
condense into single-, double-, or triple-layered loops parallel to the densest packed
planes (Fig. 2.5). After the loop diameter exceeds some critical size, the loop sides
flatten and join together. Exactly this defect is called the FrankReed dislocation.
If the flattening does not result in the joining of the boundary loops, the dislocation
originated in the Kottrell atmosphere is the boundary dislocation. If the flattening of
disk is accompanied by the tangential dislocation, then the dislocation loop is of the
partly mixed and partly boundary type.
Dislocations can move and interact fortifying or annihilating themselves. Concur-
rently the crystal full energy changes. The regularities of motions and interactions
of dislocations are considered below.
The following group of defects in the structure and in the solid bodies confirms
the Griffiths thesis that the natural crystal is most often a collection cracks or the
defects of the type of voids. These defects cover the microcracks or Smekal voids
(A. Smekalchzeck physicist) and extrinsic inclusions [1].

Fig. 2.5 Loop dislocation in


Kottrell atmosphere in alloy
Al+5 % Mg. Amplification is
4 104
2.1 Defects of Crystalline Structure 47

Fig. 2.6 Shapes of microcracks: a planar wedge-like; b wedge-like double sharpened; c elliptic
section

The microcrack is the defect in the solid body which can be characterize as a
rupture in the form of slot of micro- and submicrosize. These defects appear when
wet crystals dry under thermal or mechanical effect on the structure, disorders of
growth of crystals, foreign inclusion in the mother medium, etc.
The geometrical shape of cracks in crystalline bodies is different. Two shapes
prevail: wedge-like and elliptical (Fig. 2.6).
A high concentration of mechanical stresses endangering the solid body strength
appears at the top of microcracks. Any variable external deforming stresses pro-
mote the crack propagation encumbered with fatigue fracture of the solid body [10].
The Griffiths theory stipulates that the mechanical stress needed for the crack to
develop is

2Ed
= , (2.29)
l0.5

where d surface energy density; l0.5 crack semilength; Emodulus of material


elasticity.
The microcracks may have multiple lateral branches producing intricate geomet-
rical systems. It is believed that the integrity of communicating cracks propagates in
many cases throughout the entire solid body volume forming a single capillary sys-
tem. The latter possesses a large total surface and plays an essential role in interactions
of materials with environment. The microcapillary system changes into the bonds
following the thermal and mechanical solid body evolution. These changes have usu-
ally the progressing nature gradually making the microdefects grow and ramify into
the net. Opposite changes are possible in particular cases, such as the thermoplastic
irreversible collapse of surface microcracks discovered by A.S. Akhmatov.
The engineering materials have a special type of defects of the crystalline struc-
tures in the form of voids in the lattice. They appear due to the lack of whole group
of atoms (or ions) in the spots where hey should be according to the structural sym-
metry of the crystalline space. Thus, the crystal acquires a number of free microvol-
umes showing the known regularity in their spatial distribution. This type of defects
acquired the name of Smekal voids. The latter are not geometrically closed voids,
48 2 Crystalline Materials: Surfaces of Solid Bodies

hence, foreign substances can penetrate into them and strongly affect the physical
and chemical characteristics of the solid body.
The foreign substances in the crystalline lattice resulting from the release of
microscopic and ultramicroscopic particles in the crystal are a special type of defects.
They appear due to a broad variety of mechanisms. They can appear as the crystal
grows at the initial stages of crystallization, or due to a variety of secondary processes
like artificial introduction of foreign substances into the crystal.
Thus, the structure of crystalline materials differs from its perfect model by the
existing structural defects. They can have a decisive effect on the physical, chemical
and mechanical characteristics of materials. The engineering materials are more
active chemically than perfect crystals but their strength is many times more inferior.

2.2 Polycrystals

Polycrystalline materials together with monocrystalline materials are widely spread


in nature, metals are among them [5, 6]. Their structure and composition are inho-
mogeneous because metals consist of numerous grains of adjacent grains of crystals
called crystallites. The latter are fine monocrystals without distinctly pronounced
facets. Most often they are of micron size that is the attendant crystalline structures
are called microstructures. The microstructures can differ by size, shape and grain
orientation. Each polycrystalline material has internal boundaries separating the adja-
cent grains of crystallites. The internal boundaries can be the boundaries of grains of
one and the same phase or interphase boundaries. Boundaries in any case are parts of
substances with a higher energy than that of grains, so there is a motive force striving
to reduce the surface boundaries and to induce their dislocation correspondingly.
So, the intergrain boundary is a two dimensional defect of the crystalline lattice
separating polycrystal regions with same but differently oriented crystalline structure.
The lack of insight into the microstructure of intergrain boundaries is apparent.
If the adjacent grains are disoriented slightly, the boundary looks like a periodic
sequence of dislocations; it is a small-angle boundary. The large-angle boundary
separates the grains of the stronger (over 150 ) disoriented crystalline structure. The
spacing between the spots where the dislocations reach the examined surface of the
large-angle boundaries changes leapwise in steps at least one interatomic spacing.
Figure 2.7 shows the photo of the boundaries of both types made with the electron
microscope.
The symmetric small-angle boundaries consist of the same set of boundary dislo-
cations with the Burgers vector b and the spacing between the dislocations D, which
reduces as the angle of disagreement d grows [6] Fig. 2.8.

b
d , (2.30)
D
2.2 Polycrystals 49

Fig. 2.7 Large- and small-angle boundaries [6]

Fig. 2.8 Arrangement a and spacing between dislocations b for symmetric (100) small-angle
boundaries

It is necessary to have at lease two sets of dislocations with the orthogonal Burgers
vectors for two asymmetric small-angle boundaries (Fig. 2.9). The effect of dislo-
cations on the structure of crystals intensifies as the angle d grows (Fig. 2.9). The
boundaries this type of dislocations produce relate to the small-angle boundaries.
In addition to the mentioned small-angle intergrain boundaries, the solid bodies
manifest the small-angle twisting boundaries which contain also at least two sets
50 2 Crystalline Materials: Surfaces of Solid Bodies

Fig. 2.9 Arrangement of


dislocations over small-angle
asymmetric boundaries with
angle of inclination d

helical dislocations. The mixed boundaries comprise the dislocation grids with three
Burgers vectors.
Some crystallographic planes pass through the intergrain boundary and continue
to adjacent grains. It means that some atoms in these planes are in perfect position
in respect to the atoms in adjacent crystallites even on the intergrain boundaries.
These nodes of lattices are called the nodes of coincidence; they form the lattice of
coincidence positions. The parameter  is introduced as the measure of density of
coincidence positions in the polycrystal:

Vc. p.
= , (2.31)
Vc

where Vc. p. the volume of the lattice elementary cell with coincidence positions;
Vc the volume of the crystal elementary cell.
The small-angle boundaries can be characterized by the value of the parameter
 close to 1. For instance, for the large-angle boundaries, it is  > 1; it is  = 5
when the angle of rotation is 37 .
The energy of intergrain boundaries is the sum of energies of separate disloca-
tions; the sum depends on the orientation of adjacent grains.
The zero on the graph abscissa axis (Fig. 2.10) corresponds to the absence of
boundaries. Even slight disorientation of two adjacent grains (10 ) induces the energy
leap almost to the maximum. Then the energy reduces somewhat just up to the angle
range about 70 and remains constant afterwards [2].
It is noted above that the excessive energy of the intergrain boundaries prede-
termines the crystal drive to minimize the dimensions and the energy as a result.
2.2 Polycrystals 51

Fig. 2.10 Dependence of energy of boundaries between grains of silicon iron on angle of dis-
orientation of grains

The motion of boundaries is affected by the geometrical dimensions of grains, the


temperature, and insoluble impurities in the crystal. Most Polycrystals posses the
property of anisotropy likely due to the anisotropy of grains and presence of impu-
rities. It should be noted that anisotropy can be observed in one crystal s and not
observed in others. For instance, the common salt possesses the anisotropic crystal
mechanical properties (the elasticity differ along the ribs and along the diagonal of
the crystalline lattice), while its thermal and optical characteristics are isotropic with
a high degree of certainty.

2.3 Solid Body Surface

J. Gibbs, the outstanding American physicist, one of the founders of thermodynamics


and statistical mechanics, is honored as the forefather of the solid body surface theory.
He was the first to teat the surface as the individual system in his fundamental work
The equilibrium of heterogeneous substances in which he considered the surface
different from the thermodynamic phases it shares [1, 2, 11, 12]. The surface layer,
though being insignificantly thick, is a specific substance state with its intrinsic
energy, entropy and other thermodynamic parameters. This approach permitted to
Gibbs to develop a macroscopic surface and surface phenomena theory. Further study
of the properties of surface layers lead to realization of huge difficulties which their
existence introduced into the strict mathematical apparatus of crystallography. It was
not without reason that great physicist-theoretician W. Pauli said vividly that devil
in person created the surface [13].
The apparatus of quantum mechanics genuinely revolutionized the surface theory
evolution. It enabled to develop the electron theory of surface layers and explain
most of the effects relating to the specific structure of the solid body surface. The
surface is one of the basic defects of the crystal three dimensional structures [13].
52 2 Crystalline Materials: Surfaces of Solid Bodies

The broken surface chemical bonds alter the coordination of the sphere of surface
atoms and re-hybridize their valence orbitals. It results in new surface electron states,
the effective charges of surface atoms, the order of their arrangement and the inter-
atomic distances change. The latter results in the crystal without extra deformation
or distortion of its phonon spectrum. That is why the surface is rarely treated as the
geometrical plane. The substance layer adjacent to the surface (the surface layer)
is three-dimensional phase which has a number of physical properties different the
volumetric characteristics of the condensed body.
Nevertheless, to make further calculations easier, let us apply the model homo-
geneous surface. It may look simple but it yields often good correlation with the
experiment. Let us imagine the physical properties, the arrangement of atoms, and
the distribution of their effective charges over the homogeneous surface change only
normal to the surface. Thus, problem from three-dimensional turns into the one-
dimensional.
The atoms of the surface layer even in perfect crystals are in principally new
conditions from the solid body atoms. While the atom within the crystalline lattice is
surrounded regularly by adjacent atoms from all sides, the surface atom has the rele-
vant surrounding only from one side. It generates the force which affects the surface
atom in the direction of the crystalline lattice interior (Fig. 2.11a). The combination
of these forces causes surface tensioning particularly vividly manifested by fluids. It
is exactly due to the surface tensioning the free falling drop becomes a ball having
the least surface area.
The surface tensioning of solid bodies causes the deformation of surface layers. It
reduces the crystalline lattice period versus the period in the crystalline space deeper
regions (Fig. 2.11).
Figure 2.11a, b, leads to the conclusion that atoms in surface layers have free
(uncompensated) chemical bonds. The electrons of these atoms possess a natural
tendency to produce new links. While there are no foreign atoms on the surface, the
surface electrons have only one opportunity to couple, it is to bond to adjacent surface
atoms. It leads, in the simplest case, to pairing of the atoms and the resulting pairs
were called dimers. The atoms of each dimer approach mutually and concurrently

Fig. 2.11 Forces affecting atom within (a) and on surface of crystal (b), deformation of crystal
surface layers (c)
2.3 Solid Body Surface 53

move away from the adjacent surface atoms, the latter, in their turn, form mutually
their own dimers. This process is accompanied by changes in the original crystalline
lattice.
The crystalline lattice of surface layers (even that of the perfect crystal) is strongly
distorted and it corresponds to a high concentration of dislocations accumulating on
the surface. The solid body surface has in fact the excessive energy compared with
deeper layers of the crystal. His energy is proportional to the surface area. Among
the works dealing with the surface energy calculation, the results of Gilman are note-
worthy (J. GilmanAmerican physicist and chemist) [14] because the calculation
is relatively simple and the results feature good correlation with the experimental
results. The Gilman equation for the crystal surface energy is the following:

Ea02
Usur = , (2.32)
2 y0

where y0 the spacing between adjacent atomic planes; a0 the radius of action of
cohesive forces; Ethe elasticity modulus.
The difference between the surface crystalline structure and the volumetric crystal
structure corresponds to the same large difference in the electron structure.
Due to the crystalline lattice periodic structure, the electron density follows the
Fourier harmonic law (J. FourierFrench mathematician and physicist):

1   
e = Fhkl exp 2i h  x + k  y + l  z , (2.33)

h,k,l

where x, y, zelementary cell coordinates; elementary cell volume; Fhkl


structural amplitude [15].
Figure 2.12 shows the diagrams of distribution of the crystalline lattice energy
and the field.
A simple relation combines the electron density and the wave function of the
perfect crystal in the quantum theory of solid bodies:
x
e = ||2 cos2 , (2.34)
d
where dthe spacing between lattice nodes.
When considering the surface electron states on the solid body surface, substantial
deviations from the periodic structural order should be taken into account. Namely
the discrete periodic fields of the lattice ion carcasses in Fig. 2.12a, are replaced with
the evenly positively charged ground creating a homogeneous field averaged over
the space. This approach is vividly called the gel model [16]. According to the gel
model, the electrostatic potential created by the surface electric field is the following:


1 e (r  )
= Z e2 + dr + ex , (2.35)
|r R| |r r  |
R
54 2 Crystalline Materials: Surfaces of Solid Bodies

Fig. 2.12 Perfect crystal energy picture: a potential electron energy change; b wave function
describing electron state

where Z ion charge; Rradius-vector characterizing the atom position in the crys-
talline space; ex exchange potential.
The exchange potential is the characteristic of the quantum nature of exchange
interactions fortifying the interatomic and intermolecular bonds by increasing the
electron density between positively charged nuclei. The exchange interactions have
the non-Coulomb nature and depend on the type of wave functions describing how
the electrons behave in quantum systems.
The electron density profile for the mail state in the gel model possesses the planar
translation invariance, see Fig. 2.13. but the density e (z) variance has two particular
features: first, electrons spill out into the region z > 0 (vacuum) and produce
a surface electrostatic dipole; second, the function e (z) oscillates when its value
approaches the value compensating the volumetric charge. These are called Friedel
oscillations after their discoverer French mineralogist Ch. Friedel. The wavelength
of these oscillations is the following:

 1
, Fr = 3 2 e ,
3
Fr = (2.36)
Fr
The Friedel oscillations originate because electrons in metals can move freely
when their energy is close to the Fermi energy producing almost monochromatic
waves; here interference over the metal boundary causes oscillations. The electrons
having the constant wave vector with the value between 0 and Fr try to screen off
the distribution of the positive ground charge which contains a step at z = 0. The
Friedel oscillations contribute significantly to formation of the surface dipole layer
playing a sizeable role in surface processes [17].
2.3 Solid Body Surface 55

Fig. 2.13 Electron density profiles in gel model

The processes governed by surface phenomena are adhesion, adsorption, and


surface diffusion.
Adhesion is the binding of surfaces of unlike bodies in contact [2, 14]. Adhesion,
in narrow sense, is the ability of interfaced materials to produce strong compounds.
The atomic molecular origin of adhesive forces is the same as solid body cohesion.
Adhesion depends on the nature contacting condensed bodies, the surface property
and the contact area. Thee adhesion amplifies if the bodies are electrically charged,
if donor-acceptor bonds form in the contact, if there is capillary condensation from
the surface of one body on the other resulting in the contact chemical bonds.
The adhesive forces of interaction strongly affect the structure of crystalline
bodies. For instance, when two copper specimens with the face-centered cubic struc-
ture are brought into contact, strong enough adhesive forces appear. The densely
packed hexagonal crystals interact considerably weaker. If two specimens of unlike
metals are brought into contact, their adhesive interaction is much weaker than that
between two like metals.
The interface between the solid body and the fluid induces the wetting effect. The
double electrically charged layer and donor-acceptor bonds play a special role in
the interface near contacting surfaces. The force of adhesion is determined by the
free surface energy loss by the interfaced bodies which are equal to the equilibrium
adhesion function WA :

W A = (13 + 23 ) 12 , (2.37)

where 13 , 23 , 12 the surface tensioning coefficients between interfaced bodies


(1 and 2) and environment (3).
56 2 Crystalline Materials: Surfaces of Solid Bodies

The problems of determination of adhesive forces in metal-polymer interfaces


are due to the fact that they are strongly affected by the state of interfaced surfaces.
The adhesive forces are large during adhesive interaction between polytetrafluorine
ethylene (PTFE) and purified tungsten specimen surface in the vacuum. The elec-
tron and ion microscopy reveals that fragments of PTFE molecules on the metallic
specimen surface form groups of tightly bonded particles or clusters. The Auger
spectroscopy revealed that the fluorine atom at the polymeric chain end forms the
physical and chemical bonds with the specimen surface. The adhesive interaction
between the PTFE compounds and tungsten is much weaker in the air.
The combination of experimental methods of adhesion study has obtained the
title of adhesiometry and the instruments to implement it are adhesiometers. As a
rule, the methods of adhesiometry serve to measure the forces of fracture of adhesive
compounds.
The researchers have at their disposal the nondestructive methods which enable
to judge the forces of adhesion by variations of physical and chemical characteristics
of contact interactions. For instance, the energy of adhesion is rated by the boundary
wetting angle with the help of the Dupret-Young law (J. DupretFrench physicist
and chemist):

W A = 12 (1 + cos ) . (2.38)

Adsorption means the absorption of the condensed body (adsorbent) by the sur-
face layer from the substances in the gaseous or fluid phase. The desorption or
reverse adsorption is possible when the adsorbents are regenerated by eliminating
the absorbed substances (adsorbates). The volumetric adsorbate can be calculated
using the Freundlich formula (H. FreundlichGerman physicist and chemist):

Vad = ad Pad
n
, (2.39)

where Vad the volumetric adsorbate; ad the constant depending on the nature
of adsorbate and adsorbent; Pad the adsorbate pressure; n varies from 0.1 it 0.9
(it tends to increase as the temperature grows).
The main thermodynamic equation describing adsorption is the Gibbs equation:
 

G Gb = , (2.40)
chem T

where the tensioning coefficient at the phase interface; chem the adsorbate
chemical potential.
The Gibbs equation serves as the original to derive particular equations of adsorp-
tion. Usually the adsorbent surface is always coated by adsorbate particles; the longer
the particles stay on the surface, the thicker becomes the coat. It is due to the forces of
interactions between the adsorbent surface and the adsorbate particles. The surface
energy reduces in the process of adsorption. It proves that adsorption is an energy
gainful process with the ensuing consequences.
2.3 Solid Body Surface 57

Fig. 2.14 Hydrogen atom and


its antipode reflected in metal

There are physical or simple adsorption and chemosorption when chemical bonds
appear between molecules of substances and the sorbent. The force of attraction
should accompany the physical adsorption to bond particles to the solid body surface.
Van der Waals forces play this role (see para 1.1), their mechanism is the polarization
effect.
Electrons in the solid body are kept spaced due to the Coulomb and quantum
exchange interaction based on the Pauli principle. These forces and polarization effect
in the positively charged region around the electron called the hole with exactly the
same charge as the electron but with the opposite sign. Moving through the crystal,
the electron trails the tail of holes. When the electron quits the solid body, it leaves
a hole on the solid body surface as the polarization charge. Similar processes evolve
when atoms approach to the solid body surface. The example of interactions between
hydrogen atoms and the metal surface (Fig. 2.14) gives the idea how the symmetric
antihydrogen appears in the metal. In case of the multielectron atom the pattern
of interactions is more complicated: quantum mechanical ion exchange interaction
adds to the purely Van-der-Waals mechanism of adsorption.
The electrons are kept spaced due to the exchange components of the surface
forces. At the qualitative level, the electron to create its own exchange correlation
hole [16] with a single unaffected by other electrons. The electron moves through the
crystalline space and rails the shell of holes called figuratively the polarization coat
in [16]. The graphic presentation serves to describe vividly how the atom interacts
with the solid body surface representing it as the line of constant charge density
reflecting objectively the polarization processes.
The electron density configuration in Fig. 2.15a, corresponds to the case when
a single atom is on the electron gel surface. The electron densities in the atom
and on the solid body displace. The contour map is more demonstrative showing the
difference between charge planes in the atom-surface system and the superposition
of charge planes of the free atom and the pure surface (Fig. 2.15b). The surface
region with the deficit of electrons is well seen which can be represented as the
corresponding mirrored antiatom.
The localization of valent electrons of the atom adsorbate on the atom side
facing the adsorbent surface does not mean that bonds appear in the chemical sense.
Bonds appear due to the same cause why the energy of valent electrons reduces
because their own holes surround them. The energy of this interaction keeps atoms
58 2 Crystalline Materials: Surfaces of Solid Bodies

Fig. 2.15 Constant charge density line calculated for xenon adsorption on gel, a full charge density;
b changes in charge density induced by adsorption. Full line corresponds to excess of electrons,
dotted lines correspond the deficit of electrons

Fig. 2.16 Dependence of


energy of submersion of
atoms into gel on density of
gas of electrons

on the surface with the help of the mechanism of physical adsorption. This energy
amounts to 
e (r  )ex (r )
Wad = dr dr  , (2.41)
|r r  |

where e (r  ), ex (r )the distribution of electrons and holes in the adsorbent.


The true chemical bonds between the surface and external atoms appear as a
result of chemosorption when atoms of adsorbate and adsorbent particles blend
directly. This process is called immersion of atoms in gel. The graphic image of the
dependence of energy of interactions between free atoms in the gel of the adsorbent
surface layer has the minimum corresponding to the potential hole (Fig. 2.16). To
obtain this dependence, the electron density was assumed identical to the charge
density change in the adsorbent close to the surface. The constant charge density line
is shown in Fig. 2.17.
Figure 2.17 shows the constant charge density line (Fig. 2.17a) and the electron
density contour map (Fig. 2.17b) for the Clcopper surface (gel) system [16].
Figure 2.17b, shows clearly the surface region of fusion of electron densities of
these elements evidencing the processes of chemosorption and (atom submersion
2.3 Solid Body Surface 59

Fig. 2.17 Constant charge density line during adsorption of atoms Li, Si, Cl on gel, a full charge;
b targeted charge. Lines: designations see Fig. 2.15

in gel) and the regions of the minimal electron densities they induce when forming
the polarization coat.
The difference of electron densities appearing during chemosorption permits to
introduce the notion of the binder (the maximum electron density) and the antibinder
(the minimum density) levels which represent the system of sorbed atoms or mole-
cules called the surface cluster or surface molecule [16].
To describe molecular adsorption, the orientation of adsorbate particles should
be known in respect to the substrate. The matter of their spatial arrangement during
adsorption is omitted just because of the symmetric molecules. But these molecular
structures are capable to polarize and often acquire the charge density asymmetric
distribution. The macromolecules and molecules of lubricating materials have the
dominating chain structures. The chain structure position on the solid body surface
is characterized by two coordinates x, y and two angles , (Fig. 2.18).

Fig. 2.18 Chain molecule


orientation on plane
60 2 Crystalline Materials: Surfaces of Solid Bodies

Fig. 2.19 Orientation of chain molecules with active center on metallic specimen surface: a mole-
cule experiences torque; b nonelastic collision occurs; c absolutely elastic impact

The adsorbed molecule orientation in respect to the adsorbent surface is deter-


mined by the interaction between the surface adsorption centers and the molecule
active centers. In the simplest case, molecules are devoid of active centers at all or
have onetwo centers at the molecular chain ends or through its length.
The adsorption orientation of polar chain molecules with a single adsorption center
most worthwhile to discuss from the tribological viewpoint because they orient in
respect to the tribosurface of molecules of the lubricating material (Fig. 2.19).
When the solid body surface interacts with electrical fields, the charge redis-
tributes in the polar molecules and dipole group appear. Depending on the polar
molecule orientation in the surface force field, their interaction resembles different
impact during collision (Fig. 2.19). The mobility of adsorbed molecules is as high as
the temperature, and it is high the lower the level of saturation of the monomolecular
adsorption layer and the fewer the obstacles to molecule motion over the surface.
The multimolecular structure resulting from chemical reactions has the most free-
dom to move. The appearance of the primary molecular monolayer is accompanied
by desorption and evolves with the dynamic adsorption being in equilibrium. The
occupation of free spots with adsorbed particles on the adsorbent surface depends on
the length of chains and the position of active centers a along the chains (Fig. 2.20).
Then follow the next rows of the multimolecular layer.
Diffusion is the motion of the medium particles resulting in the transfer of sub-
stances and leveling of concentrations or the equilibrium distribution if concentra-
tions of particles of the given type in the medium [2, 17, 18]. The process of diffusion
is treated farther in more detail; now one of the features of this phenomenon is con-
sidered, it is the surface diffusion. The surface diffusion implies the displacement of
particles by random thermal fluctuations over the solid body surface within the first
layer of atoms. The energy barriers the particles overcome on the diffusing surface

Fig. 2.20 Formation of multimolecular adsorption layer of polar molecules


2.3 Solid Body Surface 61

Fig. 2.21 Schematic image of surface diffusion

are much lower than their volumetric values. Hence, the energy of activation of the
surface diffusion makes up just a part of the energy of activation of the volumetric dif-
fusion. The surface diffusion process proper is usually considered as the sequence of
elementary acts shown schematically in Fig. 2.21. When substances under the effect
of thermodynamic forces, the atoms detach from the second atomic plane, move
along smooth terraces, slide off the lattice step and get trapped in the vacancies.
Groups of bonded atoms or some regions of the surface layer as a whole can move
during surface diffusion. Usually the surface diffusion is considered as the process
altering the adsorbate concentration from the non-equilibrium to the equilibrium
distribution when there are corresponding concentration gradients.
In the general case, the solid body surface has a laminar structure (Fig. 2.22).
The topmost are the films consisting of the atoms and molecules adsorbed from
the environment. This film is differently thick: it is thicker where the defects of the
crystalline structure accumulate (the adsorption centers). The mechanical working
of the engineering materials (forging, extrusion, etc.) activates the surface layers
additionally.
The solid body surface is not absolutely smooth, at least due to the dislocations
creating dislocation steps. This type of surface deviations from the perfect state are

Fig. 2.22 Scheme of solid body surface: 1 films of adsorbed molecules, 2 surface layer proper, 3, 4
layers modified by mechanical treatment of structure, 5 solid body crystalline structure proper
62 2 Crystalline Materials: Surfaces of Solid Bodies

called the submicroroughness. Its parameters vary from fractions of a micron to


hundreds angstroms; they are essential nanotopology characteristic of the surfaces
of solid bodies. The problem of submicroroughness effect on contact interaction
between solid bodies is currently the object of thorough research.
Other surface deviations from the perfect relate to the machining of materi-
als. They are subdivided into waviness and roughness and are treated in many
publications [19].

2.4 Phase Boundaries

The surface of condensed bodies predetermines a specific type of phase boundaries


[1, 6]. The contacting phases of one substance having incompatible lattice parameters
or an entire different nature, make up a distinct interface boundary (Fig. 2.23a).
If the crystalline structure is not disoriented, the phases coalesce; all crystal-
lographic planes continue through the phase boundary (Fig. 2.23b). The coherent
boundary phase forms when the phases possess identical structures, but they have
twin mutual orientation. In his case, all lattice nodes of one crystal are lattice nodes
of the other crystal.

Fig. 2.23 Arrangement of atoms of one same substance at phase boundary: a disagreement between
lattice parameters; b latent boundary
2.4 Phase Boundaries 63

Fig. 2.24 Semicoherent boundary (nm): a scheme, b microscopic photograph of boundary between
Pt and NiO (letters indicate ruptures of crystallographic planes) [6]

As crystals grow, the disagreement between the structure parameters over phase
boundary augments and the elastic energy increases at the phase boundary. The
energy is gainful to compensate the disagreement by producing boundary disloca-
tions which relieve mechanical stresses. This type of boundaries is termed partially
coherent type (Fig. 2.24).
If both phases possess absolutely different structures, the coherence is fully lost
and the incoherent phase boundary (Fig. 2.25). Like in the case of the intergrain
boundaries, some energy can appear at the phase boundary and preferable lattice
configurations. The phase boundaries in composites are, as a rule, in the nonequilib-
rium state (Fig. 2.25).
The insufficient knowledge of the structure phase boundaries inhibits the descrip-
tion of their properties at the atomic level [6]. The phenomenological description is
usually used combing the interface properties with the physical parameters of inter-
actions between phases which can be determined experimentally. One of them is
the surface tensioning which is associated with excessive surface free energy and
determines the equilibrium type of the interface surfaces in mixed phases.
Let us consider the fluid drop shape in equilibrium on the solid body surface
(Fig. 2.26). The surface tensioning act along the contact line as the forces conditioning
the drop configuration based on the criterion of the minimum energy. The equilibrium
in vertical direction takes place if
64 2 Crystalline Materials: Surfaces of Solid Bodies

Fig. 2.25 Incoherent boundary: a schematic view; b high resolution electronic microscope image

Fig. 2.26 Scheme of fluid drop in equilibrium on solid body surface

sg = s f + f g cos . (2.42)

This equation (the Dupret equation) enables to estimate the wetting angle .
The drop spread over the solid body surface (2.42) is a particular case of the
equilibrium of three phases (Fig. 2.27). In the general case, the Herring equation is
true (C. HerringAmerican physicist theoretician):
2.4 Phase Boundaries 65

Fig. 2.27 Three phase system equilibrium. I and IIfluids, IIIgas

23 13
=
(1 + 2 3 ) sin 1 + (3 1 ) cos 1 (1 + 1 3 ) sin 2 + (1 2 ) cos 2
12
= , (2.43)
(1 + 1 2 ) sin 3 + (1 3 ) cos 3

where i = i (ln hkl ) determines the dependence of the interphase energy on the
adsorbate phase orientation in respect to the selected crystallographic directions of
the adsorbent crystalline lattice. The subline indexes correspond to Fig. 2.27.
The surface energy of boundaries of some crystallographic phases turns out very
low, for instance, that of the coherent twin boundaries. If the energy of all three bound-
aries is independent of their orientation in space, the Herring equation is simplified
to the Young equation:
12 23 31
= = . (2.44)
sin 3 sin 1 sin 2

References

1. A.S. Akhmatov, Molecular Physics of Boundary Friction (Physical-Mathematical Literature,


Moscow, 1963), p. 472
2. D.N. Lyubimov, V.A. Ryzhikov, Phys. Chem. Processes in Friction: Manual. Novochherkask:
South Russ, State Eng. Univ. 147 (2006)
3. R. Kana (ed.), Physical Science of Metals, vol. 3 (Mir, Moscow, 1968), p. 483
4. Y.D. Eshelby, Appl. Phys. 25, 255 (1954)
5. Ch. Kittel, Introduction into Solid Body Physics (Mir, Moscow, 1978), p. 791
66 2 Crystalline Materials: Surfaces of Solid Bodies

6. G. Gottstein, Physical and Chemical Principles of Science of Materials (Binom, Moscow,


2009), p. 375
7. L.S. Pinchouk, V.A. Strouk, N.K. Myshkin, A.I. Sviridyonok, Science of Materials and Struc-
tural Materials (Vysshaya Shkola, Minsk, 1989), p. 459
8. L.D. Landau, E.M. Livshits, Continuum Physics, vol. 8 (Fizmatlit, Moscow, 2005), p. 656
9. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 1 (Sov. Encyclopedia, Moscow, 1988),
p. 638
10. G.S. Pisarenko (ed.), Strength of Materials, (Vyshcha Shkola, Kiev, 1986), p. 775
11. S. Morrison, Chemical Physics of Solid Body Surface (Mir, Moscow, 1980), p. 367
12. Kh. X. Chikhos, System Analysis in Tribonics (Mir, Moscow, 1982), p. 348
13. V.F. Kiselev, S.N. Kozlov, A.V. Zoteyev, Fundamentals of Solid Body Surface Physics (INBU,
Moscow, 1999), p. 284
14. D. Bakley, Surface Phenomena During Adhesion and Friction Interaction (Mashinostroyenie,
Moscow, 1986), p. 360
15. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 2 (Sov. Encyclopedia, Mosscow, 1990),
p. 703c
16. E. Zenguil, Surface Physics (Mir, Moscow, 1990), p. 536
17. V. Jayavaya, R. Vanselova (ed.), New in Solid Body Surface Research (Mir, Moscow, 1977),
p. 314
18. D.N. Lyubimov, K.N. Dolgopolov, Difusive Processes in Friction (IP Bouryhin V.M., Shakhty
2010), p. 148
19. A.V. Chichinadze (ed.), Fundamentals of Tribology (Science and Technology, Moscow, 1995),
p. 778
Chapter 3
Dynamic Microprocesses in Solid Bodies

The method we will use in this treatise is the statement


only such objects follow from the mentioned earlier.
Galileo Galilee
There is no rest anywhere in the emptiness
inexplicable. On the contrary, the objects often, driven
by various motions, fly away far when they collide.
Roman poet and philosopher Titus Lucretius Carus
Though all this customary smell of motor oil, humans,
dogs, leather and fur, a strange alien aroma get through
weakly.
American writer D. Campbell

Abstract The main regularities of elastic vibration in crystals are discussed. There
are given elements of quantum model of elastic vibrations propagation in crystals. The
basic laws of dislocations motion and diffusive flows in solids as well as regularities
of influence this processes on solids microstructure and mechanical properties are
represented.

So, the world of crystals, like other types of solid bodies, is not stiff, frozen. Its
particular arrangement changes under the effect of external force factors. Regretfully,
these changes are least reflected in publications and are rarely considered when
constructing structural and theoretical models for solutions of engineering problem.

3.1 Elastic Waves in Crystals

The external effect on the solid body generates a response in the form of perturbations
of the crystalline space. The simplest type of these perturbations is the propagation
of elastic oscillations in the substances. Because the crystalline space is anisotropic
by nature, the oscillations appearing in the crystal depend on its structure parameters.

D. Lyubimov et al., Micromechanisms of Friction and Wear, 67


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8_3,
Springer-Verlag Berlin Heidelberg 2013
68 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.1 Action of external forces on elementary cell of cubic lattice

Let us consider how the elastic oscillations propagate in the simplest lattice of the
cubic type [1].
To displace the elementary crystal volume (Fig. 3.1) in the direction of axis x, the
equation of the second Newton law is true:

2u x Xx Xy Xz
= + + , (3.1)
t 2 x y z

where crystal density; u x the lattice dislocation in the direction of axis x; Xx ,


Xy , Xz components of the tensor of mechanical stresses.
According to the work [1]: 
Xi = cij eij , (3.2)
j

where eij singular components of the deformation tensor.


The value eij can be represented, like mechanical stress, in the form of the
following linear combinations:

exx = s11 X x + s12 Y y + s13 Z z + s14 Yz + s15 Z x + s16 X y , (3.3)

where si j = ci1
j constant elastic flexibility constants; cij elastic rigidity
constants.
Equation (3.1) is the following for the cubic crystal with the account of (3.23.3):
   
2u ex x e yy ezz ex y ezx
2 = c11 + c12 + + c44 + . (3.4)
t x x x y z
3.1 Elastic Waves in Crystals 69

Let us use the following relations [1]:

u 
ex x = x x = ; e yy = yy = ; ezz = zz = ,
x y z
u
ex y = yx + x y = + ,
y x
(3.5)

e yz = zy + yz = + ,
z y
u 
ezx = zx + x z = + ,
z x

where u, ,  displacement vector R components:

 ) = u(r )i + (r ) j +  (r )k,


R(r  (3.6)

where i, j, ksingle


 orts.
Let us use expression (3.5) and rewrite formula (3.4) as:
   
2u 2u 2u 2u 2 2
2 = c11 2 + c44 + + (c12 + c44 ) + . (3.7a)
t x y2 z 2 x y xz

The relevant equation for displacements and  are obtained from Eq. (3.7a) by
circular permutation of values, for instance:
   
2 2 2 2 2u 2
2 = c11 2 + c44 + (c12 + c44 )
+ + ,
t x y2 z 2 x y xz
 2   2 
2 2  2 u 2
2 = c11 2 + c44 + + (c12 + c44 ) + .
t x y2 z 2 x y xz
(3.7b)

One of the solutions of Eq. 3.7 has the longitudinal wave:

u = u 0 exp[i ( x t)], (3.8)

where iimaginary unity; wave vector; cyclic frequency.


The following is known:
= 2frg ,
2 (3.9)
= .

where frq frequency; wavelength.
Substitution of (3.8) and (3.9) into (3.7a) yields the following:

2 = c11 2 . (3.10)
70 3 Dynamic Microprocesses in Solid Bodies

Hence, the speed of elastic wave propagation through the crystal in the direction
(100) is equal to 
c11
cel = frg =
1
. (3.11)

The transverse wave or the displacement wave is characterized by the wave vector
directed along the cube ridge coinciding with the axis x. It displaces the particle in
the direction of axis y:
2
cel = c0 exp [i( y t)] . (3.12)

Similarly for the transverse wave in the direction (100) of the crystal:

c44
3
cel = . (3.13)

Similar expressions can be obtained too for motion of particles along axis z.
Let us consider the propagation of elastic oscillations in planes xy along the cubic
lattice diagonal, i.e., in the direction (110). The dislocation of particles  along the
axis z is   
 = 0 exp i x x + y y t . (3.14)

Substitution of (3.14) into dispersion relation (3.7b) yields the following:




2 = c44 x2 + y2 = c44 2 . (3.15)

It is apparent from Eq. (3.15) that the result is independent of the wave direction
in the plane xy.
Let us consider other waves propagating in the plane xy.
  
u = u 0 exp i x x + y y t ,
  (3.16)
= 0 exp i x x + y y t .

Application of Eqs. (3.7a) and (3.7b) yields the following system:

2 u = (c11 x2 + c44 y2 )u + (c12 + c44 )x y ,


(3.17)
2 = (c11 y2 + c44 x2 ) + (c12 + c44 )x y u.

For the wave direction (110) x = y = / 2. The condition of existence of
the non-trivial system solution (3.17) is that the determinant composed from the
coefficients at u and , is equal to zero:
 
 1 1 
 2 + (c11 + c44 ) 2 (c + c ) 2 
 12 44 
1 2 2  = 0. (3.18)
 (c + c ) 2 + (c11 + c44 ) 
2 1 2
 12 44
2 2
3.1 Elastic Waves in Crystals 71

The quadratic equation in respect to 2 has the following roots:

1 1
1)2 = (c11 + c12 + 2c44 ) 2 , 2)2 = (c11 c12 ) 2 . (3.19)
2 2
The first root describes the longitudinal wave motion, the second the transverse
wave motion. To determine the direction of displacement of particles, let us substitute
the first root into (3.17):

1 1 1
(c11 + c12 + 2c44 ) 2 u = (c11 + c44 ) 2 u + (c12 + 2c44 ) 2 , (3.20)
2 2 2
and obtain that u = . Hence, the particles dislocate along directions (110) parallel
to the vector  . Let us substitute the second root into Eq. (3.17) and obtain u = ,
i.e., the direction of particles coincides with direction (110) perpendicular to  .
Hence, the wave vector  of definite value and directions in the cubic lattice
has three types of own oscillations; their directions are neither strictly parallel nor
perpendicular to  . Thus, in accordance with relations (3.11) and (3.13), the speed
of propagation of elastic oscillations in the solid body is the following:
 1/2
elastic characteristics
cel = . (3.21)
medium density

The elasticity characteristics are determined by the prevailing deformation of the


crystalline structure in the solid body, i.e., the numerator in formula (3.21) may
contain all the Young modulus E, the shear modulus G s , the rigidity coefficient
ci j , and the Mach number M (E. MachAustrian physicist).
The Mach number M [2] serves one of the similarity criteria in the continuum
mechanics and is an essential elasticity characteristic. It is the ratio between the
medium motion speed at a given point the speed of propagation of elastic oscilla-
tions (sound) c in the moving medium:


M= . (3.22)
c
The Mach number characterizes the medium compressibility, i.e., its density vari-
ations / under the effect of comprehensive compression. With the account of
the laws of thermodynamics and the equation Bernoulli equation for perfect fluid
flow [2]:

M2 . (3.23)

It follows from relation (3.23) that the medium compressibility augments as the
Mach number grows.
72 3 Dynamic Microprocesses in Solid Bodies

3.2 Quantum Crystalline Lattice Oscillations

The energy of oscillations of the crystalline lattice or the energy of elastic waves
is a quantum magnitude. The quantum energy of the elastic wave is called phonon
similarly to the photon as the quantum of electromagnetic wave. The sound (elastic)
waves in crystals are the propagation of phonons. The thermal oscillations of atoms
in the crystals can be considered the thermal excitation of phonons similarly to the
thermal excitation of photons which the emission of the absolutely black body.
According to the assumption of M. Plank which is in the base of the matter
quantum theory, the energy of each type of oscillations of the electromagnetic field
is the proportional value h, or the energy of one photon. The energy n of photons
oscillating with the frequency f is equal to:

w = nh f ,
or (3.24)
w = n f n ,

where f n the cyclic frequency of phonons.


Numerous diffraction experiments prove convincingly that the electromagnetic
field possesses the wave properties too; on the other hand, the Plank law of distri-
bution of energy and emission based on the frequency states that the energy of the
electromagnetic field is quantized. The elastic waves manifest a similar situation.
The first experimental proof of the energy and elastic waves quantization was the
following observation: the contribution of the energy of the crystalline lattice into the
heat capacity of solid bodies tends always to zero as the temperature approaches zero.
The X-rays and neutron experience the non-elastic dissipation when they interact
with crystals. As a result, the energy and the impulse of emission correspondingly to
the appearance or absorption of one or more phonon. The precise measurement of
the effects due to these processes permits to determine the properties of individual
phonons, in particular, to establish how the frequency depends on the wave vector,
in other words, the law of dispersion.
According to the quantum laws, the phonon possessing the wave vector  and the
impulse  (quasi-impulse) interacts with other particles or fields. There are rules
of selecting the value of the wave vector  for crystals [1]. For instance, for elastic
dissipation X-ray photons in the crystal the following rule is true:

 ds = in + G
 r.l , (3.25)

where G  r.l the reverse lattice vector; in and  ds wave vectors of the incident
and dissipated X-ray phonon.
During dissipation the crystal in general experiences the recoil with the impulse
G r.l . The summary wave vector of interacting waves persists in the periodic lattice,
but only with possible addition of the vector G  r.l . The true impulse of the whole
3.2 Quantum Crystalline Lattice Oscillations 73

system is strictly constant. If the photon undergoes nonelastic dissipation when the
phonon with the wave vector  , then the rule of its selection is the following:

 ds +  = in + G
 r.l . (3.26)

If, during the process in question, the phonon is absorbed, then:

 ds = in +  + G
 r.l , (3.27)

This expression is similar to relation (3.25).


Let us consider the process of dissipation of photons using the acoustic phonons
which is essential too describe the changes in the structure of materials in friction.
Assume that the photon propagating in the crystal has the frequency f . If the crystal
is treated as the continuum with the refraction index nr f , then this photon has the
wave vector determined by the following relations:

c  pht
pht = ,
nr f
c (3.28)
pht = ,
nr f
p pht = 
pht .

where  pht the photon wave vector pht the wavelength; c the speed of light;
p pht the photon impulse.
The photon interacting with the beam of phonons in the crystal can dissipate due
to the fact that the field of elastic deformations alters the local concentration of atoms
[in accordance with (3.25)], and, hence, the crystal refraction index. Thus, the elastic
wave modulates the medium optical properties. Vice versa, the electromagnetic field
of the flow of photons alters the crystal elasticity.
If the photon undergoes dissipation in the crystal , it can result [in accordance with
relations (3.26) and (3.27)] in the appearance or absorption of the phonon. Assume
that, during the photon dissipation, a phonon appears with the wave vector phn and
the frequency phn . This process is shown schematically in Fig. 3.2.
In accordance with the energy preservation:

 pht = pht +  phn , (3.29)

pht =  pht + phn . (3.30)

If the velocity c of propagation of elastic waves is constant, then:

phn = c phn . (3.31)


74 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.2 Scheme of photon


nonelastic dissipation

Due to different speeds of sound and light, only a small portion of the energy of
the incident photon can be transferred to the phonon . For the phonon with the wave
vector  phn comparable with the wave vector of the photon  pht it can be shown that
pht = pnt ,  pht =  pht . Then, it follows from (3.26) that

phn = 2 pht sin (/2) , (3.32)

where the angle of dissipation of photons.


The frequency phonon oscillations:

2c pht nr f
phn = sin (/2) . (3.33)
c

3.3 Diffusion in Crystals

Diffusion is the most important natural dynamic process influencing the integrity
of physicist and mechanical characteristics of the solid body [1, 3, 14, 15]. The
diffusion in the material results in the leveling of the chemical potential . The diffusion
levels the concentration of each phase component by the volume of the single-phase
material at constant temperature and when there no external diffusion forces. In
case the temperature is not constant and external forces are in effect, the spatial
inhomogeneous equilibrium distribution of each component of forces sets in the
material. In the general case, the mass transfer by diffusion is described by the
differential equation in particular derivatives of the second order which is most often
the following:

dCm (x, t)
= div(D gradCm ) qm Cm + F, (3.34)
dt
3.3 Diffusion in Crystals 75

where Cm (x,t)the function describing the substance concentration at a definite


medium point at the moment of time t; Dthe diffusion coefficient; qm the ab-
sorption coefficient; Fthe characteristics of intensity of the diffusing substance.
The basis of the phenomenological description of diffusion processes are two Fick
laws (A. FickGerman physicist and chemist). According to the first of them, the
amount of the substance m passing through the cross section A perpendicular to
the direction of diffusion is proportional to the gradient of the substance concentration
C, the cross section area A and the process time t:

m = D gradC A t = D C A t,

or, in one-dimensional approximation:

C
m = D A t. (3.35)
x
The sign minus indicates that the diffusion flow moves towards the concentration
reduction, that is why the isolated system tends to disappearance of the concentration
gradients. Usually, Eq. (3.35) is used with the specific substance quantity referred to
the cross section area unity per unit of times. This physical value has acquired the
name of the density of flow of the diffusing substance J . Then the Fick equation
becomes the following;
J = D C, (3.36)

or
C
J = D .
x
The flow density J determines the substance diffusion rate. If the system has the
temperature gradient T and the pressure gradient P, Eq. (3.36) is the following:
 
T P
J = D C + DT + DP , (3.37)
T P

where DT the thermodiffusion coefficient ; D p the barodiffusion coefficient .


The Fick second law describes the substance concentration changes after diffusion
at a definite space point. When D = const, the Fick equation of the second order is
the following:
C
= D 2 C,
t
or, in one-dimensional approximation:

C 2C
= D 2. (3.38)
t x
76 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.3 Changes in concentration of two semifinite rods of diffusing substance at two different
diffusion rates

In the general case, the coefficient of diffusion D is not constant and depends
on the concentration, the spatial and time coordinates. Hence, relation (3.38) can be
re-written in the following way:
C
= [D(C)C] ,
t   (3.39)
C C
= D(C) .
t x x

The general view of the solutions of Eq. (3.39) for the medium which can be
represented as two detached semi-infinite rods (Fig. 3.3) is the following:
x

2  Dt   
C2 C1 2 C2 C1 x
C (x, t) C1 = e d = 1 + erf .
2 2 2 Dt

(3.40)
The solutions of the Fick equation of the second order are treated in [5].
The value L D = 2 Dt included into Eq. (3.40) is termed the diffusion length
in the theory of diffusion.
Different references
describe the value L differently, for
instance, L D = 2Dt in [5], L D = 6Dt in [4]. Generally, the diffusion length
or the diffusion path is proportional to Dt and it is the distance to which particles
can travel during diffusion . For further discussion, let us assume

L D Dt (3.41)

Mechanisms of diffusion. The diffusion is caused by the chaotic thermal motion


of atoms. The diffusion is not the process induced by some definite force. If mass
transfer evolves through diffusion in the medium, local deformation fields, the chem-
ical potential change or electrical forces appear, their effects is imposed on the diffu-
sion flow like it is described in (3.37). Assume that these factors produce the effect
proportional to the gradient of some generalized potential to which the force
relates that initiates the diffusion flow:

P = . (3.42)
3.3 Diffusion in Crystals 77

The diffusion flow J results from displacement of particles by the force P. When
the displacement rate is , ( C) particles passes per unit of time through unit of
area of the transverse cross section;

J f = C. (3.43)

The rate of displacement of particles is proportional to the force P:

= B  P, (3.44)

where B  is the mobility of particles which, in accordance with the Einstein equation
[1, 4] is equal to
D
B = . (3.45)
kT
Substitution of (3.42), (3.44) into (3.45) in formula (3.43) yields:

D
J = C. (3.46)
kT
With the account of (3.46) and (3.37) the equation of the first Fick law becomes
the following:  
C
Jgen = J + J = D C + . (3.47)
kT

and correspondingly that of the second Fick law:


 
C C
= D C + ,
t kT

or, in the one-dimensional form:


 
C 2C D
=D 2 + C . (3.48)
t x kT x x

The Fick laws establish the bonds between the observable diffusion flow and
the concentration gradient and show that nature tends to even distribution of con-
centrations through the whole volume of the substances. Yet, the thermodynamic
equilibrium in the substance at constant temperature and pressure is determined by
the minimum free energy and Gibbs G [6], moreover, this requirement to the
homogeneous distribution of concentrations is not obligatory.
For the system of n-components:


n
G = Wint + pV T S + i Ni , (3.49)
i=1
78 3 Dynamic Microprocesses in Solid Bodies

where i the chemical potential; Ni the number of particles of the i-th compo-
nent; ppressure; T temperature; V volume; Sentropy; Wint the internal
energy in the diffusion system.
For the mixture of components, the diffusion flow Ji of each in the thermodynamic
interpretation of open systems [7] is recorded in the following way:
 T
Ji = L ik , (3.50)
T
k

where T the gradient of the chemical potential at the temperature T .


Relation (3.36) can be re-written similarly for the first Fick law:

Ji = Dik C , (3.51)
k

By equaling the first terms of relations (3.50) and (3.51):


 T 
L ik = Dik C . (3.52)
T
k k

The values L ik and Dik are symmetric tensors of the second rank; the non-diagonal
components L ik are equal to zero in Eq. (3.52).
For the two component system, Eq. (3.52) can be represented as follows:

1 C1
L 11 = D1 ,
x x (3.53)
2 C2
L 22 = D2 .
x x

Relation (3.43)(3.45) yields L 11 = B1 C1 , L 22 = B2 C2 , or in the general form

L ii = Bi Ci , (3.54)

or
i
Di = Bi , (3.55)
(ln Ci )

Relation (3.55) reflects the dependence of the diffusion coefficient on the chemical
potential gradient.
It follows from Eqs. (3.513.52) that the diffusion coefficient is the tensor deter-
mined by the crystalline structure parameters. If the crystalline lattice is symmetric
cubic, the diffusion coefficient is isotropic, i.e., Di tensor can be replaced with scalar
value, it is true for 70 % of metals. For crystals with a low degree of symmetry,
and the diffusion coefficients depend usually on the direction and it can be fixed
mathematically by substituting tensors Di j in the matrix form [4]:
3.3 Diffusion in Crystals 79

D1 0 0 D11 0 0 D11 0 0
Dcube = 0 D1 0 ; Dhex = 0 D11 0 ; Drhomb = 0 D22 0
0 0 D1 0 0 D33 0 0 D33

One more interpretation of the tensor Di j components is shown in [5]. The diago-
nal tensor elements determine the direct processes and non-diagonal the intersecting
processes . Thermodiffusion and the Dufour effect are the intersecting processes in
the theory of diffusion (L. DufourSwiss physicist); according to this effect the
heat flow stirred by diffusion of non-reacting gases appears due to appearance of the
gradient of substance concentration. The thermal flow J Q , appears during constant
pressure because the gradient of concentration and temperature is equal to:

J Q = h f T T D  C, (3.56)

where h f the heat conductivity coefficient; the diffusant density; the dif-
fusant chemical potential; D  the Dufour coefficient.
In addition to the thermal flow, this system acquires the substance flow:

J M = C(1 C) DT T D C, (3.57)

where DT and D are the coefficients of thermodiffusion and diffusion, respectively.


In accordance with the laws of thermodynamics of the non-equilibrium systems,
(the Onzager theorem) [5, 7] it is DT = D  . In this case the following relation is
always fulfilled:
h f D
DT2 . (3.58)
T C (1 C)2
2

Kinetics of diffusion. The motion of particles in the diffusion stream can be


considered as the heat activated leapwise travel of atoms with the frequency :
 
W
fr = D exp , (3.59)
kT

where D the Debye frequency 1013 Hz (the atom frequency of oscillations);


W the activation energy; kthe Boltzmann constant.
This relation can interpreted by the number of times the atoms overcome the
energy barrier per second. The atoms can displace through the crystal following the
principle of expediency in of the following way: exchange, circular, internodal and
vacational [3, 4] (Fig. 3.4).
There is a fundamental interrelation between the frequency of jumps and the
diffusion coefficient D irrespective of the mechanism of diffusion in the crystal or
of its structure:
2 2
D = d fr = d , (3.60)
6 6td
80 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.4 Mechanisms of diffusion in solid bodies: a exchange, b circular, c internodal, d vacational

Here d designates the diffusing atom jump length, td = 1/ f r is the time interval
between consecutive jumps.
The Arrenius law is the main equation of heat-activated diffusion:
 
WD
D = D0 exp , (3.61)
kT

where Wd the diffusion activation energy.


Its value is determined by the energy of formation of vacancies Uv and by the
kinetic energy of the diffusing atom displacement E a :

Wd = Uv + E a . (3.62)

The Arrenius law serves here to describe predominantly the temperature depen-
dence of diffusion processes in condensed bodies. According to the ideas developed
by M. Eyring (the American physicist and chemist) [5, 8], the values in formula
(3.61) can be represented in the following way:

kT S
D0 = d 2 e k , (3.63)
h
where Sthe change of entropy of the diffusing substance; hthe Plank constant.
The physical sense of the value d is much identical to d , but it is also bound
by the substance structure. For crystalline bodies, it is the lattice constant, for the
amorphous and fluid bodies they are the mean spacing between molecules and the
value of the elementary jump, respectively.
3.3 Diffusion in Crystals 81

With the account of (3.60) and (3.61), let us represent the Arrenius equation in
the following way:  
Wd
D = 0 d exp
2
. (3.64)
kT

The effect of the crystal structure. For different types of crystalline lattices,
the values Wd differ slightly, though d and 0 can vary within a broad range. Rep-
resentation of the diffusion coefficient in solid bodies with the help of the rates of
the Eyring reactions yields its exponential dependence on the activation temperature
and energy. Diffusion in the unordered substitution alloys cannot be described with a
single exponent because various potential barriers are to be overcome in lattice nodes
with differently configured atoms. If the diffusion evolves through vacations or with
the help of the circular mechanism (Fig. 3.4b), the diffusion coefficients of different
mechanisms differ too. Due to this fact the substance results directed towards the
larger partial diffusion coefficient:

JR
= (D1 D2 ) C, (3.65)

where D1 > D2 are partial diffusion coefficients.


Relation (3.65) describes the regularities of diffusion follows in the crystal corre-
sponding to the Kirkendall effect .
It is rather hard to analyze how the dislocations influence the diffusion coefficient
in the polycrystal due to a large number of dislocations and uncertainties n the spatial
distribution. At present it is believed that the boundary dislocations accelerate much
the motion of atoms. Hence, the diffusion over the boundaries of crystallites having a
much more defective structure than the crystallites proper should be identified when
considering the diffusion processes in polycrystals.
The diffusion flow in the elementary volume within intergrain boundaries is a
combination of diffusion flows over the intergrain boundaries and the volumetric
diffusion. For the intergrain boundary (INB) in the direction (Fig. 3.5) of changes
in the flow density in the elementary volume, the INB (in the linear approximation)
corresponds to the equation;

dJ j
Jy+dy = Jy + ,
dy

and the equation of continuity:


    
C 1 dJx
= Jy Jx + 2 dy Jx , (3.66)
t dy dy

where the thickness of intergrain boundaries; the thickness of the diffusion


layer.
The flows along the boundaries and in the polycrystal volume are in accordance
with formula (3.36):
82 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.5 Diffusion along intergrain boundary

C C
Jy = Dbr , Jx = DV , (3.67)
y y

where Dbr and DV are boundary and volumetric diffusion coefficients.


In accordance with the second Fick law [4]:

C 2C 2DV C
= Dbr 2 + . (3.68)
t y x

It follows from expression (3.68) that the diffusion coefficients do not depend
on the localization of boundaries. With some assumptions, the following can be
obtained [4]:

  
y 2 x
C (x, y, t) = C0  1 erf . (3.69)
4
DV t Dbr DV1 2 DV t

Relation (3.69) is helpful to represent the diffusant concentration profile along


the intergrain boundary (Fig. 3.6). The deepest penetration of the substance into
polycrystal corresponds to the boundary and reduces in response to the distance
from it.
It is essential to realize that the diffusion coefficient over the boundaries between
grains is not any characteristic of the material; it depends on the boundary properties.
3.4 Motion of Dislocations 83

Fig. 3.6 Concentration pro-


file of diffusing substance
near boundaries of grains of
crystallites (C1 > C2 > C3 )

3.4 Motion of Dislocations

The motion of dislocations is the most essential dynamic process affecting the crys-
talline structure in space [4, 9]. The dislocations moves over the surfaces called the
sliding surfaces and lead to the crystal plastic deformation. The sliding surface cor-
responds to the axis of dislocation and its Burgers vector. In the particular case, the
dislocation axis lies in one plane which contains the Burgers vector.
The reason why the term the surface and the sliding plane are selected becomes
understandable when considering how the dislocations behave in the real crystal.
The dislocations in the crystal can slip along their sliding surfaces; their motion
resembles to some extent the distribution of elastic perturbations in a long spiral
spring or water ripples. As long as the dislocation travels in one direction, the upper
half of the material displaces in respect to the lone one by the value and in the Burgers
vector direction. In other words, the sliding of dislocations indicates the slipping
of one crystal part in respect to the other (Fig. 3.7a).
Crystal deformation. The vector n is directed normal to the plane of sliding of
dislocations : 
n = Sb , (3.70)


where Sdislocation 
length element; bBurgers vector.
 b)
The helical dislocations ( S//  have no definite sliding and it can be variable mak-
ing the sliding possible in the transverse direction (Fig. 3.7a). The plane of sliding
of prismatic dislocations contain the line of dislocations , correspondingly the pris-
matic dislocations are stationary. The boundary and mixed dislocations can issue
from original sliding planes (ascend) due to absorption or creation of point defects
(Fig. 3.7b). The absorbed vacancies vanish from the crystal volume; hence, the dislo-
cations serve as the sink for vacancies. The dislocations in the reverse process act
as sources of vacancies. Creation and removal of point defects changes the crystal
volume.
The sliding motion of dislocations is considered conservative because it does
not produce nor absorbs point defects or alters the crystal volume. This motion of
dislocations is called climbing. The material transfer in the process of climbing is
84 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.7 Variants of motion of dislocations: a scheme of unidirectional sliding crystal parts due to
motion of helical dislocations; b helical dislocation slides in two directions (111) and (111)

effected by the boundary dislocations . The boundary dislocations can be considered


as the dislocations resulting from appearance in crystals of additional incomplete
planes of the mother material. It requires that the boundary dislocation climbs here
and there leaving the incomplete plane unchanged in size. It is also true for the
boundary of the components of mixed dislocations .
The motion of dislocations, their appearance in the crystal induces its plastic
deformation, its numerical characteristics is the deformation displacement . Its
value corresponds to expression [9]:

= d blmd , (3.71)

where d the density of dislocations; lmd the mean distance covered by


dislocations.
The density of dislocations can be calculated with the following formula:
 2
2
d = , (3.72)
Gs b

where the mechanical stress; G s the shear modulus.


For the moving dislocation, the volumetric forces in Eq. (2.11) are represented by
finite values. They are the following for the helical dislocation parallel to the axis z:
 
2uz
z = d . (3.73)
t
3.4 Motion of Dislocations 85

The combination of the equations of equilibrium and compatibility (see Sect. 2.1)
for the dislocations moving with the velocity in the direction of axis x does not lead
to relation (2.16) like in the stationary case; it rather leads to the following equation:

2uz 2uz 2uz


Gs + G s d = 0,
x2 y2 t 2
or
2uz 2uz
Gs = 0, (3.74)
( x  )2 y2


1
where x  = (x t) 1 , = 1 c2 , cthe velocity of propagation of
2 2

elastic waves in the dislocation medium (Sect. 3.1, relation 3.21).


Equation (3.74) I solved similarly to (2.13) and it is the following:

b y
u x = u y = 0; u z = ar ctg , (3.75)
2 x
and
 G s by
zx = x z =

2 (x t)2 + 2 y 2
. (3.76)
  G s b (x t)

yz = zy =
2 (x t)2 + 2 y 2

When solving Eq. (3.4), a reduction is performed similarly to the Lorentzian


reduction in special theory of relativity limiting the velocity of dislocations to the
value c.
Deforming stresses. Integration of the density of the kinetic energy dislocations
by occupied volume yields the full energy of dislocations:

U
W = , (3.77)

where U the energy of the stationary dislocation described by formulas


(2.25)(2.27).
At the velocity of dislocations c, the parameter tends to zero, hence,
W  . Let us use the ideas of the theory of relativity [9] and determine the
following:
W  = m D c2 , (3.78)

and, respectively
m0
m D = ,

86 3 Dynamic Microprocesses in Solid Bodies

U
where m 0 = 2 .
c
The value m 0 is the equivalent mass at rest of a single dislocation length which,
by the order of magnitude, is equal to:

m 0 = b2 , (3.79)

where b is the Burgers vector modulus.


In other words, in the crystal it is approximately equal to the mass of a single
continuous row of atoms with the length equal to the length of dislocations.
Note that relations (3.74)(3.79) derived for the boundary of dislocations , possess
the principle of generality and they are applicable to other types of dislocations.
At the stress of deformation by displacement needed to displace two crystalline
blocks (Fig. 3.7a) along the crystallographic plane for the interatomic distance b (the
Burgers vector), it is calculated with the formula:

Gs b
max = , (3.80)
2 d
where dthe crystalline lattice period.
For displacement of two crystallographic planes to the distance x equal to 1/4
of the interatomic distance, the stress in the crystal should be increased to the
maximum values max . When x = b, the stress of displacement falls to zero since
this position corresponds to the crystal new equilibrium state. This dependence can
be described by the following harmonic law.

2 x
= max sin . (3.81)
b
The value max corresponds to the theoretical crystal strength in shear. How-
ever, the limit shear stress registered by testing metals and alloys is many orders of
magnitude less than max . The contradiction is due to the fact that the atoms in the
materials under load move stagewise rather than simultaneously. This type of motion
is comparable with the caterpillar motion. This mechanism in crystals corresponds
to the motion of dislocations . When the dislocation with the Burgers vector b moves
through the crystal with the thickness l, it induces the dislocation of crystal parts (the
shear deformation) by the value
b
= . (3.82)
l
The summarization of the drift of many dislocations can be recorded as (3.71).
Expression (3.71) is easily transformed into the type describing the deformation at
various moments of time:
= d b, (3.83)

where the velocity of motion of dislocations.


3.4 Motion of Dislocations 87

Let us consider the dislocation sliding in a single crystal volume under the effect of
the external force and free of internal stresses. According to (3.71), the deformation
increment by displacement d caused by the sliding of the dislocation element
with the length dl to the distance d Sd , perpendicular to the direction of motion of
dislocations, is defined by the following expression:

d = b dl d Sd . (3.84)

Simultaneously with the appearance of the displacement d , the work is fulfilled


in respect to the field of external forces:

d A1 = d . (3.85)

As a result, the force P affects the dislocation induced by the field of elastic
stresses in the crystal which sets the dislocation in motion. The work fulfilled by this
force per unit of the dislocation length is the following:

d A2 = P d Sd dl. (3.86)

The values dA1 and dA2 are equal, then, with the account of (3.86), the following
can be recorded:
b d Sd dl = P d Sd dl,

or
P = b. (3.87)

If the mechanical stress acts in the crystal, the force per unit of the dislocation
with the Burgers vector b and the linear element S (2.8) is determined by the following
formula [4]:

P =  b S. (3.88)

In order to displace the dislocation (Fig. 3.8) the force should be applied creat-
ing the critical displacement stress to its sliding plane (3.87). The critical stress or
Peierls stress p (R. PeierlsBritish physicist) can be calculated with the following
formula [4]:  
2G s 2 d
p = exp , (3.89)
1 b (1 )

where the Poisson coefficient.


Interaction between dislocations. The dislocations interact through the field
of elastic deformation. The force of interaction is described by Eq. (3.88). Let us
consider the example of interaction between two parallel boundary dislocations. The
force P12 , with which the first dislocation with its field of stresses 1 presses the
second dislocation with the Burgers vector b2 and the element of length S2 , is equal to:
88 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.8 Phases of motion of dislocation over its plane: a < p , b motion start, c crystal shear



P12 = 1 b2 S2 , (3.90)

The crystalline space of the solid body decelerates the motion of dislocations.
This effect acquired the name of locking dislocations. There are several variants of
locking (or friction by another terminology) [9].
The chemical locking occurs under the effect of defects in the crystalline lattice
packing concentrating in the layer several atoms thick. The solubility of chemical
elements in this phase differ from their solubility in the main crystalline lattice, so
the alloying element atoms can accumulate on the packing defects and multiply them.
As a result, the boundary dislocations are locked and so do the helical dislocations to
a lesser extent. For the dislocations to resume motion, additional mechanical stress
should be applied with the depended on the concentration of the dissolved element.
The chemical locking of the motion of dislocations is observed predominantly among
the crystals with the face centered lattice.
The elastic locking is typical for the crystals with the substitution atoms differ-
ing in size from the matrix atoms. The local fields of elastic deformation appear and
interact with the elastic fields of dislocations striving to reduce the lattice distortions.
The alien atoms occupy the most gainful energy positions near the dislocations. The
energy from interactions amounts to fractions of electron-volts, while the interac-
tion between helical dislocations and substitution atoms is still weaker. An extra
stress is needed to resume motion of the dislocations locked elastically; the stress is
proportional to the parameter of dimensional incompatibility of atoms :

1 dd
= , (3.91)
d dC
where dthe crystalline lattice constant; Cthe concentration of impurities.
The mechanism of elastic locking of motion of dislocations is actuated when the
Cottrell atmosphere forms around dislocations (Sect. 2.1) and acts as the peculiar
anchor. The anchor moves together with dislocations. and has either to tow the
atmosphere or get detached.
The electrostatic locking occurs in the dislocations with the stretched negatively
charged nucleus which interacts with the additional charge appearing adjacent to
3.4 Motion of Dislocations 89

the dissolved atom; the atom has the valence different from that of the solvent.
The free electrons are screened off by a highly effective braking mechanism, but in
this case its effectiveness is poor because the atomic levels can change noticeably
in case of such defects. Another mechanism of electrostatic locking is a peculiar
manifestation of the Schottky barrier observed when the Cottrell atmosphere forms
around the positively charged nucleus of vacancy dislocations. The potential barrier
on the surface of dislocations for the main charge carriers; the barrier is several
Debaye screening lengths L D . for instance, for aluminum L D is equal to 0.05 nm [10].
The locking mechanism orders the field of stresses of dislocations [11]. It is
exactly the case when the lack of precipitated atoms due to the dislocation stresses can
result in the structures of the near order around dislocations. The dislocation creates
spontaneously around itself the energy barrier (self locks), as a result, yield spots
can appear on the crystal deformation curves, while the sold substitution solutions
demonstrate the hardening effect proportional to the squared concentration of the
dissolved element.
Creep. The measured values of the critical displacement stresses in crystals are
usually smaller than those calculated according to (3.80). It is primarily to the neglect
of the factor of thermal deformation activation. The mechanical stress, at which dis-
locations can overcome obstacles, relieves due to the thermal activation. Dislocations
travel among obstacles and stop in front of them for some induction period for the
system to accumulate the thermal activation energy needed to overcome the obstacle.
With these considerations in mind, Eq. (3.80) can be represented as:
 
W  ( )
= 0 exp , (3.92)
kT

where W  the activation energy needed to overcome obstacles under the effect
of stresses . "  
4b 2 WL p
W ( ) = ,
2 (3.93)
1
WL = G s b .2
2
where p the Peierls stress calculated with formula (3.89).
Combination of Eqs. (3.92) and (3.93) yields the expression friction the critical
stress at the temperature T and the deformation rate :
 
2
= p AT ln
2
, (3.94)
0

where A4 constant.
The critical displacement stress at T = 0K corresponds to the Peierls stress and
it reduces as the temperature grows and becomes proportional to T 2 .
90 3 Dynamic Microprocesses in Solid Bodies

The speed of motion of the vacancies among the dislocations, hence, irrespective
of the mechanism of motion of dislocations, the velocity of sliding dislocations
depends on their distribution in the crystal is the following:

b
= , (3.95)
BT

where BT the temperature dependent constant. The exact value BT is determined


by the dominating mechanism of motion of dislocations.
The velocity relates to the speed of climbing of dislocations n in the direction
normal to the sliding plane by the following relation:

LD
= n , (3.96)
d

where L D the distance to which the dislocation moves in the sliding plane; d  the
distance to which the dislocation climbs in the direction perpendicular to this plane.
The motion of dislocations determines the property of solid bodies such as creep
the numerical characteristic of which K creep is approximately equal to (3.80).
Substitution of (3.80) with the value from (3.72) yields the following equation of
the mean creep rate:
4 2 2
creep = . (3.97)
BT G s

3.5 Dependence of the Crystalline Lattice Parameters


on Dynamic Factors

The dynamic processes in crystals influence significantly the physical, chemical and
mechanical characteristics of crystalline solid bodies. The processes of diffusion
alter the chemical composition of the solid body inevitably the material mechanical
properties, including the appearance of point defects due to the origination of con-
centrators of stresses in bonds (relation 2.7). The spatial and elastic waves appearing
in the solid body under the effect of external forces alter the crystalline structure. But
originating and moving dislocations alter the crystal properties most. It is apparent
that the metric of microprocesses is not a still image; it is a changing structure of the
complex world.
These phenomena are studied by a special sphere of physics of solids; it is the
solid body dynamics. This sphere comprises the classic and quantum mechanics of
collective movements of atoms in the perfect crystal, the dynamics of defects in
the crystalline lattice, the theory of interactions crystal with various physical fields,
such as thermal flows, electromagnetic and other types of emission, description of the
physical mechanisms of strength and plasticity of solid bodies [12]. The dynamics of
crystalline lattice perturbations in the solid body treat various oscillations subdivided
3.5 Dependence of the Crystalline Lattice Parameters on Dynamic Factors 91

Fig. 3.9 Methods of surface analysis [13]

into acoustic, optical and electromagnetic polarization oscillations. Each of the listed
oscillatory process has its own frequency and dispersion relations making up most
essential characteristics of the frequency spectrum.
The dynamic processes in solid bodies result in the microparameters included
into various calculation relation as constants characterizing those or those properties
materials which do not belong to them. It leads to the mismatch between the theo-
retical conclusions and the relevant experimental findings. The methods of physical
and chemical analysis of solid body surfaces serve to validate the microparameters
of substances a using fine physical experimentation [13]. Famous German tribologist
H. Czichos shows the scheme of these studies (Fig. 3.9).
The scheme of in Fig. 3.9 reflects the principle of these methods which imply
the exposure the surface of the material in question to a definite type of emission of
X-rays, electrons, ions and others. The task of the primary emission is to excise atomic
levels in the elements in question which, in their turn, emit the excessive energy. The
typical emission provides the information about the nature of the emitting atom,
their chemical environment and the specific structural features of the surface under
study. A number of authors [14, 15] point at the necessity to take into account the
effect of thermal fields on the structures of substances, as well as bonds modifying
these structures under the thermal factor effect together with the macrocharacter-
istcs of materials involved in friction. The description of temperature effect on shear
stresses in metals and allow as an example of the successful theoretical model com-
bining the crystal parameters with the effective dynamic factors [9]. The analysis
of the experimental of dependencies of displacement stresses on the temperature of
the monocrystal Mg and Li-F (Fig. 3.10) permits to assume the following simple
dependence for their description:

G s = G 0s eBT T , (3.98)

where G 0s the crystal modulus of displacement at T = 0 K; BT the calculation


constant dependent on the crystal nature.
To calculate the physical sense of the constant BT , let us consider the changes of
the Burgers vector b in response to the temperature. To this end, let us apply relations
(3.44), (3.45), (3.80) and (3.87) and obtain the following by their transformation:
92 3 Dynamic Microprocesses in Solid Bodies

Fig. 3.10 Dependence of displacement stresses on temperature: a magnesium monocrystal alloyed


with Indium and Thorium; b Li-F crystals with different magnesium content (1 slow chilling;
2 chilling in air; pure monocrystals: 3 rate of chilling 0.002 C/min; 4 rate of chilling 50 C/min)

G s b2 kT
= . (3.99)
2 d D
In accordance with equation (3.60) the parameter d of which can be assumed
equal approximately by the order of magnitude to the crystalline lattice period d, and
= d /t, expression (3.99) is simplified as follows:

Gs 2
b = kT,
2
from which it is easy to find that
"
2 kT
b= . (3.100)
Gs

Using empirical relation (3.98), let us represent formula (3.100) in the following
way: "
2 k 1/2 BT T
b (T ) = T e 2 . (3.101)
G 0s

This equation establishes the regularities of the temperature effect on the Burgers
vector. Formula (3.101) splits into two factors, the firs being constant for this crystal,
3.5 Dependence of the Crystalline Lattice Parameters on Dynamic Factors 93

while the second representing a simple enough function with only one variable T . The
simple mathematical dependence b(T ) permits to apply such relations to validation
of calculation of the mechanical, physical and chemical characteristics of solid bodies
exposed to external mechanical effects.

References

1. Ch. Kittel, Introduction into Solid Body Physics (Mir, Moscow, 1978), p. 791
2. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 3 (Sov. Encyclopedia, Moscow, 1992),
pp. 672672c
3. D.N. Lyubimov, V.A. Ryzhikov, Processes in Friction: Manual (South Russia State Engineering
University, Novochherkask, 2006), p. 147 (Phys. Chem.)
4. G. Gottstein, Physical and Chemical Principles of Science of Materials (Binom, Moscow,
2009), p. 375
5. D.N. Lyubimov, K.N. Dolgopolov, Difusive Processes in Friction (IP Bouryhin V.M., Shakhty,
2010), p. 148
6. L.D. Landau, E.M. Livshits, Statistical Physics, Part 1, (Nauka, Moscow, 1976), p. 421
7. F. Glensdorf, I. Prigozhin, Thermodynamic Theory of Structure: Stability and Fluctuation (Mir,
Moscow, 1973), p. 280
8. D.N. Lyubimov, K.N. Dolgopolov, Modern Tribology (Selected Chapters) (IP Bouryhin V.M.,
Shakhty, 2010), p. 181
9. R. Kana, (ed.), Physical Science of Metals, vol. 3 (Mir, Moscow, 1968), p. 483
10. V.F. Kiselev, S.N. Kozlov, A.V. Zoteyev, Fundamentals of Solid Body Surface Physics (INBU,
Moscow, 1999), p. 284
11. G.S. Pisarenko, (ed.), Strength of Materials (Vyshcha Shkola, Kiev, 1986), p. 775
12. A.M. Prokhorov, (ed.), Physical Encyclopedia, vol. 1 (Sov. Encyclopedia, Moscow, 1988),
p. 638
13. Kh.X. Chikhos, System Analysis in Tribonics (Mir, Moscow, 1982), p. 348
14. V.I. Kostetsky, Surface Strength of Materials in Friction (Tekhnika, Kiev, 1976), p. 283
15. V.I. Kolesnikov, Thermophysical in Processes in Metal-Polymer Triobosystems (Nauka,
Moscow, 2003), p. 279
Chapter 4
Friction Microdynamics

So far I stated the principles adopted by the mathematician


and confirmed by numerous tests.
British physicist I. Newton
It is peculiar the paradoxes the way they are, you forget the
common sense.
British physicist W. Gilbert
The reality is not a magnitude to be checked with the
touchstone.
British physicist S. Hawking

Abstract Basic notions about physical and chemical interaction of solids surface
and microprocesses initiating and accompanying this interaction are propounded.
Analysis of phenomena arising as a result of tribosurface activation is given. The
probability of Lorentz relativistic effects of local time deceleration in friction contact
is discussed. The quantity order of this effect is estimated.
American physicist P. Bridgman, Nobel Prize winner, reckons, each new theory
supposes that at last it is exactly there is the happy theory which is to prove the
right answer. Following this reckoning and the above-mentioned apparatus of
solid body mathematical physics, let us try to describe the changes occurring in the
solid body structure under the effect of friction forces and the effect of observable
changes on friction forces. Let us try to preserve the style in which the contents are
presented in the preceding chapters.

4.1 Friction Effect on Solid Body Surface

The friction forces are fundamentally different by their physical nature from the
forces of characterizing or universal gravitation. The main difference is that they
depend significantly on the configuration of contacting bodies and the velocity of

D. Lyubimov et al., Micromechanisms of Friction and Wear, 95


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8_4,
Springer-Verlag Berlin Heidelberg 2013
96 4 Friction Microdynamics

their relative displacement [1]. There are tangential forces of resistance acting along
the boundaries of interface between solid bodies [2].
The solid bodies touch with parts of their surfaces making up the real contact area
AF which is smaller than the nominal contact area. The value A F and the kinetics
of its appearance are governed by the tribological properties of materials in tribounit
and the geometry of the contact surface. The analytic expressions for calculation of
AF are quite complex, that is why simple relations are usually used in practice, such
as the equations proposed by British tribologists F. Bowden and D. Tabor [3]:
N
AF = ,
Y
(4.1)
N
AF = ,
HB
where N load; Y yield limit of softer material; HBhardness of softer material
in the interface.
The friction force FFr is proportional to the contact real area [2, 4]:

FTP = const A F . (4.2)

Expression (4.2) was named the single term friction law. In the general case, the
friction force is the function of many variables: normal load N , sliding velocity ,
contact temperature T , contact time T0 in the environment, and other parameters of
external friction. In practice, the specific friction force Fn is used often equal to the
relation between FFr and the nominal contact area AN and still more often he friction
coefficient f which the relation between the friction force and the nominal load:
FFr
Fn = ,
AN
(4.3)
FFr
f = .
N
If the tangential force is applied to the statically loaded contact between solid bod-
ies, a small relative dislocation 0.11 m appears virtually instantly in the direction
of the force in effect. These displacements preceding sliding are reversible in some
cases or irreversible in others. It means that large-scale elastic plastic processes are
already underway before the contact sliding begins.
The friction at the sliding start is traditionally likened to the result of mechanical
interactions between micron-sized asperities on the interfaced bodies. In other words,
the friction as the force of resistance is due to the dissipative processes of engagement-
disengagement of microasperities and can be represented as the sum of friction forces
over micron-sized spots. This statement has acquired the name of the friction additive
law [5]. The schemes of elementary processes during engagement-disengagement of
microasperities in the contact are shown in Fig. 4.1 [3].
The processes of formation and separation of microcontacts comprise the follow-
ing sequence of main stages: elastic and plastic deformation of asperities, sloughing,
4.1 Friction Effect on Solid Body Surface 97

Fig. 4.1 Scheme of single event in friction process: a elastic deformation, plastic deformation,
ploughing; b adhesive bonding; c is the shear of adhesive bonds, restoration of elastic deformation [3]

and adhesive bonding shear. In accordance with this friction additive law the follow-
ing can be recorded [5]:

FFr = F1 + F2 + F3 + F4 , (4.4)

where F1 the force of resistance to material elastic deformation; F2 the force of


resistance to material plastic deformation; F3 the force of resistance to
material ploughing; F4 the force of resistance to the shear of adhesive bonding
of microasperities.
The actual contact area changes in sliding. During elastic contact the values of
contact spots are equal in static and in sliding. During plastic their deformation in
relation to the actual areas in sliding and during the static contact depend on the type
of deformation micron-sized asperities determined by the normal and tangential
stresses in their contact zone. In approximation, the interactions between the soft
body and the absolutely rigid semispace can record in the following way [1]:

A FFr = ACFr 1 + f 2 , (4.5)

where A FFr the actual area of the dynamic friction contact; ACFr the actual area of
the static contact calculated with formula (4.3); the coefficient characterizing
the interface stress-strain state; f the friction coefficient.
As a rule, the coefficient varies within 3.327, still, the experimental data are
available proving that, at the friction coefficient f 0, 0.4, the value can be
considered close to 1 [3]. Hence, expression (4.5) can be written in the following
simplified form: 
A FFr = ACFr 1 + f 2 . (4.6)
98 4 Friction Microdynamics

The real contact area expands in friction due to the changes in distribution of
mechanical stresses over the friction surface. Solution of contact problems in the
theory of elasticity [3] proves that the maximum deformation of the friction piece
evolves in the layer under the asperity. In the first approximation, the deformable
layer thickness is equal to 11.5 of the contact spot diameter. When the material
slides and rubs, strain stresses appear behind the microasperity and attenuate at a
distance 510 contact spot diameters. While statically the micron-sized asperities
on the tribosurface rest on the counterbody surface forming stationary contact spots.
The sheer force initially passes the re-distribution contact spots and makes the solid
bodies approach inducing the preliminary dislocation.
Statically during plastic deformation the stress-strain state in the contact between
solid bodies in the actual contact zone corresponds to the state result from indentation
of the ball indenter into the counterbody. Plastic deformation
appears in the surface
layer of the solid body with the layer thickness h 3.1 R (Rthe indenter radius).
When micron asperities move, the surface of the viscoplastic counterbody fails to
restore the original shape; therefore the counterbody response does not coincide with
the normal to the vector of normal load N . According to the Hertz theory, maximum
compressive stresses appear on the contact surface, while the maximum one way
directed tangential stresses appear somewhat deeper [3]. For instance, when two
cylinders have the contact width R, this depth is equal approximately to 0.8R. The
sign-variable tangential stress factor in rolling is an essential factor which appears in
counterbody closer to the tribosurface than the maximum one way directed tangential
stress.
When the contact between solid bodies is exposed to friction, the zone of maxi-
mum stresses in the counterbody shifts towards the tribosurface. It is shown in Fig. 4.2
in which the field of distribution of mechanical stresses is calculated for the cylinder
and plane interface [3].
The analysis of distribution fields of mechanical stresses in Fig. 4.2 evidences that
the maximum stresses in friction displace towards the tribosurface and may appear on

Fig. 4.2 Scheme of evolution of tangential stresses (in MPa) in cylinderplane contact: a in static;
b in friction. R Indenter width; traced zoneregion of maximum mechanical stresses
4.1 Friction Effect on Solid Body Surface 99

it. The mechanical stresses distort the crystalline lattice of the solid body surface layer
originating dislocations. Thus, the region of the maximum mechanical stresses in the
interface is the zone where the dislocations concentrate. The dislocation under the
effect of friction approach to the solid body surface; it is identical if the dislocations
appear on the surface. It is noted above [relations (2.23) and (2.24)] that the growing
number of dislocations augments the plasticity and reduces the yield range Y of
materials. It expands the contact real area in accordance with the Bowden-Tabor
formula (4.1).
I.V. Kraguelskii, the soviet school of tribology founder, and associates studied
the plastic flow of copper and steel specimens and established that the distribution
of the yield limit in the surface layer is ambiguous. Two layers are identified with
regularities of deformation processes which the authors called surface subsurface
layers [6]. It is shown that the copper surface layer extends to the depth 0.51 m
and is characterized by the exceptionally high density of dislocations close to limit.
The gradient of the yield limit is also exceptionally steep an amounts on the average
to 200 MPa/m. The subsurface layer is 50100 m thick and it is characterized
by the sparser dislocations and smoother yield limit gradient amounting tentatively
to 3 MPa/m. The surface layer of steel specimens at the identical depth had much
smaller yield stresses and the yield limit gradient had the opposite sign versus he cop-
per counterbodies. The contact discreteness dictates strong initial friction moments
and strong fluctuation and stresses in the subsurface layer of the specimens. It
results in the plastic flow, inhomogeneous work hardening of the metal. The plastic
zone depth is determined by the dimension of the real contact spots and by the spot
boundary conditions (the magnitude of tangential stresses on the tribosurface or in
the lubricating material layer) [6].
The tribosurface of solid bodies has always a vividly pronounced defective struc-
ture which contains virtually all crystalline structure imperfections (Sect. 2.3). When
the deep dislocations, the situation is aggravated intensifying the tribosurface
activity. Active regions appear in the surface and subsurface layers. They feature
a strongly heterogeneous structure composition making them considerably differ-
ent from the structure and the composition of the original material. The extremely
high surface concentration of dislocations mentioned above results in the fact that
the plastic deformation during interactions can reach tens of percent and more in
the actual contact area, while its rate amounts to 102 103 s1 . So much intensive
plastic shear strain results in the energy dissipation and deformation generating heat
inducing a considerable internal heating of the surface layer [7].
When tribocontact appears, the superactive tribosurfaces approach to so intimate
spacing that forces of adhesion come into action. Because the actual contact area
grows and juvenile surfaces appear, these forces augment many times compared
with the similar forces in the static contact bringing the surfaces still closer. The
force of friction perform the work which, in accordance with formula (4.3), is equal
to
A Fr = f N t, (4.7)

where N normal load; sliding velocity; tfriction time.


100 4 Friction Microdynamics

In accordance with the findings disclosed by B.I. Kostetskii in [8], virtually all
friction work is open for heat generation Q (over 90 % of the value A Fr (4.7)).
According to [9], the thermal flow through a single asperity (the contact spot) rough-
ness is determined by the formula:

qfc = E eq fa H , (4.8)

where E eq is the mechanical work thermal equivalent; a temperature conductivity;


H contact spot microhardness.
Assume that all friction work virtually transforms into q f c which goes into the
surface layer, then it can be assumed in principle that q corresponds approximately
to the elastic energy of dislocations. In the first approximation:

1
q= E. (4.9)
2

Equate the right terms in (4.8) and (4.9). Since by definition = FAFrF , and the
value A F can be found with the Bowden and Tabor formula (4.1), then:

= f Y . (4.10)

Remembering that E = 2G s (1 + ) [10], obtain:

2Y (1 + ) G s
H = . (4.11)
E eq a

It is shown in section (3.5) that the modulus of displacement in relation (3.100)


is the function of temperature; let us represent formula (4.11) in the following way:

2Y (1 + ) G 0s BT T
H = e . (4.12)
E eq a

Thus, a considerable hardness reduction and smoothing of contacting micron-size


asperities can be expected in friction augmenting the real contact area and adhesive
interactions on the tribosurfaces. The heat flow in friction activates the motion of
dislocations in accordance with the mechanism described in Sect. 3.4, augments
their capabilities to overcome potential barriers and issue on the surface en masse.
It is accompanied with the growing plasticity of tribomaterials. It is established, for
instance, in [11] that plastics deform even diamond in friction.
4.2 Diffusive Triboprocesses in Friction 101

4.2 Diffusive Triboprocesses in Friction

The tribosurfaces acquire a combination of physical and mechanical characteristics


inducing strong adhesive bonds. Simultaneously inhomogeneous temperature field
(temperature gradients) appears in tribomaterials. All combine together as mentioned
above and in accordance with the Fick law (3.39) catalyzes the diffusion processes.
General regularities intrinsic to diffusion processes complicate in friction the
physical and chemical activation of materials in friction. The kinetics of the diffusion
processes is determined by how easily the particle detaches from the diffusing sub-
stance (3.63) in the accustomed spot and how much permeable becomes the envi-
ronment in which particles are to move.
In accordance with the main equations of the diffusion theory (3.66) and (3.68),
it can be recorded that
kT
D = d2 exp[(W D T S)].
h
Following the classic thermodynamics with the account Q = TS, it is obtained
that
kT
D = d2 exp[(W D Q)], (4.13)
h
where Q is the amount of heat generated by friction.
Equation (4.13) confirms the fact that the energy of diffusion activation because the
tribocouple absorbs the friction heat Q. The energy reduction W D according to
the Arrhenius and Fick laws (S. ArrheniusSwedish physicist and chemist) boosts
the diffusion coefficients of the diffusive flow.
The diffusion activation by friction is confirmed with the method of radioactive
indicators. It is shown that the diffusion coefficients grow in friction by 10 orders
of magnitude compared with the similar values for static conditions [8, 12]. The
example of copper diffusion into polyethylene in [13] showed that friction would
activate the mass transfer and boost the diffusion coefficient almost a thousand times.
The copper diffusion depth into the polyethylene substrate reaches 200 m under the
assigned friction conditions and the diffusion temperature dependence corresponds
to the equation:  
10343.6
D = 0.912 exp , m2 /s. (4.14)
T

When the polyethylene coating is produced by heat on the copper specimen,


the gradients of concentration and temperature of copper particles diffuse from the
substrate into the polyethylene melt due to the latters physical and chemical activity.
The diffusion intensify is much weaker than that friction activated mass transfer. The
coefficient temperature dependence of copper diffusion into the polyethylene coating
deposited at 453510 K is the following:
102 4 Friction Microdynamics
 
8659.8
D = 144.5 103 exp , m2 /s. (4.15)
T

The diffusion flow generated by friction in accordance with (3.39) and (3.49)
decomposes into three components [14]:

J = Jc + Jm + J , (4.16)

where J is the total diffusion flow; Jc , Jm , J are the diffusive flows realign to the
gradients of concentration, temperature and electric potential, respectively.
After substitution into expression (4.16) the values of particular diffusion flows
it can explicitly be rewritten in the following way:
 
C c c
J = D + QT + F z eff , (4.17)
z kT 2 kT 2

where C  the speed of light; Fthe Faraday number; z eff the effective value of
the transferred charge; Cthe diffusant concentration.
The diffusion flow value is much affected the substances structure in which the
mass transfer takes place. In accordance with the universal principle of energy expe-
diency, it is namely the structure that governs the physical and chemical mechanism
of mass transfer by diffusion. It is established in [15, 16] that the diffusion in friction
follows mainly the vacational and internodal mechanisms.
High intensity of diffusion flows makes the diffusion one of the most essential
phenomena accompanying friction and which determines the tribocouple friction
characteristics. The mass transfer influence friction and wear in a variety of ways
which are worth considering. Let us begin with the thermodynamic influence of
diffusion on friction processes.
Thermodynamics of tribosystem. The tribounit is the tribosystem configured in
a definite way. The system in the general case is a consistency of elements combined
together into a definite integrity and unity. The integrity (unity) of the object is the
capability of the latter to stand apart from the environment. The most significant
attribute implicitly included into the notion system is the structure governing its
integrity. The structure is an internal type of the system consistency determining
the unity of stable mutual links between the systems elements. L.I. Bershadskii
defined the tribosystem as the dynamic and dissipative system in which the energy
degradation is embodied and micromechanical motion occurs [17, 18]. Its carriers
are microvolumes and triboexcited materials and interphase boundaries opened to
friction activated flow of substance, heat, electricity, sound et cetera.
The apparatus of thermodynamics of inequilibrium systems are used more and
more frequently in recent decades to describe the tribocouples of various designs and
applications [19]. The reason is that the tribounit as the mechanical system is in the
state of constant energy pumping up by friction forces. Because the tribosystem
has no endless energy capacity, the tribocouple emits the accumulated energy in
some way after it achieves some critical value. The processes of energy emission
4.2 Diffusive Triboprocesses in Friction 103

is implemented by transfer phenomena, in other words, by inequilibrium processes


resulting in the spatial transfer of matter, energy, entropy, electric charges and other
physical amounts [20].
The transfer phenomena are caused by perturbations upsetting the thermodynamic
state equilibrium of tribosystems: the action of force fields, the spatial composition
inhomogeneities, the temperature or mean velocity of particles. The physical amounts
are transferred in the directions opposite to their gradients, as a result, the system
isolated from external approaches to the thermodynamic equilibrium state. If the
external effects are not supported constant, the transfer phenomena evolve stationary.
They are characterized by the irreversible flows of physical amounts Ji (the flow
of matter, heat, et cetera). In case of little deviations of the tribosystem from its
thermodynamic equilibrium, the flow depends linearly on the thermodynamic forces
k inducing the deviations from equilibrium. The deviations are described by the
phenomenological equations of the following type:

Ji = L ik k , (4.18)

where L ik are the phenomenological or kinetic transfer coefficients calculated by


soling kinetic equation [21].
The general phenomenological theory of transfer phenomena applicable to any
transfer phenomenon is provided by the thermodynamics of open systems [19]. The
open system is assumed to be capable to exchange matter, energy and impulse with
the environment. In other words, it is the thermodynamic system with transfer phe-
nomena relating to the class of open systems. Hence, the tribosystem is a typical
instance of the open thermodynamic system. The balanced energy influx and outflux
are most essential for the open thermodynamic system. Such system responds to the
upset balance changing its functions and creating or eliminating new bonds.
According to the second principle of thermodynamics [22], the entropy in the
closed system grows and tends to the maximum equilibrium value while the entropy
generation tends to zero. Unlike the closed system, the stationary state the constant
entropy and with its constant generation can exist in the open system and it cor-
responds to the tribosystem definition by Bershadskii [17]. The generated entropy
should be removed from the system and its removal per unit of time is usually cal-
culated with the following formula [23]:

S(t) = Ji i ,
i

or with the account of (4.18) with



S(t) = L ik k i . (4.19)
i,k
104 4 Friction Microdynamics

The second principle of thermodynamics for the open systems is recorded in the
following way [21]:
dS = (dS)1 + (dS)2 , (4.20)

where (dS)1 is the entropy changes under the effect of processes within the system
(dS)2 and due to the exchanges with the environment.
The values in Eq. (4.20) are the following: (dS)1 is more than and (dS)2 less
than zero. It means that due to the exchange with the environment the entropy in
the open thermodynamic system, the entropy can become negative. It proves that the
degradation of the thermodynamic system can transform into the complex behavior
of its evolution [19, 22].
The stationary states in the open thermodynamics systems can be similar to the
equilibrium states mentioned above which are described by Onsager dissipative func-
tions (L. OnsagerAmerican physicist):

1
0 = L i,k i k ,
2
(4.21)
1
0 = Ri,k Ji Jk ,
2

where Rik is the phenomenological coefficient similar to L ik .


One more thermodynamic characteristic of the stationary state is the rate of
entropy generation per volume unit:

1 dS
s = . (4.22)
V dt
This rate in the open systems is described with bilinear expressions in respect to
the flow of thermodynamic forces included into phenomenological equation (4.18):

dS  
= Ji i + Jjj. (4.23)
dt
i j

Thus, the rate of entropy variations in the open thermodynamic system is deter-
mined by the interaction between energy flows. The forces due to the interactions
affect both flows. These effects obtained the name of intersecting processes exem-
plified by the Dufor diffusion considered above. The intersecting processes imply
the non-equilibrium transfer processes during which the flows Ji and Jk are due to
the thermodynamic forces k and i at i = k, respectively. The equality L ik = L ki
corresponds to the intersecting processes.
The tribocouple is a classic example of the open thermodynamic system. First,
the tribocouple has all attributes of the notion system because all most important
tribounit characteristics, for instance, friction resistance belongs to the tribounit as a
whole rather than to its parts. The friction resistance implies the collective qualitative
4.2 Diffusive Triboprocesses in Friction 105

tribounit characteristics or the capability to achieve the state with extremely weak
friction forces while simultaneously preserving the conditions of tribostrength or the
tribounits ability to resist the destructive friction processes. Second, a combined
effect of the mechanical, physical and chemical processes within a tribounit turn it
into an integral system.
The main channels of energy dissipation in the tribosystem are transfer of matter
and heat. In accordance with the earlier described bonding, the transfer phenomena
and the gradients of respective physical amounts, thermodynamic forces c and m
in the main energy dissipation channels of the tribosystem can be recorded in the
following way [23]:

C
c = ,
C
T
T = , (4.24)
T
where C is the matter concentration; T is the temperature.
The corresponding thermodynamic flows can be represented in the following way:

Jc = L c c = m,
JT = L T T = Q, (4.25)

where m is the transferred matter mass; Q is the transferred heat amount.


Relations (4.24) and (4.25) are similar to the equations of the diffusion laws listed
in Sect. 3.3. In accordance with Eq. (4.23), the entropy variations in the tribosystem
(hence, those of the structure and main friction characteristics) are determined by the
interaction between two matter and heat flows. It is remarked in the monograph [3]
that tribomaterials fracture due to the breaks in their continuum inducing the mass
transfer towards the appearing defect in order to fill it up. Thus, the flow Jc can counter
the fracture of the abatable material. The thermal flow Q appears simultaneously and
creates defects in the crystalline structure, thus intensifying the destructive processes
in the tribomaterials.
It should be noted that other dissipative channels exist accounting for less than 2 %
of the total energy balance in addition to the mass and heat transfer in the tribosystem
and other transfer processes [8].
Summarizing the thermodynamic reasoning, it can be concluded that the matter
diffusion reduces the tribosystem entropy and so counters the fracture process by
improving its structural arrangement. Improvement of the tribosystem structure is
referred in [24] to its sophistication due to the appearance of third bodies in the
friction zone.
The third body is the zone of tribointeractions between solid bodies where the
lubricating material and debris remain. The lubricating layer is a molecular film
structure in full compliance with term nanostructure. The mechanical, physical
and chemical characteristics of the third body differ from the same characteristics
of tribomaterials. The third bodies assist to implement the principle of the posi-
106 4 Friction Microdynamics

tive gradient of mechanical behavior which is at the base of the present-day high
friction resistance of tribosystems [25, 26]. The third body appearance with the tri-
bobonds means that the friction coefficient and the wear reduce. The direct contact
between solid bodies is replaced with more favorable interaction between tribo-
surfaces through third bodies. The wear resistance as an essential characteristic of
materials is determined in many respects by the structure and duration of the third
body [2]. The tribology knows the following relation [27, 28]:

hm
I = kS , (4.26)
t
where I is the wear of materials; k S is the coefficient of filling of the tribocontact by
the third body; t is the duration of the third body; h m is the third body thickness.
The coefficient k S can be determined in the following way:

AF
kS = , (4.27)
A T.B
F

where AF is the tribocontact actual area; AT.B.F is the third body actual area.
The analysis of relation (4.26) and (4.27) leads to interesting conclusions. The
thicker the third body the sooner it and the friction parts wear away. At the same time,
the larder contact area the third body covers and the longer it exists the more favorable
conditions appear for the tribocouple to operate and the longer the steady friction
lasts the later the irreversible degradation processes begin. Thus, the tribosystem is
capable to undergo evolution with the most favorable conditions being the third body
appearance when the diffusion processes evolve to the maximum intensity.
The third body evolution proceeds though the phase of nucleation [12]. It is
shown in [29] that the third body appears initially in the tribosurface regions with the
strongest adhesion, for instance, over the boundaries between intergrain crystallites
(Sect. 2.4) (Fig. 4.3).
When the third body forms, the adsorbed atom (adatoms) jump from one adsorp-
tion center on the tribosurface to the other with the frequency exp(W/kT).
The atom stays in the bonded state in adsorption center.
The speed Jc of nucleation of the third body is determined by the surface diffusion
intensity:
Jc = c D D Nc n(t), (4.28)

where c is the dimensionless coefficient of invasion of surface centers by active


adsorbed atoms; D D is the surface diffusion coefficient; Nc is the number of active
adsorption centers; n(t) is the concentration of adsorbed atoms.
The rate of the third body adsorption area expansion during nucleation is deter-
mined by the following relation:

DD
(t) = 0.5 , (4.29)
t t
4.2 Diffusive Triboprocesses in Friction 107

where t  is the nucleation time.


After the third body first layer appears, it thickens and it can be described by the
following equation:

T.B
2C C
2
+ = 0, (4.30)
dx DD d x

where T.B is the rate of film growth on the third body on the tribosurface; C is the
concentration elements in the adsorption layer.
Thus, in accordance with Eq. (4.30), the rate of the third body growth is limited
by the diffusion processes evolving in the tribocontact zone. The time of the third
body full formation, in other words, when the film of atoms and molecules covers
the tribosurface, is determined by the following expression [14]:
 1
t 16Deff [(x i ) h m ]2 Ji , (4.31)

where Deff is the diffusion effective coefficient taking into account the variety of
forms and mechanisms of this process; i , Ji are the flows of diffusing particles; h m
is the third body thickness.
It follows from relations (4.284.31) that the kinetics of formation of the third
body, hence, the variations of friction characteristics of tribocouples are determined
in many respects by the diffusion processes. It is established in [16] how the bonds
and the diffusion activation energy WD relate to the friction coefficient (Fig. 4.4).
Hydrogen-related wear. The hydrogen transfer in metals by diffusion plays a par-
ticular role tribology [30]. On the one hand, the kinetics of hydrogen penetration into
the crystalline lattice produces its effect because it is the lightest element of the peri-
odic system of chemical elements being highly motile. On the other hand, damages in
the structure of tribomaterials favor its penetration. The elastic and plastic tribodefor-
mation affects the intensity of interactions between tribomaterials and environment

Fig. 4.3 Nuclei of third bodies over boundaries between grains of crystallites: a appearance of
polymeric film; b concentration of lubricating particles over boundaries between grains of austenite
steel [29, 30]
108 4 Friction Microdynamics

augmenting the concentration of the particles adsorbed from the environment and
affecting the crystalline structure defects. The mechanical and thermal phenomena
in friction accelerate the release of hydrogen from lubricating and tribomaterials
into the tribocontact area. Thus, the external medium enters into the interaction with
transferred dislocations activated by the surface metallic layer intensifying both phys-
ical and chemical softening and adsorptive and diffusive hydrogenation processes.
The Cottrell atmosphere mentioned in Chap. 2 in which hydrogen dominates appears
leading to local overstresses in the tribomaterial resulting in the propagation of micro-
cracks.
It is assumed in [31] in respect of the hydrogen diffusion flow in the metal that
it consists of the electroactive and electropassive components. The electroactive
component is the directed motion of the ionized hydrogen (plasma, in fact) through
the regular metal crystalline structure. The electropassive components are produced
by the atomic hydrogen penetrating into the metal through its crystalline structure
defects. The most likely hydrogen in steels is the solid solution of protons interacting
with the electron cloud. The hydrogen can be in metals in the molecular form or as
hydrates and methane in hydride producing metals.
In accordance with the Fick law, the hydrogen diffusion flow is directed in response
to the environment temperature and nature. The effect of the temperature field on the
hydrogen concentration distribution is theoretically estimated in [31]. Equation (3.39)
is the following in the one-dimensional approximation

C c T W
D0 + = J0 e kT , (4.32)
x kT 2 x

and its solution is the following:


I (z) p[ (z) (0)]


y(x) = 1 1 e p I , (4.33)
I (l)

Fig. 4.4 Dependence of


steady values of polyamide-
metal friction coefficient on
diffusion activation energy:
1 Ni, 2 Fe, 3 Cu, 4 Cr
4.2 Diffusive Triboprocesses in Friction 109

Q
where z = xl ; y = CC0 ; = kT W
;p = W ; = CC0l ;  = (l) (0); lthe
specimen thickness.
All physical values with the index 0 correspond to the tribosurface. The diffusion
flow J (x) can be represented in the following way:

(1 p)
J (x) = exp d x. (4.34)
x
0

The equations describing hydrogen flows follow from relations (4.324.34):

D0 c0 e p (0)
J0 = (1 exp( p )) ,
l I (l)
  exp[(1 p) ]
Jl = exp(1 p) 1 . (4.35)
(1 p)

The hydrogen diffusion flow follows the Arrhenius law (3.63). The mathematical
analysis of Eq. (4.35) permits to assume that there are maxima of hydrogen concen-
tration profiles in the metal [31]. Their appearance can induce secondary hydrogen
diffusion flows mentioned in [9, 14].
The motion of the secondary hydrogen diffusion can be directed towards the pos-
itive temperature gradient seemingly in contradiction to the heat and mass transfer
laws. This fact is treated in the monograph [31] as a variant of solution of diffu-
sion equations ant it was not scrupulously analyzed from the physical viewpoint.
Nevertheless, it is apparent that the negative temperature decelerates the secondary
diffusive flows. It is possible if the diffusive flow constants Jc in Eq. (4.16) excel the
similar flow values JT in the same formula.
As it follows from the Fick law in Eq. (4.17), the diffusion coefficient character-
izing the flow JT is equal to
c Q
DT = D 2 . (4.36)
kT
The distribution of hydrogen tribocouples determines the resulting flow J made
by differently directed flows Jc and JT having different signs:
 
c Q T C
J = Jc JT = D . (4.37)
kT x
2 x

Relation (4.37) can then be written in the following way:


 
D c Q
J = T C , (4.38)
x kT 2
110 4 Friction Microdynamics

where x is the diffusion layer thickness of the secondary flows; T is the tempera-
ture difference in the diffusion layer; C is the difference of hydrogen concentrations
in the diffusion layer.
The value x is actually the diffusion length which enables to rewrite formula
(4.38) in the following way:
  
D c Q
J = T C . (4.39)
2t kT 2

The redistribution of hydrogen concentration in the steel surface layers due to


the secondary flows is established from the experimental hydrogen laser sampling
analysis [32] (Fig. 4.5).
The friction is accompanied by changes in the field of mechanical and thermal
stresses due to the energy relaxation processes. It alters the hydrogen concentration
profiles in the tribopart (Fig. 4.5). It is shown in the graphs that, upon friction, hydro-
gen leaves its accumulation region in the specimen volume (the molecular micro
reservoir) and diffuses on the tribosurface augmenting the surface hydrogen con-
centration. It is exactly the secondary hydrogen diffusion flow moving from initial
accumulation regions; this flow governs, in the final account, the resulting hydrogen
concentration balance in the specimen volume (4.39). The hydrogen desorption from
the molecular microreservoirs means that the secondary diffusive flows appear
without any participation of chemisorbed hydrogen atoms.
The experimental curves in Fig. 4.5 permit to estimate the diffusion constants and
the intensity of primary and secondary diffusion flows. About 10 % of the sorbed
hydrogen relates to the secondary flows; the hydrogen concentration in the surface
flow grows approximately by the same order of magnitude (15 %). A slight rise of
the concentration is apparently due to the additional sources of hydrogen delivering
it into this specimen (from the atmosphere, for instance).

Fig. 4.5 Distribution of


hydrogen diffusion concentra-
tion through depth H on steel
piece 30 min after friction: 1
original specimen, 2 friction
path surface, 3 specimen after
friction (unrubbed surface)
4.2 Diffusive Triboprocesses in Friction 111

The above results give grounds to assert that formulas (4.374.39) provide a
satisfactory description of the processes of hydrogenation of metallic materials in
friction. The calculations and experiments have manifested that the amount of the
hydrogen diffusion grows as the function of load and test time, the hydrogen maxi-
mum concentration zone shifts deeper into the specimen. It is due to the mechanical
and thermal activation of materials in friction adding defects in the crystalline struc-
ture (particularly due to dislocations) thus accelerating considerably the diffusive
processes. Diffusive bonds as dislocations move are unambiguous and it is appar-
ent that this factor affects the kinetics of triboprocesses. Russian tribologist S.A.
Polyakov assumed that it was the main diffusion and dislocation mechanism of evo-
lution of tribosystems [24].

4.3 Motion of Dislocations in Friction

Motion of dislocations in friction is described in [24] with tensor equations:


 
dik 
+ eilm e eilm lk m = 0,
S S
(4.40)
dt
S


where jik = eilm lkS mS is the tensor of the flow density of dislocations; ik , lk
s
S
is the tensor of the density of dislocations; mS is the velocity of dislocations.
Equation (4.40) has the form of the continuum equation with the distinction that
the tensor of density of dislocations serves instead of the environment density [24].
This equation takes into account both the sliding and climbing of dislocations (see
3.8). Equation (4.40) discloses the physical sense of FrankReed-Herring dislocation
sources which have different velocity parameters and dislocation structure. Let us
consider the regularities of appearance of the dislocation structure in friction.
Tribosurface role. Large clusters of dislocations are needed to activate mechan-
ically the tribosurface. They appear at definite (critical) values of the load and speed
factor. B.V. Deryagin demonstrated that the clusters of dislocation would surface on
the solid body and induce its mechanoactivation accompanied by the exoelectronic
emission. This initial stage is believed corresponding to the friction with limited or
low-active lubrication. In these conditions the electron work depends on the normal
load and has different values during elastic deformation of the surface layer, when
unsaturated plastic deformation processes evolve intensifying the density disloca-
tions, and dynamic equilibrium is reached during saturated the plastic deformation
with nucleation and annihilation of dislocations [33].
For description of the kinetics of the dislocation displacement combination let
us see the following equation containing the velocity parameters in the implicit
form [24]:
D  
= D + 1D 2D D D D 2
, (4.41)
t
112 4 Friction Microdynamics

where D is generation of dislocations with different sources; 1D is the coefficient


of multiplication of dislocations; 2D the coefficient taking into account the exit of
dislocations from the processes of sliding and multiplication; D is the coefficient
of elimination (annihilation) of dislocations.
When the dislocations move in one direction, it is little likely that the dislocations
with opposite signs would meet and cause annihilation, therefore the term D D 2

Eq. (4.41) can be ignored. The physical sense of the coefficients D , 1 and 2
D D

can be analyzed using the diffusion and dislocation principles in Eq. (4.40). Let us
consider the stationary form of Eq. (4.41) satisfying the condition tD = 0. It yields
the expression for the density of dislocations:
 1
D = D 1D 2D . (4.42)

The coefficient D is the following in the general form:

D = 0D e mS , (4.43)

where 0D is the density of dislocations from sources at 0D = const; e mS is the


frequency characteristic of the source activation.
The Franck-Reed sources contribute insignificantly to the kinetics of displace-
ments in the dislocation structure. A considerable role in friction plays the higher
concentration of vacancies so the Herring sources should be taken into account. Their
frequency characteristic is the ratio the mean climbing rate of the dislocation loop
n growing in the source and the value of climbing hn needed for new loop. Then
(4.43) can be represented in the following simplest way:

n
= 0D . (4.44)
hn

The physical sense of the parameter 1D in (4.42) corresponds to the ratio between
the sliding velocity c and the free run l of dislocations. The value c depends
substantially on the concentration of point defects in the material. The higher con-
centration accelerates the motion of steps on dislocations overcoming obstacles by
diffusion [34]. The tribosurface is virtually unlimited source and drain of vacancies
as defects most easily appearing in metals. Thus, the velocity c and the concen-
tration of vacancies grows as a function of approach to the tribosurface. With the
account of the above-said
c [Cv (x)]
1 = , (4.45)
l
where Cv is the concentration of vacancies; x is the coordinate normal to the surface.
The physical sense of 2 is studied by revealing the causes why the dislocations
leave the sliding zone. The cause can by the climbing of boundary dislocations. The
rate and the distance of climbing h n are found in the following way:
4.3 Motion of Dislocations in Friction 113

8C0 D C
n = ,
bb C0
(4.46)
2C 
hn = ,
D bb

where  is the atomic volume; C0 is the equilibrium concentration of vacancies;


D is the coefficient of diffusion of vacancies; bb is the Burgers vector of the
boundary dislocation.
Relation (4.42) can be rewritten in the expanded form:
 1
n n c [C (x)]
D = 0D . (4.47)
hn hn l

Thus, the values D , 1D , 2D , D at each time segment change following the


dislocation and diffusion mechanisms which, in the final account, limit the rate of
relaxation of the dislocations in the structure 2D and propagation of dislocations D
and 1D . The total action of the above mechanisms is, first, to delocalize deformation
of the tribopiece and decorate its hardening due to the diffusive relaxation of stresses
(2.7), second, to increase the plasticity of the surface layer of metallic tribopieces
by creating a special polygonal cellular structure [24] the microhardness of which
changes in response to friction.
Adequacy of models. One of the consequences of regularities described by
Eq. (4.47) is the so-called scale effect. It implies that as soon as load on the fric-
tion couple drops below the critical value, the degree of hardening of contacting
microvolumes of the material augments. The value c [Cv (x)] in Eq. (4.47) changes
and increases the coefficient of propagation of dislocations 1D and augments the
degree of material hardening. The values D and 2D remain virtually unchanged
because they relate to the degree of oversaturation of the surface layer with vacan-
cies [6]. According to the terminology proposed by V.P. Alekhin [35] this mechanism
operates as the pump of vacancies.
The results of description of the dependence of density of dislocations on the
concentration of vacancies D (Cv ) in Eq. (4.47) correlate well with the experiment
and the known data. The approximating experiment dependence can be recorded in
the following way
     
1 
d d d
= 0D , (4.48)
d cl d cl d x

where (d/d)cl is the relative changes of the crystalline lattice constants when
dislocations climb; (d/d)x is the similar value at the depth x from the tribosurface.
The deviation of the theoretical dependence (4.48) from the experimental one
(Fig. 4.6) does not exceed a few percent.
Tribosurface damage. The coefficient D in Eq. (4.41) can be replaced with
expression D DD assuming that the density D grows due to the intensified
0
114 4 Friction Microdynamics

Fig. 4.6 Variations of density


of dislocations in surface
layer of copper specimen
in response to reduction
of crystalline lattice period
during friction [24]

functioning of the source of dislocations. New dislocations appear when the points
of fixing of the dislocation loop come nearer (Fig. 2.8). For this case it is advisable
to group the coefficients D /0D , 1D and 2D into one coefficient k1 . If we ignore
the D D
2 and integrate the obtained equation, the result is the following:

D = 0D exp k1 t, (4.49)

where t is the current time.


The coefficient k1 can be considered as the characteristic of durability of the
surface layer. Assume that the delamination of the actively deformable microvol-
umes from the tribosurface proceeds until some critical density of dislocations cr
is reached. The time tcr when this value is reached is identical to the surface layer
durability:
1 cr
tcr = ln . (4.50)
k1 0

All values in the coefficient k1 depend on the concentration of vacancies Cv .


The concentration can be determined with the expressions of the identical Boltzmann
or Arrhenius laws:

W (P + p) V + We
C = C0 exp , (4.51)
kT

where W is the energy of formation of vacancies; P and p are external pressure and
internal stress in the crystal; V are changes in the crystal volume when the vacancies
appear according to relation (2.1); We is the energy relating to the additional work
of formation of vacancies.
4.3 Motion of Dislocations in Friction 115

Equation (4.50) with the account of (4.51) can be rewritten in the following way:


1 cr 1 W (P + p) V + We
t p = ln exp . (4.52)
k1 0D C0 kT

If the dislocation structures formed with the help of the mechanism described
with Eq. (4.47), it can result in the polygonal cellular structure. When the dislocations
climb, their walls can rupture due to the bending causing the microcracking [10]. The
stresses in the walls are sufficiently strong and have the order of magnitude 103 G s ,
where G s is the shear modulus. The osmotic force facilitates the rupture of the bent
wall; it appears when vacancies oversaturate and reach the value 1.6 N/m per unit
of the dislocation length. This mechanism of cracking can be called the dislocation
diffusion mechanism because the material loses quicker the continuum under the
effect of the dislocation dipole due to the wall rupture.
The dislocations accelerate as their density reduces (Sect. 3.4), the c D1 . It
would seem that the propagation of dislocations in sliding would result a new stage
of hardening of the surface layer. But it is easy to show that even when the sliding
velocity of dislocations increases to the value at which they detach from the Cottrell
atmosphere (about 103 cm/s), their concentration reduces due to the climbing which
quite effectively compensates the propagation of dislocations.
Let us write Eq. (4.40) in the expanded form and assume D = 0 and separate the
dislocations into two types: those climbing from the density n and those sliding c :
   
D c c n n
= . (4.53)
t l S l n

Reduce Eq. (4.53) to the stationary form assuming D /t = 0. Then express the
total density of dislocations tot. through n = tot. 0D and assuming approxi-
mately h n /l = 1, record:
c
cym = 0D + c . (4.54)
n

Because c is larger than n , then nc > 1 and acquires the value from 103 to 5.
Introduce the relation h /l into formula (4.54) where h is the distance the dislo-
cation has to cover to annihilate on the surface. The expression is the following:

c h 
tot. = 0D + c , (4.55)
n l

reflecting adequately the experimental results.


Based on the presented data, the fundamental conclusion seems reasonable why
the plasticity of materials grows in friction. It can be assumed due to the motion
of dislocations. The process of deformation delocalizations leading to relaxation of
stresses augments the plasticity margin of tribomaterials.
116 4 Friction Microdynamics

4.4 Relativistic Effects in Friction

Continuing this rather eclectic disclosure of the regularities of motion of dislocations,


let us deal the solution of Eqs. (3.783.79). It is applicable to the swift motion of dis-
locations and is a mathematical of the equations of the special theory relativity (STR).
The contribution of the parameter into relations (3.78 3.79) is determined by the
Mach number (3.22, 3.23) which we believe should increase when the structural
damage of materials should augment due to the friction. It is caused by a consider-
able reduction of the elasticity of the surface and deep layers of the tribocouple parts
[6, 7]. The scale of these changes can by judged by the results of model experiments
when studying the elastic constants of bonds with the help of the secondary effects
produced by friction. The example of this effect is the thermal effect on the tribo-
material the strength characteristics of which are considered in Sect. 3.5. The data
mentioned in this chapter prove that the plasticity of materials grows and reaches
abnormally high values in friction [7]. In accordance with this fact and based on
Eqs. (3.22, 3.23), the Mach number should be expected to rise simultaneously. Let
us justify why the relativist effects in friction should be taken into account.
Activation of tribomaterials. The changes of the solid body structure in friction
result from the structural thermal activation of materials causing a considerable loss
of their elasticity. A critical value is reached at a definite moment due to the move-
ment of dislocations and changes in their concentration on the tribosurface a definite
moment reaches critical values, the material of the surface layer acquires a consid-
erable plasticity. The plastic deformation of this material p would be displacement
deformation (p = ). The rate of plastic deformation in shear is found with
formula (3.85). Let us add expression of the density of dislocations (3.74) into it:

4 2 2
= . (4.56)
G 2s b

The following can recorded from relations (3.46, 3.47):

D
= b. (4.57)
kT
Let us use the Hook equation = G s and transform (4.57) in the following
way:
D
= bG s . (4.58)
kT
Substitute the velocity of dislocations (4.58) into (4.56) to obtain the following:

4 2 2 D
= . (4.59)
G s kT
4.4 Relativistic Effects in Friction 117

Differential equation (4.59) can be transformed into

d 4 2 2 D
= dt. (4.60)
G s kT

Its solution is the expression

Ws
= 0 exp . (4.61)
kT
This equation characterizes the process of shear deformation of the tribopart
surface layer as the thermoactivated process with the activation energy Ws equal to:

4 2 2
Ws = Dt. (4.62)
Gs

Substitute expression (3.43) of the diffusion length L D into formula (4.62), then

4 2 2 2
Ws = L D. (4.63)
Gs

The value Ws can be determined using work expression

Ws = F D L D . (4.64)

By equating Eqs. (4.63) and (4.64), the following is obtained:

4 2 2
FD = L D. (4.65)
Gs

The dependence FD (x D ) into formula (4.65) proves that the mechanical force
affecting the dislocations has a quasielastic nature. The stress is a relation between
the load N and the real contact area A F which, from the Bowden and Tabor equation
is equal to N /Y , then = Y . In this case, (4.65) is written in the following way:

4 2 Y2
FD = L D. (4.66)
Gs

From relations (3.46) and (3.47):

kT
FD = . (4.67)
D
By equating the right terms of Eqs. (4.66) and (4.67) and taking into account
formula (3.43), the following is obtained:
118 4 Friction Microdynamics

kT 4 2 Y2
= Dt,
D Gs
or
4 2 Y2  3
= D t. (4.68)
G s kT

In accordance with formula (3.62) the diffusion coefficient can be represented in


the following way:
b2
D = f rq . (4.69)
2
Substitute the value D from (4.69) into formula (4.68) and, remembering the
f rq 1/t [where is the value coincides by order with the number of jumps of
diffusing particles (3.61)], the following is obtained:

4 2 Y2 b3 1
= t 2,
kT G s
or
4 2 Y2 b3
= f rq . (4.70)
kT G s

Let us use the numerical values listed in [24] and relations (3.22) and (3.74),
calculate the Mach number M for some tribomaterials:

8 2 Y3 b2 f rq
M= = . (4.71)
c kT G 2s Gs

The M value is of the order of magnitude 0.1 for silver; hence, the order of
magnitude of the value used in Eqs. (3.76)(3.80) to increase the accuracy of
calculation of the motion of dislocations, is equal to 0.99.
Lorentz transformations. The correctness of application of amendment in
tribological calculation elates to the solution of the following problems:
1. Would expressions (3.75, 3.76) be the Lorentz transformations (H. Lorentz
Austrian physicist, author of the classic electron theory who developed the elec-
trodynamics of the moving media) with all resulting conclusions regarding the
motion of dislocations in tribomaterials?
2. The resemblance of Eqs. (3.75, 3.76) to the Lorentz transformations is formal.
Hence the question is if equations (3.77, 3.78) are truly devoid of the physical
sense or the STR effects appear in the tribosystem?
To answer let us consider the classic conclusion about the Lorentz transformations
which begins with the analysis of transformation of the Galileo coordinates [36]:
4.4 Relativistic Effects in Friction 119

x  = x t x = x  + t
or . (4.72)
t = t t = t

Summarization of these transformations with the account of the linear interrela-


tions between x  , t  and x, t yield the following:

x  = x + t,
t  = x + t, (4.73)

where , , , are the coefficients independent of x, x  , t, t  .


for Galileo transformations

= 1; = ; = 0; = 1, (4.74)

but in this case, relations x  /t and x/t are unequal. If the x/t is the speed of light,
then it seems different in different counting systems contradiction to the structure
main postulates STR.
Einstein tries to find the solution when

x x
=  = c ,
t t
x x + t xt + c +

= = x = , (4.75)
t x + t t + c +
c +
c = .
c +

Equation (4.75) transforms into the identity if:



=  1 = 


2
1 2 c 2 2
1 2

c c
. (4.76)
=  =  1



2
1 2
2
1 2
c c

This choice of the values , , , requires to transform the coordinates what is


called the Lorentz transformation;

x  + t  t  + cx2
x= , t= . (4.77)
1 c2 1 c2
2 2

If, by analogy with (3.76), the parameter is introduced into these equations, the
mathematical identity between (4.77) and (3.78) is apparent. The canonic Lorentz
transformations are obtained due to the identical speed of light in the vacuum in
120 4 Friction Microdynamics

accordance with the arbitrary counting system [20]. The rate of propagation of elastic
oscillations c during motion of dislocations is devoid of the fundamental property
of the seed of light constancy. Hence, the transformations of coordinates (3.76) are
including disregarding this fundamental experimental fact. It should be established
if the conclusions from the Lorentz transformations are applicable to the solid body
microdynamics.
German physicist G. Minkowsky, one of the founders of the STR, commented in
1908 the result obtained by Lorentz stated, According to Lorentz each moving body
should contract in the direction of its motion; if the speed ofmotion is equal to , then
the contraction should be proportional to the multiplier 1/ 1 c2 . This hypothesis
2

seems fantastic because the contraction of bodies should not be considered the result
of ephyr resistance or any such phenomena, it can be just considered as the gift from
above or as the phenomenon accompanying the motion proper.
Whence the following conclusion. The Lorentz transformations are incompatible
with the classic ideas of the space-time. According to Lorentz, the solid body with
the length l  , stating in the system of coordinates k  , acquires a new length moving
in respect to the k  with the velocity in the system k:

2
l = l  1 . (4.78)
c2

Relation (4.78) is called the Lorentz length contraction. Absolutely hard rows
serve in the classic physics to measure distances by marking the standard length
on the rods. This length stays constant when the rod occupies a new position and
remains in it in time. Namely these postulates of the common sense are rejected
by the Lorentz transformations. Analyzing the length changes (4.78), we conclude
that it is necessary to admit its full dependence on the measured body velocity.
Entertain wrote, while the size of the ball (hence, that of any other solid body
of any shape) does not change in response
 to motion along axes y and z, the size
along axis x contracts in respect to 1/ 1 c2 stronger as the grows. All moving
2

objects observed in the system at rest flatten at = c into flat shapes. This argument
becomes senseless for the velocities above the speed of light [37].
Relativism ideology. The metric of the relativistic space relates closely to the
notions time and simultaneity. Einstein said, We see that no absolute values
should be attributed to the notion of simultaneity. Two events observed concurrently
from one of the coordinate system are already not perceived as concurrent when
analyzed from the system moving in respect to the given system. He continues, If
the point A in the space where the clock is placed, the observer in the point A can
determine the time of events immediately close to A by watching the position of the
clock arms concurrently with these events. If there is clock in another point in the
space, then the events can be determined by the observer staying immediately close
to the B. but it is impossible to compare the event A with the event B unless further
assumptions are made; we have determined only the A-time and the B-time not
4.4 Relativistic Effects in Friction 121

the common A and B time. The latter can be done by introducing the definition
that the time the light needs to proceed from A to B the light needs to pass
from A to B is the time needed for the light to pass from A to B [37].
If two events in the coordinate system k  occur in one and the same spot treated
with the time interval t  , then these two events are separated in the system k with
the time interval t:
t 
t =  . (4.79)
1 c2
2

One of the experimental checks of this conclusion is to observe the elementary


particles, such as -mesons (mu-mesons) capable to decay spontaneously. The life
time of these particles at rest is t  = 2 mcs. The mu-mesons in the space rays moving
with the velocity 0.995 of the light speed fly without decaying about 6 km, in other
words, their life t is ten time more t  from the terrestrial observer viewpoint. So
far the assumptions of the theory relativity have been checked with the accuracy up
to 1014 with the earth clock of the limited accuracy [38].
The above-listed relativistic effects follow from the Lorentz transformation equa-
tion (4.76) which are identical to Eqs. (3.75, 3.76). That is why with some certainty it
can b assumed that the application of the equations of the Lorentz transformation of
coordinates is correct enough to describe the motion of dislocations. The probability
of manifestation of the relativistic effects during motion of dislocations in the solid
body crystalline structure confirms the postulate that all laws of nature are invariant
in all inertial systems of counting in respect to the Lorentz transformations [39]. In
other words, if the Lorentz transformations in the physical system provide a con-
siderable contribution, the relativistic special effects begin automatically to play
a considerable role in reduction of sizes, alteration of the space metric, retardation
of time and dependence of mass on velocity. It means that the tribological system,
irrespective of its own velocity, can evolve in accordance with the super exact laws
of the theory of relativity.
The nature of motion of dislocations relating to the process with relativistic special
effects has deeper roots than the mathematical identity of Eqs. (3.75 and 3.76) with
(4.77). The general theory of relativity differs from the Newton classic mechanics
mainly by the estimation of the geometrical properties of space and time. It s believed
in the Newton theory that the space is Euclidean and bodies can move curvilinear
only under the effect of external forces. The theory of relativity presume that the
space is non-Euclidean and particles can move spontaneously along the trajectories
coinciding with the shortest distance between two points at a given space curvature.
Since the gravitation as he main moving force of the universe dictates attraction of
material objects, the space and time curve where the adjacent geodetic lines (by the
terminology of the general theory of relativity) converge rather than diverge. The
gravitation of massive bodies deforms space and time (Fig. 4.7); their curved metric
affect the material particles moving in space and time [38].
The light beam due to this reason curves when passing close to the massive bodies.
The angle f its deviation from the straight line is determined by the gravitation mass
122 4 Friction Microdynamics

Fig. 4.7 Curved space-time metric under gravitation of massive bodies [38]

Fig. 4.8 Relativistic effect


of light beam deviation by
sun gravitation: 1 initial beam
direction, 2 beam direction
after deviation

of the bodies and that of the sun equals to = 1.75 . This deviation was registered
during the solar eclipse of 1919 (Fig. 4.8) [39].
The external similarity between the curved space and the energy of the crystalline
space is amazing (Fig. 2.12).
American physicist and theoretician and Nobel Prize winner R. Feunman wrote,
the nature tells us that time is similar to space. The time becomes the space: they
should be measured by the same units Russian physicist theoretician O. Arsenov
tells of the mutual correspondence between the fundamental length of the physi-
cal space (the Plank length is 1.6 1033 cm) and the time quantum [40]. Einstein
asserted in general that the pas, present and future are just an illusion, though obses-
sive [37]. In other words, the STR ejects the universality of the present moment of
time and states that the simultaneity is relative and the time directions are identical.
Feunmans proof of this statement is his hypothesis stating that the antiparticle or
positron is a usual electron moving back in time for a moment. The signs of electric
charges of these particles indicate exactly the directions in time to trace the path
through Minkovskii space-time. This assumption served grounds for the mathemat-
ical apparatus which Feunman developed for the spatial time interrelation of the
quantum mechanics for which he was awarded the Nobel Prize highlighting for
the fundamental contribution into the development of the quantum electrodynamics
that had profound consequences for physics of elementary particles [41].
Relativist assessment of phenomena in tribophysics. Like the Einsteinian
space-time, the crystalline structure is different from the Newton homogeneous and
isotropic space by the assigned directions and particular points creating the power
fields which influence the motion of material particles in the solid body volume.
Chapters II and III sow that the motion of material objects in the solid body (diffu-
sive flows, vacancies and dislocations) obey both the crystal metric and act on the
4.4 Relativistic Effects in Friction 123

crystalline lattice with power fields deforming it. It is apparent the electromagnetic
quantum nature of the forces acting within and on the surface of solid bodies is
not identical to the gravitation which governs the metric of the universal space and
time. The forces of gravitation are present in the crystals, but because the subjects
of motion and the structural components of the crystalline space are so negligible,
they are small versus the forces of cohesion and they are justifiably ignored in all
calculations.
If the analogy between the SRO objects and the motion of dislocations in friction
are true, then it is justifiable to introduce relativistic amendments into calculations
of the micromechanisms of friction, wear and lubrication and to take into account
the relativistic effects when constructing the models explaining some phenomena
observed during friction. For instance, the extremely high degree of mechanoacti-
vation of tribosurfaces [8] resulting, in particular, in the appearance of self excited
states which can be explained by the tribosurface absorption of the energy of dis-
locations possessing excessive energy. According to Einstein, if the body releases
energy E than its mass drops as E/c2 . It is immaterial if the energy borrowed from
the body transforms into other types so we proceed to a more general conclusion:
the body mass is the measure of the energy the body contains [37]. This statement
provides ground for relativistic verification of formula (3.81), determining the mass
of the dislocation at rest of the singular length.
So, the time flow of the experimentation differs from the local time during which
the moving objects exist and which retards according to the STR (4.79). The time is
included into many physical laws and it is quite natural that the time dimensionality
effects as the evolution of various processes. This effect is most often insignificant
and it is justifiable to ignore this effect in the description of processes. However, there
is a category of physical phenomena when it is expedient to take the phenomenon into
account. It can be assumed that this effect intensifies the extraordinary physical and
chemical activity of the tribosurface. The super excited state of the substance from the
external observer view last longer due to the delayed local time, like mu-mesons. That
is why the tribochemical reactions manage to complete a new chemical compounds
manage to appear which would be unlikely to appear otherwise [8, 11].
The reasons are apparently equivalent to the direction of the present, future and
past time leading to the paradoxes contradicting to the causal principle. That is why
the physics employ the notion time arrow. The direction of entropy grows points at
exactly this state as the next event in the pair of adjacent events not the preceding one.
Many contemporary physicists believe that the entropy is the time arrow turning the
time into a simple sequence of states. American physicist, the Nobel Prize winner,
L. Cooper elucidates this statement in the following way [42], The idea that the
system changes during any physical process from the more ordered to less ordered
state provides us with the time direction that was missing before. The equation of
motion in the Newtonian mechanics is reversible. One way direction of motion of
planets around their orbits is in no way better than the opposite Yet I is clear that
the identified direction exists among other more complex events: it is easier to park
the car than to leave the parking, it is easier to smash glass than to fix it, it is easier to
scatter the cubes ten to assemble them, it is easier to kill tan to revive a human being.
124 4 Friction Microdynamics

If the system is ordered, its entropy is small; there are many ways to upset its order.
If the system is fully disordered (the entropy is great), it is had to find at least one
way to restore it into the original ordered state. It is easy to break the vase into tiny
fragments: any out of thousands of combinations on the rug represents the broken
vase But to restore the vase from fragments, each fragment is to be placed into its
own exact position; still it had bearing in mind the random motion of thousands of
fragments to place them into the necessary position. The assigned time direction is
explained in this way because it is hard to transfer the physical system he assigned
only possible state.
The fundamental theory of thermodynamics of open systems envisages the
possibility of reducing the entropy which we cursorily touched in Sect. (4.2). Equa-
tions (4.20)(4.23) confirm that it is possible to avoid the entropy growth by exchang-
ing the energy of the thermodynamic system with the environment. The entropy is
reduced more considerably during mass transfer processes. The more intensive the
mass transfer the less the entropy grows, and, following the logic of L. Cooper, the
more apparent the relativistic possibility of time retardation.
These abstract reasons at the first look were confirmed when S. Hawking
(S. HawkingBritish physicist) developed the thermodynamics black holes. When
the black hole interacts with the outer space, the entropy growth is zeroed in it and
around [43] virtually corresponding to the stop of time. It explains the observed fact
of the abnormal motion slowing of duct clouds around the assumed black holes.
O. Arsenov writes in this respect that the noose on the black hole cannot be tightened
into a single point. The thing is that the frozen star has so great field of gravitation
that the time nearby practically stops [40].
Thus, the regularities of time retardation during friction interaction between solid
bodies is determined b two fundamental theories: the special theory of relativity and
the thermodynamics of open systems. Te region within which these effects should be
most evident ate surface layers involved in friction. The tribologist would set them
into a particular category obeying specific laws due to the particular conditions of
their existence [3].
In fact, the dislocations accelerate in friction as much as they approach to the
tribosurface (see Sect. 4.3). It means that the Mach number of the dislocations and the
effect of the relativistic amendment effect (3.78) on the space-time parameters of
their motion intensifies. Correspondingly does the effect of friction of the relativistic
special effects, like the metric changes, time retardation and others. The surface
layers of solid bodies involved in friction exchange the energy actively with the
environment, so they become a classic example of the open thermodynamic systems
in which it is most probable to reduce the entropy growth. Based on the above said
let us consider the specific features of the processes evolving in the tribocontact.
4.4 Relativistic Effects in Friction 125

References

1. D.N Lyubimov, V.A. Ryzhikov, Fundamentals of friction theory: manual: Novochherkask:


South Russ, State Eng. Univ. 87 (2001)
2. A.V. Chichinadze (ed.), Fundamentals of Tribology (Science and Technology, Moscow 1995),
p. 778
3. Kh.X. Chikhos, System Analysis in Tribonics (Mir, Moscow, 1982), p. 348
4. I.B. Kragelskii, N.M. Mikhin, Nodes of Friction in Machinery (Mashinostroyenie, Moscow,
1984), p. 280
5. A.S. Akhmatov, Molecular Physics of Boundary Friction (Phys.-Math. Literature, Moscow,
1963), p. 472
6. N.M. Alekseyev, I.B. Kragelskii, A.M. Shapiro, Particular features of plastic flow of metals in
friction and their effect on friction and wear. J. Frict. Wear 4(3), 389398 (1983)
7. V.E. Roubtsov, A.V. Koloubayev, Effect of thermal emission within surface layer on plastic
deformation condititions in friction sliding. Modeling results, in POLYCOMTRIB-2009: Pro-
ceedings of International Science Conference, Gomel, IMMC Nat. Bel. Acad. of Sci., 2009,
p. 196
8. V.I. Kostetsky, Surface Strength of Materials in Friction (Tekhnika, Kiev, 1976), p. 283
9. D.N. Lyubimov, K.N. Dolgopolov, Modern Tribology (Selected Chapters), (IP Bouryhin V.M.,
Shakhty, 2010), p. 181
10. G.S. Pisarenko (ed.), Strength of Materials (Vyshcha Shkola, Kiev, 1986), p. 775
11. D. Bakley, Surface Phenomena During Adhesion and Friction Interaction (Mashinostroyenie,
Moscow, 1986), p. 360
12. D.N. Lyubimov, V.A. Ryzhikov, Phys. Chem. Processes in Friction: Manual. Novochherkask:
South Russ, State Eng. Univ. 147 (2006)
13. V.A. Potekha, Tribodilatometry (Gomel State University, Gomel, 2000), p. 374
14. D.N. Lyubimov, K.N. Dolgopolov, Difusive Processes in Friction (IP Bouryhin V.M., Shakhty,
2010), p. 148
15. V.A. Potekha, Study of antifriction properties of coatings of thermoplastic polymers metals.
Candidate of Science Disseration. Rostov-on-Don, 1982, p. 26
16. D.N. Lyubimov, Role of Diffusin Processes in Wear and Friction of Metal-Polymer Tribocou-
ples, Candidate of Science Disseration. Gomel, 1990, p. 22
17. L.I. Bershadskii, Interaction between structural mechanism and dissipative flows in kinetic
(non-Coulomb) wear and friction. J. Frict. Wear 10(2), 358364 (1989)
18. L.I. Bershadskii, Self-organization and concept of wearlessness of tribosystems. J. Frict. Wear
13(6), 10761094 (1992)
19. F. Glensdorf, I. Prigozhin, Thermodynamic Theory of Structure: Stability and Fluctuation (Mir,
Moscow, 1973), p. 280
20. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 2 (Sov. Encyclopedia, Mosscow, 1990),
p. 703c
21. G. Poltzer, V. Ebeling, External friction of solid bodies, dissipative structures, self-organization,
in Durability of Machinery Rubbing Parts, Issue 3. (Mashinostroyenie, Moscow, 1988), pp. 89
95
22. D.V. Sivoukhin, General Course of Physics, vol. 2 (Nauka, Moscow, 1975), p. 522
23. Y.K. Mashkov, L.N. Potseloueva, Structural energy mechanism of development of metal-
polymer tribosystems, in Durability of Machinery Rubbing Parts, Issue 3 (Mashinostroyenie,
Moscow, 1988), pp. 2835
24. S.A. Polyakov, Theoretical analysis of main mechanisms of evolution of tribosystems dur-
ing selective transfer. in Durability of Machinery Rubbing Parts, Issue 3 (Mashinostroyenie,
Moscow, 1988), pp. 327
25. I.B. Kragelskii, Friction and Wear of Machinery (Mashinostroyenie, Moscow, 1968), p. 480
26. D.N. Garkounov, I.V. Kragelskii, A.A. Polyakov, Selective. Wear in Tribounits (Transport,
Moscow, 1969), p. 104
126 4 Friction Microdynamics

27. Y.D. Eshelby, Appl. Phys. 25, 255 (1954)


28. P.V. Sysoev, M.M. Bliznets, A.K. Pogosyan, Atifrction epoxy Composite in Mahine Tool Build-
ing (Science and Technology, Minsk, 1990), p. 231
29. D.N. Lyubimov, A.E. Ivanov, V.V. Shakhbazian et al., Kinetics of formation and structures of
lubricating layers during selective transfer in metal-polymer tribocouples. J. Frict. Wear 13(3),
496500 (1992)
30. D.N. Garkounov, G.I. Souranov, Y.A. Khroustalev, Triboengineering. Hydrogen Wear of
Machine Parts. Ukhta, 2007, p. 260
31. V.I. Kolesnikov, Thermophysical in Processes in Metal-Polymer Triobosystems (Nauka,
Moscow, 2003), p. 279
32. D.N. Lyubimov, V.V. Savanchouk, V.A. Ryzhikov, Diffusion and vaction processes of hydro-
genetation in tribounits, in: Roblems of Synergetics in Tribology, Triboelectrochemistry, Mate-
rials Science and Mechanotronics, (South Russia State Technical University, Novocherkassk,
2003), pp. 69
33. A.L. Zharin, D.A. Rigney, Application of the contact potential difference technique for on-line
rubbing surface monitoring (review). Tribol. Lett. 4, 205213 (1998)
34. G. Gottstein, Physical and Chemical Principles of Science of Materials (Binom, Moscow,
2009), p. 375
35. V.P. Alekhin, Physics of Strength and Plasticity of Surface Layers of Materials (Nauka, Moscow,
1983), p. 280
36. L. Cooper, Modern Physics (Mir, Moscow 1974), p. 382
37. A. Einstein, Collection of Scientific Works (Nauka, Moscow, 1965), p. 517
38. S. Hawking, R. Penrose, Natue of Space and Time (Amfora, Saint Petersburg, 2007), p. 186
39. A.A. Barannikov, A.V. Firsov, Main Concepts of Modern Physics (Vysshaya Shkola, Moscow,
2009), p. 349
40. O.O. Arsenov, Physics of Time (Eksmo, Moscow, 2010), p. 224
41. V. Cholakov, Nobel Prizes. Scientists and Discoveries (Mir, Moscow, 1986), p. 368
42. L. Cooper, Classic Physics (Mir, Mosscow, 1973), p. 479
43. S. Hawking, Brief History of Time (Amfora, Saint Petersburg, 2010), p. 231
Chapter 5
Plasma Triboprocesses

I discovered something promising, too much more interesting


than the phenomenon which I intended to study
British physicist H. Wilson
This substance should not be considered as the hypothetic
fluid . . .
British physicist D. Maxwell
. . . it is a very fruitful subject, exactly for me, and I have not
managed far from all I intended
American writer S. King

Abstract In this chapter notions are reflected on mechanoactivation of tribosurfaces


substance and its junction to superexcited statetriboplasma. Main parameters are
represented which characterize triboplasma as well as approaches and physical mod-
els using for its descriptiondiffusional, magnetohydraulical, wave. Influence of
outer electromagnetic fields to triboplasma is discussed. Experimental setup for tri-
boplasva investigation is described.

The triborelativistic effects described in Chap. 4 govern a number of similar changes


in the physical and chemical structure of the tribosystem elements. The extreme
state of materials of these elements corresponds to their phase transformation into
the triboplasma. These effects are dealt with in the present chapter.

5.1 Mechanoactivation of Surfaces of Tribocouples

B.I. Kosteskii was the first to perform a profound physical analysis and advanced a
correct model of mechanoactivation of solid bodies [1]. The heat of friction deter-
mines the thermal activation of materials of tribocouples, which is accompanied by
nucleation, and exit of dislocations to the tribosurface. When dislocations saturate the
surface layers of tribopieces, the potential barriers, hence, energy of activation reduce

D. Lyubimov et al., Micromechanisms of Friction and Wear, 127


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8_5,
Springer-Verlag Berlin Heidelberg 2013
128 5 Plasma Triboprocesses

Fig. 5.1 Exit to surface of collapsed vacation disks

either in respect to the physical and chemical processes evolving in the materials.
The climbing of dislocations is believed the most typical surface structural tribo-
process [2]. It results from the phenomena in the Kottrell atmosphere. This process
leads to the growth of dislocation loops, which create fascinating structures like the
vacation disks on the tribosurface. Figure 5.1 shows the photographs of the disks on
the polycrystal tribosurface made with the scanning electron microscope.
The energy reduction of the barriers for sliding and climbing of dislocations
resulting from mechanothermal activation of the surface layer on the tribosurface
is similar to the weakening of bonds between structural elements of tribomaterial.
Acceleration of dislocations when approaching to the tribosurface proves indirectly
the changes in properties of the surface layer (4.3) indicating sparser dislocation
density. The metals with polycrystalline structure manifest the so-called metallurgi-
cal effects due the activating triboeffects [3]. They lead to the change of grain size of
crystallites, relaxation of stresses in friction, growth of the surface of the intercrys-
tallite boundaries, increase of the adhesion component of the force of friction. The
grains appear reducing the metal resistance to shear deformation.
Figure 5.2 shows the findings of the study of the effect of intergrain boundaries
on the copper friction characteristics. The polycrystalline copper indenter moved
in the direction perpendicular to the boundaries between grains on copper bicrystal
consisting of a pair of grains with the crystallographic orientation (111) and (210).
Since the crystallographic plane (210) possesses larger energy surface than the plane
(111), the friction coefficient grows in leaps when crossing the boundary through
the grain section. The discovered effect appears when the indenter moves in both
directions. The boundaries between grains of crystallites are characterized by the
specific friction coefficient value.
5.1 Mechanoactivation of Surfaces of Tribocouples 129

Fig. 5.2 Friction coefficient change during indenter transit through boundaries between copper
grains with crystalline orientation: a from grain (210) to grain (111), b in reverse direction [3]

According to B.I. Kosteskii, the friction heat makes up to 99 % of the total


volumetric energy which friction forces pump up on the tribosurface. Hence, in
accordance with formula (4.8), the surface stores a part of friction energy. This
energy obtained the name of deformation component of friction energy:

= 0.01qhq
W = 0.01E eq f a H (T ),
or, from (4.12): (5.1)
W 0.02 f Y (1 + ) G 0s eAT .

where, in accordance with the designation in formulas (4.9)(4.12): qhq is the heat
quantity passing through the contact spot; E eq is the thermal equivalent of the
mechanical work; f is the friction coefficient; a is the temperature conductivity;
H is the contact spot microhardness; is the Poisson coefficient; G 0s is the mod-
ulus of displacement at T = 0 K; BT is the empirical constant from relation 3.100;
T is the temperature.
Formula (5.1) has a rather simple mathematical form convenient for practical use
because all physical values in it are listed in the commonly accessible tables as well
as the experimental parameters having precise physical sense.
The energy accumulation process W is accompanied with formation of a spec-
tral structure on the tribosurface consisting of the crystalline structure surface defects
emerging on the surface. According to the terminology introduced by B.I. Kosteskii
it is named the ultradispersed microstructure, while the process of its forma-
tion is called the structural activation surface. The value of the stored energy
depends on the bonging energy released because of formation of the ultradispersed
microstructure. The division of the activation tribosurface into the structural and ther-
mal is conditional. The ultradispersed microstructure formation is determined by the
kinetics nucleation and movement of the crystalline structure defects towards the
tribosurface and relates to tribothermal activation of the tribomaterial. It is exactly
why B.I. Kosteskii introduced the term structural thermal tribosurface activation
into tribology.
130 5 Plasma Triboprocesses

The surface layer structure changes significantly because of absorption of the


deformation energy W . Though energy values W are small, it is accumulated
in the surface layers about tens of nanometers thick. The density of the absorbed
energy is equal to W/V (where V is the volumetric substances storing the
deformation energy); due to the smallness, the V can reach critical values for
the tribomaterial aggregate state. To find the exact value of the stored deformation
energy with the account of triborelativistic effects described in Sect. 4.4, the value
W determined with formula (5.1) should be multiplied by relativistic correction of
the 1/ , which is larger than a unity. Within the critical values of energy arranging
directly close other points of phase transition of tribomaterials, even its slight increase
causes considerable structural changes in substances. The extreme (critical) state of
the surface layer is the superexcited metastable state called triboplasma [4].

5.2 Triboplasma

The friction heat sources act as the actual contact spots called hot spots [4]. The
tribomaterial transition from the initial steady state into the triboplasma is similar to
the changes in substances during heating. As the temperature grows, the solid body
melts and then evaporates. If the temperature continues to grow, quite a large gas
atom ionizes. A new high-energy state or plasma appears with approximately equal
numbers of positive ions and electrons determining its quasineutrality.
Plasma physics. The term plasma is used in physics to designate the substance
quaternary aggregate state [5]. Plasma in nature has a variety of forms (Fig. 5.3). The
ionosphere exists at the altitude 70500 km; it has the concentration of particles about
1012 m3 , its thermal energy is 0.2 eV. The solar wind is the terrestrial plasma stream
with the parameters n 106 107 m3 and thermal energy kT 10 eV. The solar
crown surrounds the sun and it has plasma density n 1014 m3 and the thermal
energy 100 eV. The density of electrons in the white dwarf star is approximately
1035 m3 , the thermal energy is above 200 eV, etc. the intensive research in the sphere
of plasma physics originated in the considerable degree from the thermonuclear
synthesis problem to solve which it is necessary to create and arrest the hot (almost
stellar) plasma.
The plasma electrons and ions move together and interact through long-range
coulomb forces which diminish in accordance with he inversely squared distance
between particles. The displacement of charged particles induces electric currents
and the Lorentz forces affect magnetic fields thus the particles additionally. Various
collective processes of motion appear in the high-energy capacity system, which can
be exemplified by the wave phenomena [6].
The screening of the applied electric potential is one of the fundamental plasma
properties. If a probe with applied potential is introduced into the plasma, its
components are attracted or repulsed in response to the potential sign. Thus, the
plasma strives to screen off the electrostatic perturbation the potential sign induces.
The screening typical length can be estimated. Assume that the ion density in the
5.2 Triboplasma 131

Fig. 5.3 Various plasma types on nkT diagram [5]: n mean density of plasma particles, R D
Debye radius, W f Fermi energy

plasma is homogeneous n ions = n 0 and there is a weak perturbation of the electron


density n e of the potential . Since electrons follow the Boltzmann distribution so
their density is the following [5]:
   
e e
n e = n 0 exp n0 1 + , (5.2)
kTe kTe

where e is the electric charge; Te is the electron gas temperature.


Using the Poisson equation [7]:

or E = ,
(5.3)
(0 , E) = 0 2 = ch ,

where 0 is the electric constant; ch is the density of charges.


The charge density ch can be expressed through values n e and n i :

e2 n 0
ch = e(n e n 0 ) = , (5.4)
kTe


2 = ,
R 2D
 1  1
kTe 2 1 kTe 2
RD = = 7.45 10 3
. (5.5)
n e e2 ne e

The solution of Eq. (5.3) is the following:


132 5 Plasma Triboprocesses
 
q exp R D
r
= , (5.6)
4 0 r

where q is the electric charge; r is the current radius.


It follows from formula (5.6) that the point charge coulomb potential is screened
off due to the cloud of charges with the opposite sign accumulated around it at a
distance R D called the Debye length or the Debye radius. In fact, the R D is the
dimension of the area within which the substance is plasma. It can be written in the
following form somewhat different from (5.5):
 1
kT 2
RD = . (5.7)
4 ne2

where T is the plasma temperature; n is the mean density of particles in plasma [8].
If the initial gas volume in which plasma forms exceeds 4/3 R 3D this plasma is
considered quasilinear, if V pl < 4/3 R 3D , then the plasma is a set of independent
charged particles. The number of electrons in the Debye sphere is equal to
 3
kT 2 1
n R 3D = n 2. (5.8)
e2 e

Relation (5.8) in the theory of plasma obtained the name of the parameter of
perfection. If the plasma density grows while the temperature remains constant, the
parameter of perfection reduces.
The collective motion of charged particles reflects another fundamental plasma
property. The coherent motion of charged particles relates to the appearance of waves.
If the wave phase velocity ph exceeds slightly the thermal velocity T of particles,
then the wave propagates through plasma without attenuation or amplification. When
the values ph and T are comparable, the energy exchange between waves plasma
particles is probable. L.D. Landau discovered the specific mechanism of wave atten-
uation. Landau implied by attenuation the interaction between waves and particles,
which does require collision of plasma particles.
Triboplasma specific features. The state of the matter named triboplasma
is postulated in [4, 9, 10]. Nonetheless, the triboplasma is the fact discovered
experimentally rather than a hypothetical something. Its existence is proven by
the mechanical emission of charged particles from the tribosurface affecting the
electromagnetic fields (the field effect) in the tribocouple and thus its parameters
[11], the abnormal intensification of the chemical activity of tribomaterials [4] and
many others. The triboplasma implies in general the energy clots, which transform
the substance surface layer into the tribocontact. The triboplasma is unstable like
other plasma types and it has short life time about 108 s. Many mechanisms of
tribochemical transformations are written off to origination and evolution. For
instance, the processes of tribocoordination are described in this way; these processes
5.2 Triboplasma 133

lead to the appearance of lubricating structures like surfing films [10, 12]. The tri-
boplasma duration and the tribounit closedness to analytic instruments impede the
triboplasma research.
An effort to study the substance plasma state by measuring diffusion constants
with radioactive indicators is made in [13]. To this end, the distribution of radioac-
tive elements was investigated in time through the specimen volume. The obtained
diffusion coefficients were compared with the values established by measuring the
ion conductivity. It is found that the results obtained with both methods correlate
well thin the experiment accuracy range.
The method is advanced in [14] to determine the diffusion layer thickness on
the tribosurface with the unique method of X-ray spectral fluorescent analysis [15].
The measurement results served to find the diffusion coefficient with the help of
relation (3.43).
Triboplasma model. Let us assume that the diffusion coefficient De on the surface
layer of the abradable solid body can be determined experimentally. Assume that the
Debye length calculated with formula (5.7) governs the triboplasma parameters. It is
apparent that friction is unable to ionize the substance completely as if it occurs on
the stars or during the nuclear explosion. Hence, the triboplasma should be referred
to the so-called imperfect plasma for which Eq. (5.8) is he following [16, 17]:

n R 3D = 1. (5.9)

The diffusion process in friction is accompanied with constant changes of the


abradable surface layer structure, which occur in the solid bodies with high activa-
tion energy. Thee mechanoactivation of this layer with friction forces and the heat
they generate transforms the tribomaterial initial structure into the ultradispersed
microstructure and then into the triboplasma. Simultaneously the diffusion coeffi-
cient in the surface layer increases by several orders of magnitude. If the friction is
suspended, the reverse process occurs and the substance returns more or less fully to
the initial state.
One of the main postulates of the diffusion theory states that the diffusion
processes evolve without any volumetric changes and are quasiequilibrium. That
is why the diffusion coefficient De obtained experimentally is the value that can be
found from the mean value theorem [17]:

L
1
De = f (x)d x, (5.10)
L
0

where L is the tribolayer of the tribocouplethe diffusion length; f (x) is the function
of the change of the diffusion coefficient.
The substance active states originating in friction are characterized by the maxi-
mum diffusion coefficient D0 rendering the states unsteady. They attenuate quickly
and return to the solid body unexcited state. Hence, the friction should be accompa-
nied by the value D0 abrupt drop, the function f (x) is the following:
134 5 Plasma Triboprocesses

f (x) = D0 exp( D x). (5.11)

The substitution of the value f (x) from formula (5.11) into (5.10) yields

D0 
De = (1 e D L ). (5.12)
D L

D0 here first corresponds to the value D in the actual contact spots, second, it is
the diffusion coefficient corresponding to the triboplasma; D is the logarithmic
attenuation decrement of the diffusion coefficient.
1 D0
D = ln , (5.13)
Ld DT

where DT is the diffusion coefficient corresponding to the initial unexcited solid


body phase of the tribocouple. Its values are listed in handbooks and tables.
Let us substitute from (5.13) into formula (5.12) and obtain the following:

D0
De ln = D0 D T . (5.14)
DT

Since the plasma diffusion coefficient D0 is much times larger than the value DT
(D0  DT ), then (5.14) can be rewritten as

D0 D0
ln = ,
DT De  
or (5.15)
D0
DT = D0 exp .
De

then, according to the Arrhenius law (3.63)

D0 W
= . (5.16)
De kT

Then, using the Eyring formula (3.65) in which the structural triboplasma factor
d is substituted with the Debye radius R D , while the exponent in relation (4.13) can
be expanded into the Maclaurin series due to the smallness of difference (W Q)
only the second expansion term m can be taken. After transformations, expression
[17, 18] yields

R 2D
D0 = (Q W ), (5.17)
h

where Q = kT + Q is the value characterizing the kinetic energy of thermal


movement of triboplasma particles; T is the triboplasma temperature; Q is heat
generated by friction.
5.2 Triboplasma 135

The tribosurface is the structure oversaturated with crystalline structure defects


and weakened cohesive bonds additionally activated by friction forces. Whence the
following quite correct assumption (from the solid body physics view) can be made
at which Q  W , as a result, expression (5.17) can be simplified to

R 2D
D0 = Q . (5.18)
h
Let us return to relation of perfection parameter (5.9) from which the following
follows
1
R 2D = . (5.19)
n RD

Let us substitute the value R 2D from (5.19) into formula (5.18) and obtain the
following:

Q
D0 = , (5.20)
n h RD

whence we find the Debye length:

Q
RD = . (5.21)
h n D0

The substitution of (5.21) into (5.7) yields



2Q e
D0 = .
h nkT

By combining the constants into a separate factor, rewrite the relation for D0 :

2e Q
D0 = . (5.22)
h k nT

2e
Let us designate the constant h k made up by the fundamental physical con-
stants with a1:

2e
a1 = , (5.23)
h k

then

Q
D0 = a1 . (5.24)
nT
136 5 Plasma Triboprocesses

The substitution of (5.24) into formula (5.18) yields



Q T
De = a1 k . (5.25)
W n

Following the logic of the performed transformations, let us introduce one more
constant and designate it through a2:

2e
a2 = a1 k = k. (5.26)
h
After that relation (5.25) can be written as

i
De = a2 T. (5.27)
n

The value i is equal to the relation QW , which characterizes the degree of substance
ionization and shows how much the substance triboplasma energy is more than that
of the substance in the steady state. It assumed in [19] that the physical values i and
the performance of the tribochemical reactions are close; a quire essential parameter
but extremely hard to estimate. The performance of the tribochemical reactions (x )
determines the portion of energy of the mechanical effect (friction) expended in the
tribosystem to initiate chemical reactions. It is reported in [10] that the performance
factor ranges from 0.01 to 0.05. The calculation of the value i for metal-polymer
tribocouples yielded the magnitude 103 allowing assuming that the values i and
x are close. Thus, it can be conclude that the kinetics of tribochemical reactions
are determined by relaxation processes in the triboplasma; the application of the
energy diffusion approach apparatus to the description of triboprocesses make the
calculation simple and accurate and check the parameters hardly accountable so far.
Relation (5.27) establishes the law of proportionality of the diffusion coefficient
and the square root in response to the temperature what is typical for the substances
in the gaseous state, that the triboplasma has the properties to gases. It enables
to simplify consideration of relaxation processes by applying the apparatus of the
kinetic plasma theory and the wave physics.
Triboplasma parameters. The mentioned approach enables to resolve the para-
dox of modern tribology. It is based on the tribosurface ability to emit photons and
charged particles (electrons and protons) with the energy 15 eV remaining incon-
siderably heated. Though the tribocouple maximum temperature does not exceed
400500 C, much higher temperature localizes on contact spots existing shortly
and moving with the collinear velocity [20]. The emission with spot of electron
energy of 1 eV is equivalent to the temperature reduction by 1.1 104 K [5].
An effort is made [17] to assess the triboplasma temperature. The substitution of
the value R D from (5.7) and Eq. (4.8) into formula (5.18) yields

(kT )2 + A a f H W kT 4 e2 n h De = 0. (5.28)
5.2 Triboplasma 137

Equation (5.28) permits to find temperature T corresponding to the plasma tempera-


ture. Equation (5.28) is much simplified with the account that H is the tribomaterial
microhardness and, since it has no sense for plasma, it is assumed 0. In addition,
the last summand in Eq. (5.28) is incommensurably small versus other terms in the
equation. As a result, the temperature becomes close to the relation Wk . The plasma
energy with the introduced approximation of the value W is close to the energy of
Van der Waals interaction of the order 1019 J. Hence, the triboplasma temperature
corresponds to 104 K and well correlates with the energy of emitted electrons.
Thus, the triboplasma is a gaseous medium localized on the tribosurface and it
consists of ions and electrons. The triboplasma composition was analyzed theoreti-
cally in [2, 21] to gain insight into the nature of particles it contains.
Let us use Einstein equation (3.47) as follows:

k B  T = q D0 , (5.29)

where B  is the ion motility; q is the plasma charged particles.


Let us rewrite (5.29) in the following way

B D0
= . (5.30)
q kT

According to the molecular kinetic theory of electricity [22], the motility of


charges B  can be calculated as

q c
B  = 0.5 , (5.31)
m
where c is the free run length; is the mean velocity of particles.
For plasma in expression (5.31), let us substitute c with R D :

q RD
B  = 0.5 . (5.32)
m

The relation RD is a fragment of time t p during which the substance particles are
with the Debye radius zone; it is similar to being in the plasma state. The comparison
of Eqs. (5.30) and (5.31) yields

D0 0.5t p
= ,
kT m
or (5.33)
2D0 tp
= .
kT m
To determine the time t p let us use the relation of the diffusion length, which for
the triboplasma transforms into
138 5 Plasma Triboprocesses

RD = 2D0 t. (5.34)

With the account of relation (5.9), it can be recoded that

0.5
tp = 2
, (5.35)
D0 n 3

or, with the account of the relativistic amendment,

tp 0.5
t p = = . (5.36)
2
D0 n 3

Similarly, the triboplasma ion mass can be estimated:

0.25kT
m= 2
, (5.37)
D0 n 3

or, with the account of the relativistic amendment,

m 0.25kT
m = = . (5.38)
2
D0 n 3

The ion charge is


4D0 m
q= , (5.39)
t
where is the contact difference of potentials in the friction zone or the triboelec-
tromotive force.
The free charges q of the triboplasma particles oblige to consider the diffusion
flow in plasma as the electric current with its own magnetic field. It induces additional
forces, which act on the ionized particles. Directed diffusion of ions and electrons in
the gas consisting of neutral particles takes place in the weakly ionized plasma. The
particles in the fully ionized plasma move along the magnetic field lines. Hence, the
diffusion of particles in the plasma magnetic field is sharply anisotropic. It is caused
by different values of the Lorentz forces affecting the diffusion components making
the longitudinal and transverse diffusion coefficients different (Sect. 3.3).
Magnetohydrodynamic approach. The motion of diffusion flows through
triboplasma can be described with the equation of magnetic hydrodynamics [5, 22]:
(a) The equation of continuity describes the transfer of particles


r , t) + n (

n(

r , t) = 0,
r , t) (
(5.40)
t
5.2 Triboplasma 139

(b) The flow of particles or diffusion flow

J
= n ,

(5.41)

is determined by the equation, which is in fact the first Fick law:

r , t)
(

n (
r , t) = D (

r , t) n (

r , t) . (5.42)

The second Fick law


 
Dn (
r , t) ,
r , t) = n (
(5.43)
t
combines the diffusion coefficient D with time tp named in the plasma physics the
particle containment time

  1
Dn (
r) .
r , t) = n (
(5.44)
tp

The solution of (5.44) for the cylindrically shaped plasma with the radius r and
the constant D is the following:
   
2.4x t
n = n 0 I0 exp , (5.45)
r tp

where I0 is the Bassel function of the zero order (F. BasselGerman astronomer).
The particle containment time:

r2
tp = . (5.46)
5.8D
Relation (5.46) between the particle containment time, the diffusion coefficient
and the plasma flow radius r is actually general. The digital factor only changes
insignificantly in response to the plasma geometry.
The balance between energy and plasma corresponds to the equation derived
in [5]:
   
3 3 Vi
nkT + nkT
+ J Q = Q p
 i j . (5.47)
t 2 2 i, j x j

where
is the velocity of particles in the diffusion flow; i j is the tensor of the
diffusion flow density.
The first term in the right part of equation (5.47) is the heat Q generated per unit
of time due to the collision of particles; the second term is the work of pressure p,
the third term is the heating of the plasma substance by the internal friction forces.
140 5 Plasma Triboprocesses

The left part of the first term contains the derivative of the thermal energy in time
(the power) and its generation per volumetric unit, and the second part describes
the energy losses by convection, the third part corresponds to the energy due to the
plasma heat conductivity p . Hence, the thermal flow relates to the heat conductivity
trough the known relation [23]:

J Q = p (kT ) . (5.48)

When the convection losses are negligibly small in plasma, the heat generation in
the right part of the Eq. (5.47) equal zero:
 
3
nkT p (kT ) = 0,
t 2
   
3 p
If n = const, then kT = (kT ) . (5.49)
t 2 n

The value p /n is the temperature conductivity coefficient a (4.8). Hence, the
solution at a = const is the following:
   
2.4 t
T = T0 I0 x exp , (5.50)
r tE
r
tE = .
3.87a
The parameter t E is called the energy of particle containment time and is an
essential characteristic of mutual collisions of particles in the plasma flow.
The magnetohydrodynamic description of the plasma processes yields particularly
good results when the number of collisions between particles is large and when their
free run is short [5], i.e. in the case of the utmost suitable triboplasma description.
Assume ei designates the frequency of collisions between electrons and ions,
Te designates the thermal velocity of electrons, then

l
ei T e , (5.51)
2 R
where l is the free run length of the plasma particles; R is the radius of the power
typical portion.
If the Ohm law is adapted to the plasma hydrodynamics, then in accordance
with [22], the following relation can be written:
1
E
+
B
pi = J
, (5.52)
en

where E
and B
are the electric field intensity and the vector of the magnetic induction
of the field, respectively; pi is the gradient of pressure of the flow J
; is the plasma
5.2 Triboplasma 141

viscosity = m e ei e2 n 1
e ; e is the electron charge; n is the density of electrons;

is the velocity of electrons.


Then the speed of plasma in electromagnetic field is found from the equation
  
1
kTi m e ei p
n = nE n e
B
B e e2 B 2
    
1
kTi Ti
= nE n e
B r L ei 1 +
2
n, (5.53)
B e Te

where B,
E
are the vectors of induction and intensity of the magnetic and electric

the indexes e and
fields, respectively; e
B is the singular vector coinciding with B,
i relate to electron and ion, respectively; r L is the Larmor electron radius; k is the
Boltzmann constant.
While the first term in the right part of Eq. (5.53) an be ignored, the diffusion
coefficient of plasma particles D is in accordance with Fick law equal to
 
Ti
D = r L2 ei 1 + . (5.54)
Te

Yet, the first term of equation (5.53) is not always negligible little. For instance,
when the plasma is configured horizontally due to the different conductivity, the
separation of charges because of the drift is not fully eliminated by displacement of
charges along magnetic power lines, so the

electric
field E due to induction processes.

In accordance with Eq. (4.17), the term E e


B adds to diffusion.
The electric field induce the collective motion of particles favoring the appearance
of additional abnormal [5] energy and plasma losses.
Diffusion of particles in plasma. Assume that the plasma density n (
r , t) consists
r , t) and the term of the first order of smallness n k .
of the term n 0 (

n k (
r
k t) ,
r , t) = n k exp i(
(5.55)

where n k is the amplitude of oscillations of density of plasma particles;


is the wave
vector; i is the imaginary unity.
r , t) represents some plane wave with the vector

In other words, the value n k(

and the frequency k . Respectively, the total number of particles n per volume unit
is the following: 
r , t) = n 0 (

n(
r , t) + n k (

r , t). (5.56)
k

The perturbations n k induced by additional displacement of plasma particles. The


velocity of particles is
 

r , t) =
(
k = k exp i(

r
k t) ,
k k
142 5 Plasma Triboprocesses

and the equation of continuity of this flow is the following

n  
+ n
= 0. (5.57)
t
The frequency k is a complex value expressed by the relation

k = kr + ik . (5.58)

where kr is the real part of the complex frequency k ; k is the imaginary part k .
With the account of (5.565.58) can write:

n 0  n k  
+ + n 0 k + n k k  = 0. (5.59)
t t 
k k k,k

Separation of the terms of the first and second orders yields


 n k 
+ n 0 k = 0, (5.60)
t
k k

n 0 
+ n k k  = 0.
t
k,k 

The averaging of Eq. (5.60) in time with the account of the complex nature of the
parametrically included frequency yields

k |n k |2 + n 0 Re(n k k ) + n 0 Im(n k k ) = 0,
kr |n k |2 + n 0 Im(n k k ) n 0 Re(n k k ) = 0. (5.61)

If second Eq. (5.60) is averaged in time, obtain


 
n 0 
+ Re(n k k ) exp(2k t) = 0. (5.62)
t
k

Then the diffusion is described by equation

n 0
= ( n 0 ) = 0, (5.63)
t
and the flow of outgoing particles

J = Dn 0 = Re (n k k ) exp (2k t). (5.64)
k
5.2 Triboplasma 143

n 0 Im(n k k )
Introduce the parameter k : k = .
n 0 (Re(n k k ))
Then Eq. (5.64) becomes

k |n k |2 exp(2k t)
k
D |n 0 | =
2
,
1 + k

where the diffusion coefficient D is equal

 |n k |2
D= k . (5.65)
k
|n 0 |2 (1 + k )

The diffusion coefficient D is added into the Eq. (5.65) based on the analysis of
fluctuation losses.
Let us consider the case when the electric field E k of fluctuations of plasma
frequency has the electrostatic nature, hence, it can be expressed through the potential
gradient k in the plane wave approximation

E k = k . (5.66)

This field makes the plasma particles drift with the velocity k :

 
E
k B
i
e
B k
k = = . (5.67)
B2 B

where e
B is the singular vector with direction coinciding the magnetic field intensity.
This expression and the equation after it describe the fluctuation component of the
plasma particles normal to their flow. The substitution of (5.67) into first Eq. (5.60)
yields
 
e
B
k
n k = n 0 . (5.68)
B k

If the axis z is combined the direction e


B , the axis x with

n 0 , then, with the


account of orthogonality of the vector e
B and
n0

n 0 = pl n 0 x
, (5.69)

where x
is the singular vector, coinciding with the abscissa axis.
Then relation (5.68) can be transformed in the following way

n k pl y ek
= k = k , (5.70)
n0 B k k kTe
144 5 Plasma Triboprocesses

where k y is the y-projection of the wave vector; k pl is the value inverse to the change
in plasma density.
The value
kTe
k = y pl (5.71)
eB
is called the drift frequency.
If the complex frequency k = rk + ik is represented in the following way
k Ak exp (ik ) [5], the drift velocity of the particles is
kT n
k = i
l
B
e k
Ak exp(ik ). (5.72)
eB n 0

Then from (5.64) the equation of diffusion coefficient can obtained in the following
way in the following way:
 2
kTe  y  n k 
D= Ak sin k   . (5.73)
eB pl n0
k

The coefficient of abnormal diffusion in the tribocouple surface layer grows in time
due to the fluctuation losses. In the final account, the term k > 0 begins to dominate
in equation and (5.64) and has the maximum growth tempo. Yet, the amplitude |n k |
of wave oscillations of particles due to the non-linear effects the plasma tends to
the limit value, which is by order of magnitude, is ( pl /x )n0 . Bearing this in mind,
relation (5.65) transforms into
k
D= . (5.74)
x2

If the multiplier at kTe


eB in (5.73) acquires the maximum value equal to 1/16, the
so-called Bohm diffusion coefficient appears:
1 kTe
DB = . (5.75)
16 eB
The Bohm diffusion is the abnormally turbulent magnetized plasma transfer across
the magnetic field with the velocity exceeding considerably the classic diffusion
rate. Thus, the above transformations imply that the diffusion in plasma is sharply
anisotropic. A single leap of particles (step) along the magnetic field is equal to
the free run length; the longitudinal diffusion coefficient of electrons is much more
than that of the ions. The transverse diffusion of ions is much more than that of
electrons. The Bohm diffusion can induce the drift dissipative plasma instability due
to collisions between electrons and ions. The losses in plasma are due the motility
5.2 Triboplasma 145

of B and they are proportional to B2 , while the fluctuation losses are B1 . As the
magnetic field intensifies, the fluctuation losses decline gradually.
The anisotropy of diffusion coefficients leads to the appearance of the self-
coordinated electric field in plasma mentioned above. It is not always identical to the
electrostatic field and induces eddy currents. The latter accelerates the leveling of
the concentration of charged particles. The ambipolar diffusion is the joint transfer
of ions and electrons when their concentration is similar towards reduction of their
concentration. The diffusion in plasma is rarely ambipolar. The deviations occur due
to the spatial anisotropy and the diffusion coefficient of the diffusion flow compo-
nents. In the general case, the rule of equal divergence (the scalar field) of diffusion
flow of ions and electrons [18].
The gas plasma contains also the triboplasma in which the particles perform
longitudinal oscillations of definite frequency called Langmuir waves, in addition
to chaotic thermal motion of particles [5] (J. LangmuirAmerican physicist, the
Nobel Prize winner). The Langmuir waves propagate with the frequency p called
the plasma frequency:

4 ne2
p = , (5.76)
me

where n is the density of charged particles; e is the electron charge; m e is the electron
mass.
The separation of the value n from formula (5.76) and its substitution into
Eq. (5.27) yield 
4 e2 i
De = a2 T. (5.77)
m p

Introduce the new constant a3:


 
4 e2 4 e2 k
a3 = a2 = , (5.78)
m h m

and rewrite (5.77) in the simplified form:

i
De = a3 T, (5.79)
p

respectively:
i T
p = a3 . (5.80)
De

Thus, the equations of diffusive mass transfer together with the equations of mag-
netic hydrodynamics have yielded the relations permitting to assess the theoretically
the most essential triboplasma parameters, such as temperature, Debye radius, dif-
146 5 Plasma Triboprocesses

fusion coefficients and plasma frequency. Together with experimental data, these
equations provide an adequate physical model of the substance state.

5.3 Wave Processes in Triboplasma

The preceding section manifests that the triboplasma is a complex dynamic system of
mobile electrically compensated positive and negative charge. This physical system
is an unsteady fluctuation structure in which gradients of densities and temperatures
appear inducing various flows making the physical system to transform into the
steady state. It is identical for plasma to relax from the transformation and restore
the original aggregate state.
Therefore, let us substitute fluctuation Eqs. (5.58) and (5.57) into the equation of
continuity:
n
+ n + n = 0. (5.81)
t
Then let us linearize [22]:

n 0
i n + dr + i n 0 dr = 0, (5.82)
r

where i is the imaginary unity; is the wave vector; dr is the drift velocity of plasma
particles; is the effective (averaged) frequency.
Equation (5.82) is recorded in the approximation coincidence vector of the drift
velocity of the with the external magnetic field power lines. The magnetic field
separates the triboplasma, so in further description, it is necessary to switch over
from considering the plasma as a homogeneous electrically balanced substance to
its ion-electron structure. Since the density of electrons describes the Boltzmann
distribution, then
n e e
= . (5.83)
n0 kTe

The equation of motion of particles in the plasma is the following:

d e
ne m e = pe en . (5.84)
dt
Expression (5.84) can be rewritten in the complex form [24]:

i n e m e e = i( pe + en 0 ), (5.85)

where e is the electron charge; pe is the pulse of electrons in the perturbed state; e
is the drift velocity of electrons in the perturbed state.
5.3 Wave Processes in Triboplasma 147

If relation (5.83) is true in respect of ions, then, using the condition of quasilin-
n e n u
earity [24] = , the dispersion equation can be obtained:
n0 n0
 2   
e iT Te 5 e 2
1 + + qi = 0, (5.86)
/ Ti 3 3

(kTe ) dn e
e = ,
eBn e dr
d (ln Ti )
qi = ,
d (ln n i )

where e is the electron drift frequency; k is the Boltzmann constant; B is he magnetic


field induction.
Similarly, to (5.86), the ion drift frequency can be introduced and the parameter
q can be recorded for electrons. The obtained dispersion equations prove that the
triboplasma flows during wave processes.
The waves in plasma are propagating electromagnetic oscillations self-coordinated
with the collective motion of charged particles in plasma. There are many types of
plasma waves combined booth by the internal plasma states and external effects
of electric or magnetic fields. Dielectric properties are the most important factor
affecting the propagation of waves in plasma. The plasma dielectric permeability is
the second rank tensor i j the components of which relate to the values of plasma
frequencies.
If there is no external electromagnetic field, the longitudinal and transverse waves
exist in the charges in the system of the triboplasma components, similarly to any
cold plasma. In this case, the tensor i j changes into the scalar value. The longitudinal
oscillations in these systems are called the spatial charge or Langmuir waves (5.76).
When the thermal motion of plasma particles is taken into account, a thermal
mode of oscillations appears involving electrons and ions alike. These oscillations
are called ion-sound. Their law of dispersion is the following:
 
Ti 1
e = i 3 + ,
Te 1 + 2 R 2D

Te
i = , (5.87)
mi

where i is the ion velocity; R D is the Debye radius.


The external electromagnetic field alters the plasma wave properties: the number
of modes of number of oscillations augments, their polarization changers encum-
bering the division of waves longitudinal and transverse. The wave process becomes
so complicated that it is unfeasible to obtain the dispersion relation in the general
148 5 Plasma Triboprocesses

form. The cold plasma (triboplasma) may manifest five branches of oscillations in
the magnetic field [25]. During propagation of oscillations, there are one Langmuir
mode along the magnetic field and four transverse wave modes coexisting in different
frequency bands. The Alfvn waves in the low frequency band with the following
law of dispersion:

a
a = a 1 ,
cfr
H
a = , (5.88)
4 n i m i
eH
cfr = ,
m i c

where a is the Alfvn velocity; H is the magnetic field intensity; c is the speed of
light; cfr is the cyclotron frequency.
The Alfvn waves curve the magnetic field power lines. The external magnetic
field restores this distortion. In addition to the slow Alfvn waves, there can be fast
or magnetic sound waves.
Another example of the waves appearing in plasma during interactions with exter-
nal electromagnetic fields is electrostatic waves. The dispersion relation for this type
of plasma waves is often recorded through the dielectric permeability tensor compo-
nents:

x2 x x + 2x z x z + z2 zz = 0, (5.89)

where o , z is the projection of wave vector on the coordinate axis; x x , x z , zz is


the dielectric permeability tensor components.
The electrostatic wave phase velocity is small versus the Alfvn wave. Thus,
the electromagnetic field perturbation attenuates in the electrostatic wave just after
several periods of wave oscillations.
Other types of oscillations are observed in addition to the above listed types
of plasma waves such as super fast magnetic sound waves or whistlers and high
frequency common and uncommon waves [24].
When plasma waves interact, they dissipate due to the space-time resonance and
one wave can transform into another. The resonance conditions are:

i = 0,
i

i = 0, (5.90)
i

where i , i are the resulting frequency of the wave vectors between interacting
waves.
5.3 Wave Processes in Triboplasma 149

The dissipation of waves in plasma even at small amplitudes is an induced process;


it means the inverse bonds between original and dissipated waves should be taken
into account. This bonding induces the decay instability, which can boost the original
wave amplitude.
The dynamic processes in plasma can be considered as wave plasma induced by
the propagation of electromagnetic oscillations with different frequencies and phase
velocities. Such systems are called polychromatic in spectroscopy. Such system is
improper to represent in a number of cases as a combination of many oscillators with
their own frequencies. This set is replaced with a single oscillator having the same
effect as that of many oscillators reproducing their equivalent effect. This approxima-
tion is termed monochromatic and the relevant and equivalent frequency (wavelength,
the energy, etc.) is called effective or equivalent. This approach is broadly developed
in the X-ray spectral fluorescent analysis, which serves to produce and verify phys-
ically the monochromatic approximation [26]. The monochromatic approach was
used to triboplasma in [2] and yielded good correlation with the experiment.
Considering in the present section the description of triboprocesses during wave
propagation in triboplasma having the type of quasigaseous formation of particles
possessing the energy  p and the pulse 
pl , resonance conditions (5.90) can e
treated as the decay of quasiparticles following the laws of energy and pulse preser-
vation. These quasiparticles are called plasmons.
The so-called plasmons are long-wave oscillations in the charged surface layer
and play essential role in the triboplasma as manifestation of global in surface layers
of tribomaterials (Fig. 5.4) [27].

Fig. 5.4 Schematic image of


surface electric field

Fluctuations and density of plasma charges produce the electric field inducing
the current tending to restore plasma neutrality. Due to the inertia of charged plasma
particles, they slip into the equilibrium position corresponding to the minimum
potential energy which, in the final account, results in the collective oscillations with
some frequency p . These oscillations can be represented as individual quasipar-
ticles with the energy  p . For instance, the electron passes through plasma and
loses the energy in portions multiple of the energy and the plasmon  p rather than
continuously [8, 28].
The theory of dispersion of surface plasmons is intricate but the gel-model of the
surface enables to obtain a simple relation to calculate the frequency of plasmons:
150 5 Plasma Triboprocesses

p
pl = (1 + 0.5qd + ) (5.91)
2

where pl is the plasmon frequency; q d are the dipole moments of the plasmon
localization region.

5.4 Effect of External Electromagnetic Fields


on Triboplasma

The triboplasma stability, the kinetics of evolution of external dynamic processes


relate to interactions between triboplasma components and external own electromag-
netic fields. The triboprocesses can result in variable electric field in tribocouples. In
accordance with the laws of the Maxwell classic electrodynamics, the electric field
changes induce a weak magnetic field:

E
1 B

=  , (5.92)
x c t

where E is the electric field intensity; B is the magnetic induction; c is the speed of
light.
The motion of triboplasma particles in electromagnetic fields can be considered as
the drift or the directed displacement of charged particles under the effect of various
inducing causes superposed on the main motion. For instance, the directed motion of
triboplasma particles under the effect of electromagnetic fields is imposed on their
thermal motion. The drift velocity is calculated the following formula:

c F
H

D = , (5.93)
q H2

where F is the force affecting the charged particle; q is the particle charge; H is the
external magnetic field intensity.
Correspondingly, the drift flow is

nc F
H

jdr = , (5.94)
H2
where n is the density of charged particles.
The particles can drift in the inhomogeneous magnetic field along and across the
field directions. The transverse magnetic field inhomogeneity implying the condensa-
tion and discharge of the power lines results in the Larmor orbit rotation (precession)
(J. LarmorBritish physicist) of similarly charged particles less in the weaker field
region than in the stronger field region [7]. It is equal to repulsion of the Larmor
5.4 Effect of External Electromagnetic Fields on Triboplasma 151

circumference center across the power lines towards the reducing magnetic field.
Actually, the force appears proportional to the magnetic field intensity gradient H
.
Correspondingly the drift velocity of is equal to:

 r L H

dr = , (5.95)
H
where r L is the Larmor radius; is the normal component of the velocity of particles
(across the magnetic field power lines).
When particles move with the velocity  along the curved magnetic field power
line with the curvature radius R, the drift appears with the direction determined by
the centrifugal inertial force m2 /R. The drift velocity is equal to

 2
dr = , (5.96)
R L

where L is the Larmor frequency.


The velocities of the gradient and centrifugal drift of particles have opposite
directions for ions and electrons, thus, the oppositely directed drift electric currents
appear in the triboplasma. These drafts are displacements of Larmor circumference
centers due to the forces directed perpendicularly to the magnetic field.
On the other hand, the variable magnetic field with the induction B
produces the
Lorentz force affecting the charged triboplasma particles:
F
L = q B


(5.97)

or, in the scalar form F
L = q B sin( B).

where B is the magnetic field induction; is the charged particle velocity.


Thus, the charged particle velocity can be calculated with relation:

ldr L
= , (5.98)
2
where ldr is the drift length.
The multiplier of Eq. (5.97) is

2
1


2
sin B = sin B d B = . (5.99)

0

Substitute relations (5.98) and (5.99) into formula (5.97) to obtain the expression
to calculate the Lorentz force;

qldr B L
FLor = . (5.100)

152 5 Plasma Triboprocesses

The product q ldr is the dipole moment pd , and then relation (5.88) can be
rewritten in the following way:

pd B L
FLor = . (5.101)

The dipole moment relates to the electric field intensity through the relation:

p
d = E, (5.102)

where is the substance polarization.


Than Lorentz force expression (5.101) transforms in the following way:

L EB
FLor = . (5.103)

The total force of the electromagnetic field affecting the macroscopic system with
the mass m is equal to [5]:
m EB
FLor = L . (5.104)

Relation (5.104) can be rewritten by introducing the Umov-Poynting vector of the


electromagnetic energy flow (N.A. UmovRussian physicist, J. PoyntingBritish
physicist):
  c


S = EB, (5.105)
4
then
4m
FLor = L S. (5.106)
c

Thus, the combine electric and magnetic field effect on the particle produces the
ponderomotive force calculated with formula (5.106). This force induces the directed
displacement triboplasma making the mass transfer factor many times stronger.
This motion can be described with the magnetic hydrodynamics apparatus [22].
It is assumed that the typical distances and durations for the moving triboplasma are
much larger than the similar values (run length and time) than individual particles.
For this case, the Maxwell equation (J. MaxwellBritish physicist, founder of the
classic electrodynamics, one of the founders of statistical physics) can be written in
the following way:

div H
= 0,
5.4 Effect of External Electromagnetic Fields on Triboplasma 153

H
 
= rot
H
. (5.107)
t
Equation (5.107) are added with the equation of continuity:


+ div () = 0, (5.108)
t
where is the density of moving macroobjects.
The collective interaction of plasma components can be described with the mag-
netic hydrodynamics apparatus, which treats plasma as the fluid for which the Euler
equations are true:

Vfl
1 fV
+ Vfl Vfl = P + F , (5.109)
t

where f FV is the volumetric density of forces; P is pressure; V f l is the elementary


volume; is the fluid density.
Then
1
f FV =  [ j H ] ,
 c
1
Respectively, the force F: F =  [ j H ]d V. (5.110)
c
V

here j corresponds to the electric current density.


It follows from the Maxwell equations that

4
j = rot H,
c
Then (5.110) can be rewritten in the following way:

1
f FV = rot H
H
. (5.111)
4
Respectively, Euler equation ( 5.109) is rewritten as

V 1 1
+ (V ) V = P + r ot H
H
. (5.112)
t 4

Equation (5.112) can be reduced to the law of pulse preservation:

ik
(V ) = , (5.113)
t xk
154 5 Plasma Triboprocesses

where ik = i k + Pik is the tensor of pulse density; ik is the Kronecker


symbol.
In accordance with the vector analysis formulas
1
H
rot H
= H 2 (H ) H,
2
and, bearing in mind relations (5.1075.113), let us introduce the tensor ik which
can be written in the following way:
 
1 1 2
ik = i k + Pik Hi Hk H ik . (5.114)
4 2

Unlike the canonic form of ik , this tensor in (5.114) is supplemented with the
tensor of stresses.
The energy reservation law in the classic hydrodynamics is expressed by the
following equation:
 
2
+ Wint = divW,
t 2   (5.115)

E = + Q ,
2

Here Wint , Q = Wint + /P are the internal and thermal energy as the function
of the fluid mass unity.
If the macrovolumes of the plasma conductive fluid are moved the ponderomotive
force, then, if the magnetic field exists, the fluid energy density should be supple-
mented with the term H 2 /8 and the energy density flow (the above mentioned
Umov-Poynting vector). Than the energy preservation, law magnetic hydrodynam-
ics is rewritten in the following way:
 
2 H2
+ Wint + = divW, (5.116)
t 2 8
 
1

E = + Q + H [ H ] .
2 4

To simplify these equations, let us ignore the displacement currents:

1 E
 r ot H
. (5.117)
c t
Because the rate of displacement of the model fluid is much slower than the speed
of light, then
ldr
 1, (5.118)
c t
5.4 Effect of External Electromagnetic Fields on Triboplasma 155

Whence

H 2  c2 ,
H
= (H ) H (V ) H H div, (5.119)
t
1 V
div = .
t

Then:    
H d H H
+ V = V. (5.120)
t dt

It follows from Eq. (5.120) that the magnetic field was as if frozen the relax-
ing triboplasma ad each power line shifts together with particles in this plasma or
quasifluid. Therefore, it is with while mentioning the relevant case of the tribo-
plasma synonym or magmaplasma. The value H/ characterizing the process its
displacement changes in every space point proportionally to the stretching of the
corresponding fluid line. The fluid line in hydrodynamics implies the ideal straight
line along which the particle of theses fluid displace.
Whence it can be assumed that the triboplasma moves without intersecting the
magnetic power lines. It is worthwhile to mention the following: the triboplasma
microvolume exist a short time corresponding to the contact sot life and which can
be considered as a highly polarized fluid favorable for friction between contact spots.

5.5 The Experimental Methods of Triboplasma Study

The theoretical model of the extreme state of the tribocouple surface layer or the
triboplasma discussed in the preceding sections permits to learn its temperature,
values of charges of particles and their mass. The advanced triboplasma energy dif-
fusion model answers the problem of the triboplasma components but it does not
permit to trek the kinetics of processes governing its evolution: to determine the life
of triboplasma particle, the duration of the recombination processes, the mechanism
of interactions between the triboplasma and the tribosurface material. The deficit of
experimental data does not permit to formulate a valid theory of dissipative tribo-
processes accompanying the surface tribochemical transformations and appearance
of third bodies. The closedness of the triboplasma by tribocouples, their short life
make nearly impossible to apply common analytical methods. The researcher can
judge about the triboplasma nature only by traces left by the triboprocesses. That
is the tribologists work like paleontologists restoring the ancient pangolin appear-
ance by the imprints of fangs on the rock. At present, it is a challenging problem
to develop a method of tribomonitoring of physical and chemical transformations in
triboplasma straight when friction is underway.
156 5 Plasma Triboprocesses

Fig. 5.5 Principle scheme of monitoring of friction parameters of metal-polymer tribocouple:


1 quartz generator, 2 polymeric body holder, 3 friction body, 4 metallic counterbody, 5 electronic
frequency sensor, 6 interfacing block, 7 recorder

The dissipative processes in friction can be assumed as a combination of the


waves of electromagnetic origin from the Langmuir to the electrostatic waves (the
polychromatic approximation), or to replace the combination of waves in triboplasma
with a single effective wave (the monochromatic approximation). This approach
enables to expand the insight into the physical phenomena similarly to the triboplasma
energy diffusion model. On the other hand, the effect of external electromagnetic
fields on the triboplasma charged particles is capable both to alter the final friction
parameters of the tribosystem and to influence the plasma components selectively
and to study them in the triboprocess based on the electromagnetic response. These
points of study of triboprocesses are disclosed in the monograph [29].
The appealing results have been yielded by the method of electromagnetic mon-
itoring of the metal-polymer tribocouple described in [3033] and patented in [34].
The electromagnetic oscillations are generated in the tribocontact by the quartz gen-
erator connected to the counterbody and the metallic specimen holder (Fig. 5.5).
They generate the induced electromagnetic oscillations providing the data about the
tribocouple state.
It is shown in [35] that the quartz oscillator generates the electric field as a function
of superposition of planar waves of different amplitude and frequency. The spectrum
of oscillations is displayed to analyze the Fourier expansion of the electric signal
spectrum, which appears as the harmonics of oscillations. The friction determines
the energy redistribution leading to the new high frequency harmonics altering the
displayed data (Fig. 5.6).
The registered harmonics relate in the definite way to the intensity of diffusion
metallic particles from the counterbody into the polymeric specimen. It proves that
the diffusive and plasma processes in friction are affected considerably by the external
5.5 The Experimental Methods of Triboplasma Study 157

Fig. 5.6 Frequency spectra of electromagnetic oscillations of polymeric specimen. a before friction,
b during friction. Friction couple polyamide-6steel

electric field applied to tribocontact. Apparently, the positive potential applied to the
metallic counterbody amplifies the diffusion into the polymeric tribocounterpart [14].
It concerns, in the first place, the charged particles or triboplasma components. When
the direction of the external electric field changes, the diffusion is strongly weakened.
The electromagnetic spectra in Fig. 5.7 reflect it as the magnification of the amplitude
A of high frequency harmonics. The regularities of harmonics amplification A
in friction are shown in Fig. 5.7 in dependence of the direction and magnitude of the
external electric field in the tribocontact.
It is noted in [19] that calculation of the plasma frequency p with formula (5.80)
yields the numerical value p = 1060 MHz. During the experiment (Fig. 5.6) new
harmonics were registered in friction in the range 3040 MHz. It is good correlation
and another proof of the monochromatic approximation correctness in triboplasma
description.
This method has drawbacks due to the instability contact phenomena on the tri-
bosurface. Contact between elements of the tribocouple means fast recombination
processes strongly affecting the physical and chemical parameters of tribomaterials
and the kinetics of plasma processes. A tribometer was designed to study triboplasma
and eliminate these drawbacks (Fig. 5.8) [36].

Fig. 5.7 Dependence of


amplitudes of plasma har-
monics on value and stresses
applied to tribocouple as
shown in Fig. 5.5. 1 20 MHz,
2 30 MHz, 3 40 MHz
158 5 Plasma Triboprocesses

Fig. 5.8 Scheme of experimental stand to study triboplasma: a experimental, b scheme of plasma
trap: 1 tribosurface, 2 plasma trap, 3 triboplasma particle, H
and E
magnetic and electric field
intensities, respectively; c torricellian dependence of current power i on load F

The stand is tribotester of shaft-insert pairs. The shaft is made from polytetrafluo-
rethylene block (PTFE). The counterbody is from steel 45. The tribocouple is placed
under the vacuum cap connected to the fore vacuum pump (Fig. 5.8a).
The tribosurface and the counterbody have the slots or plasma traps to catch the tri-
boplasma particles carried away by friction. The slots contain the sensitive electrodes
of the registering means of the spectral analyzer isolated from the counterbody metal.
Constant current source supplies potentials with different signs to the counterbody;
a magnetic field is created around the counterbody. The external electromagnetic
field generated in the tribocouple maintains the triboplasma substance in the active
state [5] (Fig. 5.8b). Additionally, electromagnetic oscillations of different frequency
within the range 010 GHz are applied to the tribocouple from the external generator.
The characteristic signals from the plasma trap are delivered to the spectral analyzer
which a set of individual harmonics by frequencies.
The counterbody is made of two parts to investigate the load variations. Both
parts are pressed to the tribocouple with a calibrated spring. The friction moment is
fixed by power changes of the electric motor so that the current changes in its coils
(similarly to the friction tester Timken (Barond Co., USA)). Figure 5.8c shows how
the current power i depends on the force N of clamping of the counterbody to
the shaft. It is obvious that the dependence of i on N is linear proving that the
measurement is valid. The friction parameters were checked at the sliding velocity
1 m/s and the load 10 N, each point was checked during 30 min.
The closeness of the problems to be solved makes the tribometer like the one with
the schematic diagram in Fig. 5.8 with the common tokomak plant used in nuclear
5.5 The Experimental Methods of Triboplasma Study 159

Fig. 5.9 External view of


tokomak generator: TFC tor-
roidal field coils, MS mechan-
ical device supporting them,
PFC internal torroidal field
coils, VV vacuum chamber,
PFC external torroidal field
coils field, TC current trans-
former

physics to generate and maintain the high temperature plasma (Fig. 5.9). That is why
the tribometer is called Tribocollahedron.
The tribocollahedron checks the nominal contact area at initial stage of friction
interactions in triboplasma based on the difference in measurement o triboprocesses
[10]. The triboplasma components get inevitably into the plasma traps provided on
the counterbody; electric and magnetic fields and the analyzer are connected to the
traps.
Since triboplasma is the electrically active substance state, the external electro-
magnetic field affects the substance and reveals those harmonics of the spectrum,
which relate directly to the triboplasma. This power field effects register the spec-
trum harmonics in the frequency band 10100 MHz and 11.5 GHz. The findings in
[5] show that megahertz frequencies in plasma are due to ion oscillations, the super
high frequency band belongs to electrons. The profiles of obtained electromagnetic
spectra reveal two vividly pronounced peaks both in the Megahertz and gigahertz
regions, which, in accordance with the theory of oscillations, evidence that there are
two types of oscillators in both frequency bands. These oscillators are apparently
electrons and differently charged ions (Fig. 5.10a).
The resonance phenomenon should be observed under the external force effect
every time when the external effect frequency coincides with one of those harmonic
oscillations belonging to the system proper [37]. Since both combined systems can
oscillate with normal frequency, so the resonance occurs with one of two frequencies.
This fact permits to verify the Eigen frequency of detected harmonics by imposing
the induced electromagnetic oscillations on the tribosystem (Fig. 5.10b, c).
The results of harmonics measurement in Fig. 5.10 reveal that the following Eigen
frequencies correspond to the oscillators in the megahertz region: the oscillator gen-
erating the first harmonics resonate at 40 MHz, the second can resonate at 20 and
60 MHz. The harmonics resonate in the gigahertz band at 1.2 and 1.5 GHz. The
harmonics are numbered from low to high frequencies.
160 5 Plasma Triboprocesses

(a)

megahertz band gigahertz band


(b)

Fig. 5.10 Frequency triboplasma electromagnetic spectrum: a external electromagnetic spectrum


view in megahertz (left) and gigahertz (right) bands; b amplitude and frequency characteristics of
megahertz harmonic (left) and gigahertz (right) harmonic bands

The identified oscillators have different charged states well manifested under the
effect of electric and magnetic fields (Fig. 5.11).
The research is still underway and the curves in Fig. 5.11 and it is so far hard to
judge which harmonics relate to positively or negatively charged oscillators but can
be definitely determined how the changes in intensivity (amplitude) of harmonics
respond the friction resistance characteristics of the tribocouple.
The method we have selected to register the friction parameters the growing
current power i impairs the friction resistance characteristics of the tribocouple
because the current growth in the tribometer coils causes the amplitude of elec-
tromagnetic spectral harmonics to augment. The measurement results are shown
in Fig. 5.12.
It is established that in the megahertz spectral band the amplitude rise of the
first harmonic corresponds to the current power i and friction coefficient drop, i
demonstrate the inverse effect: its amplitude declines as the current power grows. The
first harmonic in the gigahertz band relates to the current power changes in a more
complex way: initially it demonstrates growth coinciding with the i reduction, then
this dependence becomes implicit. The second harmonic behaves within this band
like its megahertz counterpart.
5.5 The Experimental Methods of Triboplasma Study 161

(a) (b)

megahertz band gigahertz band


(c) (d)

Fig. 5.11 Dependence amplitudes of harmonics on directions of electric and magnetic fields:
a and b in megahertz and gigahertz spectral bands in dependence on b in megahertz and gigahertz
spectral bands in directions; c and d, b in megahertz and gigahertz spectral bands in dependence on
magnetic field directions. 1 and 2 harmonics at direct field direction (plus on metallic counterbody;
NS magnetic field), 1 and 2 at inverse electric field polarity and inverted magnetic poles

The following conclusions can be made from the above said:


The triboplasma in the tribocouple in question is a complex formation consisting
of positively and negatively charged ions and electrons.
One type of ions becomes manifest in the megahertz and gigahertz band elec-
tromagnetic spectral bands. The oscillators corresponding to these ions have the
Eigen oscillations 40 MHz and 1.4 GHz, respectively.
The oppositely charged particles manifest themselves in the megahertz band at the
frequencies 20 and 60 MHz and at the gigahertz spectral band at the frequencies
1.5 GHz.
different harmonics affect differently the tribocouple wear resistance characteris-
tics: the triboplasma components 40 MHz and 1.4 GHz improve the wear resistance
characteristics, on the contrary, the components 20, 60 MHz and 1.5 worsen these
characteristics.
To conclude it can be stated that the triboplasma is real state of the tribocouple
surface appearing at the initial friction stages. It is a complex electron-ion system
162 5 Plasma Triboprocesses

Fig. 5.12 Dependence of amplitude of harmonics and friction resistance characteristics of tribo-
couple: a megahertz band; b gigahertz band (indexes 1 and 2 correspond to numbers of harmonics)

strongly influencing the tribocouple friction characteristics. At present the established


effect cannot be determined unambiguously due to triboplasma may components,
charge components each producing is own effect on the kinetics of evolution of
triboprocesses.

References

1. V.I. Kostetsky, Surface Strength of Materials in Friction (Kiev, Tekhnika, 1976), 283 pp.
2. D.N. Lyubimov, K.N. Dolgopolov, Modern Tribology (Selected Chapters) (Shakhty, IP Boury-
hin V.M., 2010), 181 pp.
3. D. Bakley, Surface Phenomena During Adhesion and Friction Interaction (Moscow, Mashinos-
troyenie, 1986), 360 pp.
4. G. Heinke, Tribochemistry (Mir, Moscow, 1987), 635 pp.
5. K. Miyamoto, Fundamentals of Plasma Physics and Controllable Synthesis (Moscow, Fiz-
matlit, 2007), 424 pp.
6. L. Cooper, Modern Physics (Mir, Moscow, 1974), 382 pp.
7. V.L. Ginzburg, Propagation of Electromagnetic Waves in Plasma (Moscow, Fizmatlit, 1967),
210 pp.
8. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 3 (Sov. Enc., Moscow, 1992), 672 pp.
9. D.N. Garkounov, Triboengineering (Moscow, Mashinostroyenie, 1985), 424 pp.
10. A.S. Kouzharov, Coordination Tribochemistry of Selective Transfer. Doctor of Engineering
Science Dissertation, Rostov-on-Don, 42 pp., 1991
11. D.N. Lyubimov, V.A. Ryzhikov, Phys. Chem. Processes in Friction: Manual (South Russian
State Engineering University, Novochherkask, 2006), 147 pp.
12. A.A. Polyakov, Terminology of selective transfer, In: D.N. Garkunov (ed.) Durability of
Machinery Rubbing Parts, Issue 4. (Mashinostroyenie, Moscow, 1989), pp. 1115
13. D.N. Lyubimov, V.V. Levkin, A.E. Ivanov, Energy Diffusion Mechanisms of Evolution of Tri-
bosystem. Radiotekhnika. Service Equipment and Technology. Collection of Scientific Works,
Issue 26 (Part 1) (DGAS, Shakhty, 1997), pp. 6369
14. D.N. Lyubimov, Role of Diffusion Processes in Wear and Friction of Metal-Polymer Tribocou-
ples. Candidate of Science Dissertation, Gomel, 22 pp., 1990
15. A.S. Vershinin, D.N. Lyubimov, V.P. Krasnoloutskii, Calculation of X-Ray Spectral Fluorescent
Analysis of Film Structures with Gradient of Concentrations through Depth, Transfer of Higher
Schools, Series Physics: Deposited in VINITI No. 235283, 8 pp., 1983
References 163

16. D.N. Lyubimo, V.A. Ryzhikov, Fundamentals of Friction Theory: Manual (South Russian State
Engineering University, Novochherkask, 2001), 87 pp.
17. D.N. Lyubimov, V.V. Levkin, A.E. Ivanov, Determination of Contact Temperature Using Energy
Diffusion Ideas About Friction. Household Service Equipment and Machinery. Collection of
Scientific Works (DGAS, Shakhty, 1998), pp. 106109
18. D.N. Lyubimov, K.N. Dolgopolov, Diffusive Processes in Friction (Shakhty, IP Bouryhin V.M.,
2010), 148 pp.
19. K.N. Dolgopolov, Improvement of Performance of Household Equipment by Intensifying Lubri-
cation of Tribocouples. Candidate of Engineering Science Dissertation, Shakhty, 27 pp., 2009
20. P.N. Bogdanovich, V.M. Belov, P.N. Sysoer, Thermal Processes in Sapphire-Metal Contacts.
New Materials and Technologies in Tribology, pp. 6566, 1992
21. D.N. Lyubimov, V.V. Levkin, A.E. Ivanov, Nature of Triboplasma Components. Collection
of Scientific Works. Improvement of Service Engineering and Technology (DGAS, Shakhty,
1999), pp. 151155
22. L.D. Landau, E.M. Livshitz, Continuum Electrodynamics, vol. VIII (Moscow, Fizmatlit, 2005),
656 pp.
23. D.V. Sivoukhin, General Course of Physics, vol. 2 (Nauka, Moscow, 1975), 522 pp.
24. L.S. Polak, Plasma Chemical Reactions and Processes (Nauka, Moscow, 1977), 240 pp.
25. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 1 (Sov. Encyclopedia, Moscow, 1988), 638
pp.
26. N.F. Losev, Quantitave X-Ray Spectral Fluorescent Analysis (Nauka, Moscow, 1969), 336 pp.
27. S. Morrison, Chemical Physics of Solid Body Surface (Mir, Moscow, 1980), 367 pp.
28. Ch. Kittel, Introduction into Solid Body Physics (Mir, Moscow, 1978), 791 pp.
29. L.I. Sviridenok, N.K. Myshkin, T.F. Kalmykova, O.V. Kholodilov, Acoustic Electric Methods
in Tribology (Science and Techn., Minsk, 1980), 980 pp.
30. D.N. Lyubimov, V.V. Shakhbazyian, A.E. Ivanov et al., Electromagnetic spectra and physical
chemistry of metal-polymer tribocontact. J. Frict. Wear 11(6), 10841086 (1990)
31. D.N. Lyubimov, V.V. Shakhbazyian, M.P. Bitkov et al., Oscillatory Electromagnetic Spectra
of Friction Metal-Polymer Contact. Deposited in VINITI No. 227190. Rostov-on-Don, 7 pp.,
1990
32. N.K. Vershinin, D.N. Lyubimov, Effect of frictional interactions on metal-polymer tribointer-
face surface on electric and chemical properties, in Conditions and Prospects of Development
of Eastern Donbas, vol. 1 (Novocherkassk University, Novocherkassk, 2001), pp. 274277
33. D.N. Lyubimov, K.N. Dolgopolov, N.K. Vershinin, A.E. Ivanov, Evolution of super excited
states of substance tribosurface. in POLYCOMTRIB-2011: Proceedings of International Sci-
ence Conference (IMMC National Belgium Academy of Science, Gomel, 2011), p. 144
34. V.I. Kolesnikov, D.N. Lyubimov, V.V. Shakhbazyan et al., Method f Monitoring of Friction
Conditions of Metal-Polymer Tribounit. Invention Certificate No.1589131 (1990)
35. R. Feinman, R. Leiton, M. Sands, Feinman Physics Lectures, vol. 6 (Mir, Moscow, 1978), 346
pp.
36. D.N. Lyubimov, N.K. Vershinin, K.N. Dolgopolov, Fine triboplasma structure in fluoroplast
steel contact and its relation to frictional parameters of tribointerface. Frict. Lubr. Mach. Mech.
7, 3741 (2011)
37. S.E. Khaykin, Physical Fundamentals of Mechanics (Nauka, Moscow, 1971), 751 pp.
Chapter 6
Wave Tribochemistry of Post-Plasma States

Still, we believe that the thinking saving principle is not the


best physical research method.
American physicist L. Cooper
What are waves? They are not soil, nor water, or steel, or string
strand, or quartz, they just propagate in these substances.
American physicist D. Pierce
The sea traveler believes the ocean consists of waves rather
than water.
British physicist A. Eddington

Abstract The postplasma period of tribosurface substance transformation is


discussed. The problem of concentration autovibrations of reagents in lubricant is
studying theoretically. Mathematical apparatus of wave mechanics and some princi-
ples of tribochemistry are used for description of these processes.

Summarizing the above said, note that virtually each triboprocess has a correspond-
ing wave, starting with Friedel waves in gel-models of surface states ending with
Alfvn waves in triboplasma. The processes of propagation of elastic deformation in
crystals evolve as undulations. The dislocations move in friction under the effect of
quasielastic force (4.66) and in accordance with the laws of classic mechanics and
generate waves with the frequency

2 Y
d = , (6.1)
GM

where M is the mass of dislocations.


It is naturally tempting to imagine the friction microdynamics as the combination
of wave processes. It is quite acceptable from the viewpoint modern physics because
even the theory of chemical bonding observes the laws of quantum (wave) mechanics.

D. Lyubimov et al., Micromechanisms of Friction and Wear, 165


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8_6,
Springer-Verlag Berlin Heidelberg 2013
166 6 Wave Tribochemistry of Post-Plasma States

6.1 Interaction Between Triboplasma with Tribosurface

The idea that the triboplasma is the physical system consisting of many charged
oscillators permits to construct a rather correct theoretical model of its internal evo-
lution. The model is applicable to solution of these problems in which the triboplasma
corresponds to a definite wave process having the frequency p and the energy
W [1]. In accordance with the laws of wave mechanics, to describe the changes
taking place in a given physical system, let us introduce those values which remain
constant or very little in the given system state but they would acquire any value.
These values in the theoretical physics came to be called the adiabatic invariants I
[2]. The following relation can serve as this value: I = Wp .
The condition of these expressions in accordance with the adiabatic invariant is
the following:  
dI d W
= = 0. (6.2)
dt dt p

Equation (6.2) is easily reduced to

ln W + C1 = ln p + C2 , (6.3)

where C1 and C2 are integration constants.


Equation (6.3) is better to represent in the following way:

W = p A5 exp C3 , (6.4)

where A5 is the normalization coefficient ensuring similar dimensionality of the left


and right arts of Eq. (6.4); C3 are the combination of constants C1 and C2 .
It follows from formula (6.4) that the dimensionality coefficient A5 is equal to
(J s). The fundamental value or work Spossesses this dimensionality in physics
[2]. The work is the function of variables describing the physical system state, so it
determines properly the dynamics of the object under study. The least pressure princi-
ple is implemented during evolution of physical systems implying that the evolution
proceeds in such a way that the work assumes the minimal value. The work is con-
sidered in the modern theoretical physics when formulating virtually any theory as
the main physical value. The difference between kinetic and potential system energy
determines the value S at a given time moment. Whence the quantum value of
work is apparent. The work dimensionality coincides with the dimensionality of
the Plank constant h as the world physical constant. Thus, the Plank constant acts
as the work quantum once again corroborating the work fundamentality as the
physical notion.
Thus, relation (6.4) can be rewritten by expressing A5 and exp C3 as the value
dependent explicitly on the work, for instance: A5 exp C3 = B7 S; where B7 is a
new dimensionless constant:
W = B7 p S,
6.1 Interaction Between Triboplasma with Tribosurface 167

whence
W
S= . (6.5)
B7 p

In accordance with the least work principle and based on relation (6.5), when tak-
ing into account the small dissipation energy corresponding to the contact spot con-
ditions, the triboplasma evolves towards the value p . Similar results are obtained
when studying how the ionization and substance relaxation time depends on the tri-
bocouple surface, which, using the Hamiltonian equation [3] can be written in the
following way:
T  
dq d qd p
Tt = dt = dH
= , (6.6)
dp
dH
0

where Tt is the ionization period of relaxation processes; p and q are summarized


impulse and coordinate; H is the Hamiltonian constant.
But, in accordance with the Heisenberg uncertainty principle dqd p h, relation
(6.6) can be represented in the following approximate way:

Tt . (6.7)
H
The physical sense of the value H is identical to the energy change from the
least work principle. In other words, the less H different, the more stable is the
physical system. The analysis of relations (6.5) and (6.6) proves that the tribomuta-
tion in the plasma from the viewpoint of wave physics reduces to the appearance of
high frequency oscillators. The motion of charged particles in triboplasma creates
truly the electromagnetic emission with the spectral lines (harmonics) evident in
the 102 104 MHz. The monograph [4] relates this frequency band to the emission
of ions and electrons by the plasma components, respectively. Meanwhile, the tri-
boplasma gigahertz band displays the harmonics relating both to the electrons and
positively charged particles. The superposition of electron and ion oscillations causes
the wobbling, which is described by the following wave function [2]:

1 2 1 2
(x, t) = 20 cos t sin t, (6.8)
2 2
where 0 is the amplitude of oscillations; 1 , 2 is the frequency of harmonics; t is
the time characterized by the frequency w = 1 2 with the period T = 2/w .
The representation of plasma oscillations as the wobbling described by the func-
tion (x, t) agrees with the measured charge densities of the thermonuclear plasma
shown in the monograph [4] (Fig. 6.1) because the value (x, t) characterizes the
density of charges in particular.
Thus, the wobbling frequency w is at the same time is the plasma frequency that
can be calculated with the known formula [4]:
168 6 Wave Tribochemistry of Post-Plasma States

Fig. 6.1 Image of contours of equal charge densities in thermonuclear plasma [4]

4q 2 N
w2 = 2p = , (6.9)
Mp

where q is the charge of plasma particles, N is the number of plasma particles; M P is


the reduced mass of plasma particles; M P1 = m 1 1
e + m i ; m e is the electron mass;
m i is the ion mass.
Let us represent the triboplasma as the charge wave appearing between tribo-
surfaces having the frequency p calculated with formula (5.80). The steady state
products appear on the tribosurface due to the relaxation processes and chemical
interaction between triboplasma reactive groups. They are nuclei of chemical com-
pounds with the properties dependent on the tribocouple friction characteristics. The
compounds require an excessive charge density q to form in the triboplasma and
tribosurface interface [5]. From the viewpoint of wave physics, this condition is iden-
tical to the known relation friction different maxima of electron beams dissipating in
the crystal and bearing information about interaction between atoms [6]. Figure 6.2
shows schematically the charge wave in the triboplasma; the wave as the charge
maxima on the tribosurface.
In this case
h pl
= C  , (6.10)
p

where h pl plasma layer thickness; p plasma wave length; C  integer number,


C  = 1, 2, 3. . .
Relation (6.10) can be transformed in the following way:
p
t = C  , (6.11)
2
6.1 Interaction Between Triboplasma with Tribosurface 169

Fig. 6.2 Distribution of


charge densities q initi-
ating tribochemical synthesis
(hpl plasma layer thickness,
Stribosurface)

h pl
where t = is the actual plasma relaxation time.

Hence,
2C 
p =
t
or, in accordance with (5.80), it is the following:

i T 2C 
a3 = .
De t

Then:
De
i =2 C  . (6.12)
a3 T t

Let us introduce one more constant a4 equal to



h me
a4 = 2 = 2 . (6.13)
a3 2e k

With it in mind, Eq. (6.12) can be rewritten in the following way:

De
i = a4 C  . (6.14)
T t

The parameter i (the degree of substance ionization) reflects the kinetics of trans-
formation of the tribomaterial aggregate state.
Relation (6.14) determines the optimum triboplasma characteristic ensuring the
process of post-plasma tribochemical reactions. Their evaluation with formula (6.14)
permits to establish the relations between the most essential parameters governing
chemical transformations: the transferred substance mass depends in accordance with
formula (3.37) on the diffusion coefficient, the relaxation time t and corresponding
plasma temperature T . The temperature affects both time t and the diffusion coeffi-
cient De . It can be assumed that there is some optimal value I at which the evolution
170 6 Wave Tribochemistry of Post-Plasma States

of post-plasma states leads to the appearance of third bodies with the relevant low
friction coefficients ad low wear levels. It follows from (6.14) that the I compa-
rable before with the efficiency of tribochemical reactions that the latter grow the
shorter the substance stays in the triboplasma state. In other words, the efficiency
of tribosystems rows as soon as the triboplasma transforms into steadier substance
state.
Let us consider the triboplasma on the solid body surface from the quantum stand-
point. Let us represent the triboplasma as the planar wave propagating in the semiin-
finite space. It is shown in [7, 8] that the frequency oscillations of plasmons reduces
after interaction with the solid body surface: s2 = 0.52p the solid body-evacuated
medium interface solid body-rarefied medium and 2s = 0.5(2p1 + 2p2 )the
condensed media designated with indexes 1 and 2. The frequency of oscillations
reduces because the plasma part of its energy is lost during interaction with the tribo-
surface. In the particular case, the frequency of surface oscillations s coincides with
the frequency Friedel oscillations (2.36), the surface chemical activity can augment
substantially due to the resonance effects.

6.2 Post-Plasmatic States and Wave Processes Due


Tribochemical Synthesis

Like any plasma, the triboplasma is a non-equilibrium system. The chemical reaction
in it evolves differently from the traditional chemistry canons [9]. The average energy
of molecules reagents and their products amount to 0.150 eV, therefore, the role
of nonelastic collision increases sharply in the plasma chemical system. The typical
durations of the physical and chemical processes come close to make impossible to
treat them independent. So it is better to consider multichannel alternatives of evo-
lution of tribosystems taking into account the interaction between channels and the
environment effect on the reacting subsystems [10]. We observed the same hierarchy
when analyzing Eq. (6.14) noting the interaction between three channels: diffusion,
temperature and time. This interaction creates optimal conditions by the criterion
of initiation of tribochemical reactions in order to distribute the charge densities
(q max), (Fig. 6.2).
The kinetics of plasma chemical reactions is described using the Pauli equation
taking into account the quantum structure of the energy of atoms and molecules [11]:

Ni  
= i pi j p ji N (t) ch.r
i
Ni (t) + Ri (t), (6.15)
t
i

where Ni is the concentration of reacting particles in the energy state i; pi j and p ji


are the probabilities of transition of atoms and molecules from one to another energy
state during collision; ch.r
i is the coefficient of the rate of chemical reactions at the
6.2 Post-Plasmatic States and Wave Processes Due Tribochemical Synthesis 171

ith energy state; Ri is the speed of excitation of the energy level; i is the frequency
of collisions between particles.
In the general case, the coefficient of the rate of chemical reactions depends on
the section of the reaction, its energy and the function of distribution of reacting
particles by energy [10]. In order to initiate appearance of chemical compounds due
to the recombination phenomena in the triboplasma, the kinetic energy of plasma
components should be reduced in order to increase the plasma section of chemical
reactions. The latter is identical to the areas of denser electrons shown in Fig. 6.2 and
determined for plasma by wave equations (6.10) and (6.14).
The nonelastic collision between plasma components leads to the appearance of
complex ions and radicals; the dynamics of their appearance is described by the
kinetic equation [5]:

d
N (t  , t) = k N 2 (t  , t) km N (t  , t)N (t),
dt
N (t  , t) = N0 ,

t dS (6.16)
N (t  , t) dt 
0 dt
N (t) = ,
S(t)

where N (t  , t)the surface density of ions and radicals at the moment of time t;
Sthe chemical compound appearing on the surface within the moment of time t  ;
N0 the initial density of ions and radicals at the moment of time t = 0; N (t)the
observed concentration of ions, radicals and clusters.
The plasma ions and radicals vanish combining into molecules. This process is
limited by the rate of delivery of plasma components to the tribosurface active centers.
Two delivery mechanisms are most likely:
the diffusion which in Eq. (6.16) is characterized by the constant k D proportional
to the coefficient of diffusion De ;
the mechanical agitation of reacting substances taking place when the tribosurfaces
slip and cause the recombination of ions and radicals detached from different sur-
faces. The constant km proportional to the shear deformation takes this process
into account.

Thus, the analytical descriptions of the initial state of tribochemical interaction


are the equations of relaxation processes [5, 12]:

t
Wa (t) = Na0 (t)Wa (t) + Va ( )a (t )Wa (t )d , (6.17)
0

where Wa the adhesion energy; Na0 the initial density of adhesive bonds; a (t)
the function of distribution of the adhesive bonds; Va ( )is density of formation of
the adhesive bonds; Wa (t)is average the energy of the adhesive bonds.
172 6 Wave Tribochemistry of Post-Plasma States

The process of formation of the adhesive bonds on the tribosurface is to some


extent identical to ionization of triboplasma particles; the process contributes to the
adhesion energy Wa . The number of active centers appearing in some way on the
surface is described by the system of the following linear equations:

dNa = (W1 Na W2 Na ) d + 1 (Na ) d,


(6.18)
dN = (Na N ) p1 d [W1 N1 (1 b) W2 ] d + 2 (Na ) d,

where W1 , W2 is density of probabilities of appearance (recombination) and rupture


(ionization) of adhesive bonds; Na is the number of active centers Na identical in the
sense to Ni from (6.15) is the time of relaxation calculated from relation (6.14);
b is the probability of rupture of adhesive bonds; 1 (N ), 2 (N ) are the function
and the rate of changes in the number of adhesive bonds and active centers in the
tribocouple in operation.
On the tribosurface:

1 (Na ) = 1 f 1 (x)(Na A6 + Na2 B8 ),


(6.19)
2 (Na ) = 2 f 2 (x)(Na A7 + Na2 B9 ),

where 1 and 2 are constants of reaction intensities; f 1 (x) and f 2 (x) are functions
taking into account the tribocouple design; A6 , A7 , B8 , B9 are the empirical constants.
The constants of rates of tribochemical reactions are determined from the
Arrhenius equation:
Wch
ch.r = 0 e kT , (6.20)

where 0 is the pre-exponential multiplier; Wch is the energy of activation of the


chemical reaction; k is the Boltzmann constant; T is the temperature.
Equation (6.19) is multipurpose. To follows from the nuclear kinetic theory of
molecules, providing there are tribochemical changes in the tribosurface properties
that the density of the probability of appearance (rupture) of adhesive bonds depends
only on the sliding velocity of contacting bodies and on the tribocontact geometrical
configuration. These bonds can be written in the following way:

W1 = 1 Na s ,
(6.21)
W2 = 2 (1 2 Na )s ,

where 1 and 2 are the tribocontact geometrical parameters; s is the sliding velocity.
With the account of (6.19), (6.21), system (6.18) is reduced to the following linear
equation:
d2 N d Na
2
+ 2A + B Na = B Na |t , (6.22)
d d
1
where 2A = ,
+ W1 + W2 1 f 1 (x)A6
6.2 Post-Plasmatic States and Wave Processes Due Tribochemical Synthesis 173

W2
B= .
+ bW1 W2 1 W1 A7 1 f 1 (x)A6 1

d Na
If the condition A2 > 0, Na (0) = Na0 , d 0 = bW2 Na2 is fulfilled, the
=
solution of Eq. (6.22) is the following:

t t
Na ( ) = Na [ +c1 e 1 + c2 e 2 , (6.23)

where t11 = A6 + A26 B8 , t21 = A7 A27 B9 ,
 
t1 (N Na0 ) + t1 t2 W1 Na0 b t2 (N Na0 ) + t1 t2 W2 Na0 b
c1 = , c2 = ,
t2 t1 t2 t1
N = Na |t .

The changes in time of the mean energy of adhesive bonds depend on the nature of
tribochemical processes. The kinetics of generation and interaction of active centers
play a governing role.
For instance, when substance on the tribosurface is in the post-plasma state, the
dynamic equilibrium is typical between ionization of particles (appearance of active
centers) and their recombination (interaction) of the centers. The transition of the
triboplasma to the steady state is accompanied by the equilibrium disorder and dislo-
cation shift towards recombination processes. The dynamic equilibrium between ion-
ization and recombination at the post-plasma stage of evolution of materials ensures
a long enough stage of tribosystem evolution.
The chemosorption of reactive triboplasma agents precedes the tribochemical
interaction between surface layers. The analytical expression for the chemosorption
can be written in the following way [5]:
 
dC Wch
= k0 C(1 S) exp , (6.24)
dt kT

where C is the concentration of the chemisorbed substance; k0 is the coefficient


depending on the material; S is the surface portion occupied by the chemisorbed
substance close in the sense to the parameter ks from formula (4.27); Wch is the
energy of activation of chemosorption processes approximately corresponding to
Wch from relation (6.20).
Relation (6.24) does not disclose the chemosorption mechanisms described in
Sect. 2.3. It can be done by involving the quantum chemistry apparatus, but it is
beyond the scope of this section.
Using the second Fick law, rewrite the value dC
dt in the following way:

dC 2C
= D 2 . (6.25)
dt x
174 6 Wave Tribochemistry of Post-Plasma States

Let us equate the left parts of Eqs. (6.24) and (6.25):


 
2C Wch
D 2 = k0 C(1 S) exp . (6.26)
x kT

According to the Arrhenius law, the diffusion coefficient D can be represented as


(3.63), and then relation (6.26) is written as

Wd 2C Wch
D0 e kT = k0 C(1 S)e kT , (6.27)
x2
where Wd is the energy of activation of diffusion processes.
The activation energies Wd and Wch have close absolute values [5, 13]; hence,
relation (6.27) can be rewritten as

D0 dC2 = k0 C(1 S),


2

or (6.28)
2C k0 (1S)
x2
+ D0 C = 0.

Formula (6.28) characterizes the harmonic oscillations with the frequency x :



k0 (1 S)
x = . (6.29)
D0

Equations (6.28) and (6.29) describe analytically the experimental data called in
[14] the friction oscillations of metal ion concentration in friction in the selective
transfer mode (ST) (Fig. 6.3).
Equations (6.28) and (6.29) are not limited by the selective transfer phenomenon;
they are rather general proving the universality of the effect of friction oscillations
of the concentration of chemisorbed substance in friction.
The solution of differential Eq. (6.28) can be written in the following way:

C (x) = C0 cos (x x) . (6.30)

Fig. 6.3 Oscillations of con-


centration of copper sali-
cyliden anilinate ions during
friction of copper alloys on
steel during friction in ST
mode [14]
6.2 Post-Plasmatic States and Wave Processes Due Tribochemical Synthesis 175

The concentration of chemisorbed substance becomes equal to zero at the argu-


ment x equal to diffusion length L D , in other words:

C0 cos (x L D ) = 0,

what is possible only when



x L D = ,
2
or (6.31)

x = .
2L D

Substitution of (6.31) into the formula to find the diffusion length L D yields:

x = . (6.32)
8De t

Let us equate the right parts of Eqs. (6.29) and (6.32):



k0 (1 S)
= . (6.33)
8De t D0

With the account of expressions (5.16) rewrite equality (6.33):



k0 kT pl (1S)
= ,
8De t De Wd
or
k0 kT pl (1S) (6.34)
= Wd t = 1.12,
8

k0 kT pl (1S)
Wd t = 1.26.

The chemisorbed substance enters into the chemical reaction; its kinetics is
described in the general form by the equation [5]:
 
dC Wch
= k0 C exp
n
, (6.35)
dt kT

here n is the number indicating the reaction sequence.


By equating the left parts of Eqs. (6.24) and (6.35), the following is obtained:

C n1 = 1 S, (6.36)

 is the mean chemisorbed substance concentration.


where C
Expression (6.36) confirms the hypothesis about the governing role of diffusion
processes on the actual contact area formation in [15, 16] limiting considerably the
176 6 Wave Tribochemistry of Post-Plasma States

alternatives of the nature of tribochemical processes. In particular, it is evidently


that the chemical reactions are prohibited in friction with the sequence n = 1 in the
conditions at which S = 1. Really, if a third body coats (passivates) the tribosurface,
the chemical interaction of the tribomaterial with the environment is impossible. The
substitution of the result from (6.36) into (6.34) yields

k0 kT pl C n1
t = 1.26 (6.37)
Wd

The triboplasma energy according to [17] can be represented in the following


way:
3
W pl = kT. (6.38)
2
In accordance with (6.38), formula (6.37) can be rewritten in the following way:

k0 W pl C n1
t = 1.89,
Wd
or (6.39)
W pl C 1n
= 1.6 .
Wd k0 t

It follows from these equations that the relation between the triboplasma energy
and the diffusion activation energy (actually the energy of cohesive bonds) is constant
for the given friction conditions. Relation (6.39) is useful for the next theoretical
constructions.

6.3 Apparatus of Wave Mechanics for Description


of Tribocouple Micromechanics

The wave phenomena during tribosystem plasma states contribute considerably to


the tribosystem structure at the microlevel. It can be expected that a general algorithm
will be developed to describe the tribophysical processes based on the wave physics
apparatus [2].
The main equation of the wave physics is the relation called the wave
equation [3]:
1
 = 2 2 , (6.40)
c t
6.3 Apparatus of Wave Mechanics for Description of Tribocouple Micromechanics 177

where  is the D Alamber operator (J. D AlamberFrench mathematician); is


the Laplace operator (P. LaplaceFrench mathematician and physicist); t is time, c
is the speed of light.
Equation (6.40) is a linear homogeneous equation in partial derivatives. Its solu-
tion is the function and time coordinate (x, y, z, t) called the wave function. This
equation is invariant (it preserve its form and structure) in respect to the linear trans-
formation of the coordinates and time combined into the parametric Poincare group
(J. PoincareFrench mathematician and physicist). These transformations are more
general than the Lorentz transformations (4.77) proving the fundamentality of the
solutions obtained in the wave physics. L. Cooper remarked that, the properties
of the solutions of the equations seem certainly more important and fundamental
than the equations themselves [6]. The solution of the wave equations possesses
the property of superposition. If 1 (r, t) is one solution of Eq. (6.40), 2 (r, t) is
another solution, whence 3 (r, t) = 1 (r, t) + 2 (r, t) is also the solution of the
wave equation under the same conditions. This principle reflects the fundamental
property of waves implying that the elementary wave can be identified from among
all wave movements; complex wave phenomena are reduce to this wave movement.
This technique in the mathematical physics is called physicist the harmonic analysis.
The wave functions are the objects without any definite physical sense. They
can be scalar values (for instance, pressure in gas or electric field potential) vectors
(velocities of particles, intensities of electric and magnetic fields). It is reflected in
the quantum mechanics and its main law, the Schrdinger equation:

i = H . (6.41)
t
The Schrdinger equation is the wave equation with the wave function comprising
numerous physical values which it assumes when describing physical process and
also the sense of the amplitude of probability, in other words, the value which when
squared equals the event probability [18]. Wave Eq. (6.40) can be represented in the
Cartesian coordinates as a set of four equations of oscillators (the oscillating system);
three coordinates are space and one time:

2
+ x2 = 0,
x2
2
+ y2 = 0,
y2
(6.42)
2
+ z2 = 0,
z 2
2
+ 2 = 0.
t 2

The coefficients x , y , z are the wave vector components . The solution of


Eq. (6.40) is the following planar wave:
178 6 Wave Tribochemistry of Post-Plasma States

= exp i (t r ) , (6.43)

providing:
2
2 = x2 + y2 + z2 = c2 . (6.44)

The simplest planar wave is the following:


 
i
= exp (it) = exp W t . (6.45)


Let us use wave functions (6.45) to consider the essence of the realistic amend-
ments verifying the theoretical calculation of tribophysics. Let us assume the sta-
tionary wave functions of particles have all coordinates x, y, z the same but depends
on the time and equals . The second system of coordinates, x  , y  , z  , t  , moves in
respect to the first one in the direction of abscissa axis with the velocity . Then
the coordinates x  , t  relate to some space point through Lorentz x, t transformations
(4.77). The Lorentz transformations for the velocity is directed in the negative
direction of the axis x, time t relates to t  with formulas (4.77), hence, the wave
function (6.45) is written in the following way [19]:
    
i i W0 t  W0 
exp W t = exp , (6.46)
  c2

where is the relativistic amendment.


The changes in the wave function in space and in time in the second system of
coordinates can be written in the following way:
 
i    

exp W p t p x ,

W p (6.47)
 W0 
Wp = , p = 2 .
c

where p  is the relativistic impulse.


Respectively, the wave function for particles with the impulse p is a proportional
value  
i 
exp W p t  p  x  , (6.48)


where W p = ( p  c )2 + W0 .
In the non-relativistic problems, the energy W can be calculated with the following
formula:
W = mc2 + W  , (6.49)
6.3 Apparatus of Wave Mechanics for Description of Tribocouple Micromechanics 179

2
where W  = Wint + 2m p
is the energy excess or deficit versus the energy at rest
W0 = mc2 .
In the general case, W p should include the kinetic atom energy and its energy
bonds
or excitation. It can be called the internal energy with the amplitude
exp i W p x px [19].
The dispersion equation considered above for particular cases of description of
waves in plasma are used together with Eq. (6.40) those from the wave physics.
The dispersion relations imply the integral solutions of the response functions
describing the response of the equilibrium stationary physical system to the external
effect. The dispersion relation reflects the analytical properties of the response func-
tions in the complex plane frequency registering it as the (the energy) dependence
setting a number of its limits. To derive a dispersion relation, it is unnecessary to
know the structure or the physical system dynamics. Such relations are based on the
general principle of causality: no physical event can affect the event in the past. In
wave physics the dispersion relation is the frequency dependence included into wave
Eqs. (6.43) and (6.45), from the wave vector , i.e., = ( ). Actually, Eq. (6.46)
belongs to the dispersion relations.
The key notion when analyzing the dispersion of waves are the phase and group
velocities of waves ph and g which coincide only at = c when ph = g = c.
These values in the anisotropic media differ both by magnitude and direction.
The KleinGordon equation is of definite interest because it is relativistic invariant
(O. Klein and W. GordonSwedish and German physicists) which describes the free
scalar physical field that in one-dimensional approximation is the following [2]:

2 (x, t) (S D)2 2 (x, t)


= 02 (x, t) , (6.50)
t 2 m x2
where (SD) is the characteristic of the stress-strain state; 0 is the system own fre-
quency of oscillations.
Equation (6.50) is one of the fundamental equations of the quantum physics
equally true for the de Broglie waves and the relativistic particles. It turns into the
classic wave equation when 0 is equal to zero. Assume that all tribosystem elements
move in the steady mode and oscillate with the frequency of the motive force, while
the phase constants of all moving elements are the same. In this case:

(x, t) = A (x) cos (t + ) ,


2 (x, t)
= 2 A (x) cos (t + ) , (6.51)
t 2
2 (x, t) d 2 A (x)
= cos (t + ) .
x2 dx2
The substitution of Eq. (6.51) into (6.50) and reduction by cos (t + ) yields the
differential equation for the space configuration A(x) of the oscillators corresponding
to the steady mode of the effective force with the frequency :
180 6 Wave Tribochemistry of Post-Plasma States

d 2 A(x) m  2 
= 0 2
A(x). (6.52)
dx2 (S D)2

The solutions of differential equation (6.52) differ strongly at > 0 and < 0 .
In the first case, Eq. (6.52) describes the sinusoidal wave and is the following:

d 2 A(x)
= 2 A (x) , (6.53)
dx2
 2
where 2 = m
(S D)2
02 is the wave number.
The expression for 2 is a relation for the sinusoidal wave. The general solution
of wave equation (6.53) is the following:

A(x) = A8 sin( x) + B10 cos( x) , (6.54)

where A8 and B10 are the constants determined from the boundary conditions.
Exponential wave appear in the tribosystem at the frequencies < 0 . Let us
introduce the positive constant 2 :
m  2 
2 = 0 2
. (6.55)
(S D)2

Then Eq. (6.52) becomes the following:

d 2 A(x)
= 2 A(x) . (6.56)
dx2
The general solution of Eq. (6.56) is a superposition of two exponential functions:

A (x) = Ae x + Be x , (6.57)

and the wave function satisfying (6.56) is the following:



(x, t) = Ae x + Be x cos (t + ) . (6.58)

The constant , which characterizes the reduction of the amplitude per length
unit, is called the absorption coefficient:

(x, t) = A(x) cos t = Ae x cos t. (6.59)

The value 1 is called the penetration depth. There is a similarity between the
values and the wave number for sinusoidal waves; the like similarity is observed
between the penetration depth and the wavelength . Expressions (6.53) and (6.55)
are a full dispersion relation for tribosystem elements.
6.3 Apparatus of Wave Mechanics for Description of Tribocouple Micromechanics 181

Fig. 6.4 Matching of two wave functions 01 and 02 in tribosurface region

Let consider the instances of application of the wave physics apparatus to treat-
ing plasma and post-plasma processes in tribosystems. Let us consider the substance
transition into the triboplasma, then the post-plasma processes, and finally restoration
of the initial state. Let us equate the wave function describing one of the substance
aggregate states and to the wave function corresponding to the state into which sub-
stance transits the initial steady state.
Assume that the dislocations describe the behavior by some wave function 1 ;
it is represented by the plane waves with the frequency disl existing within the
interval of time 0 to t1 . Under the effect of friction forces, the substance surface turns
into the triboplasma; its existence is described too by the plane wave with the wave
function 2 and the frequency p within the interval of time t1 to t2 :

1 = 01 exp(idisl t1 ) ,
 (6.60)
2 = 02 exp i p t2 ,

where 01 , 02 are the wave functions at t = 0.


At the moment when the transformed dislocations are delivered by the friction
forces to the tribocontact surface, both waves are blended into the plasma (Fig. 6.4).
As a result, the wave functions equalize.
At the moment of time t, 1 = 2 :

01 exp(idisl t) = 02 exp i p t

considering only real numbers of the obtained equation, the following can be
recorded:
X 1 cos disl t = X 2 cos p t
or
(6.61)
X1
cos disl t = cos p t.
X2

where X 1 and X 2 are the real parts of the amplitudes of wave functions 1 = 2 .
182 6 Wave Tribochemistry of Post-Plasma States

It is noted above that only an insignificant part of the friction energy turns the
tribocouple surface into the triboplasma corresponding to the condition X 1
X 2 .
Hence, the relation between the amplitudes X 1 / X 2 is rather large. It follows from
relation (6.61) that cos disl t is small. Therefore:

disl t ,
2
or (6.62)

disl .
2t
In accordance with relation (6.1):

2 Y
= ,
GM 2t
whence (6.63)
GM
t= .
4Y

The value cos p t is close to unity, p t 0, or, in accordance with (5.80) and
(6.63):
GMT
a3i 0.
De

It is in the extreme case


a3i G M T = 0. (6.64)

Since the values a3, G, M, T differ from zero at the moment of time t, i = 0. In
other words, t corresponds to the particular moment of time (the singularity point),
after which the tribosurface substance leaves he initial steady state without transiting
yet in the plasma state. The substance begins to turn into plasma at the moment of
time t  = t + dt.
The spontaneous concentration of the wave energy within a small space region
is the wave collapse. The latter appears in the plasma and sound waves and make
the process probable in friction. The plasma collapse is accompanied by the appear-
ance of the cake-like cavern or the reduced density in which the oscillating electrical
field gets locked with its maximum being in the center. The system becomes like
a black hole [18]. The triboplasma wave collapse alters significantly the tribo-
couple parameters. To assess its probability and aftermath is one of the challenging
tribophysical problems.
References 183

References

1. V.L. Ginzburg, Propagation of Electromagnetic Waves in Plasma (Fizmatlit, Moscow, 1967),


p. 210
2. F. Crawford, Wave (Nauka, Moscow, 1974), p. 521
3. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 1, (Sov. Encyclopedia, Moscow, 1988), p.
638
4. K. Miyamoto, Fundamentals of Plasma Physics and Controlable Synthesis (Fizmatlit, Moscow,
2007), p. 424
5. D.N. Lyubimov, K.N. Dolgopolov, Modern Tribology (Selected Chapters) (IP Bouryhin V.M.,
Shakhty, 2010), p. 181
6. L. Cooper, Classic Physics (Mir, Moscow, 1973), p. 479
7. E.A. Stern, R.A. Ferrell, Phys. Rev. 120, 130 (1960)
8. C.Z. Kunz, Physics 196, 311 (1966)
9. L.S. Polak, Plasma Chemical Reactions and Processes (Nauka, Moscow, 1977), p. 240
10. Y.V. Tsvetkov, S.A. Panfilov, Low-Temperature Plasma and Restoration Processes (Nauka,
Moscow, 1980), p. 180
11. A.M. Prokhorov (ed.), Physical Encyclopedia, vol. 3, (Sov. Encyclopedia, Moscow, 1992), pp.
672672c
12. D.N. Lyubimov, V.A. Ryzhikov, Processes in Friction: Manual (South Russia state Engineering
University, Novochherkask, 2006), p. 147 (Phys. Chem.)
13. G. Gottstein, Physical and Chemical Principles of Science of Materials (Binom, Moscow,
2009), p. 375
14. A.S. Kouzharov, Coordination tribochemistry of selective transfer. Doctor of Engineering Sci-
ence, Disssertation. Rostov-on-Don, (1991) p. 42
15. V.A. Potekha, Study of antifriction properties of coatings of thermoplastic polymers metals.
Candidate of Science Disseration, Rostov-on-Don, (1982) p. 26
16. D.N. Lyubimov, Role of diffusion processes in wear and friction of metal-polymer tribocouples.
Candidate of Science Dissertation, Gomel, p. 22, 1990
17. D.V. Sivoukhin, General Course of Physics, vol. 2 (Nauka, Moscow, 1975), p. 522
18. A.A. Barannikov, A.V. Firsov, Main Concepts of Modern Physics (Vysshaya Shkola, Moscow,
2009), p. 349
19. R. Feinman, R. Leiton, M. Sands, Feinman Physics Lectures, vol. 6 (Mir, Moscow, 1978), p.
346
Chapter 7
Antifriction Micromechanisms

Such modifications were clumsy, but it was still worse that


none of them managed to make the theory work. They all
contradicted to the experiment
American physicist R. Price
In order not to undermine in this way the science reputation,
we preferred to select just some subjects to work with them
thoroughly
American physicist E. Rodgers
The thing is that most appealing and elegant scientific results
would always possess the feature of being dull and obscure to
the profane
Soviet writers A. and B. Strugatskiis

Abstract Notions about micromechanisms and structure of lubricating layers


forming under friction, causation of this process with electron structure of lubri-
cant components are presented. Quantum model of F. London complex conjuga-
tions formation and notions of Russian tribologists A.S. Kuzharov, D.N. Garkunov,
A.A. Poljakov concerning unwearing effect are invoked for lubricants antifric-
tional properties description. Much attention is devoted to experimental results on
microstructure and antifrictional properties lubricating layer.
The present chapter is based on the systemized context of six preceding chapters
to endeavor to describe friction free tribosystems. The proposed model uses the
micromechanisms leading to third bodies in the interface as specific lubricating
structures passivating the tribosurface. The passivation makes the tribosurface passive
reducing the friction force component to diminish thus the friction coefficient and
wear of materials in most cases.

D. Lyubimov et al., Micromechanisms of Friction and Wear, 185


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8_7,
Springer-Verlag Berlin Heidelberg 2013
186 7 Antifriction Micromechanisms

7.1 Lubricating Layer Formation

In accordance with relation (5.106), the triboplasma effect on the electromagnetic


field results on the displacement macrovolumes in the polarized interface. It is noted
in [1] that atoms and molecules in the polarized tribocouple can live between frag-
ments of a second to hundreds of hours enabling to consider the displaced macro-
volumes as stable bodies virtually.
The electric field generated in the tribosystem affects considerably the adhesion
and the structure of lubricating layers. It can be assumed that the polarized substance
serves as a construction material for lubricating layers.
Let us consider how they appear. The external electric polarization creates the
gyroscopic moment in molecules (multiples) stabilizing them [2]. The potential
energy U and the probability p of orienting at the angle to the tribosurface are
combined and governed by the following relations:

U0 (1 cos n)
U () = ,
2
  (7.1)
U ()
p () = p0 exp ,
kT

where n is the multipole order of symmetry; p0 is the pre-exponential multiplier.


The external electric field orders the structure of molecules deposited on the
tribosurface. The correlation is established between the maximum orientation of
multipoles and the minimum friction coefficient [2].
Really, the multipole approaches to the solid body surface and becomes exposed
to the surface energy effect. Charges redistribute in the polar molecules resulting
in the effect of multipole groups the torque making them fully or partly oriented in
respect to the tribosurface (Fig. 2.19).
Because of the inhomogeneity of the electromagnetic field on the tribosurface
(see Sect. 2.3) and the field gradient along molecules, the interaction is known [2]
not to be limited only by the orientation of molecules. The polar group molecules
would displace approaching to the solid body surface. The molecule accelerates as
they approach to the surface and impact it. Like in classic mechanics, the impacting
can be considered non-elastic, if the molecule gives away to the surface a major part
of the energy, which is adsorbed by its center of attraction. If the molecule faces
the surface with its adsorbent non-polar end or the impact contacts the region where
the field s weak, or the impact is close to the elastic, the molecule rebounds from the
surface and return into external environment.
If non-elastic impacts prevail, the number of molecules fixed and steadily oriented
on the tribosurface grows gradually. The adsorbed molecule loses considerably their
motility which they would posses otherwise in the lubricating layer and transforms
into the bounded bidimensional state. As the tribosurface energy augments, the orien-
tation of the adsorbed multipoles intensifies normal to the tribosurface; the multipoles
get ordered following the fundamental principle of energy minimization. Figure 7.1
7.1 Lubricating Layer Formation 187

Fig. 7.1 Dependence of surface potential S p and absorption on number of adsorbed particles and
scheme of orientation of their electric moments [2]

shows the curve how the tribosurface potential S p depends on the number of adsorbed
multipoles together with the scheme of their orientation.
The above reasons have significant limits determined by the complex structure of
the electromagnetic field on the tribosurface. Therefore, it is virtually impossible to
calculate the full surface potential energy with the account of the surface topology
excepting several simplest cases. The kinetics of appearance of molecular layers fol-
lows the quantum adsorption mechanisms treated in Chap. 2 based on the interaction
between adsorbate electron shells and the solid body gel-surface. Interaction between
atoms and molecules in the surface layer alters the surface structure and topology
created by the power field on this surface. The example in [3] shows the interaction
between the golden monocrystal and the common salt monocrystal surfaces where
interatomic forces begin to generate clusters on the golden surface consisting of
chlorine and sodium atoms. The forces of boning but the surface of gold and chlo-
rine and sodium atoms are different. After the electron structure change due to the
clusterization, the interaction between gold and chlorine atoms becomes dominating
when the chemical bonding takes place as a result of overlapping of gold d-orbitals
and p-orbitals with four chlorine atoms on the cluster surface (Fig. 7.2) [3].
The activation of the solid body surface by friction forces and transition of the
material into the triboplasma state expand opportunities of exclusively various chem-
ical transformations. Paper [4] shows that the frictionexcited surface acquires the
active enter or the mesoregions of abnormally intensive adsorption, the centers are
analogs of the cluster depicted in Fig. 7.2. The energy model of active centers is
treated in works [4, 5]. The adhesive center appears in the spots where dislocations
188 7 Antifriction Micromechanisms

Fig. 7.2 Contour map of binder orbitals of AuCl in AuNaCl (001) system. Full (dotted) lines
correspond to positive (negative) wave functions

emerge on the tribosurface and serves the region of intensive transfer of the material
producing the lubricating layer.
The motility of molecules in the lubricating layer stays longer the higher the
temperature, the lower the degree of saturation of the monomolecular layer on the
tribosurface and the fewer the hysteric obstacles to the molecules movement. These
obstacles may be, first, the micron-sized asperities on the tribosurface, and, second,
the adsorbed molecules. The motility maxima can be inhibited by chemical reactions
in the monomolecular layer resulting in complex molecules and their cluster.
The adsorption of the molecular structure on the tribosurface corresponds to the
boundary friction conditions in the tribosystem. The boundary lubricating film is
thin compared with its linear dimensions. It can be assumed that the rate of the
displacement of the lubricating fluid particles normal to the tribosurface is negligibly
slow versus the velocity  along the surface. The pressure is constant throughout
the film thickness, the dependence of its material viscosity on temperature and it can
generally be represented in the following way:

= 0 exp [ (T T0 )] . (7.2)

If the lubricating layer is assumed the incompressible fluid, then the relation
between viscosity and pressure in the isothermal:

(P) = 0 e P (7.3)

where 0 is the lubricating fluid viscosity at the air pressure and normal temperature;
is the viscosity piezo coefficient; P is pressure.
The hydrodynamic elastic lubricating film bears the load on the micron asperities
of the tribosurface. The friction force is determined in the general case by the viscous
friction of the film and the contact spot friction.
7.1 Lubricating Layer Formation 189

German tribologist G. Vogelpohl summarized in 1954 [6] the friction coefficient


expressions for the hydrodynamic radial bearing:

1/2
f fr = 3r S0 , (7.4)

where r is the relation between the radial clearance and the shaft radius; S0 is the
Zommerfeld number.

Pr2
S0 = , (7.5)

where is the angular velocity; is the lubricating medium viscosity.


The proposed the following formula for the mixed lubrication conditions:
    1/2
Dr
f fr = f 0fr 1 1 3 , (7.6)
Pr2 h min P

where f 0fr is the friction coefficient at rest; Dr is the radial clearance; h min is the
film thickness corresponding to the Stribeck minimum curve.
German tribologist R. Stribeck studied sliding and rolling bearings and obtained
the results in Fig. 7.3 that the friction coefficient f fr depended on the lubricating layer
thickness in the following way:

x= , (7.7)
P
where is the sliding velocity.
According to Stribeck, the abscissas to the left of x1 correspond to the zone of
boundary contact friction with the characteristics described by relation (7.6); the
abscissas to the right of x3 correspond to the hydrodynamic or rheodynamic friction;
mixed friction dominates within the interval x1 x2 ; the Vogelpohl relations can
describe it.

Fig. 7.3 Stribeck diagram [6]


190 7 Antifriction Micromechanisms

The boundary friction mechanisms and parameters are determined by the following
main factors:
the physical nature of tribomaterials, the condition and properties of tribosurfaces;
the structure, physical and chemical characteristics of the lubricating material;
the regularities of the adsorption of lubricating molecule, the physical and mechan-
ical characteristics of the layer of adsorbed molecules;
the pressure, the lubricating layer effective thickness, the sliding velocity, the tem-
perature and other parameters of the tribounit;
the tribochemical transformations of the lubricating tribomaterial molecules, the
changes in the boundary layer structure in friction: redistribution of the lubricating
material over the profile of micron asperities, the adsorption of polar molecules on
the tribosurface, the thermal disintegration of molecules, catalytic processes and
chemical reactions, oxidation, in the first place.

Thus, the boundary friction during adsorption and appearance of lubricating mate-
rial polarized molecules and active oriented film structures they produce on the
tribosurface are not stationary processes. The analysis of the lubricating layer back-
ground after its molecules contact the tribosurfaces until the layer degenerates
permits to trace the chain of events modifying the triboprocess radically during
their evolution. During the run-in stage, boundary layers nucleate, then the third
body appears and govern the steady friction; any damage of the third body lands
to degradation and fracture of boundary layers [2]. Famous Russian tribologist
A.A. Polyakov treated the process of evolution of lubricating layers as manifesta-
tion synergic processes representing them by the dissipative structure. According to
Polyakov, the dissipative structure implies the space or space-time structure existing
and circulating indefinitely long in time provided there is constant energy and matter
supply. The dissipative structure is an open, reversible unending structure appear-
ing in a relatively homogenous environment [4]. D.N. Garkunov and A.A. Polyakov
admitted the fact of space-time mechanism of evolution of lubricating layers as an
essential argument in favor of the relativistic effects in friction. The dissipative struc-
ture consists of intersecting reacting flows. For instance, a substance arrives into the
system activated by friction and with excessive free energy and leaves it after the
energy is expended. A.A. Polyakov believes that the dissipative structure results from
kinetic transitions from conservative movement of dislocations to their climbing cor-
responding to the conditions, which lead to the relativistic dislocations.

7.2 Chemical Interaction in Friction

The chemical interaction between substances in friction possesses specific features


determined by the plasma chemical nature when generating the third body. At the
stage of triboplasma appearance, the chemical interaction between tribomaterials
7.2 Chemical Interaction in Friction 191

Fig. 7.4 Scheme of evolution


of tribochemical reactions:
reaction intensivity versus its
duration [7]

follows Eq. (6.15), and then chemical transformations proceed further as described
in Sect. 6.2 when the dynamic equilibrium between ionization and recombination
of ionized triboplasma particles evolves. The relations between tribochemical trans-
formations of tribomaterials with plasma and post-plasma processes governs the
kinetics of tribochemical reactions. They quickly reach the maximum intensity after
the induction period is over, the reactions evolve in steady friction, and their tribo-
chemical intensity falls as quickly to zero as soon as friction ceases (Fig. 7.4).
The variety and specific features of products characterize the tribochemical reac-
tions since the tribosystem is the open thermodynamic system in which entropy can
reduce, German physicist and chemist G. Heinicke asserts [8] that chemical trans-
formations can evolve in friction contact with positive entropy changes, for instance:

SiC + 2H2 Si + CH4 ,


2Cu + CO2 2CuO + C.

Unique tribochemical reactions of oxidation of noble metals can take place that
are unlikely under usual conditions:

3 1 3
Au + CO2 Au2 O3 + C.
4 2 4
There is a class of chemical transformations, which govern the specific tribo-
chemical feature of the selective transfer [8, 9]. It is found that the d-elements in the
tribocontact zone or with the electron d-shell vacant (iron, cobalt, nickel, copper) or it
is just filled (zinc, cadmium, hafnium) reduces the friction coefficient and tribocouple
wear (Fig. 7.5).
One the explanation of this phenomenon is that many d-elements are typical
acceptor electrons forming easily complexes with organic and inorganic ligands. The
processes of transitions of electrons and formation of complex compounds belong to
the category of plasma chemical transformation of tribomaterials. The chemical and
active components of lubricating materials, surfactants, are chemisorbed by active
surface centers. It can lead to the surfing film as one of the most effective factors of
wear and friction reduction. The theory of complex compounds can help understand
the mechanism how the surfing film appears.
192 7 Antifriction Micromechanisms

Fig. 7.5 Dependence of


friction coefficient f on
degree of filling of d-orbitals
in atoms of metals during
friction interaction between
metals and silicon carbide [10]

7.3 Complex Compound of Tribomaterials

The complex chemical compounds in the chemistry of complex (coordination) com-


pounds are those with form from simpler ions or molecules without generating a
larger number of electron pairs than the original substances contain. These com-
pounds are capable to exist both in the crystalline and in the dissolved states; their
components are bound by the donor-acceptor links. Reputable quantum chemists
E. Cartmell and G. Fowles insist, . . . it is hard to define the term complex com-
pound formally [11]. It is used to designate the compounds in which one of the
atoms forms more bonds than its valence permits. Schematically the complex origi-
nation can be written in the following way:

M + L M L, (7.8)

where M is the metal central atom (ion); L is the ligand; ML is the complex.
The term ligand serves to designate groups of atoms or molecules surrounding
the central (complex forming) atom. The fields enervate by the ligand perturb the
energy and the wave function of the central atom and provide the complex with its
specific chemical properties. Various types of links are effective in complex com-
pounds: from inclusion of molecules into the crystal lattice to covalent or coordination
links. The coordination links are most often covalent because they are implemented
by combining the electron pair between two atoms. The term coordination means
that the combined electrons belonged originally to one of the atoms. That is why the
synonym of complex compounds is the term coordination compound.
The mathematical apparatus of the theory of perturbations [12] serves to describe
the chemical properties of complex compounds; this theory is the base of the theory of
fields of ligands [1]. Its author is British physicist J. Van Vlech, Nobel Prize winner,
who fulfilled in the 1930s the spectroscopic study of ions of transitional metals into
crystals. The transitional elements differ from the elements in the main subgroups
of the periodic system because they produce many stable compounds with electron
shells partly left vacant. These shells, as it is noted in Sect. 7.2, are d-orbitals and
free atoms, in the first approximation. The triangular, octahedral or more intricate
7.3 Complex Compound of Tribomaterials 193

complexes can form in the transitional elements [11]. Though the coordination links
are most often the property of transitional metals, there is not any prohibition limiting
the bonding among other elements.
The preceding chemical models of the properties of complex compounds explained
the fact by electrostatic interaction between ions of metals and ligands. The com-
bination of fields generated by ligand charges is called the ligand field. This model
explained how complex molecular configuration appeared. The electrostatic theory
was substantially modified to answer the problem of stability of complex compounds
and the effect of the ligand fields on the d-electrons of the central ion [13]. It is
apparent that the energy of d-electrons is determined by two main perturbations:
the electron repulsion and the effect of fields of ligands. The complex compounds
among which the ligand field dominates are called the strong field complexes. The
complexes in which the determination of d-electron levels the electron repulsion
plays main role are named the weak field complexes [1].
The complex formation sis described using the quantum mechanics models of
chemical interaction making up the theory of nucleus and valence. Based on the
content of Sect. 7.2, let us treat the model of transitional states founded by German
physicist F. London in 19291930.
The model of the transitional state is based on the assumption that complex
compounds pass through the stage of the so-called activated complex or through the
non-equilibrium or quasimolecular state. The non-equilibrium activated complex
does not comply with the minimal potential energy. However, estimates with the
methods of statistical physics and probability manifest that the components of the
chemical reaction of generation of complexes are very likely to undergo the activated
complex stage. The process is shown schematically in Fig. 7.6.
The changes in the concentration of substances participating in the reaction
X, Y, Z during chemical interaction are plotted along the abscissa axis called the
reaction coordinate. The ordinate axis corresponds to the variations of the poten-
tial system energy as a result chemical transformations. It follows from Fig. 7.6 that
the activated complex is a less gainful energy state versus the initial state of reagents
and even in worsened state of reaction products which are in the so-called saddle
point [1]. The following formula [1] represents the rate of complex XY generation

Fig. 7.6 Dependence of


energy and reaction coordi-
nate on complex generation:
X + YZ XY + Z
194 7 Antifriction Micromechanisms

during interaction between ions, atoms or radical X with the molecule or ligand YZ:

d
[X Y ] = ch.r [X ] [Y Z ] , (7.9)
dt
where ch.r. is the constant of the rate of reactions calculated with relation (6.20).
In accordance with the Arrenius law, to implement the chemical reaction, the
reagents X, Y, Z should overcome the energy (potential) barrier with the activation
energy Wx (6.20). However, it possible to bypass this barrier. Transformations of
chemical elements follow the laws of quantum mechanics the regularities of which
appear in the so-called tunnel effect. The effect determines the wave function
omitting the zeroing in the points where the full energy of particles W is less than the
potential barrier. This behavior of free particles
  is described in quantum mechanics by
W
the wave function proportional to ex p i t , and, when the particles pass through
  
W U
the potential barrier, the wave function reduces to ex p i t , where U is

the potential barrier value. Due to his fact, even if U > W , the probability of the
particles to be beyond the potential barrier is ||2 > 0 and that is what the tunnel
effect means.
F. London explained the nature of this barrier in the quantum mechanics terms.
During the above reaction, two electrons with opposite spins initially link the atoms Y
and Z with simple bonds, in other words, while the atom X has an unpaired electron.
When the X to the YZ system, the interaction between these three electrons weakens
the YZ bond and atoms Y and Z tend to depart. Hence, the approach of the atom X
to the system YZ increases its potential energy, which is compensated just partly by
new bonds XY. In the final account, the X reaches the point where the attraction of
XY becomes dominating. Then the Z leaves the system reducing its potential energy
(Fig. 7.6). The appearance of the above-described relativistic effects (6.46) boosts
the tunnel effect probability accelerating the complex generation.
Quantum mechanics calculation of the energy of the states of chemical bonds in
the complex compounds is performed with the valence routines based on the electron
wave function found by the main state with the expression [1]:

= N+ [ M (1) L (2) + L (1) M (2)] [ (1) (2) (1) (2)] , (7.10)

where N+ is the normalization multiplier; M , L are the wave functions of the


central ion and the ligand, respectively; , are their spin functions; 1, 2 are the
numbers of electron basic states.
The energy of this system is estimated in the following way:



E= H d V = N2 M (1) L (2) L (1) M (2) H M (1) L (2)
V

L (1) M (2) d V1 d V2 , (7.11)


7.3 Complex Compound of Tribomaterials 195


where H is the Hamiltonian function included into Schrdinger equation (6.41).
If Eq. (7.11) is used to plot successfully the energy surface, it does not mean
that the absolute speed of chemical reaction can be calculated, because they involve
substances with a broad spectral energy range of thermal excitation. Therefore, the
statistical physics should b used in calculation taking into account hat that the reactive
substances are in equilibrium with activated complex.
The mechanism of generation of complexes. During generation of complexes,
compounds have different geometrical configurations depending on the ligand field
type. The ligand field effect on the electron (most often) shell and the central ion
leads to the fission of their energy spectra (Fig. 7.7).
The fission parameter
depends on the ligand and metal ion nature. The ligand
field parameters are determined by the central ion perturbation the external electron
ligand shells. That is why it is justifiable to expect that the factors affecting the value

are comparable with metal ion and its charge. The drawback of this model is that
the treated interactions possess the electrostatic nature and the model ignores the
quantum exchange forces affecting the redistribution of electrons among the energy
levels and formation of covalent bonds. These forces should affect the value
, which,
in the final account, depends on the total intensity of ligand fields, irrespective of
their nature. The analysis of Orgel diagram (L. OrgelBritish physicist) confirms it
for most common octahedral complexes (Fig. 7.8).
The Orgel diagram shows how the relative energy of low- and high-spin config-
urations of electron shells depends on the parameter
. The energy is counted from

Fig. 7.7 Scheme of fission of


d-orbitals of central ion M in
octahedral ligand field L

Fig. 7.8 Dependence of


energy of octahedral com-
plexes with coupled spins
(low spin) and free spins (high
spin) on fission value

196 7 Antifriction Micromechanisms

the energy zero; the mean values are selected of five out of d-orbitals surrounding the
central ion. When the values of the
configuration of the complex compound are
small, the configuration with free spins (the high-spin) configuration is considerably
more stable than the low-spin. As the
grows, the difference between energies of
configurations diminishes and when the value
 is critical, both configurations have
the same energy. When
continues to grow, the low-spin configuration becomes
still more stable than configuration with free spins. Thus, irrespective of distinct
division of the central atoms by the number of coupled electrons, the theory ligand
fields ignore the difference between types of bonds because the orbitals of the close
energy and symmetry participate in the bonding.
Russian tribologist A.S. Kuzharov established [14] that in friction the complex
compound form following the mechanism similar to that in the theory of the acti-
vated complex. The role of the activated complex in the Kuzharov model is played by
the super-excited plasma and post-plasma state of the substance on the tribosurface.
The tribochemical complex compounds are synthesized directly from the substance
active long-living post-plasma states resulting from relaxation in triboplasma. The
tribocontact particles in microvolumes making up the triboplasma interact with tribo-
surfaces and the lubricating material [9]. The microasperities comprising the contact
spots on interfaced tribosurfaces intensify the relaxation processes in the post-plasma
producing coordination compounds on rubbing surfaces. This process is illustrated
in work [9] and in Fig. 7.9. Two complexes being at the apexes of asperities on the
actual contact spot approach with the velocity . The surface is activated in the actual
contact spot producing the triboplasma with the encircle center. The microasperities
leave the direct interface; the triboplasma relaxes and produces in volume new mole-
cular products. The new LCuL complex attaches to the surface by donor-acceptor
interaction and yields a complex compound and pure metal fragments.
The activated complex state corresponds to the ideas of A.A. Polyakov who wrote,
nature contains the equilibrium structures results from competition between order
and disorder and the dissipative structures generated by energy flow remote from the
equilibrium [15]. Polyakov believes that the dissipative structure means that the
tribocouple as reached the mode of self-organization corresponding to the tribounit
structural complication. Referring these structures to the systems of the space-time
type, famous Italian physicist D. Carrierri wrote, The main difference of the closed
in the internal equilibrium state from the system open to the flows of substance and
energy is how it behaves in time. In the equilibrium state, the system stays in

Fig. 7.9 Schematic diagram


of metal complex formation
on steel surface during friction
in selective transfer mode [9]
7.3 Complex Compound of Tribomaterials 197

respect to the time revolution. This symmetry gets immediately disordered if the
effect of external flows moves the system away enough from the equilibrium so that
this new state is apparently more ordered [4]. Carrieri believes thus that after the
transit to dissipative structures is accomplished, the space-time continuum changes
and so does the course of time [16].
The tribosynthesis of complex compounds relates to the activated complex with
the dissipative structure living microsecond fractions and needing the energy to be
continuously pumped up by friction forces. The tribocontact between rough surfaces
and tribocouple materials lasts 104 103 s [16], still, the abnormal intensity of
the reaction of tribochemical synthesis of coordination compounds is emphasized by
many tribologists [8, 9, 14]. One of the models of tribochemical processes attributes
the high activity of reagents to the appearance of inverse population of transitional
states [17]. The inverse population is the substance non-equilibrium state when the
population of upper among a pair of energy levels with one type of atoms, ions or
molecules in the composition exceeds the population of the lower level. It is little
likely; the structural thermal substance activation by friction would the order of
population of atomic shells from smaller to larger energies. Moreover, the inversely
populated media spontaneously emits coherent electromagnetic waves. This fact was
registered experimentally. Versus this model, the hypothesis about the intensification
of tribochemical activity due to the relativistic contribution to the tunnel transition
probability X + YZ XY + Z (Fig. 7.6) is more justifiable.
In addition to formation of complexes from reaction capable substance fragments
on the tribosurface of the lubricating medium, friction initiates also more compli-
cated chemical processes. The central atom of the complex can be substituted with
the counterbody metal. This process of ovecomplication was registered during fric-
tion between the epoxy composite and salycylaldodecylamine copper on the zinc
counterbody using the IR-spectroscopy and X-ray spectral analysis (Fig. 7.10).
The phenomena of ovecomplication are noted within the band 12001800 cm1
as the dislocation with the maximum absorption bands at 2030cm1 within the
long wave spectrum portion. The ovecomplication is confirmed by the coincidence
between the tribosurface copper specimen R-spectra (3) and the models specimen (4),
containing the salycylaldodecylamine zinc complex. The copper substituted with zinc
in the original complex appears on the counterbody as isolated fragments.
The X-ray spectral analysis manifested that copper deposits most on the zinc
counterbody tribosurface during first minutes of friction and it corresponds to the
uppermost local loads and temperature, or the most intensive tribochemical transfor-
mations. Afterwards, the substituted copper surface concentration stabilizes proven
by the constant intensity of the characteristic X-ray emission.
The energy of complex formation reactions. The intensity of tribochemical
interaction is determined by the mechanical energy supply into the tribocontact. It is
worthwhile to estimate the energy consumption efficiency during complex formation
tribochemical reaction. The estimate criterion is the substance quantity gained or
lost due to the friction energy absorption by the tribosurface. Like radiation, sound
and photo chemical reactions, it advisable to introduce the notion of tribochemical
output or the value formally independent of the friction interaction parameters.
198 7 Antifriction Micromechanisms

Fig. 7.10 Findings of spectral study of friction interaction between epoxy composite filled
is the copper complex and zinc counterbody. a IR-spectra: 1 original epoxy resin (ER);
2 ER + copper complex; 3 ditto after friction; 4 ER + zinc complex. b dependence of intensiv-
ity of characteristic copper X-ray emission (CuK ) on zinc counterbody on friction duration

The tribochemical output is the number of particles in the tribochemical reaction


gained or lost per tribosurface unit when the tribosystem absorbs 100 eV of the
mechanical energy [14]:
N 100 eV
T (Y ) = , (7.12)
WFr An Fr

where N is the number of particles involved in the tribochemical reaction; W Fr is


the friction interaction energy; An is the nominal contact; Fr is the tribochemical
efficiency.
Assume that the tribochemical complex forming reactions are due to the post-
plasma effects on the contact spots, it would seem that the actual contact area A F
should be used instead An in expression (7.12). Nevertheless, it is remarked above
that the tribochemical reactions occur over the entire nominal contact area.
Let us call the efficiency and designate Fr the part of the friction energy consumed
directly by the tribosystem to perform tribochemical reactions. The experimental
determination of the tribochemical efficiency faces considerable difficulties, but the
T can be calculated by the parameter i from formula (5.27). The authors of [14]
assess the value Fr 50 times more than the efficiency of other mechanochemical
reactions. The considerably different triboengineering and mechanochemical effi-
ciencies are apparently due to the fact that tribochemical reactions evolve in small
space volumes limited by friction surfaces. The reactions localize in films with the
thickness about the Debye radius R D . It dictates a more efficient interaction between
the metal and the ligand than, for instance, during volumetric reaction activation
7.3 Complex Compound of Tribomaterials 199

Fig. 7.11 Auger electron spectra of epoxy composite specimen: a before and b after friction on
zinc counterbody, relative units

with ultrasound. It is repeated in the work in question [9] that the triboplasma and
post-plasma friction effects accelerate the tribochemical reactions 3008000 times.
Structure of tribochemical complexes. Chemical transformations are studied
in [18] during friction of the epoxy composite on the galvanized steel by Auger
spectroscopy. Figure 7.11 shows the spectra with the absorption bands typical for
zinc oxide (A), the zinc counterbody surface with the chemisorbed oxygen layer
(B) and the original zinc (C). During friction of the epoxy indenter against the zinc
counterbody, the component B augments in response to the reduction of the A
spectrum component. The B kinetics in the Auger electron spectrum corresponds
to the formation of complex compounds.
The structure of these compounds is established in [15]:

It is typical that the epoxy resin is transferred in friction to the counterbody.


The friction tracks under the polarization optical microscope manifest the frag-
ments in the ultraviolet optical spectrum, which cluster near the intergrain boundaries
(Fig. 7.12) [5].
This phenomenon was later in [14] attributed to generation of ions or radicals
clustering on the tribosurface when they are deposited from the ion post-plasma
phase registered with the atom force microscopy (Fig. 7.13).
This structure appear in friction as a regular stage in the evolution of relaxation
processes in triboplasma leaden to reduction of the friction coefficient and wear of
rubbing bodies. The nucleation of clusters during friction interaction can be discov-
ered in the electron spectrum registered with the method disclosed in Sect. 5.5 by
variations of the harmonics intensity in the gigahertz band (Fig. 7.12b): one harmonic
relates to electrons, the other to particles with the positive charge. in accordance with
relation (5.38), the mass of this particle is of the order of magnitude 1016 kg, the
charge determined from relation (5.39) and the condition of the triboplasma charge
200 7 Antifriction Micromechanisms

Fig. 7.12 Photographs of zinc counterbody tribosurface: a in usual light; b in polarized light

Fig. 7.13 Photographs of nuclei of third bodies (nanoclusters) during selective transfer [19]

neutrality, corresponds to 106 Coulomb, i.e., to 1013 elementary charges. It is con-


cluded that this system is rather a cluster than the molecular ion. The cluster moves
under the effect of force (5.104), while the combined motion of clusters is described
by magnetic hydrodynamics equation (5.120).
Depending on the grease composition, the tribosurface chemical properties, the
duration and conditions of friction interaction, both inorganic products and organic
tribopolymers can deposit on the tribosurface when they are produced by reactions
of mono- and poly-nuclear coordination of metallic compounds (Fig. 7.14).
A.S. Kuzharov calls the complex formation in friction the tribocoordination mean-
ing the orientation of products of synthesis in respect to the tribosurface. It is estab-
lished in [9] that the active hydrogen atoms are the essential condition for formation of
complex compounds in friction. If they are replaced, for instance, with alkyl or alkyd
radicals, the complex formation stops. The ligand diffusion towards the tribosurface
is a limit tribocoordination stage.
7.3 Complex Compound of Tribomaterials 201

Fig. 7.14 Appearance of polynuclear coordination compounds of metals in friction. Designa-


tions of elements in structural formulas X = O, S, Se NH; Y = O, N; Z = CH, N; R = Al, Ar;
M = Cu, Ni, Fe [9]

The practical independence of the complex formation rate on the lubricating mate-
rial temperature and zero activation energies corresponding to the formal Arrhenius
law application to the description of interaction between ligands and the metallic
counterbody, prove that the energy needed for the chemical reaction is imparted to
substances during dynamic contacting between tribosurfaces. The tribocoordination
accelerates as energy that is more mechanical is supplied up to some limit value
depending on the ligand concentration. It agrees with the known tribochemical rela-
tion between the tribochemical reaction rate and the initiating mechanical energy [8].

7.4 Tribocoordination or Surfing Effect

It is shown in Sect. 4.2 that the dynamic evolution complicates the tribosystem,
reduces the entropy factor and forms third bodies. They appear due to the physical
and chemical processes in the tribocontact, the complex formation laying a par-
ticular role [20]. The film of complex compounds demonstrates high lubricity and
202 7 Antifriction Micromechanisms

easy sliding of tribounits like the surfboard sliding. This effect has a prototype in
fundamental physics under the surfing effect term or limited gravitation wave velocity
during interaction between gravitation and electromagnetic fields [21].
Physics and chemistry of complex compounds. The tribosurface is free of oxide
films in the selective transfer mode (ST), while the dislocations of atoms of these
active centers exit to the surface and possess additional bonds. Almost all metals are
acceptors of electrons, while the majority of additives or products of tribodestruction
of tribomaterials are donors of electrons. They act as ligands favoring the formation
of complex compounds. It is a typical attribute of ST formation of chemical bonds
between the lubricating material ligands and metal atoms on tribosurfaces [15].
G.D. Gromyko investigated the lubricity complex compounds [22]. He discovered
the positive effect o oxidation processes on the lubricating function of complexes
and explained it by the growing thermodynamic stability of the appearing complex
compounds. The effect of the oxidation processes on the surface kinetics of the chem-
ical reactions is well known [6]. The surface reactions often involve the intermediate
structures, which are insignificant at the chemosorption stage, but the affect the chem-
ical reaction rate and can favor the appearance of intermediate reactive groups. The
surface chemical reactions evolve predominantly (or exceptionally) in one group of
active centers [3]. The metals with stable oxides have typically the oxygen contain-
ing ligands. These complexes agglomerate near the intergrain boundaries being the
active centers with high adhesive energy. If the lubricating medium is free of products
of oxidation of hydrocarbon, the metal organic compounds appear in which the metal
bonds to the hydrocarbon through the carbon atom. These compounds are unstable
and disintegrate at rather low temperatures.
The authors of [18, 23] perfumed one of the fundamental experimental research
of the tribostructure based on complex compounds and linked the processes of their
formation to oxidation reactions. They studied the friction of epoxy compounds on the
zinc counterbody. The compositions of the epoxy specimens contained the mercupral
(the copper tetraethyl tiuramdisulfide complex) together with its ligand and catalysts
or inhibitors of oxidation processes. The inhibitors of oxidation reactions were KBr
and Vaseline oil, the catalyst was KMnO4 . The triboengineering tests manifested that
introduction of the additives suppressing the oxidation processes in the tribocontact
improves the tribocouple friction resistance. It is particularly apparent when the
Vaseline oil is used because it possesses its own lubricity when it is used together
with the ligand (Fig. 7.15a). Even a sight amount of the KMnO4 oxidant increases
the friction coefficient and intensifies polymeric material wear (Fig. 7.15b).
The measurements of the tribological parameters couple in question are due to
the effect of additives on the kinetics of third body appearance. This process and
relevant tribochemical transformations were studied with X-ray and Auger electron
spectroscopy [7, 24]. The X-ray electron spectrum permits to judge about the sur-
face qualitative and quantities elementary composition and the valence state of the
elements in question (Zn, Cu, S). The radiation transition of the elements in question
into the p-shell (zinc Zn2p, copper Cu2p, sulfur S2p) permits to judge about their
valence state by changes in the typical changes of spectral lines and the relevant type
of the appearing chemical bonds. The most informative are the long wave zinc L
7.4 Tribocoordination or Surfing Effect 203

Fig. 7.15 Dependence of wear I (1, 2) and friction coefficient f (1 and 2 ) on concentration of
introduced oil (a) and oxidants (b)

and M-spectra (ZnL3 M45 M45 ) taken from the metallic counterbody surface. These
lines are produced by transition of electrons to their external shells close to the atom
boundary so that the spectral lines become extremely sensitive to changes in the
chemical environment. The spectral factorization into characteristic components
provides information about the type of chemical compounds and heir concentration
on the surface [25].
The conclusion from the results of the X-ray electron analysis of the copper
and sulfur spectra is that the ratio of their concentrations on the tribosurface changes
constantly and they do not corresponds to the ratio between molecules in the complex.
The complex compound appearing in friction has the copper univalent unlike the
bivalent copper in the mercupal. It is the experimental proof of ovecomplication,
which, in this case, is caused by copper substitution in the complex and a proof also
that the bonds of mercupral molecules rupture.
The Auger spectrums of the electrons leaving the film with the thickness equal
to the length of the free run of electrons have the energy about 992 eV. When the
specimen turns, the angle changes between the direction in the analysis and the
direction normal to the specimen (Fig. 7.16) permitting the spectral analysis of film
layers located at different depth.
The free run length of zinc electrons according to [23] is under 1.3 nm. Taking into
account that there is polymeric transfer layer on the zinc counterbody tribosurface,
the free electrons run to a longer distance than in metals; it can be assumed that the
total analysis does not exceed 3 nm.

Fig. 7.16 Arrangement of


X-ray tube (1), slot into
analyzer (2) and analysis of
counterbody (3)
204 7 Antifriction Micromechanisms

Fig. 7.17 Tribosurface Auger spectra: a, b surface indenter; c, d counterbody; a, c = 30 ;


b, d = 80 (in conditional units)

The Auger spectra of transfer films depend little on the composition of the poly-
meric specimen material (Fig. 7.17). In most cases, two components can be identified
in the spectra: A corresponds to the zinc oxide, B correspond to the zinc oxida-
tion intermediate stage. Sometimes the spectrum contains the third component C
corresponding to the Zn metallic state.
The structure of the zinc counterbody surface layer before friction can assume the
following: the top layerzinc oxide, the layer beneath is the oxidation intermediate
stage and then follow the zinc proper. The total thickness of the system in question
does not exceed 1.3 nm. The thickness of each sublayer of the proportional area of
the relevant spectral components changes weakly over the specimen surface. The
tribosurface layer corresponding to the relevant oxidation intermediate stage does
not always carry the zinc oxide layer. It is established by changing the angle of
installation of specimens as shown in Fig. 7.17. The spectra registered at = 30
show that the deeper layers contribute to the Auger spectrum, while the surface layers
contribute at = 80 . The spectral component B intensifies when the counterbody
is at = 80 , corresponding to the zinc oxidation intermediate stage. It means that
this layer lies on the zinc oxide surface layer. If it is taken into account, the zinc in
7.4 Tribocoordination or Surfing Effect 205

the complex compound has exactly the oxidation intermediate stage, the component
B proves that a zinc complex compound is under formation.
Numerous experiments have permitted to establish the following regularity: the
antifriction characteristics when the Auger spectral band of absorption widens char-
acterizing the zinc oxidation intermediate stage and proving that the metal complex
concentration augments. When the complex compound layer doubles, the friction
coefficient reduces four times to the value typical for hydrodynamic friction (0.06).
It is shown in [26] that the metal complexes appear both on the metallic tribocouple
surface and on the polymeric surface, including on that made from most chemically
inert fluoroplast-4. The copper positively effect is observed on the triboengineering
characteristics of materials containing PTFE fibers. This effect is apparently due to
the appearance of copper coordination compounds together with products of copper
destruction of macromolecules in the friction zone according to the following scheme:

It can be assumed that adsorption and chemical fixation (chemosorption) of copper


complex compounds occur on the polymeric specimen surface during the friction
interaction, and then they are transferred to the metallic counterbody. The following
alternative are proposed in [26] how the boundary lubricating layer appears on the
steel surface (Fig. 7.18): athe complex compound adsorbed on the fresh (juvenile)
steel surface, bthe juvenile surface is coated in advance with the monolayer of
copper atoms on which the complexes are adsorbed.

Fig. 7.18 Scheme of complex formation in tribocouple PTFE-steel: a formation of adsorption


between iron atoms, b with atoms of copper transfer layer
206 7 Antifriction Micromechanisms

Surfing film. The complexes adsorbed on the tribosurface are associated with
stable structures named the surfing films. The complex molecules are held in the
surfing films by intermolecular forces demonstrating an extremely high organization
of coordination saturated particles called the supermolecule [27]. The surfing film
is a protective adsorbed based on the associated complex compounds. The latter result
from the chemical compound of free metal atoms in the spots where dislocations
exit to the tribosurface with lubricating material ligands [28]. The surfing film per-
forms the functions of mass exchange between the third body and the environment
and possesses the property of self-restoration. Being destroyed under the effect of
friction forces and reappearing again, it remains stable withstanding the processes
of tribodestruction and oxidation. If the grease consisted only of the surfing films,
it would be eternal needing no replacement. Regretfully, it is impossible so far to
make the material with surfing films in advance so that the films would be supplied
externally into the tribocouple. In practice the oil soluble complex compound and
the materials capable to produce complexes in the lubricating medium, are used as
additives to lubricating materials.
Let us consider the mechanisms of formation of surfing films. It is noted above
that the ST creates favorable conditions for producing coordination compounds,
because the dislocations exit to the tribosurface. The dislocations bear considerable
energy providing grounds for development of the dislocation theory of seizure of
metals. However, a paradoxical reaction of chemosorption of lubricating transfer
layers occurs during the ST; this effect was named the surfing effect.
The chemical and active components of the lubricating material or ligands appear
in the spots where the dislocations exit to the tribosurface of metal complexes. When
the coordination bonds appear, the bonds between metal atoms and crystalline lattice
atoms weaken or rupture. The bonds between the metal and the complex atoms
transform into the chemisorbed state. The complex compound molecule becomes
able to diffuse over the tribosurface combining with other molecules into the film
islands schematically shown in Fig. 7.19a. The processes of film appearance and
tribodestruction mentioned above evolve with approximately the same rate dictating
the dynamic stability of the surfing structure. The analysis of images made with
the scanning tunnel microscope (STM) [29] show that the surfing films possess the
quasicrystalline structure different from the tribosurface structure on which they
appear. Figure 7.19b, shows the photograph of the counterbody surface after friction
of the epoxy specimen containing mercupral.
The surfing film has the structure of the ordered grid of atoms with atom spacing
1.9 nm.
The surfing films are similar to those appearing in friction in the ST mode. They
produce collective adsorption function plastifying the metal on the tribosurface usu-
ally inhibited during common friction by oxide layers. Therefore, the friction pro-
duces denser (doubled or several times higher) concentration of vacancies sharply
localizing he charge carriers charge and determining the variety of chemosorption
mechanisms. The chemosorption accelerates as the temperature and pressure grow
resulting in a higher reliability of tribocouple within the range of working tempera-
tures reducing the wear rate and boosting life of lubricating materials.
7.4 Tribocoordination or Surfing Effect 207

Fig. 7.19 Surfing film portion: a scheme of formation; b STM-photograph of surfing film formed
by copper complex [30]

The range of velocities and loads during ST remains the most essential and still
outstanding problem because this range governs the appearance of dissipative struc-
tures, which reduce the friction and wear parameters by several orders of magnitude.
The servovit film dissipative structure evolves following the vacancy dislocation
mechanism usually producing the surfing structure. The surfing film results from
reaction (7.8) implying the activity of ligands initiated by the atoms on the tribosur-
face with free bonds. The ligand can interact with several atoms in the lattice at once
making the total interaction very strong. It favors clustering adding to the process
special volumetric and surface characteristics that are absent among other forms
of existence of solid bodies. The processes of clustering of complex compounds
relate to the interaction between triboplasma ions and motion of dislocations. The
more dislocations surface, the stronger precipitation of ions, the more intensive is
the clustering and so are the processes of formation of surfing films.
The clustering of complex compounds is possible only when the tribosurface is
free of oxide films. The complex would rather appear on the juvenile surface than
the oxide. The chemical structure of the new complexes lacks the component of
interatomic bonds, irrespective the considerable oxygen content. When there are no
hydrocarbon ligands, oxygen plays their role, which produces bridges binding lig-
ands to the tribocouple metal [31]. The oxygen in the surfing film combines with
metals without oxidizing them because extremely strong covalent bonds are pro-
duced confirmed by the Auger spectroscopy results shown in [23]. The oxide films
208 7 Antifriction Micromechanisms

inhibit evolution of the surfing structure impairing the antifriction characteristics of


tribocouples.
The oxide layer is known to reduce the probability of overcoming by the electron
of the potential barrier in chemical and inhomogeneous systems, for instance, in
the surfing filmsolid body substrate interface. The metal oxide on the surface
created by the oxide layer divides the lubricating film from the surface executing
the tunnel or over barrier mechanism [32]. It decelerates the rate and reduces the
probability chemical reaction between the complex forming ligand and the surface
layer atom. The surfing film sliding (its propagation in fact, due to formation of
chemical bonds with the adjacent film molecules or the surface layer atoms) over the
solid body polycrystalline surface demands to overcome the potential barrier across
intercrystallite boundaries. The thickness of intercrystallite boundaries is comparable
with the de Broil wavelength in the friction excited metal structures with mobile
intergrain boundaries. It can be assumed, as a rule, that the energy of activation
of electron motility when overcoming the potential barrier is reduced following the
thermal mechanism and additionally to the tunnel mechanism. Generally, it augments
the probability of surface chemical reactions. It can be assumed from the above-said
that the tunnel effects contribute quite considerably to the tribochemical synthesis of
complex compounds and to the entire tribosurface coating.
The structure of the complexes appearing on the tribosurface is quite variable
and not all possible configurations can be enumerated because o the large amount of
ligands. The configuration of molecules in complexes affects substantially the friction
parameters of tribocouples constituting the essence of the tribocoordination notion
in the general sense. The least friction corresponds to the orientation of molecules
in coordination compounds, which makes the plane face the tribosurface instead of
the angles and ribs of molecules, as Fig. 7.19a, shows.
It should be noted that, in many cases, the surfing film is added with the tribopoly-
meric, colloid and other third bodies. The lubricating behavior of colloid particles
is due to their ability to co-exist together with the adsorbent on the tribosurface.
The colloid particles are one order of magnitude thicker than the surfing film; they
attach loosely to the friction surface producing an additional sliding plane in the
tribocontact zone. Due to this fact, the surfing films had been doubted for a long
time. The surfing film was assumed the servovit film making it hard to estimate the
surfing effect on the reduction of friction coefficient and wear of the tribocouple. The
metal masking the surfing film as it were the servovit film followed from formula
(7.9), Figs. 7.6, and 7.10. The structural hierarchy of lubricating layers produced on
the tribosurface during selective transfer of the steel on copper alloy tribocouple is
shown in Fig. 7.20.
Similar structures were observed during friction of tribocouples metal-complex
containing polymeric composite. Their structure is similar to the multilayered het-
erochemical structure shown in Fig. 7.19b. The identity of structures of third bodes
resulting from the surfing effect can a manifestation of some fundamental mechanism
of self-adjustment of friction and wear processes.
7.5 Additives Based on Complex Compounds 209

Fig. 7.20 Servovit film


adsorption layers: 1
copper alloy, 2 servovit film,
3 chemisorption layers on
servovit film, 4 lubricating oil
colloid layer in tribopolymer,
5 steel

7.5 Additives Based on Complex Compounds

Most of the additives based on the coordination compounds of metals are referred
to the metal cladding additives because the lubricating behavior is based on the
interaction between the metallic quasimetallic films in the interface. The doping
of lubricating oils with metallic organic complex compounds produces a positive
effect [33]. The results of numerous works in the sphere of chemmotology prove
that the lubricating compositions containing metallic organic or complex com-
pounds possess much better triboengineering properties than the base lubricating.
The following cladding amine complexes of transitional metals are advisable for
the base lubricating oils: [M(L)x Am] [33], where M is the transitional metal cation;
LH2 NRNH2 , H2 N(RNH)n H, Hp N(CH2 CH2 OH)3p ; RCH2 or (CH2 )8 ; A
Cl , SO2
4 , CH2 COO , n = 1. . .4; p = 0, 1, 2; x = 2, 3; m = 2, 4.
The introduction of 10 % of this complex into Lithol-24 reduces the linear wear
of metallic bearings almost to zero and the friction coefficient drop lower than
0.06. Similar effects are observed when the complex with nitrogen-, oxygen- and
sulfur-ligands (RR R N)y MXz are introduced in the base lubricating oil, where
R = C6 C30 , R , R = H or C1 C30 ; M = Sn, Cu, Cd, X = Cl, J, Br, y = 16,
z = 24.
The additives were developed in the USA based on the so-called binuclear com-
plexes of unlike metals or blended complexes of these metals M1 M2 (A)x m(B Ry ),
where M1 and M2 are Cb, Cu, Ag, Zn, Hg, Ni, Co, Fe, Pb etc.; are chloride, bro-
mide, azide, thyocyanate, sulfate, carbonate, etc.; BR are oxides, sulfides, selenides,
tellurides, saturated, unsaturated, cyclic or aromatic hydrocarbon radicals contain-
ing from 4 to 20 carbon atoms. The example of the wear resistance additive is the
following copper complex of the acetylsalicylic acid (Fig. 7.21):
The effect of this additive on the antifriction characteristics of the low-temperature
grease CIATIM 201 is shown in Fig. 7.21 [33].
The chelates of metals have manifested their effective antifriction property many
times. When this complex of free ligands is added with the above chelates, the results
become the best because then the constant metal complex concentration is maintained
210 7 Antifriction Micromechanisms

Fig. 7.21 Chemical formula of Cu complex of acetylsalicylic acid

Fig. 7.22 Dependence of friction coefficient on working time (h) of tribocouples steel 40 Kh
cast iron SCh-15 for lubricating materials: 1 CIATIM 201, 2 CIATIM 201 + 5 % copper salt (II)
acetylsalicylic acid c

in the friction zone. The lubricating compositions of the base oil with additives of
2-oxybenzaldodiecylamine (a) as the ligand and its chelate compounds (b) exemplify
the improved lubricating:

The mixed powdered d-element and the complex former are often added to the
lubricating material. In this case, the metallic complex forms directly in the friction
zone due to the tribochemical transformations.
Chelates of metals containing the molybdenum as the central atom are particularly
effective. They are added into the basic grease in the concentration just 0.52 % by
volume. Molybdenum dithiocarbonates resulting from tribodisintegration of molyb-
denum disulfide possess excellent lubricating properties, for instance, molybdenum
disulfide tridithiocarbonate MoS2 (SCNR2 )3 , where R is the radical.
The antioxidizing additive based on the mixture of copper and molybdenum
dithiocarbonates manifests good results.
7.5 Additives Based on Complex Compounds 211

Due to the book-limited content, it is impossible to list just the main additives
from among their numerous varieties based on the metallic complexes. It is worth-
while to highlight that many of them, in addition to their main purpose, are effective
additive to polymeric self-lubricating compositions. One of the best home plastics
for triboengineering purposes is KGM-3 containing the mercupral as the antifriction
additive [34]. Complex compounds of metals yield a stable antifriction effect in all
types of tribocouples. It is because surfing films and buffer layers [33, 35] appear
containing tribopolymers.

References

1. D. Marrel, S. Kettle, D. Tedder, Chemical Bonds (Mir, Moscow, 1980), p. 382


2. A.S. Akhmatov, Molecular Physics of Boundary Friction (physical and mathematical literature,
Moscow, 1963), p. 472
3. E. Zenguil, Surface Physics (Mir, Moscow, 1990), p. 536
4. D.N. Garkounov, Scientific Iscoveres in Trboengineering. Wearlessness Effect. Hydrogen Wear
of Metals (Moscow Agrarian Academy, Moscow, 2004), p. 384
5. D.N. Lyubimov, G.G. Chigarenko, A.G. Ponomarenko et al., Study of formation of lubricating
layer by dry friction of metal-polymer systems. J. Frict. Wear 13(2), 368372 (1992)
6. Kh.X. Chikhos, System Analysis in Tribonics (Mir, Moscow, 1982), p. 348
7. D.N. Lyubimov, V.A. Ryzhikov, Processes in Friction: Manual (South Russia State Engineering
University, Novochherkask, 2006), p. 147 (Phys. Chem.)
8. G. Heinke, Tribochemistry (Mir, Moscow, 1987), p. 635
9. A.S. Kouzharov, Coordination tribochemistry of selective transfer. Doctor of Engineering Sci-
ence, Disssertation. Rostov-on-Don, p. 42, (1991)
10. D. Bakley, Surface Phenomena During Adhesion and Friction Interaction (Mashinostroyenie,
Moscow, 1986), p. 360
11. E. Cartnell, G.V.A. Fowls, Valence and Structure of Molecules (English translation) ed. by
M.V. Bazilevsky (Chemistry, Moscow, 1978), p. 360
12. L.D. Landau, E.M. Livshits, Quantum Mechanics. Non-Relativistic Theory, vol. 3 (Nauka,
Moscow, 1974), p. 464
13. L. Orgel, Introduction into Chemistry of Transitional Compounds (Theory of Field of Ligands)
(Mir, Moscow, 1967), p. 452
14. A.S. Kouzsharov, Y.I. Reboukhin, Complex formation in friction. J. Frict. Wear 12(1), 99107
(1991)
15. A.A. Polyakov, Self-orgnization of selective transfer structure, in Durability of Machinery
Rubbing Parts, vol. 3 (Mashinostroyenie, Moscow, 1988), pp. 4595
16. D.N. Lyubimov, K.N. Dolgopolov, Friction and Theory of Relativity: Time Abnormalities
Tribosystems (Fbgou Vpo Yurgues, Shakhty, 2011), p. 126
17. V.I. Kolesnikov, S.B. Boulgarevitch, M.V. Boyko, K.E. Korets, E.E. Akimova, Population
of transtional states during friction activated chemical reactions and probale causes of poor
temperature dependence on the rate of these reactions, in Proceedings of the X International
Conference Tribology and Reliability, (Saint Petersburg State University of Communications,
Saint Petersburg, 2010), pp. 4149
18. A.T. Kozakov, D.N. Lyubimov, A.V. Nikolskii et al., Study of frictional interactions between
surfaces of materials during selective transfer in epoxy compositemetal couple. J. Frict. Wear
13(5), 918924 (1992)
19. Y.P. Kosogova, Nano electric chemical technologies of implementation of wearlessness effect
in wateralcohol media. Candidate of Engineering Science. Rostov-on-Don, p. 22, (2009)
212 7 Antifriction Micromechanisms

20. Ch. Kittel, Introduction into Solid Body Physics (Mir, Moscow, 1978), p. 791
21. A.G. Polnarev, D. Bascaran, Surfing effect in the intersection of electromagnetic and gravita-
tional waves. Phys. Rev. D. 77(10), 3343 (2008)
22. G.D. Gromyko, The effect of oxydation processes on severe boundary friction, in Theory of
Friction, Wear and Lubrication, vol. 3, pp. 124125, Tashkent, 1975
23. A.T. Kozakov, D.N. Lyubimov, A.V. Nikolskii et al., Effect of third body structure on frictional
characteristics of epoxy compositemetal couple. J. Frict. Wear 13(6), 10321038 (1992)
24. D.N. Lyubimov, K.N. Dolgopolov, Difusive Processes in Friction (IP Bouryhin V.M., Shakhty,
2010), p. 148
25. A.T. Kozakov, Physical fundamentals of electron spectroscopy of charged surface of solid
bodies. Rostov-on-Don: YUFU, p. 406, 2009
26. V.N. Kovalev, V.O. Grechko, V.G. Redchenko, Mechanism of lubrication. Development and
study of triboengineering properties of modified materials based on PTFE fibers, in Special
Purpose Antifrictional Plastic (NPI, Novocherkassk, 1987), pp. 110117
27. Sh.M. Len, Supermolecular ChemistryScale and Prospects, in Nobel Lecture. Chemistry,
vol. 2 (Moscow) (1987)
28. A.A. Polyakov, Terminolgy of selective transfer, in Durability of Machinery Rubbing Parts,
issue 4, pp. 1115, Mashinostroyenie, 1989
29. D.N. Lyubimov, K.N. Dolgopolov, Modern Tribology (Selected Chapters), (IP Bouryhin V.M.,
Shakhty, 2010), p. 181
30. D.N. Lyubimov, K.N. Dolgopolov, Micromechanisms Controlling the Structural Evolution
of Tribosystems. Nanocomposites with Unique Properties and Applications in Medicine And
Industry, ed. by Dr. John Cuppoletti, (Tech Open Access Publisher, Croatia, 2011), pp. 83-112.
ISBN 978-953-307-351-4
31. D.N. Garkounov, Triboengineering (Mashinostroyenie, Moscow, 1985), p. 424
32. V.F. Kiselev, S.N. Kozlov, A.V. Zoteyev, Fundamentals of Solid Body Surface Physics (INBU,
Moscow, 1999), p. 284
33. A.S. Kouzsharov, N.Y. Onischuk, Metal cladding lubricatng materials, in Durability of Rubbin
Parts of Machins, issue 3, (Mashinostroyenie, Moscow, 1988), pp. 96144
34. D.N. Lyubimov, Role of diffusion processes in wear and friction of metal-polymer tribocouples.
Candidate of Science Disseration. Gomel, p. 22, 1990
35. E.S. Boushkova, A.D. Garnovskii, Coordination compounds of metals containing organic lig-
ands as effective additives to materials, in Wearlessness, pp. 151170, Rostov-on-Don, 1990
Afterword

Authors used mathematical apparatus and methodologies of great physical


theoriesgeneral and special relativity, quantum mechanicsto triboprocesses
description. It provides a way of estimating a lot of analogies between tribology
processes and phenomena described in physical theories confirming the unity of
material world.
We have endeavored in this book to present the tribology fundamental
principles relating to the main theories of physics condensed state and to explain
the still rather obscure triboprocess stages using the well-known ideas about the
physical effects and mechanisms never so far applied virtually in the science of
friction. The tribophysics can develop from its stage of nucleation it has been
staying in since the mid of the twentieth century exactly involving the fundamental
laws and theories embracing all laws of physics and chemistry.
We paid particular care to the investigation of the substance microstructure and
the mathematical physics theoretical apparatus. It has enabled, in accordance with
the Galileo postulate about validation of true points issuing the said earlier, to
construct models of transformations of substances during triboprocesses from the
simplest electron-ion triboplasma forms to the highly organized third body
supermolecules like surfing films. We have endeavored to apply adequately the
mathematical apparatus and the methods of great physical theories: STR, quantum
mechanics, relating to the description of triboprocesses. It enabled to identify
many analogies between the described tribological processes and the physical
theory of the phenomena confirming the material world unity that the present work
touches may be too cautiously. In general, we are quite glad with the results and
hope that our work will motivate colleagues to develop the ideas in the sphere
where tribology, magnetic hydrodynamics, statistical physics, electrodynamics,
relativistic quantum mechanics, principle of symmetry and the laws of
preservation, meet. It can be expected that this crossing between sciences will
yield the results distinguishing the tribophysics and contribute gloriously to
tribology.

D. Lyubimov et al., Micromechanisms of Friction and Wear, 213


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8,
 Springer-Verlag Berlin Heidelberg 2013
Index

l-mesons Boundary friction, 16, 188, 190


Burgers vector, 4142, 45, 4850, 8387,
9192, 113
A
Activated complex, 193, 195, 196197
Adhesiometer, 56 C
Adhesion, 14, 18, 5556, 99, 106, 128, 186 Ceramic material, 15, 24
Adhesive center, 187 Chelates of metals, 209
Adiabatic invariants, 166 Chemical composition inhomogeneity, 38
Adsorption, 56, 5761, 106107, 187188, Chemical locking, 88
190, 205, 206, 208 Chemical potential, 19, 57, 74, 76, 7879
Adsorbed atom (adatom), 106, 107 Chemosorption, 57, 58, 173, 202, 205206
Airy function of stresses, 44 Climbing of dislocations, 111
Alfvn wave, 148, 165 Cluster, 56, 59, 111, 171, 187188, 199200,
Ambipolar diffusion, 145 207
Angle of disagreement, 46 Coherent (semicoherent, incoherent) boundary
Anisotropy, 22, 51, 145 phase, 6263
Arrhenius equation, 172 Cohesion, 1, 1718, 55, 123
Atomic orbitals, 9, 15 Complex chemical compounds, 192
Atomic planes, 21, 27, 53 Compressibility, 14, 1718, 35, 71
Atomic radius, 1920 Conformations, 26
Atomic volume, 14, 19, 113 Connective, 9
Atom nucleus, 6, 1921, 35, 89 Connective orbital, 9
Contact spot, 9899, 129, 130, 134, 136, 155,
167, 188, 196, 198
B Contact spot microhardness, 129
Barodiffusion coefficient, 75 Coordination links, 192
Black hole, 124, 182 Coordination number, 19
Body-centered cubic (BCC), 24 Creep, 8990
Bohm diffusion, 144 Crystal, 5, 1015, 1733, 3742, 44206
Bohm diffusion coefficient, 144 Crystal volume, 31, 38, 4546, 68, 8183, 87,
Bohr radius, 35 114

D. Lyubimov et al., Micromechanisms of Friction and Wear, 215


Springer Series in Materials Science 176, DOI: 10.1007/978-3-642-35148-8,
 Springer-Verlag Berlin Heidelberg 2013
216 Index

C (cont.) Electrons of conductance, 13


Crystalline lattice, 14, 20, 2223, 2830, 33, Electrostatic locking, 88
3739, 48, 5153, 65, 72, 78, 81, 86, Elementary (primitive) cell, 19, 23, 2628, 50,
88, 90, 92, 98, 107, 113114, 122, 206 53, 68
Crystallites, 2627, 33, 48, 50, 81, 82, 106, Energy of Gibbs, 21
128, 129, 208 Energy spectra, 195
Crystallization, 15, 2021, 22, 40, 47 Entropy, 2021, 51, 78, 80, 102105,
Current conductive, 13 123124, 191, 201
Euler equations, 153
Exchange forces, 5
D Extrinsic inclusion, 46
Debye radius, 130132, 134, 137, 145, 147,
198
Debye sphere, 132 F
Defects, 38 Face-centered cubic (FCC), 24
Deformation, 25, 39, 4345, 52, 68, 71, 73, 76, Far order, 37
8389, 9699, 107, 111, 113, 115117, Fermi energy, 31
128130, 165, 172 Fermi level, 32
Deformation component of friction energy, Fermi temperature, 30, 54, 131
129 Field effect, 132
Degree of substance ionization, 136 Fission, 195
Density, 68 Fluid line, 155
Density of the absorbed energy, 130 Forces, 3
Desorption, 56, 60, 110 Forces of gravitation, 123
Dielectric permeability, 16, 147148 FrankReed dislocations, 39
Diffusion, 6061, 74, 7682, 90, 101102, Frenkel defect, 42, 46, 111
104115, 117, 118, 129, 133139, Friction additive law, 96
155156, 169171, 176, 201 Friction coefficient, 96, 97, 106, 107, 128, 160,
Diffusion flow, 75, 7678, 81, 102, 108110, 185, 186, 188189, 191192, 199,
138139, 145 202203, 205, 208, 209
Diffusion length, 76, 110, 117, 133, 137, 175 Friction force, 95, 96, 102, 105, 129, 133, 135,
Dimers, 53 139, 181, 185, 187, 188, 197, 206
Dipole, 2, 54, 60, 115 Friction oscillations, 174
Dipole moment, 2, 150152 Friction resistance, 104, 107, 160162, 202
Dipole-to-dipole interactions, 2 Friedel oscillations, 54, 165, 170
Dislocation dipole, 115
Dislocation loops, 128
Dislocations, 22, 38, 4046, 4853, 61, 67, 81, G
8290, 98100, 107, 110120, 121, Gel model, 5354
122, 126127, 165, 181, 187, 190, 201, Gibbs energy, 21, 56, 77
206, 207
Dislocation steps, 61
Dispersion, 3 H
Dispersion equation, 147, 179 Habitus, 22
Dissipative structure, 190, 197, 207 Hamiltonian, 6, 8, 3031, 167, 195
Donoracceptor bonds, 55, 192, 196 Hardness, 12, 28, 96, 100, 113, 129, 137
Drift frequency, 144, 147 Hartree-Fok method, 8
Dufour effect, 79 Heat conductive, 13
Dupret equation, 64 Herring equation, 64
Hexagonal densely packed (HDP), 130
Homogeneous-chain and heterogeneous-chain
E polymers
Elastic locking, 88 Hot spots, 130
Electron clouds, 35, 1213, 28, 108 Hybrid orbital, 12
Index 217

Hydrogenation processes, 108, 111 Magnetic sound waves, 148


Hydrogen-related wear, 107 Magnetic susceptibility, 34
Mechanisms of diffusion, 80
Melting point, 12, 24, 33
I Metallurgical effects, 128
Incoherent phase boundary, 63 Method of a linear combination of atomic
Induced moment, 2 orbitals (LCAO), 9
Intergrain boundary, 48, 4950, 63, 8182, Method of valent schemes, 9
106, 128, 199, 202, 208 Microcrack, 4647, 108, 115
Intermetallides, 15 Miller indexes, 27
Intersecting processes, 79, 104, 155, 190 Mixed boundaries, 50
Inverse population, 197 Model of the transitional state, 193
Ion motility, 137 Molecular orbitals, 910
Ionization potential, 3 Monochromatic approximation, 149, 156, 157
Motion of dislocations, 84
Multipole, 186187
J
Juvenile surfaces, 99
N
Near order, 37, 89
K Nodes of coincidence, 50
Keesom effect, 2 Nominal contact area, 96, 159, 198
Kinetic energy, 14, 28, 31, 32, 80, 85, 134, 171
Kirkendall effect, 81
KleinGordon equation, 179 O
Kottrell atmosphere, 4546, 128 Open system, 78, 103104, 124, 191
Orgel diagram, 195
Orientations forces, 12
L Ovecomplication by friction, 197, 203
Langmuir waves, 145, 147
Large-angle boundary, 48, 50
Larmor frequency, 151 P
Larmor orbit, 150 Parameter of perfection, 132
Latent boundary, 62 Particle containment time, 139
Lattice constant, 17, 80, 88, 113 Pauli equation, 170
Ligand, 13, 191196, 198, 201202, 206, Pauli exclusion principle, 3, 5, 7, 10, 1213,
207208, 209 2829, 57
Linear element, 41, 87 Peierls stress, 87, 89
Linear, ramified and reticular structures, 25 Perfect crystal, 23, 29, 3738, 48, 5254, 90
Local time, 95, 123 Performance of the tribochemical reactions,
Locking mechanism orders the field of 136
stresses of dislocations, 89 Perturbation theory, 7
Loosening orbital, 9 Phase boundaries, 64
Lorentz force, 130, 138, 151152 Phenomenological transfer coefficients, 103
Lorentz length contraction, 120 Phonon, 51, 7274
Lorentz transformations, 118121, 177178 Photon, 7274, 136
Lower thermal expansion coefficient, 12 Pinpointed defects, 38
Lubricating layer, 105, 185186, 188190, Plasma, 130
205, 208 Plasma frequency, 33, 143, 145146, 157, 167
Plasmon, 149150, 170
Poisson coefficient, 38, 43, 87, 129, 131
M Polarization coat, 57, 58
Mach number, 71, 116, 118, 124 Polarization coefficient, 2, 34
Madelung energy, 1011, 29 Polarization effect, 57
218 Index

P (cont.) Submicroroughness, 61
Polychromatic approximation, 155 Subsurface layer, 99
Polygonal cellular, 113, 115 Surface diffusion, 55, 6061, 106
Polymeric chain (linear, ramified and reticular Surface energy, 2122, 31, 33, 47, 53, 55, 56,
structures, homogeneous- and hetero- 65, 186187
geneous), 2526, 56 Surface layer, 33, 5153, 56, 58, 61, 9899,
Polymers (tribopolymer), 15, 25, 2526, 33, 110, 111, 113117, 124128, 130,
200, 211 132133, 144, 149, 155, 173, 187, 204,
Ponderomotive force, 152 208
Positron, 122 Surface tensioning, 52, 55, 6364
Potential barrier, 21, 81, 89, 100, 127, 194, 208 Surfing films, 133, 191, 202, 206209
Potential energy, 2, 4, 14, 30, 33, 150, 186,
193194
Primary emission, 91 T
Principle of the positive gradient of mechani- Temperature conductivity, 100, 129, 140
cal behavior, 105 Term, 3
Theory of perturbations, 192
Thermal activation surface, 89, 111, 116,
Q 127128, 129
Quantum number (main, orbital, magnetic), Thermal expansion coefficient, 12, 1718
67, 9 Thermodiffusion coefficient, 75, 79
Quantum object, 56 Thermodynamic, 5, 51, 56, 71, 101, 102105,
Quantum state, 4, 5, 29 124
Third body, 105106, 176, 190, 202, 206
Transfer phenomena, 102103, 105
R Transitional metals, 15, 17, 19, 192, 209
Real contact area, 96, 98, 100, 117 Transitional state model, 193197
Relativistic (triborelativistic) effects, 120121, Tribochemical output, 197
127, 130, 190, 194 Tribocollahedron, 159
Rigidity coefficient, 71 Tribocoordination, 132
Triboplasma, 127
Triboplasma temperature, 136
S Triboplasma wave collapse, 182
Schrdinger equation, 57, 9, 30, 177, 195 Triborelativistic effects, 127
Secondary hydrogen diffusion, 108, 109110 Tribosystem, 102107, 111, 118, 136, 156,
Section of the reaction, 171 159, 170, 173, 176, 179180, 185, 191,
Shear modulus, 42, 71, 84, 115 198, 201
Sliding plane, 8387, 90, 208 Tribounit, 102
Small-angle boundary, 4850 Tunnel effect, 194, 208
Solar crown, 130 Typical emission, 91
Solar wind, 130
Spatial groups, 23
Specific friction force, 96 U
Spin, 7, 9, 12, 28, 194195 Ultradispersed microstructure, 129, 133
Steady states determined, 8 Umov-Poynting vector, 152, 154
Strength, 12, 13, 19, 45, 86, 90, 104, 116 Uncertainty principle, 56, 167
Stronger, 12
Structural activation surface, 129
Structure type (body-centered cubic (BCC), V
face-centered cubic (FCC), hexagonal Vacancy, 39, 89, 207
densely packed (HDP)), 23 Vector of main translations, 22
Sublimation, 17, 33 Vector of state, 5
Index 219

Vector of the magnetic induction, 140 Wave vector, 7, 8, 28, 54, 69, 71, 7274, 143,
Viscosity, 21, 28, 141, 188189 146, 148, 178, 179
Voids, 38, 4647 Wear resistance, 106, 161, 209
Volume modulus of elasticity, 29 Work, 166
Volumetric diffusion, 61, 8182

Y
W Youngs modules of elasticity, 15
Wave equation, 176 Yield strength, 45
Wave number, 31, 180

Vous aimerez peut-être aussi