A Workbook of

Electrochemistry
A Workbook of
Electrochemistry
John O'M. Bockris
Professor of Physical Chemistry
The Flinders University
Adelaide, Australia
and

Ronald A. Fredlein
Lecturer in Physical Chemistry
The University of Newcastle
New South Wales, Australia

PLENUM PRESS· NEW YORK-LONDON· 1973

 Library of Congress Catalog Card Number 72-83606
ISBN-13:978-1-4613-4564-O e-ISBN-13:978-1-4613-4562-6
DOl: 10.1007/978-1-4613-4562-6

i0 1973 Plenum Press, New York

Softcover reprint of the hardcover 1st edition 1973

A Division of Plenum Publishing Corporation

227 West 17th Street, New York, N. Y. 10011

United Kingdom edition published by Plenum Press, London

A Division of Plenum Publishing Company, Ltd.
Davis House (4th Floor) 8 Scrubs Lane, Harlesden,
London NW 10 6 SE, England

All rights reserved

No part of this publication may be reproduced in any form

without written permission from the publisher
 PREFACE

In this book, the objective has been to set down a number of questions,
largely numerical problems, to help the student of electrochemical
science. No collection of problems in electrochemistry has previously
been published. The challenge which faces the authors of such a book
is the breadth of the material in modern electrochemistry, and the
diversity of backgrounds and needs of people who may find a "problems
book" in electrochemistry to be of use.
The general intention for Chapters 2-11 has been to give the
first ten questions at a level which can be dealt with by students who
are undergoing instruction in the science of electrochemistry, but have
not yet reached graduate standard in it. The last two questions in
Chapters 2-11 have been chosen at a more advanced standard, corre-
sponding to that expected of someone with knowledge at the level of
a Ph.D. degree in electrochemistry.
Chapters I and 12 have a different character from the other
chapters in the book. Chapter 1 is for the generalist who has some
knowledge of the impending developments and rapid changes currently
going on in electrochemistry, which results from the realization that
we cannot continue with a fossil-fueled economy, and must turn
increasingly to one which centers itself upon atomic energy and
electricity. Chapter 12, on the other hand, presents a broad spectrum
of problems in which no reservations have been made about the knowl-
edge of the student. The questions in Chapter 12 should be answerable
by an experienced graduate in electrochemical science -one who
regards as his province the whole field, with its widespread applications.
In Chapters 2-1 I, there has been a further breakdown of the
problems. The first two or three problems are particularly simple and
generally need only some direct application of one or two equations.
v
vi PREFACE

The presence of simple, direct questions of this character seems to be

necessary to help the student obtain familiarity with the units, symbols,
and basic ideas of the field. Any student who finds continued difficulties
in obtaining satisfactory answers to Problems 1 and 2 of Chapters 2-11
is not sufficiently familiar with the basic concepts of modern electro-
chemistry and should undertake further course work or individual
study before proceeding.
A word must be said about the use of the book in courses of
electrochemistry, electroanalytical chemistry, and electrochemical
engineering. The book has been written predominantly for students
in electrochemical science, i.e., that broad interdisciplinary field
concerning the electrical properties of interfaces, and their contribu-
tions to the fields of metallurgy, energy conversion and storage,
electrosynthesis, electrobiology, and so on. Students who wish to
specialize in electroanalytical chemistry or electrochemical engineering
have to go through a fundamental course in modern electrochemical
science (which must of course be oriented in terms of quantum mechan-
ics and solid state physics) as the foundation of their specialization.
After such a fundamental course, they should be able to do at least
Problems 1- \0 of Chapters 1-11.
The time needed to work out most of the problems in Chapters 2-11
varies from a few minutes each for Problems 1 and 2 (so long as the
basic material is well understood beforehand and the relevant formula
immediately recognized), to times of the order of one to two hours
each for the Problems 3-10, to times of the order of a day for Problems
11 and 12. These latter problems are intended as "take-home exam"
type problems, and the availability of libraries and computing machines
is assumed.
With respect to the necessary data, this has been given in
Problems 1-10 of Chapters 1-11. In Problems 11 and 12 of Chapters I-
11, and in all problems of Chapter 12, little or no information has
been given, to simulate the actual conditions of problem-solving in the
laboratory or office.
Books of electrochemical constants are readily available and we
may here mention the classic book by Latimer, Oxidation Potentials
(2nd ed., 1952). Also, Electrochemical Data by Conway (1952) and the
Handbook of Electrochemical Constants by Parsons (1959). Volumes 9a
and 9b of Physical Chemistry by Eyring, Henderson, and Jost (1970)
have many tables of constants, as do the books mentioned below.
For a time, a great paucity of books on modern electrochemistry
 PREFACE vii

existed, but the field has had the benefit, in the last few years, of several
new books. Of these must be mentioned Electrode Processes by
B. E. Conway (1965), The Double Layer and Electrode Kinetics by
P. Delahay (1967), Electrochemical Kinetics by K. Vetter (1967), Modern
Electrochemistry by J. O'M. Bockris and A. K. N. Reddy (1970), the
two volumes of electrochemistry in the Physical Chemistry volumes
mentioned above, and Electrochemical Science by J. O'M. Bockris and
D. Drazic (1972).
Solutions have been provided to about half of the numerical
problems of Chapters 2-11. No attempts have been made to give
solution to the problems of Chapter 12. Here, the discussion of a full
answer might sometimes give rise to an original publication and the
"correct" answer to some of the applied problems may be regarded
as a matter of options and be subject to changing social and economic
pressures.
This book was written in the Electrochemistry Laboratory of the
University of Pennsylvania. The authors acknowledge the assistance of
many of their co-workers, especially: Dr. Bonciocat (Chapters 4 and
7), Dr. Buck (Chapter 6), Dr. McHardy (Chapters 9 and 10),
Mr. O'Grady (Chapter ll), Dr. Paik (Chapters 2 and 3), and Mr. Sen
(Chapters 5 and 8).

Adelaide J. O'M. Bockris

Newcastle R. A. Fredlein
August 1972
 CONTENTS

Preface . . . . . . . . . . . . . . . . . v

1. General Questions about Electrochemistry

2. Ion-Solvent Interactions 5
3. Ion-Ion Interactions. . 19
4. Ion Transport in Solutions 31
5. Protons in Solution 45
6. Ionic Liquids 61
7. The Electrical Double Layer 85
8. Basic Electrode Kinetics . . 105
9. More Basic Electrode Kinetics 121
10. Some Electrodic Reactions of Interest 135
11. Some Aspects of Electrochemical Technology 151
12. Miscellaneous Problems . . . . . . . . . . 167

Appendices

1. Values of Some Important Physical Constants 173

2. Some Useful Conversion Factors . . . . . . 174

ix
 CHAPTER 1

GENERAL QUESTIONS ABOUT

ELECTROCHEMISTRY

r7l Make up three definitions of electrochemistry. Discuss the

t...!J appropriateness of the definitions to the future applications of
electrochemistry. Do you think the field should embrace both the
physical chemistry of ions in solution and the properties of charged
interfaces, or be limited to the latter?
What about analytical chemists who devote themselves to the
development of new electrochemical methods in analysis and the
underlying electrochemistry? Should they be called electrochemists?
What of technicians who work in plating shops? What of academics
who specialize in quantum mechanical calculations of electrochemical
kinetics? What of men who work at the electrochemistry of surgery,
or people who design anticorrosion systems for ships? Should they all
be called electrochemists or electrochemical scientists, or what should
they be called?

~ Outline the subjects grouped under the heading of "materials

L...:..J science." What fields of endeavor and what applications does
it embrace? Mention several aspects of materials science which rest
upon electrochemical science.

r;-] Attempt to rationalize, in terms of the thermodynamics of ions

~ in solids and in the liquid phase, the statement: "All surface
problems involving one conducting phase in contact with another
conducting phase have properties which can be discussed in terms of
electrochemical theory."
2 CHAPTER 1: GENERAL PROBLEMS IN ELECTROCHEMISTRY

r::;-) What evidence exists to support the statement: "Some hetero-

~ geneous chemical reactions occurring on solids in contact with
gases may in fact have an electrochemical mechanism?" Illustrate the
type of mechanism which may be involved.

r;-J The theory of electron emission from metals to a vacuum has

t2.J been treated classically. Why is it that for electrochemical
reactions only a purely quantum mechanical treatment is significant?
Make numerical calculations to illustrate your answer.

r7l Mention a dozen examples of practical phenomena in which

~ electrochemical science is involved, and describe what that
involvement is. Give the main reason why electrochemical science has
such a wide involvement in many practical sciences.

r-:;-l Consider the future of the field of electrochemistry. Should it

l!J continue as a branch of physical chemistry, and, at universities,
should academics interested in contributing to the field work in
chemistry departments? Or would you advocate the growth of a new
discipline in universities ("electrochemical science") outside depart-
ments of chemistry? Discuss the pros and cons of establishing separate
departments of electrochemical science and draw a conclusion.

101 To what degree can electrodic reactions be thought of as an

L!J aspect of heterogeneous catalysis? To what extent may hetero-
geneous catalysis be looked at in electrodic terms?

f9l Discuss three major antipollutional developments, affecting

l2J the whole community, which you would expect to see accom-
plished by, say, 2000 A.D., and which depend upon the development
of electrochemical technology.

The German physical chemist Nernst must be reckoned as one

of the greatest chemists of all time. Estimate his effect on
 PROBLEMS 3

electrochemistry and the results of that influence on the quality of life

in the 21st century.

f11l Weigh the statement: "Electrochemical science is a new applied

~ science." How do you think electrochemical science compares
with materials science, energy conversion, and communications, as a
new science? Estimate the circumstances which led to the identification
and "coming out" of these new sciences. Consider in detail the three
new sciences, and estimate the date of their identification and decide
which early discoveries and formulations constitute their roots. Estimate
the time which it took for these sciences to become accepted in their
own right as university topics.

'12l Increasing pollution of the atmosphere and the waters, diminish-

~ ing reserves of raw materials and fossil-fuels, and increasing
world popUlation are some of the problems which beset man's existence
on this planet. Abundant electricity, at a cost of about one-fifth the
present cost (in present dollars) is technologically feasible from very
large breeder reactors. To what degree (and in what ways) would the
picture of plenty for all mankind, held out by the advocates of breeder
reactors, be dependent on the extensive development of a new electro-
chemical technology? To what extent are there alternative technological
paths which do not involve electrochemistry, but which would be
entirely free of pollutant emissions, in particular, carbon dioxide?
 CHAPTER 2

ION-SOLVENT INTERACTIONS

Gl The experimentally determined heat of sublimation of rubidium

~ iodide is + 148.6 kcal mole- I and the corresponding heat of
solution is +6.2 kcal mole-I.
Calculate (a) the heat of hydration of the salt from the experimental
data; (b) the same quantity, in terms of the Born model, utilizing the
following data: rRb+ = 1.48 A; rI- = 2.16 A; Ew = 78.3 (25°C);
dEwldT = -0.356 deg- I .

Using the data of Problem I and rCI- = 1.81 A, calculate the

entropy of hydration of the Cl- ion, in terms of the Born model.

IJl The determination of the absolute heat of hydration of the

L.:J proton by Halliwell and Nyburg (1963) (LJHH +- h = -266 kcal
mole-I) permits the absolute heat of hydration of other ions to be

TABLE 2.3.1. Integral Heats of Solution

of NaCI at Various Concentrations·

103 Concentration JR, ,

(molal) cal mole- 1

1.45 952 ± 33
2.88 931 ± 18
5.31 941 ± 15
6.39 945 ± 47
8.98 959 ±2
9.93 937 ±6
20.56 965 ±5

* Data from Criss and Cobble (1961).

5
6 CHAPTER 2: ION-80LVENT INTERACTIONS

calculated. From the integral heats of solution of NaCI at 25°C

(Table 2.3.1), calculate the absolute heats of hydration of Na+ and Cl-.
The relative heat of hydration of Cl-, L1H~r~_h , is -347.5 kcal mole-1
and the heat of sublimation of NaCI is +184.7 kcal mole-I.

r:l Consider a dilute solution of NaCI at 25°C. Calculate for each

L..iJ ion: (a) the Born heat of solvation; (b) the ion-dipole heat;
(c) the ion-quadrupole heat; (d) the induced ion-dipole interaction;
(e) the heat of structure breaking in the solvent.
rw = 1.38 A; r CI- = 1.81 A; rNa+ = 0.95 A
The dipole moment of water, f1-w = 1.87 X 10-18 esu cm.
The quadrupole moment Pw = 3.9 X 10-26 esu cm2 •
The polarizability G\:w = 1.46 X 10-24 cm3 •

Make clear what assumptions are made for each calculation, and
state what you think of the accuracy. Make an estimate of the heat of
solvation of Na+ and Cl- ions arising thus, and then discuss the likely
meaning of discrepancies with the experimental values determined in
Problem 3, in terms of modelistic considerations.

r;l In Table 2.5.1, data are given on the adiabatic compressibility

~ of solutions of certain salts. The adiabatic compressibility of
water at 25°C is 44.730 x 10-6 bar-i.

TABLE 2.5.1. Adiabatic Compressibility and Density (p)

of Electrolyte Solutions as a Function of Concentration (C) at 25°C

c, p, Adiabatic compressibility,
Salt moles Iiter-1 gml- 1 x 106 bar-1

NaI 5.046 1.5617 28.783

2.053 1.2290 36.831
1.013 1.1120 40.838
0.100 1.0086 44.278
0.0500 1.0028 44.481
4.053 1.3244 22.900
2.019 1.1673 30.940
1.005 1.0843 36.907
0.101 1.0062 43.823
0.0510 1.00176 44.224
 PROBLEMS 7

Utilizing the relation of Passynski (1938) between the number of

incompressible water molecules around an ion in solution and the
adiabatic compressibility of the solution, calculate the total hydration
number of NaI and CaCI 2 at each concentration. Discuss to what
extent the result you obtain represents the hydration number of the
salt concerned. Define carefully the hydration number to which you
refer.

Ionic vibration potential measurements permit the calculation

of
(2.6.1)

for an electrolyte in solution, where n+ and n_ are the individual

solvation numbers of the cation and anion, respectively, and t+ and L
are the transport numbers at infinite dilution. Using the equations
presented by Zana and Yeager (1967) and the data of Tables 2.6.1
and 2.6.2 together with the results of Problem 5, calculate the individual
ionic hydration numbers of the ions. The radii of hydrated ions,
estimated by Nightingale (1959), are given in Table 2.6.3.

TABLE 2.6.1. Transport Numbers at Infinite Dilution

of Ions in NaI and CaCI 2 Solutions

Electrolyte

Nal 0.395 0.605

CaCl. 0.438 0.562

r:;l Consider the water in the first layer around an ion, e.g., Na+,
L2J in aqueous solution. Write an expression for the total electro-
static interaction energy of the ion with a water molecule. Explain
the physical meaning of all the terms. Deduce the electrostatic field
strength at the center of the water molecule.
Discuss the basis on which one could conclude whether the water
molecules are dielectrically saturated. Use the equations of Booth
(1951), which relate dielectric constant to the field strength to estimate
the degree of saturation of water in the first layer near the sodium ion.
rNa+ = 0.95 A; r". = 1.38 A.
8 CHAPTER 2: ION-SOLVENT INTERACTIONS

TABLE 2.6.2. Ionic Vibration Potential Terms (tPo/ao)*

for NaI and CaCl 2 Solutions at Various Concentrations

Concentration, 106(<polao),
Electrolyte M V cm-1 sec
NaI 0.05 -6.4
0.10 -6.4
1.0 -5.7
5.0 -5.7
0.05 I.1
0.10 1.1
1.0 1.0
5.0 0.8

* <Po is the ionic vibration potential and ao , the velocity

amplitude of the solvent, arises from the equation Vo = a o
exp[ -i(wt - ax)] which describes the periodic velocity of
the solvent. <Polao is obtained directly from experiment.

TABLE 2.6.3.
Hydrated Radii of Some Ions

Ion Yt-h, A
Na+ 3.58
1- 3.31
Ca2+ 4.12
CI- 3.32

Distinguish, in terms of a physical argument, the external and

internal fields in a dielectric. Calculate the ratio of these fields
for benzene. €c 6 H 6 = 2.274.
In Kirkwood's theory of the dielectric constant of associated
liquids, the factor g is usually taken as four. Draw a diagram to show
how this factor arises from concepts of the structure of water. Stress
the physical, rather than the formal, meaning.
What degree of applicability would you expect Kirkwood's theory
to have to liquid ammonia?

Describe qualitatively why dielectric constants of ionic solutions

become less with increasing concentration of dissolved ions.
 PROBLEMS 9

The empirical relation

(2.9.1)

where Es and Ew are the dielectric constant of an electrolyte solution

and water, respectively, C is the concentration, and 8 a constant, has
been found to be obeyed up to concentrations of ......,1 M. By considering
the solution as a suspension of spheres of dielectric constant Eions in
a medium of dielectric constant Ew , the relation

(2.9.2)

has been deduced (Fricke, 1924). Here, p is the volume fraction of the
solute. Derive (2.9.1) from (2.9.2) and show that

~ [ Ew - Eions V Ew - Ewoo ]
o = 1.5 Vions 1000 + w 1000 nil (2.9.3)

where Vions and V;" are the actual molar volumes of solute and water,
respectively, Ewoo is the deformation polarization of water, and nil is the
total primary hydration number. Make clear the assumptions and
approximations of your derivation. Could the hydration number of
an electrolyte be obtained from (2.9.1) and (2.9.3)?

r:jQl What is the definition of the "primary solvation number"?

~ How does it differ from the "solvation number"? Distinguish
clearly between the coordination number and the primary solvation
number of an ion in solution.
It has been suggested that the essential difference between solvation
number and coordination number resides in the time an ion remains
stationary in a given site in solution, 71' compared with the time it
takes to orient a water molecule out of the surrounding water structure
and into a position of maximum interaction with the ion, 72' If
72 < 71 , the water molecule concerned becomes part of the hydration

shell, and counts in the solvation number; if 71 < 72, the water
molecule concerned remains only a coordination water.
From the equation
Ll2 = 2Dt (2.10.1)
10 CHAPTER 2: ION-SOLVENT INTERACTIONS

where LI is the average distance traversed in one hop, calculate 7"1.

(Take D '" 2 X 10-5 cm 2 sec- 1 .) What sort of a value would you
suggest for T2 ? How would you go about making (a) a quick estimate
of T2 , (b) a more thorough estimate of 7"2 ?
Utilize approximate values of T2 to consider the rationality of the
distinction between solvation number and coordination number.

Some experimental entropies of hydration of ions are listed in

Table 2.11.1.

TABLE 2.11.1. Entropies of Hydration

of Individual Ions

Ion So, cal deg- 1 mole- 1

H+ -31.0
Na+ -26.3
Rb+ -15.6
TP+ -16.4
Sr +
2 -56.6
Cd H -66.5
Be -14.4

Give an interpretation of the negative sign of this entropy in terms

of continuum theory.
Suppose that the model for the region near an ion is given as
follows: The coordination number varies from ion to ion and for Na+
it may be taken as six. The solvation number (the number of water
molecules oriented toward the ion in positions of maximum interaction)
is four for Na+. The second layer consists of a structure-broken region
and outside the second layer the structure is that of the bulk.
Calculate the entropy of Na+ in the gas phase. Formulate the
various structural, structure-breaking, and Born contributions to the
solvational entropy, allowing for the difference of coordination number
and solvation number.
See how near your calculations can bring you to the measured
values of N a + and other ions listed in Table 2.11.1 (you can be satisfied
with discrepancies up to about 33 %).
 ANSWERS 11

Work out the quadrupole interaction between an ion and a

water molecule and show that, for a positive ion, it is given by

(2.12.1 )

where Pw is the quadrupole moment of water, {J-w is the dipole moment,

and r is the distance from the center of the ion to the center of the water
molecule when the ion and water molecule are in contact, and the
water molecule is oriented for maximal interaction. In your deduction,
be very careful to make a clear and diagrammatic definition of the
quadrupole moment. Show that it consists of several contributions of
the type of X2, where x is a distance.
Calculate the error introduced into the calculated heat of hydra-
tion of a univalent cation and anion by neglecting the quadrupole
character of water. Make an estimate of the uncertainty of the heats
of hydration of individual ions calculated from the ion-quadrupole
model.

ANSWERS

r.;r (a) The heat of solution .:JHs of an ionic crystal is equal to the
L2.J heat of dissociating the crystal into its individual ions (i.e., the
heat of sublimation .:JHsubl) plus the heat of hydration of the ions .:JHh •
Thus,
.:JH, = .:JHsubl + .:JHh (2.1.1)

Upon substituting the numerical values, the heat of hydration of RbI

is found to be
.:JH Rbl - h = 6.2 - 148.6 kcal mole-1
= -142.4 kcal mole 1

(b) From the Born equation, the heat of hydration of an ion i,

.:JHi _ h , of radius ri and charge z is given by

.:JH I c~c - NAz2eo~ (1 _ J.- _ ~ OEw) (2.1.2)

1-' 2ri Ew Ew 2 aT
12 CHAPTER 2: ION-SOLVENT INTERACTIONS

where N A is Avogadro's number and €Ow is the dielectric constant of

water. Substituting given values yields

. __ 6.03 X 10 23 x (4.8 X 1O-1())2 ( _ _1_ 298 x 0.356 )

iJH,_h - 2ri 1 78.3 + (78.3)2
6.54 x 104
iJHi _ h = - esu 2 cm-1 mole- 1
r,.

when ri is expressed in centimeters,

iJH RbI _ h = iJHRIJ +_ h + iJH I -_ h

= -6.54 X 104 (_I

r
_ + _1_)
r -
Rb + I

Substituting the given values for rRb+ for rI- yields

iJH RbI - il = -7.45 X 1012 ergs mole- 1
= -178 kcal mole- 1

The relative heat of hydration of an ion i, iJH[~~, is related

to the absolute heat of hydration by

From the given data

iJHCl -._ h = -·347.5 -+- 266 kcal mole-1

-81.5 kcal mole- 1

The heat of solution of sodium chloride, iJHNacl-s , can be found by

extrapolating the integral heats of solution to infinite dilution; a rea-
sonable straight line plot, suitable for extrapolation, being obtained by
plotting against the square root of the molality. From Fig. 2.3.1,
iJHNac1 -., = 0.91 kcal mole-1
iJHNacl-il =~ iJHNacl-s - iJHNacl-subl
= 0.91 - 184.7 kcal mole- 1
= -183.8 kcal mole-1
 ANSWERS 13

,... 1000
"0
E
0
2
(J')

u
I
900
0
z
::c
<l

800
0.5 I 1.0 1 1.5
(mNoCI )'2 (molal)'2

Fig. 2.3.1. Determination of the integral heat of solution of sodium

chloride at infinite dilution.

Since
LJHNaCl _ h = LJHNa +_ h + LJHCl -_ h
LJHNa +_ h = ~183.8 + 81.5 kcal mole-l
~ 102.3 kcal mole- l

r7l The Passynski (1938) relation between the total hydration

~ number of an electrolyte nh and the compressibilities {Jsoln and
{JSOlv of solution and solvent, respectively, is

(2.5.1 )

where Nl is the number of moles of solvent molecules in the solution

containing N2 moles of salt.
For one liter of solution,

N2 = molar concentration
Nl = (lOOOp ~ N2 M2 )/M1

where M2 is the molecular weight of the solute and Ml = 18.02 for

water.
14 CHAPTER 2: ION-80LVENT INTERACTIONS

TABLE 2.5.2. Calculation of the Total Primary Hydration Number nh

C,M 1 - f3.o1n/f3.o1v N, nh

NaI 5.046 0.3565 44.7 3.16

2.053 0.1766 51.1 4.40
1.013 0.0870 53.3 4.61
0.\00 0.0100 55.1 5.58
0.0500 0.00557 55.2 6.16

CaCl. 4.053 0.4880 48.5 5.84

2.019 0.3083 52.3 7.99
1.005 0.1749 54.0 9.40
0.\01 0.0203 55.2 11.1
0.0510 0.01 13 55.3 12.3

Table 2.5.2 summarizes the results of the intermediate steps of the

calculation with the final result, nh , in the last column.
The calculation is based on the assumption that the solvating
water molecules are totally incompressible while all other water
molecules retain their normal compressibility. These assumptions have
been reviewed by Bockris and Saluja (1972) and they conclude that
the solvating water molecules are incompressible but that coordinating
nonsolving water molecules are partly ( <30 %) compressed. This leads
to the conclusion that nh calculated from (2.5.1) is too large by about
0.5 water molecules.

r]l The interaction energy consists of the ion-dipole interaction

~ term and the ion-quadrupole interaction term. For a positive
ion, it becomes

(2.7.1)

where flow and Pw are the dipole moment and quadrupole moment of
the water molecule, respectively.
The first term is the interaction energy of a dipole of water with
the field due to the ion. The second term, the quadrupole-ion inter-
action energy, stems from the fact that the distribution of charge in
the water molecule is such that the electrostatic interaction energy
between the charges in a water molecule and the ionic charge cannot
 ANSWERS 15

be accounted for by only the dipole-field energy. The quadrupole

energy accounts for most of this difference. The field due to the ionic
charge eo is
Xion = eO/(ri + r w)2
= (4.8 x 10-10)/(2.33 x 10-8)2
= 0.86 X 106 statvolt/cm (2.7.2)

The expression for the dielectric constant is (Booth, 1951)

(2.7.3)

where ex = 28/3 V73 and f3 = V73/6. Xion is the field due to the ion
in the absence of a medium and No is the number of molecules per
cubic centimeter. It should be noted that EO is the result of both atomic
and electronic polarization and is equal to six [Rampolla et al. (1959)] .
.P(x), the Langevin function of x, is given by

2(x) = coth x - l/x

With Xion ~ 106 statvolt cm-1 , the argument of.P becomes

x ~ v73/6 (1.8 x 10-18)(8 x 106)/(4.2 x 10-14)

~ 13.5 x 102
Thus
2(x) ~ 1
and

Substituting numerical values yields the dielectric· constant in the first

layer:
_ + 2877 x 6.023 X 1023 x 8 x 1.87 X 10-18
E - EO 3 X (73)1/2 x 18 x 0.86 x 106
= EO + 1.5
Since the contribution from orientation polarization to the
dielectric is small (1.5), this means that the dielectric saturation of
water in the first layer around an ion is essentially complete.
16 CHAPTER 2: ION-SOLVENT INTERACTIONS

r;l Water molecules near an ion are under the influence of a

L!.J strong electric field due to the ion; therefore, they are completely
aligned to that field, fixed and insensitive to the external field. Hence,
only the nuclear and electronic contributions to the dielectric constant
are important for the hydrating water molecules. The dielectric constant
arising from these contributions is ",6.
The water molecules in the secondary region are also affected by
structure breaking, which leads to a smaller value of the number of
water molecules which turn as a group and hence to smaller E.
Rearranging (2.9.2) yields

and

Rearranging gives
HEs/Ew) - 1][(Eions/Ew) + 2] -.lL
(Eions/Ew) - (Es/Ew) 1- P

and since Eions ~ 2 while Ew = 78.3, we have

Es - Ew = l.5[p/(l - p)](Eions - Es) (2.9.4)
Rearranging (2.9.4) gives

Es = Ew + [1.5p(Eions - Ew)/(l + O.5p)] (2.9.5)

For solutions that are not too concentrated, O.5p ~ 1, where p is the
volume fraction of the hydrated ion, and (2.9.5) becomes
(2.9.6)

Equation (2.9.6) holds for ions that are not hydrated. For hydrated
ions (2.9.6) becomes
(2.9.7)

where EionS-h is the dielectric constant of hydrated ions and Ph, the
volume fraction of hydrated ions, is given by

(2.9.8)
 ANSWERS 17

To a first approximation,

(2.9.9)

where j; is the volume fraction of the ion in a hydrated ion and fw is

the volume fraction of restricted water in a hydrated ion. EwCXJ is the
dielectric constant of restricted water. Consequently, (2.9.9) can be
written as

(2.9.10)

Substituting (2.9.8) and (2.9.10) into (2.9.7) yields

(2.9.1 )
where
" I5[
a = . V!ons
Ew -
1000
Elons
+ V w Ew 1000
- EwCXJ
nh
]
(2.9.3)

Because (2.9.9) is only an approximation and because the total

primary hydration number nh of the electrolyte is assumed to be
independent of concentration, application of equations (2.9.1) and
(2.9.3) would not constitute a reliable method of finding nh .
 CHAPTER 3

ION-ION INTERACTIONS

Calculate the radius of the ionic atmosphere in aqueous potas-

sium bromide solution. CKBr = 10-3 mole liter-I, Ew = 78.3.

~ Calculate the number of ions comprising the ionic atmosphere

L..=.J around a K + and also a Br- in a solution of 10-3 M potassium
bromide. What is the average distance apart of such ions?

r)l Calculate the self-energy (kcal mole-l) of Na+ in dilute aqueous

L...:.J solution. Correspondingly, calculate the energy of the ionic
atmosphere for the same ion, in 10-3 M sodium chloride solution.
YNa+ = 0.95 A; Ew = 78.3.

~ Derive in two different ways (i.e., using the Debye treatment

~ and the Guntelberg treatment) an expression for the reversible
charging of an ion in a solution of constant ionic strength, taking into
account the radius of the ion. Explain clearly the difference in the
physical assumptions of the two methods. In deriving expressions for
the free energy of ion-ion interactions in solution, what value is there
in comparing the results obtained by using the two treatments of
reversible ion charging?

f5l Derive equations to relate the root mean square activity coeffi-
~ cient on the mole fraction scale to those on the molarity and
molality scales. The root mean square activity coefficient of the ions
19
20 CHAPTER 3: ION-ION INTERACTIONS

in 3.2 M aqueous calcium chloride solution is 1.55. The density of the

solution is 1.246 g ml-i. Calculate the root mean square activity
coefficient on the molality scale and on the mole fraction scale.

r(;l Explain fully and clearly what is meant by a standard state in

~ problems connected with solutions. What standard states are
frequently used for ionic solutions? Which is normally used? Is the
standard state for the solute a hypothetical or a real state? Distinguish
a reference state from a standard state in solution theory.

r]l Utilize the Fuoss approach to ion pair formation to find the
L..:.J ionic association constant of NaCl (a) in water at 5 M; and
(b) in benzene at 10-6 M. Ew = 78.3; EC 6 H 6 = 2.27.

In the calculation of the free energy of interactions of ions in

solution, the mathematical approximation of linearizing the
Boltzmann term is made, i.e.,

(3.8.1)

where Zieo{;r is the Coulombic potential energy of an ion of charge ZieO

in an electrostatic potential {;r. Show that it is pointless to attempt
a more rigorous solution by avoiding this approximation.

f9l The spectroscopic determination of the association constant of

~ copper sulfate in water is carried out by measuring the difference
in optical densities of two solutions. Solution (1) contains Cu(CI04)2 ,
Na 2S04 , and NaCl0 4 to maintain constant ionic strength, while
solution (2) contains NaCI0 4 and the same amount of Cu(CI0 4)2 as
solution (1). Assuming the Beer-Lambert law, show that at wave-
lengths at which only Cu++ and associated CuS0 4 absorb,

(3.9.1)

where Di and D2 are the optical densities of solutions (1) and (2),
Ccuso 4 is the concentration of associated CuS0 4 , I is the length of the
 PROBLEMS 21

cell, and ECUSO, and ECU++ are the molar absorptivities of CUS04 and
Cu++, respectively.
The data in Table 3.9.1 pertain to three approximately constant
ionic strengths II , 12 , and 13 . Calculate
(3.9.2)

at each ionic strength and roughly estimate the association constant

K = acuso,/acu++aso--,
, using Debye-Hiickel theory with parameters
A and B equal to 0.5115 and 0.3291 x 108 respectively.

TABLE 3.9.1. Optical Density (D 1 - D 2)*

for Various Solution Compositionst

103CCu(CIO')2 ' 103CNa2s0, ' 103CN&CIO, ' Dl - D2

M M M
II
4.05 8.00 56.5 0.143
4.05 12.0 45.2 0.206
4.05 16.0 33.9 0.260
4.05 20.0 22.6 0.312
4.05 24.0 11.3 0.352
4.05 28.0 0.0 0.389
12
5.00 4.74 42.6 0.134
5.00 7.10 36.4 0.187
5.00 9.46 29.3 0.247
5.00 11.83 22.7 0.293
5.00 14.21 17.3 0.340
I.
5.00 1.66 41.1 0.046
5.00 3.34 26.9 0.087
5.00 6.66 28.5 0.160
5.00 8.34 24.2 0.193
5.00 10.00 20.0 0.226

* Dl - D2 is the optical density of the solution of interest minus that of a blank.

t Data from Hemmes and Petrucci (1968).

~ Write elucidatory notes on the following:

(a) The reason for the term potential electrolyte.

22 CHAPTER 3: ION-ION INTERACfIONS

(b) The theory which gives the range In which the Poisson-
Boltzmann equation can be linearized.
(c) The significance of negative values of the ion size parameters.
(d) A relation which may exist between the water molecules
counted in the Stokes-Robinson model for concentrated electrolyte
solutions and the primary solvation number.
(e) Approaches by which individual activity coefficients may be
estimated.
(f) The modelistic meaning of activity coefficients greater than
one.

r.j1l Essentially, the theory of ionic solutions is successful up to

~ about 0.005 N, whereafter there is a dependence upon empir-
icisms to obtain agreement with experiment. Review the rationale of
these empiricisms. In particular, examine the validity of the Stokes-
Robinson model for solutions of higher concentrations. How valid is
the use of the ion-ion cloud interaction term in equations for the
activity coefficients when applied to very concentrated solutions?
Discuss the implications for models of the agreement with experiment
of the log y± against concentration relation which Stokes and Robin-
son obtain.
Use the data of Table 3.11.1 and the Stokes-Robinson relation
to calculate the energy of the ion-ion cloud interaction. Examine the
relative importance of this term in the Stokes-Robinson treatment of
concentrated solutions of electrolytes and examine its concentration

TABLE 3.11.1. Data for Calculation of the Ion-Ion Cloud

Interaction Energy for Concentrated Sodium Chloride Solutions at 25°C

Concentration y± Osmotic coefficient Total hydration

(molal) (molal scale) (molal scale) number

0.1 0.778 0.932 6.6

0.5 0.681 0.921 6.1
1.0 0.657 0.936 5.6
2.0 0.668 0.983 4.7
5.0 0.874 1.192 3.8
 ANSWERS 23

dependence. Assume that the primary hydration number is the appro-

priate quantity to use in this calculation.

r;:;-]Synthesize the position of the 1970's on the physical aspects of

~ the theoretical treatment of concentrated ionic solutions,
paying attention to the work of Frank and Thompson (1960) and
Frank (1966) on the dependence of log y on Cl/3, and to that of
Friedman (1971) on the cluster theory in electrolytes. Discuss possible
links between these theories of concentrated aqueous solutions and
models of simple molten salts associated with the names of Angel,
Bloom, Rice, Richards, Riess, and Stillinger.

ANSWERS

r:tl Since the radial distribution of ionic charge in the ionic atmo-
~ sphere attains its maximum value at a distance K-1 from the
central ion, where K is the Debye reciprocal length, the radius of the
ionic atmosphere is defined as K- 1 • Hence,

r = K- 1 = (EkT/417 L nizi2e02f/2 (3.1.1)

•
where ni is concentration of the ionic species and E is the dielectric
constant of the solvent. The other symbols have their usual meanings.
From a consideration of the units, K- l has the units centimeters when
ni is in ions per cubic centimeter, i.e. (with CKBr = O.ooIM),

nK + = nBr- = N A CKBr /1 000

= 6.023 X 1017 ions cm-3

Substituting this and given values into (3.1.1) yields

[ 78.3 x 1.38 x 10-16 x 298 ]1/2

r = 2 X 417 x 6.023 X 1017 x (4.8 X 10-10)2
= 9.7 X 10-7 cm
24 CHAPTER 3: ION-ION INTERACTIONS

r.;-, The self-energy W of an ion in a medium of dielectric constant E

L2J is approximated by the energy to charge a sphere of radius ri
equal to the crystallographic radius of the ion. Thus, for Avogadro's
number of ions,

(3.3.1)

Substituting numerical values yields

W = (4.8 X 10-1°)2 X 6.023 X 1023

2 X 0.95 X 10-8 X 78.3
= 0.933 X 1011 ergs mole-1
= 2.23 kcal mole-1
The energy of the ionic atmosphere is the extra chemical potential due
to the ion-ion cloud interaction, .1/-'i_1. From the Debye-Hiickel
theory,

(3.3.2)

where K-l, the Debye-Hiickel reciprocal length, has been evaluated

for a 10-3 M solution of a 1: 1 electrolyte in Problem 1 as
K- 1 = 9.7 X 10-7 cm
Substituting numerical values in (3.3.2) yields
6.023 X 1023 X (4.8 X 10-1°)2
.1/-'i_1 = - 2 X 78.3 X 9.7 X 10-7
= -0.914 X 109 ergs mole-1
= -2.18 X 10-2 kcal mole-1

f"5l For an electrolyte that gives rise to v+ cations and v_ anions in

~ solution, the mole fraction Xi of the ionic species i is related
to the molarity C and the molality m by

X+ = p _ 0.001CM2 + 0.OOICM1v (3.5.1)

0.00lmM1v+
(3.5.2)
1 + 0.001mM1v
 ANSWERS 25

where
(3.5.3)

and P is the density of the solution. Ml and M2 are the molecular

weights of solvent and solute, respectively. At very low concentrations
of electrolyte, (3.5.1) and (3.5.2) become

X+o = 0.OOICoM1v+/po (3.5.4)

= O.OOlmoMlv+ (3.5.5)

where Po is the density of the solvent. Since the ratio a+/a+o, where
a+o is the activity a+ of the cation at infinite dilution, is independent
of the chosen standard state, and since the activity coefficients are
unity at infinite dilution, we have

(3.5.6)

where f+ ,y+ , and y + are the activity coefficients on the mole fraction,
molar, and molality scales respectively. Substituting (3.5.1), (3.5.2),
(3.5.4), and (3.5.5) into (3.5.6) yields

f+ = y+(p - 0.OOICM2 + 0.OOICM1v)/po (3.5.7)

f+ = y+(1 + O.OOlmMlv) (3.5.8)

Analogous equations exist for the anion. Since

(3.5.9)

then
(3.5.10)
and
(3.5.11)

Substituting numerical values in (3.5.10) yields

f± = 1.55[1.246 + 0.0032(3 X 18 - 111)]/1.0
= 1.65
From (3.5.10), (3.5.11), and

m = l000Cf(IOOOp - CM2) (3.5.12)

26 CHAPTER 3: ION-ION INTERACTIONS

we can relate the molal to the molar activity coefficients and

(3.5.13)

and substituting numerical values into (3.5.13) yields

y± = 1.55[1.246 - (0.0032 x 111)]/1.0

= 1.38

The ionic association constant K for the process

Na+ + CI-? NaCI (3.7.1)

is given by
K = 8(1 - 8)-2 C-l (3.7.2)

where 8 is fraction of associated ions and C is the total concentration

of associated and unassociated Na+ or Cl-. The Fuoss treatment
yields the relation

8 = [47T(1 - 8)2 CNAa3/3000] exp[b/(l + Ka)] (3.7.3)

where
(3.7.4)

and K, the Debye reciprocal length, is given by

(3.7.5)

where ni is the concentration of ionic species of charge Zi in ions cm-3 •

The distance of closest approach a is not simply the sum of the
crystallographic radii and is best estimated semiempirically. Since the
extended Debye-Hiickel equation best fits the experimental data with
a = 4 A in the case of NaCI solutions (Falkenhagen and Ebeling, 1971)
this value will be used here. Here, € is the dielectric constant of the
solvent.
Substituting (3.7.3) into (3.7.2) yields

(3.7.6)
 ANSWERS 27

Evaluating band K from (3.7.4) and (3.7.5), respectively, yields, for

aqueous 5 M NaCl solution,

b = 1.79, K = 7.36 X 10 7 cm-1

Substituting numerical values into (3.7.6) yields

K _ 417 X 6.023 X 10 23 x (4 X 10-8)3 1.79

- 3000 exp 3.92
= 0.254 liter mole- I

The units of the association constant arise from the units of C in

Eq. (3.7.2).
For 10-6 M solution of NaCI in benzene, the following values are
calculated in the same way:

b = 61.7, K = 1.93 X 105 cm-1

which give K = 2 X 1025 liters mole-I.

Equation (3.9.2) can be rewritten in the form

K' = x/(a - x)(b - x) (3.9.3)

where a and b are the total concentrations of Cu++ and SO:t-, respec-
tively, and x is the concentration of associated CuS0 4 . Assuming the
Beer-Lambert law, we can write

(3.9.4)

(3.9.5)

and subtracting (3.9.5) from (3.9.4) yields (3.9.1), which may be more
conveniently written as

LID = x! LIE (3.9.6)

where LID and LIE are written for DI - D2 and ECUS0 4 - ECU++ ,
respectively. Substituting (3.9.6) into (3.9.3) yields

K' = LlD/(I LIE) (3.9.7)

[a - (LiD/! LlE))[b - (LiD/! LIE)]
28 CHAPTER 3: ION-ION INTERACTIONS

which rearranges to

ab a+b-x 1
(3.9.8)
JD = I J€ + K'I J€

and K' can be found from the slope and intercept of a plot of ab/JD
against (a + b - x). Since x is unknown, the slope and intercept
must be found by an iterative procedure. An initial plot ab/JD is
made against (a + b), i.e., x is initially assumed to be zero. From the
slope of this initial plot, Sint , approximate values of x are calculated.
Since the slope of the initial plot is approximately (I J€)-t, we have,
from (3.9.6),
xapprox = (JD) Sint (3.9.9)

A more precise estimate of the slope and intercept can be obtained by

plotting ab/JD against (a + b - xapprox) and the procedure repeated
until a constant slope and intercept are obtained. In the present case,
the first plot of ab/JD against (a + b - xapprox) gives satisfactory
results, and details of the calculation are shown in Table 3.9.2.
The value of K' calculated from the first plot of ab/JD against
(a + b - Xapprox) is 18.1 liters mole-I, which is close to the final com-
puter-calculated value of 18.9 liters mole-1 . Values of K' together with
the average ionic strength of the solutions are given in Table 3.9.3.

TABLE 3.9.2. Details of Calculation of K' for CUS04

in Aqueous Solution at Ionic Strength 11

.1D 1000(a + b), 10"ab/.1D 103:1:approx ,* lO3x ,t

M M M

0.143 12.05 227 0.47 0.52

0.206 16.05 236 0.68 0.75
0.260 20.05 249 0.86 0.95
0.312 24.05 260 1.03 1.13
0.352 28.05 276 1.16 1.28
0.389 32.05 292 1.28 1.42

* Calculated from an initial slope of 3.3 x lO-3.

t Computer solution.
 ANSWERS 29

TABLE 3.9.3. Values of K' at Different Ionic Strengths*

K'
I 1 liters mole-1

II 0.090 18.9
12 0.070 28.4
13 0.058 31.5

*I is the average ionic strength.

The ionic strength of each solution is given by

1= t L CiZi2 (3.9.10)
i

= 3a + 3b + CNaclO 4
- 4x (3.9.11)

The ionic association constant K is related to K' by

(3.9.12)

2·6.-------------------.

,..
+1

-
CI
0
N
I
X:
CI
0
2·2

0·02 0·04 0·06 0·08

I
Fig. 3.9.1. Estimation of the association constant of copper sulfate
by extrapolation of values of log K' - 2 log y± to infinite dilution.
30 CHAPTER 3: ION-ION INTERACTIONS

where y is the molar activity coefficient of undissociated CuS0 4 and y±

is the molar mean ion activity coefficient of dissociated CuS0 4 ,

log K = log K' - 210gy± (3.9.13)

where log y has been neglected since the activity coefficient of the
uncharged species, CuS0 4 , will be close to unity.
To obtain an estimation of K, y± can be calculated from the
Debye-Hiickel law:
(3.9.14)

However, there is a problem in deciding what value to use for the

distance of closest approach, a, in a mixture of electrolytes. The
procedure adopted here has been to select two extreme values of a
(0 and 5 A), plot log K' - 2 log y± against I, and extrapolate both
plots to infinite dilution (Fig. 3.9.1). The two values of K so obtained
may be considered to be the extremes of the range of values of K
indicated by the data. Figure 3.9.1 shows that K lies in the range
210-260 liters mole-I.
The most appropriate value of a would yield a plot of
log K' - 2 log y± against I that is parallel to the I axis. When sufficient
data are available, this value of a can be found by an iterative procedure
and K obtained.
 CHAPTER 4

ION TRANSPORT IN SOLUTIONS

CD
0.01 P.
Calculate the conventional and the absolute Stokes mobilities
of a univalent ion of radius 2.0 A in a solution of viscosity

~ According to Faraday's law, passage of a current I for time t

L2..J leads to the deposition or dissolution of a number of ions
NAltfnF, where n is the number of electrons involved per ion and NA
is Avogadro's number.
However, in ionic solutions, each ion carries a different fraction
of current. Explain the apparent paradox between Faraday's law and
the concept of transport numbers. Formulate the essence of your
explanation in an equation.

rJl What is the root mean square distance traversed by chloride

L...:..J ions in aqueous solution in 1 min?
Chlorine is produced in a 10-3 M solution of chloride ions,
containing a high concentration of indifferent electrolyte, by applica-
tion of a current step of 1 rnA cm- 2 • Calculate the concentration of
chloride ions 50 fLm from the surface of the electrode 1 sec after the
current is switched on. D C1 - = 2.03 X 10-5 cm 2 sec-i.

r:jl State the differences among Einstein's equation, the Stokes-

L2..J Einstein equation, the Nernst-Einstein equation, and the
Nernst-Planck equation. Why does the factor of 300 appear in the
latter two equations? If the diffusion coefficient of a divalent ion is
31
32 CHAPTER 4: ION TRANSPORT IN SOLUTIONS

5 X 10-5 cm 2 sec-I, calculate its mobility. To what degree do you

think your answer will be correct?

r;-, Calculate Walden's product for a univalent ion of radius 3 A at

L2..J 25° C for water. At infinite dilution, what would be the equiva-
lent conductivity of the ion in methanol? 7]CHaOH = 5.46 X 10-3 P.

f6l Calculate the concentration gradient established at an electrode

~ solution interface when a univalent ion migrates toward the
electrode in a solution of concentration 1 M at 25°C. The field in the
diffusion layer to which the ion is subject is 105 V cm-l •

Gl The potassium chloride salt bridge is often used to minimize

~ liquid-liquid junction potentials. Calculate the liquid-liquid
junction potential when saturated KCI solution is in contact with
(a) 0.01 M HCI, (b) 5 M CaCl 2 solution using the data of Table 4.7.1.

TABLE 4.7.1.
Transport Numbers of Cations

Electrolyte t+

KCl 0.491
HCI 0.821
CaCl 2 0.438

rol The relaxation effect on the asymmetry of ionic clouds in

L.!J solution is related to the time it takes for the atmosphere to
form. Show that this is given by

where K-1 is the thickness of the ionic atmosphere.

At what frequency of an electric field applied to 0.001 N LiCI
solution would the ionic atmosphere relaxation effect become unde-
tectable?
DUCl = 1.27 X 10-5 cm 2 sec-l
 PROBLEMS 33

~ Some physical properties of I-propanol-water mixtures are

L.:..J given in Table 4.9.1 and the equivalent conductivity of sodium
chloride at various concentrations in I-propanol-water mixtures is
given in Table 4.9.2. Using the Fuoss approach, estimate the association
constant of sodium chloride in each solvent mixture (Goffredi and
Shedlovsky, 1967).

TABLE 4.9.1. Some Physical Properties

of I-Propanol-Water Mixtures at 35°C

Wt.% Dielectric 103 Viscosity, Density,

CaH,OH constant P g ml- 1

60.0 34.49 19.40 0.8766

80.0 24.70 18.02 0.8350
90.0 21.34 16.62 0.8140

TABLE 4.9.2. Equivalent Conductivities and Mean Ion Activity

Coefficients of NaCI in I-Propanol-Water Mixtures at 35°C

Wt.% 104 eNaCl , A,

CaH,OH M mho cm' equiv-1 y±

60.0 5.0869 45.337 2.63

11.130 44.269 2.25
18.452 43.352 2.03
24.239 42.759 1.92
33.823 41.959 1.79
80.0 4.448 35.081 2.08
10.986 33.205 1.80
17.716 31.865 1.67
24.607 30.812 1.59
34.212 29.652 1.51
42.774 28.809 1.46

90.0 1.4379 33.229 1.73

4.0318 31.433 1.48
6.6842 30.150 1.38
8.6972 29.379 1.33
11.524 28.465 1.28
14.322 27.702 1.24
34 CHAPTER 4: ION TRANSPORT IN SOLUTIONS

Ions migrating in high applied electric fields can outrun their

~
10 . . .. ..
lomc atmospheres, I.e., III the relaxatIon tIme ofthe atmosphere,
the ions migrate further than K- 1 . This is known as the Wien effect
(Bockris and Reddy, 1970) and causes the conductivity of the electrolyte
solution to increase with increasing field. Calculate the field strength
required to bring about a 5 % deviation from Ohm's law by the Wien
effect for 0,01 N KCI solution at 25°C. Ai'<.CI = 149.8 ohm-1 cm 2 equiv-l;
TJw = 8.90 X 10-3 P; Ew = 78.3; D KC1 = 1.92 X 10-5 cm 2 sec-I.
Deviations from Ohm's law can also result at high fields since it
is then no longer a good approximation to relate the drift velocity
linearly to the field strength. For the above solution, what field strength
is required to bring about a 5 % deviation from Ohm's law by this
effect?

r:j1l What, in general terms, is the primitive of Fick's second law?

~ Why are three initial and boundary conditions needed for the
solution of this differential equation?
Use Laplace transformations to deduce the concentration response
in the solution adjacent to soluble metal electrode when a constant
anodic current is switched on at t = O. Correspondingly, show that,
if a sinusoidal current is applied about the reversible potential, then

C(x AnF (DW)-1/2 cos (7T

= 0) = Co - Imax wt - 4 ) (4.11.1)

where Co is the concentration of the metal ions of charge n in the

bulk of the solution, Imax is the amplitude of the current wave of
frequency W/27T, A is the area of the electrode, and D is the diffusion
coefficient of the metal ions in aqueous solution.

Radiotracer techniques have been used to measure the diffusion

coefficients of Cl- and Na+ ions in fused sodium chloride.
These values are given by

DCI - = 1.9 X 10-3 exp( -7400/RT)cm2 sec-1

DNa+ = 2.1 X 10-3 exp( - 7100/RT) cm 2 sec-1

Calculate the equivalent conductivity of fused sodium chloride at

 ANSWERS 35

935°C using the Nernst-Einstein equation and compare the result with
the experimental value of 156.4.
Discuss this discrepancy in terms of the Onsager reciprocity
relations. The Nernst-Einstein equation is a phenomenological relation.
Can it be incorrect? Suggest structural hypotheses for its inapplicability.

ANSWERS

The absolute Stokes mobility of an ion Uab is given by

(4.1.1)

where r is the radius of the ion and 1] the viscosity of the solvent.
Substituting numerical values into (4.1.1) yields

Uab = (67T X 2 X lO-B X 0.01)-1

= 2.7 X lOB cm sec-1 dyne- 1

The conventional mobility Uconv is given by

Uconv = ze o/18007Tr1]
4.8 X 10-10
18007T X 2 X IO-B x 0.01
= 4.31 X 10-4 cm 2 sec-1 V-I (4.1.2)

The root mean square distance Xrms traveled by an ion executing

random walk motions in time t is given by

Xrms = (2Dt)I/2 (4.3.1)

where D is the diffusion coefficient. Substituting given values into

(4.3.1) gives

Xrms = (2 x 2.03 x 10-5 x 60)1/2

= 4.9 X 10-2 cm
36 CHAPTER 4: ION TRANSPORT IN SOLUTIONS

The concentration of chloride ions C at any time t and distance

from the electrode x is given by the solution of Fick's second law. The
initial condition

C[t = 0] = Co (4.3.2)

and the boundary conditions:

C[x- 00] = Co (4.3.3)

and
(OC/ox)rIJ=o = -JID (4.3.4)

must be satisfied. Condition (4.3.2) implies that, before the current

step is switched on, the concentration of chloride ions is uniform and
everywhere equal to that in the bulk of the solution, Co. The implication
of condition (4.3.3) is that, far from the electrode, the concentration
of chloride ions remains unchanged, while the implication of condition
(4.3.4) is that the flux J of chloride ions at the electrode is described
by Fick's first law. The solution of Fick's second law, under these
conditions, is given by

J [2t1/2 ( X2) X (X2 )1/2] (4.3.5)

C = Co - D1/2 7(1/2 exp - 4Dt - D1/2 erfc 4Dt

where the flux is given by

J = ilF mole cm-2 sec-1

and i is the current density ofthe step. A consideration of units in (4.3.5)
shows that the concentration terms must be expressed in moles cm-3 •
Substituting numerical values into (4.3.5) yields

_ 10-3 [2 ( (5 X 10-3)2 )
C = 10 6 - 96,500(2.03 X 10-5)1/2 7(1/2 exp - 4 x 2.03 X 105 x 1

5X 10- 3 (5 X 10- 3)2 )1/2]

- (2.03 X 10-5)1/2 erfc 4 x 2.03 X 105 x1
= 10-6 - 2.3 x 10-6 [0.830 - 1.11 erfc(0.554)] (4.3.6)

Since
erfc(X) = 1 - erf(X) (4.3.7)
 ANSWERS 37

by finding the error function erf(0.554) from tables, we arrive at

C = 10-6 - 2.3 x 10-6 (0.830 - 1.11 x 0.435)

= 2.0 x 10-7 mole cm-3
= 2.0 x 10-4 mole liter-1

~ Walden's product A7], where A is the equivalent conductance

~ of the ionic species and 7] the viscosity of the solvent, can be
approached theoretically by eliminating the diffusion coefficient D
from the Nernst-Einstein equation,

A = (Zf2/RT)D (4.5.1)

and the Stokes-Einstein equation,

D = kT/67Tr7] (4.5.2)
The result is
(4.5.3)

Substituting numerical values into (4.5.3) gives

A _ 4.8 X 10-10 x 96,500

7] - 67T X 3 X 10-8
= 82 esu coulomb cm-1 equiv-1 (4.5.4)

The conventional units of Walden's product are mho cmS equiv-1 P.

One way of converting the units is by using

esu coulomb esu sec V mho

cm equiv cm equiv
0.333 x 1O-2(esu)2 sec mho
cm2 equiv
= 0.333 x 10-2 dyne sec mho equiv-1
= 0.333 x 10-2 mho cm2 equiv-1 poise
Consequently, Walden's product in conventional units is obtained by
multiplying (4.5.4) by 0.333 X 10-2 ,

A7] = 0.272 mho cms equiv-1 poise

38 CHAPTER 4: ION TRANSPORT IN SOLUTIONS

In methanol, the individual ionic equivalent conductivity is found

from Walden's product to be

A = 0.272/5.46 X 10-3 mho cm2 equiv-1

= 49.8 mho cm 2 equiv- 1

,.-:;-, The potential difference between two solutions in contact is

L2J the diffusion potential resulting from the diffusion of all the
ions in the two solutions. For two solutions designated I and II, we
have from Planck (l890a,b) (cf. Bockris and Reddy, 1970):

ljJI - ljJII =
RT
- I III -ti d(ln ai) (4.7.1)
F i I Zi

The transport number of the ith species, t i , is defined in terms of the

concentrations Ci (in equivalents per unit volume) and conventional
mobilities Ui of the species in the two solutions by

ti = CiUi FX /I CiUiFX
i
(4.7.2)

where X is the electric field. Since X is constant, (4.7.2) becomes

ti = CiUi/I CiUi (4.7.3)

i

Assuming ideal solutions, we can write d(ln Ci ) = dCi/Ci for

d(ln ai) in (4.7.1); substituting (4.7.3) for ti yields

ljJ _ ljJ
I II
= RT I
F
III IiUi/ZiCiUi dC·'
I
(4.7.4)

Making the assumption that the concentration of the ith species at

some point in the diffusion layer, Ci(x), is given by

C;(x) = Ci,l + (Ci,II - Ci,l) x

where x is a positive fraction, (4.7.4) becomes
 ANSWERS 39

Evaluating the integral of (4.7.5) yields

(4.7.6)

and (4.7.6) is known as the Henderson equation (Henderson, 1907,

1908).
For a solution of pure potassium chloride, we have, from (4.7.3),

(4.7.7)

and rearranging yields

(4.7.8)

and substituting for t K + the value of 0.491 (Table 4.7.1) into (4.7.8)
gives
(4.7.9)

Similarly, we have, making use of the data of Table 4.7.1,

UH + = 4.760u K + (4.7.10)
u ca2 + = 0.808u K + (4.7.11)

Consider now the liquid-liquid junction

I II
KCI(sat)IIHCI(O.OI M)

The concentration of saturated KCl solution at room temperature is

4.2 M; substituting the values for concentrations and ionic mobilities
given by Eqs. (4.7.9) and (4.7.10) in (4.7.6) yields

ifi ~ ifi ~ RT [0.01(4.760 ~ 1.037) ~ 4.2(1 ~ 1.037)

I II ~ F 0.01(4.760 + 1.037) ~ 4.2(1 + 1.037)
I 0.01(4.760 + 1.037)]
X n 4.2(1 + 1.037)
= 0.059(0.197 / ~8.502) log(0.058/8.56)
= 0.0029 V
40 CHAPTER 4: ION TRANSPORT IN SOLUTIONS

Similarly, for the liquid-liquid junction

I II
KC1(sat)IICaCI 2(5 M)

we have, by substituting (4.7.9) and (4.7.11) into (4.7.6),

if! - iflI = 0.059( -6.18/9.89) log(l8.45/8.56)

= 0.012 V

For the association of sodium chloride

Na+ + Cl- +t NaCI (4.9.1)

the association constant KA is defined by

(4.9.2)

where CNac1 ' CNa +, and CCl- are the actual concentrations of the
species NaCl, Na+, and Cl-, respectively, existing in the solution.
The Fuoss approach to the estimation of association constants
from conductivity data leads to the relation

(4.9.3)

where AO is the equivalent conductivity at infinite dilution, A that at

concentration C, and y± the mean ion activity coefficient on the molar
scale. The function Z is given by

(4.9.4)
where
(4.9.5)
and
A = (zeoF/9007T'Y})(87Tz2eo2NA/l000ekT)1/2 (4.9.6)
B = (eo2w/6ekT)(87Tz2eo2NA/l000ekT)1/2 (4.9.7)

where 'Y} is the viscosity and, for a 1:1 electrolyte, w = 0.5859.

 ANSWERS 41

From Eq. (4.9.3), it is seen that the association constant can be

found from the slope of the plot of Z/11 against Cy±2/Z(110)2. However,
it is first necessary to evaluate 11° and Z. Recalling Kohlrausch's law

(4.9.8)

where A * is a positive constant, it can be seen that 11° can be obtained

from the extrapolation of a plot of 11 against Cl/2 to infinite dilution.
The procedure does not yield highly accurate values of 11° since the
degree of association changes with concentration and A* is not strictly
a constant under these circumstances. Figure 4.9.1 shows a plot of 11
against Cl/2 for the data of Table 4.9.2. The equivalent conductances
at infinite dilution at 60, 80, and 90 % C3 HPH are 47.4, 38.0, and
35.4 mho cm 2 equiv-l, respectively.
The term Z is evaluated making use of Eqs. (4.9.4)-(4.9.7). Values
of A and B for various solvent compositions are listed in Table 4.9.3.
Values of Z and intermediate values in the calculation of Z are listed
for various salt concentrations in 60 % C3 H 70H in Table 4.9.4.
Table 4.9.5 lists values of Z/11 and I1Cy±2/Z(110)2 for 60 %C 3 HPH

50

Fig. 4.9.1. Determination of ,10 by extrapolating plots of A against Cl/2 to infinite

dilution for sodium chloride in 60, 80, and 90 % CaH70H solutions.
42 CHAPTER 4: ION TRANSPORT IN SOLUTIONS

.~
.0240
0/
0/
:::J
0-
w

0/
N
I

E
0

/
E .0230
..c:
0

/'
<
'"
N

.0220 5
1.0 1.5 2.0
104 A Cy~ / Z (AO)2 (ohm cm- 2 equiv.M)
Fig. 4.9.2. Plot of ZjA against ACy±2jZ(AO)2 for sodium chloride in 60%
CsH,OH at 35°C.

TABLE 4.9.3. Values of A and B for I-Propanol-Water Mixtures

Wt.%
I-propanol A B

60.0 41.2 0.75

80.0 52.5 1.24
90.0 61.2 1.56

TABLE 4.9.4. Calculations of Z for Various Salt Concentrations

in 60% I-Propanol-Water Mixture at 35°C

104 CNaC1 ,

M 10z* ! cos- 1 [t(3s/ 2z*)] Z

5.089 11.28 29° 57' 0.9986

11.130 16.45 29° 55' 0.9993
18.452 20.92 29° 54' 0.9995
24.239 23.97 29° 53' 0.9998
33.823 27.97 29° 52' 1.0002
 ANSWERS 43

TABLE 4.9.5. Values of Z/A and ACy±2/(AO)2Z

for 60% C 3 H,OH at 35°C

Z,'11, ACy± 2/Z(AO)2,

ohm em" equiv ohm em- 2 equiv M

0.02203 0.711 x 10- 4

0.02257 1.11 x 10- 4
0.02306 1.46 x 10-4
0.02338 1.70 x 10- 4
0.02384 2.02 x 10- 4

at 35°C and these values are plotted in Fig. 4.9.2. From the slope of
this plot, the association constant KA is found to 13.4 liters mole-I.
A more precise method of finding KA from the given data is the
numerical solution of (4.9.3). Values of AO are assumed and values of
KA calculated at each concentration until the value of AO which yields
a value of KA independent of concentration is obtained. Table 4.9.6
lists computer calculations of AO and KA obtained in this way (Goffredi
and Shedlovsky, 1967) for 60, 80, and 90 % C 3 H 70H. It can be seen
that there is close agreement between our values of AO and KA and
the values obtained by the numerical method. However, this would
probably not be the case for less precise data.

TABLE 4.9.6. Association Constants and Values of AO

for NaCI in I-Propanol-Water Mixtures at 35°C

Wt. % C 3 H,OH AU, mho em 2 equiv- 1 KA , liters mole- 1

60 47.38 13 ± I
80 37.96 45 ± 1
90 35.34 158 ± 3
 CHAPTER 5

PROTONS IN SOLUTION

r;l H 90 4+ is often given as a reasonable configuration for the

L.!J proton in solution. Using the Born equation and the equations
for ion-dipole and ion-quadrupole interactions, show that the proton
affinity of water must be about 180 kcal mole-1 . The absolute individual
heat of hydration of protons in water is -263 kcal mole-1 .

Ew = 78.5; dEwldT = -0.356 deg- 1

/Lw = 1.87 x 10-18 esu cm; rH

3
0+ = 1.05 A

rw = 1.38 A; Pw = 3.9 X 10-26 esu cm 2

Using the data of Table 5.2.1, deduce the percentage Grotthuss

conduction, or anomalous proton conduction, of each solvent.

TABLE 5.2.1. Equivalent Conductance at Infinite Dilution of Hydrogen

and Lithium Chlorides in Water, Methanol, Ethanol, and Propanol at 20°C

Solvent
Property investigated
Water Methanol Ethanol n-Propanol

11 of hydrogen chloride 426.2 192 84.3 22

11 of lithium chloride 115.0 90.9 38.0 18

The proton affinity of water is -182 kcal mole-1. Using the

data of Table 5.3.1, calculate the deuteron affinity of water.
45
46 CHAPTER 5: PROTONS IN SOLUTION

TABLE 5.3.1.
Zero-Point Vibrational Frequencies
(em-I) for H 30+ and DH20+ *

H30 + DH 2O+

3500 3500
3500 3500
3500 2600
1520 1520
1520 1320
1520 1320

* Data from Conway (I958).

r7l (a) The equivalent conductance of HCI in methanol-water

~ mixtures plotted against the mole per cent methanol gives
a minimum at a certain concentration of methanol. Why?
(b) Why is anomalous proton mobility most pronounced in water?
(c) Give examples of amphiprotic solvents.
(d) Define proton free energy levels.
(e) Explain with examples: "primary solvent effect on acid
strength. "

r;l From classical mechanics, show that the frequency of free

~ rotation of a water molecule is about 1013 sec-I. However, the
dielectric relaxation time is 0.55 x 10-11 sec-I, with an activation
energy of 3.8 kcal mole- l at 300 o K. Interpret these two results qualita-
tively. The moment of inertia of a water molecule is 2 X 10-40 g cm 2 •
Show that the rate of proton transfer per water molecule is about
5 M x 1011 sec-I, where M is the molarity of protons in the solution.
The actual rate of proton exchange transfer in either direction has
been shown to be 7.5 x 105 cm sec- l (Conway et al., 1956). Take the
radius of a water molecule as 1.4 A.
The rotation of a water molecule in solution requires the breaking
and reformation of a number of hydrogen bonds to neighboring water
molecules in the water structure. An approximate value of the minimum
activation energy for this process may be taken as that estimated for
 PROBLEMS 47

the rotation of a single water molecule, 10 kcal mole-i , which neglects

the cooperative rotation of neighboring water molecules (Conway et al.,
1956). Calculate the maximum rate of thermal reorientation of water
molecules. What considerations lead you to a decision as to whether
this breaking is rate-determining in aqueous solutions? Calculate your
decision.

f6l In solution, the vibrational levels in HaO+ are changed by the

~ interaction of HaO+ with the solvent. Several IR and Raman
studies have been done on aqueous solutions of strong acids, but
assignment of frequencies from such work is difficult because the
hydroxonium ion bands appear as broad shoulders on the H 20 bands.
O'Ferrall et al. (1971) have made a detailed vibrational analysis of the
hydroxonium ion in solutions and found that their vibrational fre-
quencies agreed with the values experimentally obtained for the crystal
of H 90 4+Br- but not with the experimental values obtained for the
hydroxonium halides salts in solutions of sulfur dioxide. This seems
to indicate that the hydroxonium ion in solutions exist more like an
H 904+ than HaO+.

TABLE 5.6.1. Calculated and Observed Frequencies* for Solutions

of HaO+ Salts in SO. and Crystalline H 90, +Br-

HaO+/SO• HgO,+Br-
IRobs Raman Calc. IRobs Calc.

vI(a) 3405 3415 3410 2060 2640

va(c) 3470 3447 3455 2630 2060
v.(a) 785 1313 1320
v,(c) 1670 1700 1690 1845 1840
vL(a) 902 870
(c) 738 920
.,.(a) 569 405
(c) 556 570
Vb 78 97-27

* Frequencies in units of em-I. VI, V. , Va, and V, are the bending and stretching
frequencies, while VL , Vr, and Vb have been assigned to Iibration, hindered trans-
lation, and hydrogen-bond bending, respectively.
48 CHAPTER 5: PROTONS IN SOLUTION

The wave numbers for the different vibrations are tabulated in

Table 5.6.1. Levich (1970) asserts that a significant rate of transfer of
electrons from an electrode to a proton in HaO+ cannot occur because
(a) the difference in successive energy levels in HaO+ is so large that
Tafel lines would not be smooth; (b) there will not be a sufficient
occupancy of higher vibrational or rotational levels to maintain the
observed rate. Calculate the energy interval (in meV) between rotational
states in HaO+. Calculate the needed number of states to accommodate
typical reaction rates (e.g., 10-10 A cm- 2, with a heat of activation of
20 kcal mole- 1). Thus, examine whether the thermal equilibrium of
H30+ with water provides a situation consistent with this mechanism
of the activation.

r:;-] The pKw values for pure ice and pure supercooled water at
L2J -10°C are 20.6 and 15.1, respectively, and the specific con-
ductivities are practically the same with a value of 1.4 x 10-9 mho cm- I •
Calculate the proton mobility in ice and water at -lOoC taking the
transport number of the proton in ice to be the same as that in water,
0.64.
Give a qualitative discussion of the effects of proton concentration
on the mechanism of proton transfer. Using the results of Problem 5
and the information that the spontaneous thermal rotation rate of
water molecules in ice is 106 sec-l, show that the proton mobility
in ice is consistent with proton tunneling as the rate-determining step
for proton transfer in ice.

Assess the relative contributions of Wynne-Jones, Gurney,

and Bell to the theory of the strength of acids.

f9l Proton free energy levels (pfel) are shown in Fig. 5.9.1. Use
~ these data to calculate the pKA values for acetic and chloro-
acetic acids. What is the relative strength of chloroacetic acid with
respect to acetic acid? Estimate the relative strength of chi oro acetic
acid with respect to .B-chloropropionic acid in methanol. Take
Ew = 78.5 and EMeOH = 32.5 and the effective radii of the .B-chloro-

propionic and chloroacetic ions as 4.2 A and 2.9 A, respectively.

KA for .B-chloropropionic acid is 1.0 X 10-4 •
 PROBLEMS 49

I
e:
0
-'" IJ
~

u-
CII-
-0
w>
--~O+ 0.0 HzO--
1.0

-0.1
0.9

-0.2
H5~-
0.8 504 --
HPO-;= 'If04-
CH~ICOOH- CHzCICOO-
-0.3
HCOOH HCOO-
0]
CeH!l NH 3!.......- CJ-i!lNHz
CH!COOH -0.4 CHj:OO-
III 0.6
CII
> HZC0 3- HC0 3- III
.!! -0.5 CII
e: >
0
HzP04- - 0.5 HP04-- .!!
e:
0 0
~ -0.6 0
-
~
~
'0 NH4+ 04 NH3
.!! BO z- e:
~ HBOl 0
::J u
u N~+CHz C 0 0 - - -0.7 NH 2 CH2 COO- 0
u :>
0 HC03-- C05--
0.3

-0,8
02

-09
0.1

-1.0
H2 O 0.0 OW
I
c:
-1.1 ~III
u:=
ClIO
i;j>

Fig. 5.9.1. Proton free energy levels for a selection of acids in aqueous
solution.
50 CHAPTER 5: PROTONS IN SOLUTION

~ Proton transfer from an H30+ to a suitably oriented H 20 in

~ solution takes place by tunneling. Write the Schrodinger
equation, in one dimension, appropriate to this situation and find the
probability that a proton which presents itself at the barrier will
tunnel. Assume an Eckhardt barrier and express the probability that
the proton will tunnel at an energy W, in terms of the barrier width,
the energy of activation, and the difference in the initial- and final-state
energies. Express the initial- and final-state energies in terms of the
applied field.
Show that for the low-field approximation and assuming a classical
distribution of protons,

tV - tv = (No/kT) If
(J
Pwexp[-(W -1>x)/kTJdw

(5.10.1)

~ .
where Nand N are the numbers of particles that can permeate the
barrier in the forward and reverse directions, respectively, in unit
time; No is the number of particles approaching the barrier in unit
time that may permeate it; 1>x is a function of the applied field; and

cosh 2KWI/2 - cosh (1)x/ W) KWI/2

cosh 2KWI/2 - cosh 7T{[4(E* - 1>",) - C]/C}1/2
cosh 2KWI/2 - cosh (1)x/ W) KWI/2
cosh 2KWI/2 - cosh 7T{[4(E* + 1>x) - CJ/Cp/2

where K = 7T/(2m)1/2/h, with m the mass of the proton and I the barrier
half-width; 1> is the electric potential energy in the activated state;
C = (h 2 /8) m1 2 ; W, the total energy, is a function of the position x
across the barrier; E* is the activation energy.

f11l Use the Morse equation quantitatively to construct potential

~ energy curves to represent the transfer of a proton from H30+
to H 20 in solution. Take the radius of the hydrogen atom as 0.3 A,
the OH bond distance as 1.05 A, and the effective radius of a water
molecule in a hydrogen-bonded situation as 1.4 A. The force constant
of the OH bond is 5.31 x 10-5 dyne cm-I, the dissociation energy of
 PROBLEMS 51

a proton from H30+ is 180 kcal mole-I, and the dissociation energy
of an H30+ in water is 263 kcal mole-l. The resonance energy at the
activated complex may be approximated as about half the hydrogen
bond energy.
Make a numerical calculation of the fraction of protons that tunnel
through the barrier using the considerations expressed in Problem 10
and assuming a classical distribution of protons.
Eigen and deMayer (1958) rightly suggested that a quantized
distribution of proton energies should be used to calculate the fraction
of protons tunneling through the barrier. By numerical calculation,
show that the fraction of protons tunneling through the barrier cal-
culated in this way is about twice that calculated using a classical
distribution. Calculate the net rate of proton permeation through the
barrier. What conclusion concerning the mechanism of proton migra-
tion can be calculated from this value?

f12l Figures 5.12.1 and 5.12.2 show the partial molar heat capacities
~ of the proton in an aqueous solution, and its partial molar
volume over the same temperature range.
Derive a statistical mechanical equation for the heat capacity of
the proton in solution in terms of the partition functions of the various

+20

I
0>
'0
CII 0
CU
0..
u -10

20 40 60 80 100 120
TEMPERATURE "C
Fig. 5.12.1. Experimental values of the apparent
molal heat capacity of the proton in aqueous solution as
a function of temperature. (Data from Conway. 1964.)
52 CHAPTER 5: PROTONS IN SOLUTION

- a
I
Q)
"0

~:2o~
u 4 ~o
I>
6

20 40 60 80 100
TEMPERATURE DC
Fig. 5.12.2. Experimental values of the apparent molal
volume of the proton in aqueous solution as a function of
temperature. (Data from Conway, 1964.)

bonds which H+ forms with its surrounding water sheath. Attempt to

make the model which you suggest consistent with the data given.
Make a plot of your final relation between heat capacity and temperature.
Interpret the results in such a way that information on the vibra-
tional frequencies of the solvated proton is obtained. Discuss whether
the O-H bond has a continuum of vibrational energy levels in its
interaction in HaO+ and with other waters.

ANSWERS

r:jl When a proton dissolves in water, there are essentially three

~ contributions to the heat of hydration of the proton LJHH +_ h •
These contributions are:
(a) The heat of the reaction:
H+ +H0 2 - H30+ (5.1.1)

which is the proton affinity of water.

(b) The heat of interaction of H30+ with three water molecules,
LJHH 3 OLw, to form the hydrated species H 90 4 +. This interaction
 ANSWERS 53

energy is given by the sum of the interactions of the ion with three
water dipoles of moment I-'w, and the interaction with three water
quadrupoles of moment Pw :

(5.1.2)

(c) The heat of interaction of the species H 90 4 + with the solvent,

LJHBC' Regarding the solvent as a continuous dielectric medium, this
term is given by the heat of Born charging a sphere of radius equal to
that of H 90 4+ to a charge +eo , in a medium of dielectric constant Ew :

LJHBc = _ NAe 02 (I __1__ ~ OEw) (5.1.3)

2(r HaO ++2rw) Ew Ew20T
Consequently,
LJHH+_ h = LJHH +_ H20 + LJHHaoLw + LJH BC (5.1.4)

Substituting numerical values into (5.1.2), we have

LJH = _ 6.023 X 1023 x 3 x 4.8 X 10-10 x 1.87 X 10-18

Ha O +- w (1.05 + 1.38)2 x 10-16
6.023 X 1023 x 4.8 X 10-10 x 3.9 X 10-26 x 3
+ 2(1.05 + 1.38)3 x 10-24
= -27.46 x 1011 + 11. 79 x 1011 ergs mole- 1
= -65.62 + 28.16 kcal mole- 1
Substituting numerical values into (5.1.3), we have

__ 6.023 X1023 x (4.8)2 X 10-20 ( _ _1_ 298 x 0.356 )

LJHBC - 2(1.05 + 2.76) 1 78.5 + (78.5)2
= -18.29 X 1011 ergs mole- 1
= -43.72 kcal mole-1
Substituting for LJHH+_ h , LJHHaO+-w, and LJHBC in (5.1.4) yields

LJHH +_ HaO = -263 + 65.6 - 28.2 + 43.7 kcal mole-1

= -181.9 kcal mole- 1
54 CHAPTER 5: PROTONS IN SOLUTION

The proton affinity of water is the heat change of the reaction

(5.3.1)

when it occurs in the gas phase. Similarly, the deuteron affinity of water
is the heat change of the reaction

(5.3.2)

when it occurs in the gas phase.

At OOK, the difference between the deuteron and proton affinities
of water, LJH(D+-H+)H 20 , is given by

(5.3.3)

where the summations are the total zero-point energies of the sub-
scripted species. Equation (5.3.3) is valid because the electronic contri-
butions to the total energies of H30+ and DH 2 0+ are constant and, at
OOK, the rotational and translational contributions to the total energy
vanish.
Equation (5.3.3) can be written

(5.3.4)

Substituting numerical values in (5.3.4) from Table 5.3.1 yields

LJH(D+_H+)H.O = 0.5 x 6.623 x 10-27 x (13.76 - 15.06) x 103

= -4.30 X 10-24 erg sec cm-l

Multiplying by Avogadro's number and the velocity of light (in cm

sec-I), we have

LJH(D+-H+)H2 0 = -4.30 X 10-24 x 6.023 X 1023 X 3.0 X 1010

= -77.7 X 109 ergs mole- l

= -1.85 kcal mole-l
 ANSWERS 55

since LlHH +_ H 20 is equal to -182 kcal mole-I, then

If we assume a water molecule to be a free, classical rotator,

then
(5.5.1)

where I is the moment of inertia of the water molecule and w the

angular frequency of rotation. Substituting the given value for I in
(5.5.1), we have

w = (1.38 X 10-16 x 298/2.0 X 10-40 )1/2

= 1.43 X 1013 rad sec- 1

The fact that the measured dielectric relaxation time is much

longer than the time of rotation of a free water molecule confirms that
liquid water does not consist of free molecules. Hydrogen bonding
between water molecules gives liquid water a structure similar to a
somewhat broken-down and slightly expanded form of the ice structure.
Consider an aqueous solution, of molarity M, of a completely
dissociated monobasic acid. The number of protons in a lamina 1 cm 2
in cross-sectional area and one water molecule thick is

where rw is the radius of a water molecule in centimeters. Since the

actual rate of proton exchange transfer (velocity) in either direction is
given as 7.5 x 105 cm sec-I, the time fOl: the protons to pass through
the lamina is 2r w/(7.5 x 105 ) sec. Consequently, the number of protons
passing through the lamina in 1 sec is
2rwMNA x 7.5 x 105 /1000 x 2rw
Since there are 2rwNA x 55.5/1000 water molecules in the lamina,
then the number of protons passing through one water molecule per
second is

which on substitution for rw becomes

M x 7.5 x 105 /55.5 x 2.8 x 10-8 = M x 4.87 X 1011 sec-1
S6 CHAPTER S: PROTONS IN SOLUTION

If water rotation has an energy of activation of 10 kcal mole-1

resulting from the energy to break hydrogen bonds, then the angular
frequency of rotation of a water molecule in the structured liquid is
given by
Wb = W,r exp( -10,0001 RT)

where Wb and W'T are the angular frequencies of the bound and freely
rotating water molecules, respectively. Substituting 1.43 x 1013 rad sec-1
for W'T' we have

Wb = 1.43 x 1013 exp( -10,000/1.987 X 298)

= 8.82 x 105 rad sec-1
= 1.40 X 105 sec-1

Except at low proton concentrations, about 10-7 or less, the rate

of proton transfer, M X 4.87 X 1011 sec-I, is much greater rate of
rotation of the water molecule. Consequently, it is the latter step that
must be rate-determining for proton migration.

In pure water or ice, the hydrogen ion concentration is given

by

(5.7.1)

Substituting the given values of the pKw for ice and water at -lO°C
into (5.7.1), we have

CW - 1ce = 5.0 X 10-11 mole liter- 1

and

CH+-water = 2.8 X 10-8 mole liter-1

The hydroxide ion concentrations are, of course, equal to the hydrogen
ion concentrations.
The specific conductivity J of a 1: 1 electrolyte solution is given
by
(5.7.2)

where u-,- and u_ are the conventional mobilities of the cation and
 ANSWERS 57

anion, respectively, and C is the concentration of the electrolyte in

moles cm-3 • Substituting for J and C in (5.7.2), we have

(U+ + U-)ice = 1.4 x 10-9/5.0 x 10-14 x 96500

= 0.290 cm 2 sec-I V-I

and
(U+ + U-)water = 1.4 x 10-9/2.8 x 10-11 x 96500
= 5.2 x 10-4 cm 2 sec- I V-I

Since, for a 1: 1 electrolyte, the transport number t+ is given by

(5.7.3)

we have, by making use of the given value of t+ ,

U+_ice = 0.290 x 0.64

= 0.185 cm 2 sec- I V-I

and
U+_water = 5.2 X 10-4 x 0.64
= 3.3 x 10-4 cm2 sec-I V-I

Consequently, although the specific conductivities of ice and

water are almost the same, the mobility of the proton in ice is about
103 times that in water.
In Problem 5, it was shown that the rate of proton transfer per
water molecule is M x 4.87 X 1011 sec-I, where M is the molarity
of the protons. Hence, as the concentration of protons is lowered,
any given water molecule then has passing across it, along a given
coordinate, a decreasing number of protons per second. The number
of times per second that a typical water molecule undergoes field-
induced rotation, so that it can receive the tunneling proton, is reduced
as the concentration of protons is reduced. At low concentration of
protons, the rate of rotation of the water molecule needed to pass on
the proton drops to less than the rate of thermal rotation of the
molecule. Proton tunneling is then the rate-determining step.
The rate of proton transfer per water molecule in ice is
4.87 x 1011 .x 5 X 10-11 sec-I. Since this value is much smaller than
the rate of spontaneous thermal rotation in ice, about 106 sec-I, proton
tunneling is rate-determining in ice.
58 CHAPTER 5: PROTONS IN SOLUTION

r;-J Looking at the left-hand scale of Fig. 5.9.l, we have that the
~ pfel of acetic acid and chloroacetic acid are -0.385 and
-0.275 eV, respectively. The left-hand scale of Fig. 5.9.l actually
gives the standard free energy of an occupied proton level in the acid
with respect to the standard free energy of the occupied proton level
in H30+. Consequently, for the process

(5.9.l)

where HA is an acid and A-is its conjugate base,

(5.9.2)

where UHA is the proton free energy level of HA and JG~A is the
standard free energy change of (5.9.1). Using the conversion factor

1 eV molecule- 1 = 23.06 kcal mole-1

we have that

JG~COH = 8.87 kcal mole-1

JG~l_ACOH = 6.63 kcal mole-1

The acid dissociation (5.9.l) is traditionally written in the form

HA~A-+H+ (5.9.3)

and the dissociation constant KA is defined as

(5.9.4)

The equilibrium constant K of (5.9.1) is given by

(5.9.5)

Neglecting activity coefficients, (5.9.4) and (5.9.5) may be written in

terms of the corresponding concentration terms. Consequently, KA and
K will be numerically equal only for dilute solution and when the
concentrations are expressed as mole fractions. In terms of the usual
molar units of concentration, we have

KA = KCH 2 0 (5.9.6)
 ANSWERS 59

since the protons referred to in equation (5.9.3) are bound to water

molecules and therefore
(5.9.7)

Since
K = exp( -LJGo/RT) (5.9.8)
KA = C H 0 exp( -LJGO/RT) (5.9.9)
2

Substituting for LJGo and taking C H20 as 55.5 moles liter-I, we have

(KA)acetic = l.75 x 10-5 M

(KA)chloroacetic = l.26 X 10-3 M

and since
pKA = -loglo KA
(pKA)acetic = 4.76
(pKA)chloroacetic = 2.90

The relative strength of one acid HAl with respect to that of

another HA2 in a given solvent is defined by the equilibrium constant
K' for the reaction
(5.9.10)

The larger the value of K', the greater the relative strength of HAl
with respect to HA2 . Now,

(5.9.11)
(5.9.12)

Substituting the given values for chloroacetic and ,B-chloropropionic

acids into (5.9.12) yields

K' = 1.3 x 1O-3 /l.O X 10-4

= 13

The equilibrium constant K' can be written in terms of the standard

free energy change of (5.9.10), LJG0 as 1
,

-RTln K' = LJGo, (5.9.13)

60 CHAPTER 5: PROTONS IN SOLUTION

Now, LlGo, can be separated into two terms, a chemical term LlG c
and an electrostatic term. Since A1 - and A2 - are the only charged
species participating in (5.9.10), the electrostatic term is simply the
difference between the charging energies of A1 - and A2 -. Hence,

In K' = _ LlG c _ NAe 02 (_1_ _ _1_) (5.9.14)

RT 2ERT rA 1 - rA • -
If K' is known in one solvent, LlG;/RTcan be evaluated and then
K' calculated in another solvent. Evaluating LJGc/RT for the given
data in aqueous solution, we have

_ LlG c = In K' + NAe 02 (_1_ _ _1_) (5.9.15)

RT 2ERT rA - rA -
1 •

and substituting numerical values into (5.9.15) yields

LlG c 6.023 x 1023 x (4.8)2 X 10-20 X 10-7

- RT = In(l3) + 2 x 78.5 x 8.314 x 298 x 10-8

x (2~9 - 4~2)
= 2.56 + 0.38
= 2.94

Substituting this value in (5.9.14) and taking the dielectric constant

of methanol as 32.5 yields

,_ _ 6.023 X 1023 x (4.8)2 X 10- 20 X 10-7 (_1_ _ _1_)

In K - 2.94 2 x 32.5 x 8.314 x 298 x 10-8 2.9 4.2
= 2.94 - 0.92
K'= 7.5

Therefore, the relative strength of chloroacetic acid to that of ,B-chloro-

propionic acid in methanol is 7.5. It should be noted that the above
procedure is valid only for similar acids and similar solvents; only then
c
can the chemical term LlG be regarded as constant.
 CHAPTER 6

IONIC LIQUIDS

The probability that a hole in a liquid will have a radius

between rand r + dr is P dr and is given by

(6.1.1)
where
a = 41TyjkT (6.1.2)

and y is the surface tension of the liquid. Show that the mean hole
volume Vh is given by

Vh = O.68(kTjy)3/2 (6.1.3)

Use the data of Table 6.1.1 to calculate the mean size of holes in
molten salts and compare these values with the volumes of the ions
concerned. Comment on the conclusion.

TABLE 6.1.1. Surface Tension and Ionic Radii

of Various Molten Salts

T, y, r+ , r_ ,
Salt DC dynes cm- 1 A A
NaCl 820 115 0.95 1.81
NaBr 785 97 0.95 1.95
NaT 670 87 0.95 2.16
KCl 800 97 1.33 1.81
KBr 750 88 1.33 1.95
KI 720 76 1.33 2.16
CaCl 2 585 83 1.21 1.81

61
62 CHAPTER 6: IONIC LIQUIDS

r:;r Table 6.2.1 lists the molar volume change on fusion of various
~ salts together with the free volume per mole calculated from
ultrasonic compressibility measurements. What model of molten salts
do the data support?
The free volume and hole volume of some molten salts obtained
at various temperatures are listed in Table 6.2.2. Plot the free volume
against the hole volume and make a physical interpretation of the result.

TABLE 6.2.1. Molar Volume Change of Fusion, 4V,

and the Molar Free Volume, V" of Various Salts

LI V, v, ,
Salt mlmole-1 ml mole- 1

LiCI 5.9 1.0

NaCl 7.4 0.75
NaBr 8.2 1.15
NaI 8.7 1.15
KCI 7.1 0.85
KBr 8.0 1.02
Kl 9.4 0.95
CsCI 13.3 0.78
CsBr 14.3 0.86

TABLE 6.2.2. Molar Free Volume V,

and Molar Hole Volume V. for Molten Salts

V, , V. ,
Salt ml mole- 1 ml mole- 1

LiCI 0.36 10.8

0.63 17.5
0.78 21.7
0.96 25.9

NaBr 0.51 20.5

0.57 25.3
0.75 31.3
0.93 38.5

KI 0.45 28.3
0.63 40.3
0.90 56.6
1.29 71.7
 PROBLEMS 63

r;l Assuming the Stokes-Einstein relation, calculate the anionic

~ transport numbers from the data listed in Table 6.3.1. Discuss
the results.
Derive a relation, making use of the Stokes-Einstein relation, for
the anionic transport number of a fused 2:1 electrolyte taking into
account the dissociation equilibria

MX 2 +t MX+ + X- (6.3.la)
MX+ +t M2+ + X- (6.3.lb)

Using the Pauling radii (rpbCl+ = 4.83 A; rpb2+ = l.21 A; rCl- = l.81 A)
and the experimental anionic transport number of 0.8, determine
whether Pb 2+ or PbCI+ is the predominant cationic species in fused
lead chloride.

TABLE 6.3.1. Experimental Anionic Transport Numbers

and Ionic Radii

Salt r+, A r_, A L

LiCI 0.60 1.81 0.25

NaCl 0.95 1.81 0.38
KCl 1.33 1.81 0.38
RbCI 1.48 1.81 0.42

r:j'l In the Temkin model of an ideal solution of molten salts,

L2..J each salt is considered to be completely ionized. Further,
if the physical properties of all the cations and all the anions, respec-
tively, are considered to be exactly the same, the total enthalpy and
energy of the solution will be that of the pure components. Show that
the entropy of mixing L1Smix is given by

(6.4.l)

where ni+ and ni - are the numbers of gram ions of cations and anions,
respectively, from component i of the solution, and N i + and N i - are
the ion fractions defined by

N·+
z = n.+/'"
z L n·+
z,
64 CHAPTER 6: IONIC LIQUIDS

~ In fused NaN0 3-KN03 at 263°C, Cd++ and Br- form a series

~ of complexes. Transition time measurements can, in this case,
be used to obtain information on the transformation of these com-
plexes. The transition time measurements are made by applying
cathodic galvanostatic pulses to a hanging mercury drop electrode
in the melt. The overall process is known to be

CdBr3 - ~ + Br-
CdBr 2 (6.5.la)
CdBr 2 ~ CdBr+ + Br- (6.5.lb)
CdBr+ ~ Cd++ + Br- (6.5.lc)
Cd++ + 2e- ~ Cd(Hg) (6.5.ld)

The formation constants Kl , K2 , and K3 corresponding to steps

(a), (b), and (c) are known to be 8 ± 3, 65 ± 33, and 100 ± 50
kg mole-I, respectively.
For processes represented by the reaction scheme (6.5.1), the
transition time T is given by

(6.5.2)

60

_IN
U
QJ
en 40
E
..-iN
f...

20

1.5
r-' (mA)-'
Fig. 6.5.1. Results of transition time experiments, T'/2 plotted
against [-1 for a NaNOa-KNOa melt containing 4.82 x lO- a mole
kg- I Cd2+ and 9.21 x 10- 2 mole kg- I KBr.
 PROBLEMS 65

where Z = (k+ + LCBr~ )1/271 / 2, with CBr - the bulk concentration of

Br- and C the sum of the bulk concentrations of all species containing
Cd++; i is the cathodic current density, and K, k+ , and k_ refer respec-
tively, to the formation constant, the forward rate constant, and the
reverse rate constant of the rate-determining step. Devise a means of
calculating K, k" , and L . To which step in the overall process do
these quantities refer?
Results of experiments are shown in Fig. 6.5.1, for which CBr - =
9.21 X 10- 2 mole kg- 1 and C = 4.82 X 10-3 mole kg-i.

r-;-l (a) Transition times for silver deposition from fused NaN0 3 at
~ 330°C are listed for various melt compositions in Table 6.6.1;
a cathodic pulse of constant current density is being used. Calculate
the solubility product of silver chloride under these conditions.
(b) The data of Table 6.6.2 refer to a constant melt composition
of 2.01 x 10- 2 M AgN0 3 in fused NaN0 3 at various temperatures.
Calculate the activation energy and preexponential factor for diffusion
of silver ions in fused NaN0 3 over the temperature range 280-317°C.

TABLE 6.6.1. Transition Times for Silver Deposition

from NaN0 3 Melts of Various Compositions at 330°C

IO'CAgN03 ' IO"CNaC1 , T,

molal molal msec

2.49 0 805
0.52 805
0.93 650
1.83 400
2.44 275
2.93 210
3.77 155

~ In the "density fluctuation hole model" of molten salts, the

L2J holes are thermally distributed in size. Hole formation in these
liquids can be formally treated as a chemical reaction and
(6.7.1)
66 CHAPTER 6: IONIC LIQUIDS

TABLE 6.6.2. Transition Times for Silver Deposition

from NaN0 3 Melts of Composition 2.01 x 10-2 M AgN0 3
at Various Temperatures and Current Densities

Temp., Current density, Transition time,

°C rnA cm- 2 msec

317 \0 470
7 960
5 1880
3 5220
297 \0 383
7 780
5 1530
3 4250
280 \0 324
7 660
5 1290
3 3600

where Nh and N are the equilibrium numbers of holes and ions per
mole of liquid, respectively, and LlG h 0 is the standard free energy of
hole formation. The diffusion process can be visualized as the net result
of a density fluctuation in the salt which momentarily provides a local
cavity and the jumping of a sufficiently activated ion into the cavity.
If the probability of a successful collision between an ion and an
adjacent hole is given by exp( -LlG//RT), where LlG;* is the free
energy for jumping, show that
(6.7.2)
where Ep is the activation energy for diffusion at constant pressure,
LlHh o is the standard enthalpy of hole formation, and LlE;* is the
activation energy for the jump of an ion into a hole.
Utilizing the information that the change in volume of a melt with
temperature is predominantly due to the change in the number of
holes and only in a small part due to the change in the most probable
hole volume, rationalize the relation
(6.7.3)
where Ev is the energy of activation for diffusion at constant volume.
 PROBLEMS 67

..
-;
0
E
....
E

..
u

E
::>
(5
>
...
0
0
~

44

430~------~~~------~~~----~12~OO~--------~
Pressure (0 tm.l
Fig. 6.7.1. Plot of the molar volume of molten NaN0 3 as a function of pressure
at various temperatures. (Data from Nagarajan and Bockris, 1966.)

Using the data of Fig. 6.7.1 and Table 6.7.1, calculate E1I and Ev
for sodium diffusion in fused sodium nitrate. Discuss whether or not the
values you calculate are consistent with the density fluctuation hole
model.

101 Electrical conductance, molar volume, surface tension, UV

~ spectra, vapor pressure, and self-diffusion studies on molten
CdCI 2- KCl mixtures all indicate the formation of complex ions.
Tetrahedral CdCl;-, planar triangular CdCI 3 -, and pyramidal CdCI 3 -
68 CHAPTER 6: IONIC LIQUIDS

TABLE 6.7.1. Effect of Pressure and Temperature

on the Diffusion Coefficient of 22Na in NaN0 3 *

10' Diffusion coefficient, cm 2 sec'

Pressure,
atm 350°C 370°C 390°C

I 2.14 2.38 2.61 3.04

400 1.91 2.16 2.32 2.65
600 2.21
800 1.74 1.99 2.08 2.50
1200 1.58

* Data from Nagarajan and Bockris, 1966.

have been suggested as the most probable complex ions in this melt.
The complex ion formation has been studied by Raman spectroscopy
and Table 6.8.1 summarizes the Raman spectra attributed to the
complex ion.
What are. the point groups for the complex ions suggested above
and how many Raman frequencies would you expect for each complex
ion and what is the state of polarization of each? Compare the results
of these considerations with the data of Table 6.8.1 and decide which is
the predominant complex ion formed under the conditions of the
Raman study.

TABLE 6.8.1. Summary of Raman Spectroscopic Data

for Molten CdCI 2-KCI Mixture

Composition, Temperature Frequency State of

mole%KCI range,OC Llv, cm~l Intensity polarization

50 482-708 257 ± 2 Strong Polarized

245 c::: 3 Weak Depolarized
211 =6 Weak Polarized
177 +5 Very weak Depolarized

Show that for a diffusion process in a molten salt,

Ll V* -RT(o In DloP)T (6.9.1)

 PROBLEMS 69

where D is the diffusion coefficient and Ll V* the activation volume.

Further, show that for the density fluctuation hole model,
Ll V* = Vh + Ll V/ (6.9.2)
where Vh is the most probable molar volume of holes in the melt and
Ll V/ is the activation volume for the jumping process.
Using the data of Tables 6.7.1 and 6.9.1 and Fig. 6.7.1, calculate
the activation volume for diffusion of Na 22 and CSI34 in fused sodium
nitrate. Calculate Ll V/ for each diffusion process and comment on the
numerical values of these terms. Are they consistent with the density
fluctuation hole model? Take the surface tension of fused NaN03 as
110 dynes cm-I at 390°C.

TABLE 6.9.1. Effect of Pressure and Temperature

on the Diffusion Coefficient of 134Cs in Fused NaNOs *

105 Diffusion coefficient, em" sec-1

Pressure,
atm 350°C 370°C 390°C 420°C

2.3\ 2.62 2.90 3.39

400 2.05 2.35 2.64 3.03
800 1.83 2.07 2.42

* Data from Nagarajan and Bockris, 1966.

~ The following facts are known about simple liquid silicates:

(a) The heat of activation for viscous flow is very high (134 kcal
mol-I) for pure Si0 2 (Fig. 6.10.1). However, during the addition of the
first 10 mole % M 2 0, the heat of activation for viscous flow drops to
about 40 kcal mole- I and thereafter decreased only slightly on the
addition of further M20 down to ",,50 mole %M20.
(b) Compressibility and expansivity of simple liquid silicates are
negligible from 0 to 10 % M20. At this composition, they undergo a
sudden inflection and rise sharply with increasing M 20 content
(Fig. 6.10.2).
(c) The partial molar volume of Si0 2 in all compositions of M 20-
Si0 2 is independent of composition.
Discuss in depth whether these facts are more consistent with the
70 CHAPTER 6: IONIC LIQUIDS

Pure silica =134 keal

60

I
!!
0
e
0
40
.-
u

II"
LIJ

20
NazO-SiO z

o
Malar % NazO
Fig. 6.10.1. Variation of the energy of activation for viscous flow of
a Na 20 + Si0 2 melt as a function of mole percentage of Na 20 .

classical three-dimensional theory of glasses, due to Zachariasen (1932),

or with a model in which, at > 10 % M 20, discrete silicate anions
become the predominant structural entity of silicates and glasses.

~ Figure 6.11.1 shows a plot of the heat of activation at constant

~ pressure for viscous flow, E of a number of ionic liquids
Tl ,

plotted against melting point T", . Rationalize the relation

(6.11.1)

according to some model of fused salts.

 PROBLEMS 71

60

50
•0
X 40
u
0
Q)

'0

.......E
E 30

-
u
'n.
20
~~
>I~ /0
coCO

---- 0
10

o 40 50 60
mole percent MfJ

Fig. 6.10.2. The sharp change in the expansivity of M 20-SiO, melts

around 10 mole % M 20 composition ; ( .6 ) K,O + SiO,; ( 0 ) Na,O + SiO, ;
(0) Li 20 + SiO, ; ( \I) SiO•.

G2l A microdensitometer trace of the Raman spectrum of molten

~ cryolite at 1030°C is shown in Fig. 6.1 2.1 (a), and in Fig. 6.1 2.1 (b),
it is analyzed into its thermal radiation, instrumental background, and
Raman scattering components. Only two Raman frequencies, one
centered at 633 and the other at 575 Llcm- 1 , are observed, both being
polarized. The band at 575 Llcm- 1 is very weak.
The spectrum of solid cryolite at room temperature, which is
independent of crystal orientation, is shown in Fig. 6.12.2. X-ray
diffraction has shown that crystalline cryolite contains finite AlF:-
 72 CHAPTER 6: IONIC LIQUIDS

CoCI 2 3' 74 RTm

10
o

.
o
E 6
.....
'0
u

"F"'
w 4

o
o 200 400 800 1000 1200

Fig. 6.11.1. The dependence of the experimental energy of activation for viscous
flow on the melting point. (Data from Nanis and Bockris, 1963.)

octahedra. However, there is probably some distortion of the ideal

cubic structure at room temperature.
From symmetry considerations and comparison with the Raman
spectra of tetrahedral and octahedral fluorides, show that it is reasonable
to assign the bands at 633 and 575 Llcm- 1 to AIF4 - and AlF~-, respec-
tively. On the basis that the principal dissociation in the melt is

(6.12.1)

roughly estimate the dissociation constant.

 PROBLEMS 73

(a)

,-
- --- -- - - - - -- - -- - ~- - -~­
(b)

:::::=-~~-:~~~:::::.::!..::::-.::==~. ,.\
I I ! !
700 600 500 400 300 200 100
Fig. 6.12.1. (a) Spectrum of molten cryolite at 1030°C. (- - - -) Halation.
(b) Analysis of the spectrum into Raman scattering (- -), thermal radiation
(- - -), and instrumental background radiation (- - -). (Data from Solomon et al.,
1968.)

554 A em-'
YoI ,.
(AIF, )
1

600 500 400 300 200

AV (em -I )

Fig. 6.12.2. Raman spectrum of single-crystal cryolite at 30°C. (Data from

Solomon et at., 1968.)
74 CHAPTER 6: IONIC LIQUIDS

ANSWERS

Assuming spherical holes, the mean hole volume is given by

(6.1.4)

where, from (6.1.1), <r 3 ) is given by

(6.1.5)

and making the substitutions t = ar2 and K = 16a7 / 2/157r 1 / 2 in (6.1.5),

r
we have

<r 3 ) = (KJ2a 5 ) t 4 e- t dt (6.1.6)

o

Since the integral in (6.1.6) is a gamma function of value 4!, (6.1.6)

becomes
(6.1.7)

Substituting for Kand a [Eq. (6.1.2)] in (6.1.7) yields

<r 3 ) = (8J57r 2)(kTJy)3/2 (6.1.8)

and substituting (6.1.8) into (6.1.4) yields (6.1.3),

Vh = (32/157r)(kTJy)3/2
= 0.68(kTjy)3/2

For NaCl, the mean hole volume is, by substitution of numerical data
into (6.1.3),

Vh(NaCll = 0.68(1.38 X 10-16 X 1093/115)3/2

= 32.3 X 10-24 cm3
= 32.3 A3
 ANSWERS 7S

Table 6.1.2 lists the mean hole volume of the salts together with the
volumes of the individual ions. It is apparent that the mean hole
volume of a molten salt is much larger than the volume of the cation
and in general slightly larger than the volume of the anion. In the hole
theory of fuged salts, migration or diffusion of ions takes place by an
ion hopping into a hole. Consequently, it is necessary for the holes to
be large enough to accommodate an anion or a cation.

TABLE 6.1.2. Mean Hole Volumes Vh Compared to the Volumes

of Anions and Cations in Certain Molten Salts

Salt Vh, As V+, As v_, As

NaCI 32.3 3.59 24.8

NaBr 39.7 3.59 31.1
NaI 39.3 3.59 42.2
KCl 40.7 9.85 24.8
KBr 43.5 9.85 31.1
KI 52.1 9.85 42.2
CaCl 2 34.6 3.82 26.8

The Stokes-Einstein equation relates the mobility Ui of an ion i

to the viscosity of the medium 7J by
(6.3.2)
'i
where is the ionic radius. The transport number of the ith species is
defined by
(6.3.3)
where the current Ii due to a particular species is given by
Ii = ZiFCi Ui X (6.3.4)
where Ci is the concentration and X is the electric field. For a 1: 1
electrolyte, the transport number of the anion L is, by substitution of
(6.3.4) into (6.3.3),
L = u_/(u+ + u_) (6.3.5)
and substituting (6.3.2) into (6.3.5) yields

L = '+/('+ + ,_) (6.3.6)

76 CHAPTER 6: IONIC LIQUIDS

TABLE 6.3.2. Experimental and Calculated Anionic

Transport Numbers for Some Fused Salts

Salt c(calc) L(exp)

LiCl 0.25 0.25

NaCl 0.34 0.38
KCl 0.42 0.38
RbCl 0.45 0.42

For fused lithium chloride, the transport number of Cl- is

L = 0.60/(0.60 + 1.81) = 0.25

Table 6.3.2 compares experimental anionic transport numbers with
those calculated on the basis of the Stokes-Einstein relation. The
agreement is seen to be good, particularly when the experimental
difficulties of measuring transport numbers of fused salts is recalled and
when it is further recalled that the Stokes relation, from which (6.3.2)
is derived, is valid only when the moving particle is large compared to
the sizes of particles that compose the medium. This is obviously not
the case for ionic migration in a fused salt. However, this agreement
shows that this approach can be used to obtain an approximate value
of the transport number in unknown systems.
For a 2:1 electrolyte undergoing the dissociation (6.3.1), we have
from Eqs. (6.3.3) and (6.3.4)

(6.3.7)

where the subscripts -, +, and 2+ refers to the species X-, MX+, and
M2+, respectively. Substituting (6.3.2) into (6.3.7) yields, with re-
arrangement,

(6.3.8)

Lead chloride must be at least partly dissociated to be conducting and

we may consider three distinct cases.
Case 1: C+ ~ C2+ and therefore C+ ~ C_ and (6.3.8) becomes

(6.3.9)
 ANSWERS 77

and, substituting numerical values, we get

c = 0.73

Case 2: C+ R:j C2+ and therefore C_ R:j 3C+ and (6.3.8) becomes

(6.3.10)

and, substituting numerical values, we get

t R:j 0.32

Case 3: C2+}> C+ and therefore C_ R:j 2C2+ and (6.3.8) becomes

(6.3.4)

and substituting numerical values, we get

c = 0.25

Since the experimental value of tCI- is 0.8, it may be concluded that

case 1 best represents the available data and that PbCl+ is the pre-
dominant cationic species in the melt.

f5l Provided the term (k+ + k_CBr _)1/2 in (6.5.3) has a suitable
L...:J value, the error function in (6.5.2) can be approximated, when 7
is large, as
erf(Z) = 1 (6.5.4)

and when 7 is small, as

erf(Z) = 2Z/7T 1 / 2 (6.5.5)

and Eq. (6.5.2) becomes under these limiting conditions

7 large (6.5.6)

7 small (6.5.7)

Consequently, a plot of 7 1 / 2 against i-I, for a wide range of current

densities, can be expected to exhibit two linear portions corresponding
78 CHAPTER 6: IONIC LIQUIDS

to the limiting conditions (6.5.6) and (6.5.7). The slopes of these two
linear regions are

(6.5.8)

and

(6.5.9)

and the ratio of the slopes is given by

From the slopes of the linear portions of Fig. 6.5.1,

(1 + KCBr -) = 17.7(3.8
and
K = [(17.7(3.8) - 1](0.0921
= 39.7 kg mole-1

From the intercept of the plot of 7 1 / 2 (7 large) against ;-1, we have

Therefore,

and substituting for K and CBr - yields

k+ + 0.0921k_ = 16.1 sec-1

Since K, the formation constant, is the reciprocal of the dissociation
constant,
K = k_(k+ = 39.7 kg mole-1
Consequently,
k+ = 3.5 sec- 1
L = 137 kg mole-1 sec-1

Since the value of K is within the range of experimental values for step
(b) of (6.5.1), this step is the rate-determining step.
 ANSWERS 79

r:;-, The diffusion coefficient is proportional to the product of the

~ number of holes times the probability that an ion will jump into
a hole, exp( -LJG;*jRT). From Eq. (6.7.1) and the probability of
jumping, we have

D = Do' exp( -LJGhOjRT) exp( -LJG;*jRT) (6.7.4)

= Do' exp[( -LJHho + T LJSho - LJH;* + T LJS;*)jRT] (6.7.5)

where the subscripts hand j refer to hole formation and jumping,

respectively. Since there is no PV work performed in the jumping of an
ion into an already created hole,

(6.7.6)

Substituting (6.7.6) into (6.7.5) yields

D -- D0 ' exp ( LJSho +

R LJS * ) exp
j (- + LJE;*)
LJHhoRT (677)
. .

and comparing (6.7.7) with the empirical relation

(6.7.8)

where Do =1= f(T) and Ep is the experimental energy of activation at

constant pressure, reveals that

(6.7.2)

Since the mean hole volume Vh is approximately temperature-

independent, we can write

(6.7.9)

and substituting Eq. (6.7.1), we have

but (6.7.10) can only be true at constant volume (Le., VT 2 - VT 1 = 0)

if, at constant volume, LJHh ° = o. Consequently, we find (6.7.3),
Ev =:= LJE;*
Calculation of Ep: From Eq. (6.7.8), it is seen that Ep can be
obtained from the slope of the plot of log D against T-l at constant
80 CHAPTER 6: IONIC LIQUIDS

4.0
";' 3.5
u

0;;::0
Q)
If)
3.0 0

E
u 2.5 o ~
0 ~oo~o
..,
O~,p=latm.
o
0 2.0
~p=400atm.
P=800atm.

1.5 1.4 1.5 1.6 1.7

1d r 1(OK- 1)
Fig. 6.7.2. Plots of D against T- 1 at various constant pressures
for Na 22 diffusion in fused NaN0 3 •

pressure. Figure 6.7.2 shows such plots for I, 400, and 800 atm. From
the slope of the plot,
-2.303[dlogDjd(T-I)] =2.17 x 103 deg
and
Ep = 1.987 x 2170 cal mole-I
= 4.31 kcal mole-I
Calculation of ElJ: As shown in Fig. 6.7.1, the pressure, at
different temperatures, at which the melt has a constant volume VI
is found by drawing a horizontal line through the molar volume-
pressure plots. Plotting the data of Table 6.7.1 (Fig. 6.7.3), the diffusion
coefficient corresponding to VI at different temperatures can be inter-
polated. Ev is then found from the slope of the plot of 10g[D<v)] 1
against (T-I) as shown in Fig. 6.7.4:

-2.303{d log[D(v1)]jd(T-I)} = 393 deg

Ev = 1.987 x 393 cal mole-I
= 0.78 kcal mole- I

From Eqs. (6.7.2) and (6.7.3), we have

J H" = 4.31 - 0.78 kcal mole- 1
0

= 3.53 kcal mole- I

 ANSWERS 81

3.5r-------------------------------,
I~ 3.0r-----_________
2.5~ ~o
---0____0_____ ------- 420.C
N

5 ----------0
=-----------0___
-----=:::~ ________
_________
ono

Q
2.0
---0 - - . . .:
______0
390· C
370· C
350.C
1. 5'-----4-:'-0-::-0---8-=-0~0:-----1-l20":-0---.J
P (atm)
Fig. 6.7.3. Plots of D against pressure at various temperatures for
Na 22 diffusion in fused NaN0 3 •

-;- 3.0
u
Q)
Cf)

E 2.5r-
u t- -0-

0
on 2.0r-
0

1.5
1.4 1.5 1.6 1.7

Fig. 6.7.4. Plot of D (at constant volume) against T-l for Na 22

diffusion in fused NaN0 3 •

and
Ej * = 0.78 kcal mole- 1

The expectation of the density fluctuation hole model that the

standard enthalpy of hole formation should be large compared to the
activation energy of jumping is thus borne out by experiment. Since
the microjump model assumes that no holes comparable in size to the
82 CHAPTER 6: IONIC LIQUIDS

particle are present, the expectation from this model would be that
E;* ~ Ep.

[!] The activation volume for any rate process is given by

JV* = [o(JG*)/OP]T (6.9.3)

where JG* is the free energy of activation for that process. Since
diffusion is a rate process, the diffusion coefficient D can be written in
the form
D = Do exp( -JG*/RT) (6.9.4)

where Do is independent of pressure and temperature. Substituting

(6.9.4) into (6.9.3) yields
J V* = -RT[o(ln D)/OP]T (6.9.1)
On the basis of the density fluctuation hole model, we may consider
that
(6.9.5)
where JG" ° is the standard free energy of hole formation and JG;*
the free energy of activation for jumping. Differentiating (6.9.5) with
respect to pressure at constant temperature yields
(6.9.6)
Now, LI V" = V", the most probable molar hole volume, since the
volume of a hole before it is formed is zero. Consequently, (6.9.6)
becomes (6.9.2):
J V* = V" + J V;*
Equation (6.9.1) shows that J V* can be obtained from the slope
of the plot of log D against P at constant temperature (Fig. 6.7.3 for
the diffusion of Na22 in molten NaN03). From the slope of the plots
at 390°C, we have
2.303[(d log D)/dPlr = -1.95 x 10-4 atm-1
and from (6.9.1),
J V* = 82.06 x 663 x 1.95 x 10-4 cm3 mole-1
= 10.6 cm3 mole-1
 ANSWERS 83

For L1 V* to be expressed in cms, the gas constant R must be expressed

in cms atm deg- 1 mole-l.
Similarly for the diffusion of 134CS in NaNOs at 390°, we have, by
plotting the data of Table 6.9.1 (Fig. 6.9.1),

L1 V* = 82.06 x 663 x 2.74 x 10-4 cmS mole- 1

= 14.9 cms mole-1

From Eq. (6.1.3), the molar mean hole volume Vh is given by

Vh = 0.68NA (kT/y)3/2 (6.9.7)
where NA is Avogadro's number. For fused sodium nitrate at 390°C,
we have

Vh = 0.68 x 6.023 x 102S(1.38 x 10-16 x 663/110)3/2

= 9.8 cm3 mole-1
Substituting the values of L1 V* for Na+ and Cs+ in fused NaNOs and
the value of Vh into (6.9.2) permits the values of L1 V;* to be calculated
(Table 6.9.2).
For Na+ diffusion, the fact that the magnitude of the volume
change associated with the diffusion process L1 V* is approximately
equal to the most probable hole volume is a clear and direct demon-
stration of the importance of the part played by holes in diffusive

1. 5'------4-0
......0----8-':-0-0---1--'2-00----'
P (atm)
Fig. 6.9.1. Plot of D against pressure at constant temperature for
134CS diffusion in fused NaNOs.
84 CHAPTER 6: IONIC LIQUIDS

TABLE 6.9.2. Activation Volumes

for Diffusion Processes in Fused NaNO a

Na+ Cs+

LI V*, em 3 moie- 1 10.6 14.9

Vh , em 3 moie- 1 9.8 9.8
LI V;*, em moie-
3 1 0.9 5.1

transport. The fact that L1 Vj * is close to zero for Na+ and smaller
alkali cations is in direct contradiction to a microjump model. For
the diffusion of Cs+, L1 V/ is not close to zero. This is to be expected
on the basis of the density fluctuation hole model for the diffusion
of an ion with a volume larger than the mean hole volume. The
crystallographic radius of Cs+ is 1.69 A.
 CHAPTER 7

THE ELECTRICAL DOUBLE

LAYER

r:jl Calculate the electric field, in esu per square centimeter and
~ also in volts per centimeter, in the double layer at a metal
electrode-aqueous solution interface when the excess surface charge is
10 and 40 p,C cm- 2 (microcoulombs per square centimeter). Assume
there is no specific adsorption but that the solution is concentrated.

G'l The viscosity of water is 0.01 P at 25°C. Calculate the velocity

L..:J of a colloidal particle in a field of 1 V cm-1 if its zeta
potential is 13 mV. Ew = 78.3.

r;-, Calculate the charge density on the metal as a function of

L2.J potential and then the capacitance as a function of potential
from the electrocapillary data of Table 7.3.1 (see page 86).

f4l Table 7.4.1 (see page 86) lists the interfacial tension as a func-
L:.J tion of potential and solution concentration for the mercury-
lithium chloride solution interface. Table 7.4.2 (see page 87) lists the
charge on the mercury electrode under the same conditions, and molar
activity coefficients are listed for various concentrations of lithium
chloride in Table 7.4.3 (see page 87). All potentials are referred to that
of a reversible calomel electrode in the same solution, which is reversible
to chloride ions. Calculate the cationic and anionic Gibbs surface
excesses for each potential and solution concentration.
85
86 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

TABLE 7.3.1. Variation of Interfacial Tension

of a Mercury-l.0 M CsCI Interface with Potential Difference·

Potential difference, mV vs Interfacial tension,

reversible calomel electrode dynes cm-1
-0 345.0
-100 376.4
-200 397.1
-300 410.5
-400 418.7
-500 422.6
-600 422.9
-700 419.9
-800 414.0
-900 405.6
-1000 395.1
-1100 382.9
-1200 369.2
-1300 353.6

* Data from Kovac, 1964.

TABLE 7.4.1. Interfacial Tension at Various Potentials

and Concentrations of Lithium Chloride·

Interfacial tension, dynes cm-1

Potential, m V vs
R.C.E. 3.0M l.OM O.30M O.IOM 0.03M 0.01 M

-100 372.7 379.2 383.6 385.7 383.1 382.9

-200 391.2 397.6 401.1 402.5 398.7 397.9
-300 405.2 410.8 413.4 414.3 410.6 409.4
-400 414.9 419.3 421.1 421.8 419.0 417.6
-500 420.4 423.5 424.8 425.6 424.1 423.1
-600 421.8 423.9 425.2 426.1 426.3 426.0
-700 419.4 421.0 422.5 424.0 425.6 426.5
-800 413.6 415.2 417.3 419.5 422.4 424.7
-900 405.1 407.0 409.9 413.1 417.0 420.7
-1000 394.3 396.9 400.6 405.0 409.6 414.8
-HOO 381.8 385.3 389.8 395.4 400.4 406.9
-1200 367.0 372.2 377.7 384.3 389.7 397.4
-1300 352.7 357.8 364.1 371.9 377.7 386.3
-1400 335.1 341.6 348.8 357.9 364.3 373.8

* Data from Kovac, 1964.

 PROBLEMS 87

TABLE 7.4.2. Electrode Charge Density at Various Potentials

and Concentrations of Lithium Chloride*

Charge density, J-LC cm- 2

Potential, mY vs
R.C.E. 3.0M 1.0M 0.3M O.IM 0.03M 0.01 M

-100 20.8 21.3 20.5 19.6 17.6 17.1

-200 16.2 15.8 14.7 14.1 13.7 13.1
-300 11.9 10.8 9.9 9.5 10.1 9.8
-400 7.6 6.3 5.6 5.5 6.7 6.8
-500 3.4 2.2 2.0 2.1 3.6 4.2
-600 -0.5 -1.3 -1.2 -0.8 0.7 1.7
-700 -4.1 -4.5 -4.0 -3.3 -1.9 -0.7
-800 -7.2 -7.1 -6.4 -5.4 -4.4 -2.9
-900 -9.7 -9.2 -8.4 -7.3 -6.5 -5.0
-1000 -11.6 -11.0 -10.1 -8.9 -8.4 -6.9
-1100 -13.0 -12.4 -11.5 -10.4 -9.9 -8.7
-1200 -14.3 -13.4 -12.8 -11.8 -11.3 -10.3
-1300 -15.9 -15.3 -14.3 -13.2 -12.7 -11.8
-1400 -18.7 -17.4 -16.4 -14.8 -14.1 -13.2

* Data from Kovac, 1964.

TABLE 7.4.3. Mean Ion Activity
Coefficients in Molar Units
for Lithium Chloride Solutions

CLiCI ,

M y±

0.010 0.89
0.030 0.84
0.10 0.78
0.30 0.72
1.0 0.72
3.0 0.96

r;-, The data listed in Table 7.5.1 were obtained from electro-
~ capillary measurements on the mercury-O.l M sodium thio-
cyanate solution interface. Calculate the amount of contact-absorbed
CNS-, expressed in units of charge, at each potential. Construct a plot
of contact adsorbed thiocyanate against charge on the electrode.
Comment on the result.
88 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

TABLE 7.5.1. Cationic and Anionic Surface Excesses

for 0.1 M Sodium Thiocyanate Solutions*

Potential,
mY Fr+,!-,C cm- 2 Fr_,!-,C cm- 2

-300 8.2 -28.5

--400 8.7 -23.3
-500 8.5 -17.9
-600 8.2 -13.0
·-700 8.0 -8.7
-800 8.1 -5.3
-900 8.5 -2.7
-1000 9.3 -0.8
--1100 10.4 0.4
-1200 11.9 1.0

* Data from Kovac, 1964.

r(;l Express the Volta potential difference in terms of the total

~ charge on the metal electrode.
Consider water dipoles attached to an electrode surface. Suppose
that they (a) fully occupy the surface; (b) can have only two positions:
these are symmetric and can be designated to as t and t; and (c) are in
equilibrium with water molecules in solution.
Show that the potential difference at the interface due to the water
molecules is given by
LlX = 47TfLN t tanh [fLX -
E kT
UC (()I
kT"
- ()t)] (761)
..

where Nt is the total number of water molecules in the electrode

surface, E is the dielectric constant of the oriented water layer, fL is the
dipole moment, X is the electric field strength due to the electrode,
U is the interaction between two neighboring dipoles, c is the coordina-
tion number, and () t and () t are the fractions of the two types of water
in this model.
Plot LlX against the electrode charge qM over the range 0 to
+20 fLC cm- 2• Pay particular attention to Nt; the value to be used
should be consistent with a fairly small LlX (e.g., ±100 mV) at qM =
±20 fLC cm- 2 . Note that since U = f(N t ), there is feedback between
the U values chosen and Nt. What average distance between water
molecules on the surface is consistent with LI X < 100 m V ?
 PROBLEMS 89

f7l Consider the interaction energy between anions in contact

~ with an electrode interface, taking into account the image
energy of the ions with the electrode. Show that the total lateral
interaction energy is given by

(7.7.1)

where ri is the ionic radius, nCA is the number of contact-adsorbed ions

per cm 2 on the surface, and E is the dielectric constant in the first layer.

r.::l Prove that if contact-adsorbed ions interact coulombically

L.!J with the electrode of charge density qM, then, at fractional
e,
coverage the isotherm is

(7.8.1)

where r i , Zi , and a i are the radius, charge, and activity in the bulk
of the solution of the contact-adsorbed ion, LlC c o is the chemical
component of the standard free energy change associated with the
contact adsorption of the ions, and Xl and X 2 are the distances from the
metal surface to the inner Helmholtz plane (IHP) and the outer
Helmholtz plane (OHP), respectively.

I9l Show that the isotherm based on Coulombic repulsion, (7.8.1),

~ gives a relation between the contact adsorbed charge qCA and
the charge on the metal qM which passes through a point of inflection.
e,
Further, this point of inflection is given in terms of the fraction of
the inner Helmholtz plane (IHP) occupied by contact adsorbed ions, by

e= (4/3B)2/3 (7.9.1)
where
B = 777/2e 0 2r,2
z ( q max N A )3/2
(7.9.2)
4EkT F
90 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

and ri is the ionic radius, E is the dielectric constant in the IHP, N A is

Avogadro's number, and qmax is the charge density associated with
complete coverage by contact-adsorbed ions.
Table 7.9.1 lists the contact-adsorbed charge, for a number of
anions, corresponding to the electrode potential at which the hump
is observed on the capacitance-potential curve. These values of the
contact adsorbed charge qCA-hump were experimentally determined on
mercury. Calculate the values of qCA at the inflection of the qCA versus
qM curve and compare them to qCA-hump . Discuss the results.

TABLE 7.9.1. Contact-Adsorbed Charge at the Capacitance

Hump qCA-hump for Various Anions of Radius 'i
Concentration, ri, qCA-hump,
Anion M A ,..Ccm- 2

1- 0.25 2.00 -18.3

CI- 3.0 1.81 -11.1
CN- 0.3 1.95 -6.7
BrO.-- 0.3 1.6 -13.2
ClO.- 0.3 2.5 -6.7

Q A metal electrode M is placed in a solution in experiment A,

~ and in experiment B, the metal is placed immediately outside
the solution. In each case, the electrode M is connected to another
electrode in solution that is reversible to metal ions in the solution.
The potential between the electrodes is adjusted for zero current and
the potential measured as 0.389 and 0.452 V for experiments A and B,
respectively.
Deduce equations which permit the calculation of the metal-
solution Volta potential difference. What is its numerical value in
this case?

rtil A platinum wire passes through a platinum ring. The wire is

~ strung tautly between two fixtures and the whole is immersed
in solution. The potential of the wire and ring can be varied by conven-
tional electrochemical circuits and measured against a reference
electrode in the solution. By gradually tilting the wire from the hori-
zontal, the coefficient of sliding friction can be obtained.
 PROBLEMS 91

Show that the coefficient of friction will vary with potential in

a parabolic way and that it passes through a maximum at the potential
of zero charge. Plot, numerically, the coefficient of friction against
electrode charge over the range qM = - IO to + IO fLC cm- 2 •

Consider a model of organic adsorption on electrodes in which

the following assumptions are made:
(a) There is equilibrium between the adsorbed organic material
and the electrode.
(b) The organic does not dissociate as it transfers from the electrode
to the solution and vice versa.
(c) Only electrostatic interactions between the electrode and
(i) water molecules, (ii) organic dipoles, are significant.
On this basis and with the modelistic assumption that water
exists either in the t or t position, show that (for an organic without
electrical interaction with the surface)

Borg
-- -- = Corg A exp [ - -nY (47T
-- fLqM
- - YUc )] (7.l2.l)
I - Borg kT E

where A is a constant for a given system and temperature, n is the

number of water molecules displaced by one organic molecule, fL is the
dipole moment of water, U is the lateral energy of interaction between
water molecules, c is the coordination number of water molecules on
the surface, and Y is given by
Y = Bt - Bt
Calculate and plot Borg as a function of qM for A = 10.
Examine the effect on this equation of taking into account the
electrical interaction of the organic molecule with the electrode.
By considering the change in the double-layer energy that results
from the interaction of large organic molecules with the electrode,
it can be shown that Borg should be decreased as the potential is changed
in either direction from the p.z.c. (Frumkin, 1926; Frumkin and
Damaskin, 1964). Make numerical calculations to compare the relative
importance of the effect with that calculated on the basis of the model
presented in the first part of this problem.
92 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

ANSWERS

Q In concentrated solutions, the thickness of the diffuse double

L!J layer become negligible and, in the absence of specific adsorp-
tion, the double layer approximates a parallel plate capacitor with
charge equal to the surface charge on the electrode. For such a capacitor,
the electric field X is related to the charge q by

X = 47Tq/E (7.1.1)

where E is the dielectric constant of the material between the electrode

and the outer Helmholtz plane (OHP). E is certainly not that of bulk
water. For the totally restricted water molecules adsorbed on the
surface, E """ 6, but its value increases rapidly with distance from the
electrode to the value of bulk water. Taking a value of, say, 20 for E and
substituting numerical values into (7.1.1) yields

X = 47T x 10-5 /20 coulomb cm- 2

= 47T X 10- 5 X 3 X 10 9/20 esu cm-2
= 1.9 X 104 esu cm- 2
= 1.9 X 104 X 300 V cm-1
= 5.6 X 106 V cm- 1
when
q = 40 fLC cm- 2
X = 7.5 X 104 esu cm- 2
= 2.3 X 10 7 V cm-1

The charge density qM on the mercury electrode is given by

the Lippmann equation:

(oy/oV)1' = -qM (7.3.1)

where y is the interfacial tension, V the potential, and fL the chemical

potential of species in the solution. For y in dynes per centimeter and V
in volts, the units of qM are dynes per centimeter per volt. The conver-
 ANSWERS 93

sion to coulombs per square centimeter can be derived in a number of

ways:
dyne/cm V = dyne A/cm VA
= dyne A sec/cm J
= 10- 7 A sec cm- 2
= 10- 7 coulomb cm- 2

Figure 7.3.1 shows a plot of the data of Table 7.3.1 and the evalua-
tion of qM from the slope of the plot. The charge density on the metal
is plotted against potential in Fig. 7.3.2. The differential capacity C
is given by
(7.3.2)

and is the tangent to the curve of qM against V (Fig. 7.3.2). At

-600 mV RCE, the differential capacitance is seen to be 32.5 fLF cm- 2 •
Capacitances at other potentials are similarly calculated.

Irl A good deal of evidence exists to show that cations, with the
~ exception of large cations such as Cs+, N(CH 3)4 + etc., are not
specifically adsorbed. Consequently, the entire cationic excess charge
can be considered to be cationic diffuse layer charge qd +. From Gouy-
Chapman theory, the cationic charge in the diffuse layer is related to
the potential at the OHP, 1(;0' by

(7.5.1)

Hence, by equating qd I to z+Fr+, where r+ is the experimentally

determined Gibbs surface excess of the cation, we can calculate 1(;0
from (7.5.1). Knowing 1(;0' the anionic charge in the diffuse layer qa-
can be calculated from

qd - = -(wRT/27r)1/2[exp( -zJ!fo/2RT) - I] (7.5.2)

For the charge to have units of coulombs per square centimeter,

the term on the right of(7.5.1) and (7.5.2) must have these units. For n,
the bulk concentration of the ion, in moles per cubic centimeter andR
94 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

420

410

400

390

360

'~ 370
.."
:>.
~ 360
>-..
350
( oor)
V fL
95
= -:e5
d
yne cm
-1 V - I
340
QM = - 11.3 fLcou l omb cm - 2
330

-1.6 -1.2 - 0 .8 -0.4 o

ELECTRODE POTENTAl (V, RCE)

Fig. 7.3.1. Evaluation of the electrode charge from the plot of

surface tension against potential for the mercury...{}.1 M CsCI
interface. (Jy,'J V)I' = 95 /0.85dynes cm- I V- I; qM = -11.3 ftC cm- 2.

in ergs per mole per degree, the units can be converted in the following
way:
(erg/cm 3 )1 /2 = (I 0- 7)1 /2(V A sec/cm 3)1 /2
= (I 0- 7/300)1 /2(statvolt coulomb/cm3)1 /2
= (l0-7j300)1 /2(esu coulombjcm4)1 /2
= (10- 7/3 x 109 X 300)1 /2(coulombjcm 4)1/2
= 0.333 x 10-9 coulomb cm- 2

Evaluating the term (wRTf27T)1 /2 for a 0 .1M solution of a 1:1

electrolyte, we have

(E;~T (2 = 0.333 X 10- 9 [78.5 X 10- 4 x ~!14 X 107 X 298 (2

= 1.85 X 10- 6 coulomb cm- 2
 ANSWERS 95

+24

+20

+16

~+12
N

'E
<.> +8
..0
-6
E C _ 26 x 10 coulomb
~ +4 -~vcmr
o<.> = 32.5 ,.Fcm- 2
3- 0
:.
CT -4

-8

-12

-16
-1.6 -1.2 -0.8 -0.4 0
ELECTRODE POTENTIAL (V,RCEI
Fig. 7.3.2. Electrode charge plotted against electrode potential
for the mercury-O.I M esCi interface. C = (26 x 10- 6 /0.80)
coulomb/V-em' 32.5 fL F em- 2 .

The potential in the OHP, tPo, can now be evaluated from (7.5.1)
for the case when the cationic excess is 8.2 J1-C cm- 2 (see the first line
of Table 7.5.1):

tPo = - (2R T/z+F) In[(8.2/ 1.85) + I]

- (2 X 8.314 x 298/96,500) In(5.43)
= - 0.086 V

Substituting for tPoin (7.5.2) yields

qd- = - 1.85 X 1O-6[exp(0.086 X 96,500/2 X 8.314 X 298) - 1]
= + 1.50 J1-C cm- 2
96 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

The positive sign indicates a deficiency of anions in the diffuse

double layer. Since the contact-adsorbed charge qCA is related to the
total anionic excess charge by

(7.5.3)
then
qCA = -28.5 - 1.5
= -30.0 f1-C cm-2

The value of the electrode charge qM corresponding to this value of the

contact-adsorbed charge is computed from

(7.5.4)

and from Table 7.5.1,

-qM = -28.5 + 8.2

= -20.3 f1-C cm-2

Using the data of Table 7.5.1, values of qCA are computed for the
entire range of values of qM. The results are plotted in Fig. 7.5.1.
Careful examination of Fig. 7.5.1 shows that the plot passes through
a point of inflection. The point of inflection is a characteristic feature
of such plots and consequently it is essential that any isotherm derived
for contact adsorption on electrodes predict the existence of the
inflection point. In Problems 7-9 of this chapter, an isotherm based
on Coulombic repulsion, taking into account image charges, will be
developed and tested. Further, it will be shown that the point of
inflection corresponds to the "capacitance hump" exhibited by
capacitance-electrode potential plots.

I7l Let us divide a unit area in the IHP into as many cells as there
L:.J are contact-adsorbed anions, and make the assumption that
the cell centers, which correspond to the time average positions of the
ions, are arranged according to a hexagonal pattern. Choosing one
ion as the reference ion, there will be a ring containing six ions around
the reference ion (Fig. 7.7.1). The nth ring of anions contains 6n ions
and, if we consider only univalent anions, it will have a total charge of
-6neo ·
 ANSWERS 97

/
301-

"IE
u
20 /'
/
u
0-
c 10 /
.
o / inflection

/
...........
.....,...
o 10 o 10 20
qM (fLcoulomb cm- 2 )

Fig. 7.5.1. Plot of contact-adsorbed charge against electrode charge

showing the point of inflection for the mercury-D.I M NaCNS interface.

As a mathematical convenience, we will neglect the discreteness

of the charges in the ring and consider it to be a continuum of charge
with a charge density per radian, I n , given by

(7.7.2)

The potential", at the reference ion due to the charge on the nth ring is

(7.7.3)

where r is the distance between the centers of the cells (Fig. 7.7.1).
Consequently, the interaction energy between the reference anion and
the charge of the nth ring, -",eo (which is repulsive and thus of positive
sign), is given by

(7.7.4)

Consider now the interaction energy between the reference anion

and the nth ring of image charge (Fig. 7.7.2). If the electrode-to-
98 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

meared - out charge

d ensi ty

Fig. 7.7.1. Contact-adsorbed ions arranged in a hexagonal pattern.

Here, nr in the distance between the reference ion and the nth ring
of charge (shown as a continuum).

reference-ion distance is ri , then the distance between the reference ion

and the ring of image charge is [(nr)2 + (2r;)2)1 /2. Consequently, this
interaction energy, - eorpim age (which is attractive and hence of negative
sign), is given by

(7.7.5)

Hence the interaction energy between the reference anion and the
nth ring of charge and its image is given by the sum of (7.7.4) and
(7.7.5), namely

(7.7.6)

The total lateral interaction energy Wu is therefore the sum of the

interactions due to all the rings and is given by

Wu = L (27T I i n I eo/Enr){J - [J + (2r;/nr)2]-1 /2} (7.7.7)

n= l
 ANSWERS 99

Ring of Charged ring

image charge on IHP

Location of
electrode surface

Fig. 7.7.2. The image of the contact-adsorbed charge is induced behind

the electrode surface. The sign of the image charge is opposite to that of
the contact-adsorbed charge but equal in absolute value.

Since (2ri/nr)2 is less than unity, [1 + (2rdnr)2]-1/2 may be replaced

by its binomial expansion:
(7.7.8)
where higher-order terms are neglected. Substituting (7.7.8) into (7.7.7)
yields

Wu = L (47T I In I eOri2/a3n3)[1 - (3ri2/n2r2)] (7.7.9)

n~l

Since r is the distance between cell centers (i.e., the ion's diameter),
each cell has an area 7T(r/2)2 and hence there are 4/7Tr 2 ions per unit
area. However, nCA is the number of contact-adsorbed ions per unit
area, and consequently,
(7.7.10)
100 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

Substituting (7.7.2) and (7.7.10) into (7.7.9) yields

w =
'" 3ri 2eo 2 1T3/2 nCA
'\' \ 1 _ 31Tri2nCA _ I!
3/2 .
(7.7.11)
Ll L..
n~l
2m2 l 4 n2

Inspection of (7.7.11) shows that it contains two summations:

(7.7.12)

and
00

L n- 4 = 1T4 /90 (7.7.13)

n~l

Substituting (7.7.12) and (7.7.13) into (7.7.11) yields the desired

expression (7.7.1):

~ The number of contact-adsorbed ions per square centimeter in

L!J the IHP, nCA, is related to the total density of sites for
contact adsorption nT and the fractional coverage () by
n CA = ()n T (7.9.3)
= ()nTeeNAIF (7.9.4)
= () 1 qrnax 1 NA/F (7.9.5)
Introducing (7.9.5) into (7.8.1), and since

(7.9.6)

Eq. (7.8.1) can be written

In[()/(l - ()] = const + In ai + AqM - B()3/2 (7.9.7)

where
const = In(2ri F/l000 1qrnax I) - (LJGco/RT) (7.9.8)
A = 41Te o(x 2 - X1)/EkT (7.9.9)
B = (1T7 /2e o2ri2/4EkT)(1 qrnax 1 N A/F)3/2 (7.9.10)
 ANSWERS 101

Equation (7.9.7) will pass through points of inflection where the second
derivative d 2()jdqM2 equals zero. The differentiation may be performed
as follows:

(7.9.11)

and inverting,
(7.9.12)
where
P = [lj()(l - ()] + IB()1/2 (7.9.13)
Now,

(7.9.14)

(7.9.15)

(7.9.16)

where
dP -I + 2() + !B()-1/2()2(l - ()2
d() ()2(1 _ ()2 (7.9.17)

"'" -I + 2() + !B()3/2 _ !()5/2

(7.9.18)
~ ()2(1 _ ()2

if the term in ()7/2 is neglected as small (see below). Substituting (7.9.18)

into (7.9.16) yields

d 2() _ _
dq M2 -
A2
2p3()2(l - ()2
(-2
+ 4() + ~B()3/2
2
_ 3B()5/2) (7919)
. .

The mathematical condition for the inflection point is satisfied when

(7.9.19) equals zero. That is, when

2 - 4() - !B()3/2 + 3B()5/2 = 0 (7.9.20)

Equation (7.9.20) can be factorized as follows:

(1 - 2()(1 - !B()3/2) = 0 (7.9.21)

102 CHAPTER 7: THE ELECTRICAL DOUBLE LAYER

and consequently has two roots ()l and ()2 given by

()l ~ (4/3B)2/ 3 (7.9.22)

()2 C~ 112 (7.9.23)

The first root ()1 is smaller than ()2' and for the first root the error
introduced by neglecting ()7/2 [see Eqs. (7.9.17) and (7.9.18)] is negligibly
small.
For contact-adsorbed anions, the charge density corresponding
to the inflection point, qCA-infl , is given by

(7.9.24)
(7.9.25)

Substituting (7.9.lO) into (7.9.25) and taking into account that

(7.9.26)
we have
(7.9.27)

The dielectric constant E in (7.9.27) refers to the IHP value. The

dielectric constant in the double layer increases from 6 in the saturated
water layer on the electrode to some intermediate value between 6 and
80 in the OHP. Since contact-adsorbed ions extend a little further
from the electrode than the water molecules, a value a little larger
than 6 is appropriate for the calculation of qCA-infl' The following
numerical calculations are based on E = lO.
Substituting numerical values in (7.9.27), we have

~ 4 (4 X lO2 x (1.38)2 X lO-32 X (298)2)1/3

qCA-lnfl =,
r:/3(10)32/3 97T 7 x 4.8 x 10-10

--6.93 X lO4M3 esu cm- 2

~23.1!rt3 fl-C cm- 2

when ri is expressed in angstroms.

Table 7.9.2 lists values of qCA-infl calculated from the ionic radii
in Table 7.9.1 using Eq. (7.9.27). Also listed for comparison in Table
7.9.2 are the values of qCA-hump, i.e., the experimentally determined
contact-adsorbed charge corresponding to the hump on the capacitance-
 ANSWERS 103

TABLE 7.9.2. Comparison of qCA-lnfi and qCA-hump

for Various Anions Contact-Adsorbed on Mercury

qCA-lnfi , qCA-hump,

Ion (!-,Ccm- 2 ) (!-,Ccm- 2 )

]- -S.9 -IS.3
CI- -10.5 -11.4
CN- -9.5 -6.7
Br0 3 - -12.5 -13.2
ClO.- -6.S -6.7

electrode potential curve. In all cases except iodide, there is good

agreement between qCA-infl and qCA-hump and this provides strong
support for the Coulombic repulsion model of contact adsorption and
the contention that the capacitance hump is due to contact adsorption.
The poor agreement in the case of iodide may, in part, be due to
distortion of the large, easily polarized iodide ion in the electric field
of the electrode.
 CHAPTER 8

BASIC ELECTRODE KINETICS

f1l An electrode reaction for which the exchange current density

~ is 5 x 10- 7 A cm- 2 is proceeding at an overpotential 'YJ =
-0.199 V under the control of a potentiostat. What is the current
density? Using a fast switch, the potentiostat is disconnected from
the electrode and a cathodic current density of 2.05 x 10- 4 A cm- 2
imposed on the electrode. What is the new steady-state overpotential
and approximately how long would it take to attain it? Take the
electrode area as 0.1 cm 2 , the double layer capacitance as 50 f-tF cm- 2 •

f2l An electrode-solution interface has a capacitance of 50 f-tF cm- 2 ,

L..:J and a low-amplitude (,-.....,5 mY) ac signal of 1000 Hz is applied
to it. If the exchange current density for the reaction is 3 rnA cm- 2 ,
calculate the Faradaic resistance. Also work out the Faradaic reactance
and impedence (in ohms cm- 2).

f3l What is Bronsted's law? Suppose that H2 is being evolved on a

L.:J cathode at an overpotential of -0.160 V for a rate of 2 x 10-7
mole cm- 2 sec- 1 of H2 at 25°C. The rate-determining step is

(8.3.1)

and the coverage by Had is assumed to be low. An organic substance

is now added to the solution and adsorbs, reducing the average heat
of adsorption of H by 12 kcal. Calculate the new rate of the reaction
if the overpotential is now -0.222 V at 25°C.
105
106 CHAPTER 8: BASIC ELECTRODE KINETICS

r:;---J Draw some current-potential lines for both positive and

~ negative overpotentials to illustrate graphically the mechanism
of Faradaic rectification. Suppose (3 = 0.495; calculate the percentage
rectification of a square-wave signal of 200 mV amplitude imposed on
the reversible potential (io = I X 10- 3 A cm- 2 ).

According to Levich (1970), the energy of activation E*

of a charge-transfer process at an electrode is given by
(8.5.1)
where E8 is the solvent reorganizational energy required to trigger a
quantum mechanical transition from the initial to the final state and Q
is the heat of reaction. By considering appropriate potential energy
curves, show, however, that the relation (8.5.1) does not depend on
the nature of the model for charge transfer provided that the energy
profiles of the initial and final states can be represented by those of
simple harmonic oscillators.

Using Eq. (8.5.1) and the relation

(8.6.1)

where YJ is the overpotential and Q the heat of the reaction, derive an

equation for (3. Why is such a relation incompatible with experiment?
What is the essential point about the consideration expressed in
Problem 5, which leads Eq. (8.5.1) into error?

ryl Do image forces affect electrode-kinetic calculations? Examine

~ this point by comparing the fluctuations in image energy
experienced by an ion in the OHP with typical values of energies of
activation.
Assuming that the only forces acting on an electron emitted from
a metal during a charge-transfer process in the gas phase are the electron
metal image interactions and the electron-ion attraction, calculate the
potential energy-distance relationship. Then, estimate, roughly, the
probability of electron penetration of the barrier in a cathodic reaction.
Assume the barrier to be a Gamow type and assume the availability
 PROBLEMS 107

of suitable acceptor states in the ion. Take the electron work function
of the metal as 5.0 eV, and the metal-ion distance as 8 A.

Iol In an early theoryt of homogeneous charge-transfer processes

L.!J developed by Marcus (1963), the free energy of activation for
charge transfer LlG* is given by

LlG* = LlG o * + LlGc * (8.8.1)

where LlG c * is the Coulombic contribution given by Coulomb's law as

(8.8.2)

for two species, I and 2, a distance r apart. Estat is the static dielectric
constant of the medium. The organizational contribution Ll Go * is
given by

(8.8.3)

where

,\ _ (ne)2 ·_1_ + _1___q(_l___1_) (8.8.4)

- 2 (2a l 2a 2 r. Eopt Estat

and Ll G is the standard free energy charge of the reaction. Here, n is

C

the number of electrons transferred in the reaction, a l and a 2 are the

effective radii of the two reactants, r = al + a 2 , and Eopt is the optical
dielectric cunstant of the medium. The effective radius a of a reactant
is given approximately by the sum of the radius of the central ion and
the diameter of the ligand.
For bimolecular homogeneous reactions, the rate constant k
can be written as
k = Z exp( -LlG*/RT) (8.8.5)
and the frequency factor Z has a value of approximately 1011 liters
mole- l sec-I.
Using Eq. (8.8.5) and the data of Table 8.8.1, calculate experi-
mental LlG* values for the processes given in the table. From Eqs.

t In later publications, this author took into account the thermal activation of
the bonds in inner shells in complex ions, as well as the energy changes concerned
with the electrostatic fluctuations of outer water layers.
108 CHAPTER 8: BASIC ELECTRODE KINETICS

(8.8.1 )-(8.8.4) and the data of Tables 8.8.1 and 8.8.2, calculate the
earlier theory values of LlG*. Test this theory by plotting the calculated
LlG* values against the experimental values and discuss the result.

TABLE 8.8.1. Rate Constants and Standard Free Energy Changes

of Some Homogeneous Charge-Transfer Reactions

Temp., lIG', k,
Reactants °C kcal mole- 1 liters mole- 1 sec-1

Ce3+ + CeH 25 0 1.1 x 10-4

C02+ + Co3+ 25 0 0.75
Co(NH 3)!+ + Co(NH3)~+ 64.5 0 <10-"
Cr 2+ + Cr3+ 24.5 0 2 x 10-5
Cr 2 + + Fe3+ 25 -27.21 --2 x 10-3
Cr2 + + Co(NH3)~+ 25 -11.76 9.0 x 10-5
Fe2 + + CeH 25 -15.45 7.3 x 10'
Fe 2+ + Co'" 0 -24.67 10
Fe2+ + Fe3+ 25 0 4.2
Fe2+ + Fe(phen)~+ 25 -7.15 3.7 x 104
Fe(CN):- + CeH 25 -17.52 1.9 x 10·
Fe(CN):- + Fe(CN)~- 4 0 3.5 X 102
Fe(CN)!- + IrCI!- 25 -5.76 3.8 x 105
Fe(CN):- + Mo(CN):- 25 -2.77 3 x 10'
Fe(dipy)~+ + CeH 25 -8.99 1.96 x 105
Fe(phen)~+ + Co'+ 25 -17.99 1.6 x 10'
Fe(phen)~+ + Fe(phen)~+ 0 0 > 105
MnO:- + MnO,- 0 0 7.0 X 10'
Mo(CN):- + Ceo, 25 -14.76 1.4 x 107
Mo(CN):- + irCI!- 18 -3.0 > 105
Mo(CN):- + Mo(CNn- 25 0 3 x 10'
V2+ + Co(NH 3J!+ 25 -11.76 9.0 x 10-5
W(CN):- + CeH 25 -20.75 > 10"
W(CN):- + W(CN):- 25 0 7 x 104

The so-called electrostatic model of electrode processes predicts

that

(8.9.1 )
 PROBLEMS 109

TABLE 8.8.2. Radii of Some Central Ions and Diameters of Some Ligands

Radius, Diameter,
Central ion A Ligand A
Ce H 0.92 CI~ 3.62
C02+ 0.72 CN- 3.84
Co3+ 0.62 H 2O 2.76
Cr 2+ 0.89 NH3 3.38
Crh 0.63 0-- 2.64
Fe2+ 0.74 0- 3.52
Fe 3+ 0.64 Phenanthroline ~5.77

IrH 0.68 Dipyridyl ~5.21

Mn6+ ~0.50

Mn'+ 0.46
MOH ~0.66

Mo5+ 0.70
WH 0.70
W5+ 0.66

where khom and k het are the rate constants for exchange reactions of
the homogeneous type,

(8.9.2)

and the heterogeneous type,

(8.9.3)

respectively; Zhom and Zhet are the respective frequency factors.

Experimental data obey Eq. (8.9.1), and this was at one time thought
to be good evidence in favor of the solvent activation model. Show
however, that (8.9.1) holds regardless of the type of activation for
models of e'lectrode processes.

The expression of time-dependent perturbation theory for

the probability of transition between two states i and f is

(8.10.1 )

where H is the Hamiltonian for the transition, State clearly the kind
110 CHAPTER 8: BASIC ELECTRODE KINETICS

of transition to which this equation is applicable and give two examples

outside electrode kinetics.
Calculate the wavelength of an electron which has an energy of
1 eV. Consider the magnitude of this wavelength and then discuss
whether it would be justified to apply (8.l0.l) to a calculation of the
preexponential factor of the rate expression for an interfacial redox
reaction.

r;:;-] A physical interpretation of the Levich (1970) theory of charge

L.!!J transfer can be made in the following way. Surrounding the ion,
water dipoles oscillate at a frequency Vo ~ 1011 Hz. These oscillations
induce energy into the solvated ion, and when this energy has reached
a value such that the energy of an electron level in the ion is that of the
Fermi energy of the substrate, a quantum mechanical transition of an
electron from the metal to the ion takes place.
Examine this model assuming the necessary energy of activation
of the solvated ion is about 1.0 eV. Calculate the energy of an individual
oscillator and determine the number of oscillators that would be
involved if the ion were to be activated in this way. Approximately
how far from the central ion does this number of oscillators extend?
Determine the probability that a given central ion is activated 1 eV
above its ground energy in a solution as a result of electrostatic
fluctuations to which it is subject from the surrounding solvent.
Compare this probability with that due to thermal excitation of the
rotational-vibrational levels of the ion.

Consider hydrogen evolution with a rate-determining step

(S.12.1)

Consider the potential energy-distance relations of the initial and final

states to be represented by two intersecting Morse curves. Deduce:
(a) An expression in terms of Morse constants for the symmetry
factor f3 and examine the value of the coefficient df3ld'T} over a range
of 2 V. Utilize the values: the heat of activation LJH* = 23 kcal mole-1 ;
the heat of dissociation of H30+ to H 20 and H+ (in the gas phase) is
ISO kcal mole-I; the heat of dissociation of Had is 45 kcal mole-1 ; the
 ANSWERS 111

enthalpy of reaction (8.12.1) is 55 kcal mole-I; the Morse constant a

for H30+ and Had may be taken as 3 x 108 cm-I; the separation
between the zero-point energy states on the reaction coordinate for
H30+ and Had may be taken as 3.3 x 10-8 cm.
(b) The condition for barrierless electron transfer.

ANSWERS

For a simple electrode process, the current density i is given

by the Butler-Volmer equation:

i = io{exp[(I ~ f3hF/RT] ~ exp[ ~f3"7F/RT]} (8.1.1)

where io is the exchange density and f3, the symmetry factor, is always
close to 0.5. In the form (8. I.I), i has positive or negative values for
anodic or cathodic processes, respectively. The overpotential "7 is
defined by

(8.1.2)

where V and VR are the electrode potential and the reversible potential
of the electrode process, respectively.
When I "7 I > so mV at room temperature, one of the exponential
terms in (8.1.1) becomes negligible compared to the other. Substituting
given values into (8.1.1) yields

i = ~S X 1O-7[exp(0.S X 0.199 X 96S00/8.314 X 298)]

= ~2.41 X 10- 5 A cm- 2

After the imposition of the cathodic current step of 2.05 X 10-4 A

cm- the steady-state overpotential is given by the solution of
2,

~2.0S X 10- 4 = ~S X 1O-7[exp(0.S X 96S00"7/8.314 X 298)]

'fj = ~0.309 V

Since the anodic contribution to the net current is negligible

under the present conditions, the charge transfer resistance ReT may be
112 CHAPTER 8: BASIC ELECTRODE KINETICS

obtained by differentiating the cathodic part of (S.l.l) with respect to "1,

thus
di/d7J = (io{3F/RT) exp( -(37JF/RT) (S.1.3)
= i{3F/RT (S.l.4)

Therefore, if A is the area of the electrode,

RCT = (d7J/di)/A = RT/i{3FA (S.1.5)

Substituting numerical values into (S.l. 5) yields

R _ S.314 X 29S
CT - 2.05 X 10-4 X 0.5 x 96500 x 0.1
= 2505 ohms

The situation at the electrode is approximately represented by the

equivalent circuit in Fig. 8.1.1, and the time constant T of the circuit to
respond to the imposition of a constant current is given by

(8.l.6)

where Rs is the total solution and internal circuit resistance. Substituting

numerical values for C DL and RCT yields

T = 5 X 10-6 x 2505
= 12.5 msec

Since 7" is the time required for the system to proceed through l/e
of the change, steady-state conditions essentially prevail after 47" have
elapsed; in this case, ",,50 msec.

COL
Fig. 8.1.1. Equivalent circuit for a charge-transfer process at an electrode.
 ANSWERS 113

r;-, Bronsted empirically proposed an equation to correlate the

~ specific rate constants k of a series of acid-base reactions with
the equilibrium constants K of the reactions. Thus, for reactions of
the type
AIH +H 0 2 ~ AI- + H30+
A2H +H0 2 ~ A 2- + H 20+ (8.3.2)

the Bronsted relation takes the form

(8.3.3)

GA is a constant for the structurally similar series of acids, AI' A2 , ... , Am,
and f3 is a positive fraction, which is constant for a given reaction series
and a given solvent.

Equation (8.3.3) can be written in the form

Z exp( -iJG*jRT) = GA exp( -f3 iJGOjRT) (8.3.4)

where Z is the frequency factor, iJG* the free energy of activation, and
iJGo the standard free energy of the reaction. Taking logarithms and
rearranging (8.3.4) gives

(8.3.5)

For the reactions involving the homologous series of acids AI, A2 ,... , Am,
GA and Z are constant and we have

(8.3.6)

Thus a change in the standard free energy of the reaction produces a

fractional change in the free energy of activation of the reaction.
Horiuti and Polanyi (1935) proposed that the same relation should
exist for hetrogeneous charge-transfer process at electrodes. Conse-
quently, considering reaction (8.3.1), a change in the heat of adsorption
of hydrogen iJ(iJH), will produce a corresponding change f3 iJ (iJ H) in
the free energy of activation of the reaction. The unknown entropy
terms are neglected. f3 is a symmetry factor analogous to the electro-
chemical symmetry factor and may be taken to have the same value.
114 CHAPTER 8: BASIC ELECTRODE KINETICS

The rate of the simple cathodic process (8.3.1) can be given by the
Butler-Volmer theory as

VI = (kT/h) CH+(l - 0) exp( -.dG c*IRT) exp[ -f3(.dcp) FIRT] (8.3.7)

where VI is the rate, C H + the concentration of hydrogen ions in the

OHP (moles per square centimeter), 0 the fraction of the surface
covered by Had, .d Gc * the chemical part of the free energy of activation,
and f3(.dcp)F the electrical contribution to the free energy of activation,
with .dcp is the potential difference between the metal and the OHP.
Writing

(8.3.8)

where .dcpR is .dcp when the electrode is held at the reversible potential
and 'T] is the overpotential, Eq. (8.3.7) becomes

VI = k' exp( -.dGc*IRT) exp( -f3'T]lIRT) (8.3.9)

where k' is a constant.

Applying the Bronsted law, the new rate V2 after the adsorption of
the organic material and changing the overpotential to a value 'T]2 will be

V2 = k' exp[ -(.dGc * - f3 .d(.dH)IRT] exp( -f3'T]2FIRT) (8.3.10)

Comparing (8.3.9) and (8.3.10), we have

Va = VI exp[f3 .d(.dH)IRT] exp[ -f3(.d'T])FIRT] (8.3.11)

where
.d'T] = 'T]2 - 'T]I
= -0.222 - (-0.160)
= -0.062 V

Taking the symmetry factor f3 as 1/2 and substituting numerical values

in (8.3.11) yields

V2 = 2 X 10-7 exp( -0.5 x 12,000/1.987 x 298)

x exp(0.5 x 0.062 x 96,500/8.314 x 298)
= 2.7 x 10-11 mole cm-2 sec-I
 ANSWERS 115

f5"""l The potential energy of a simple harmonic oscillator is parabolic

L::J with respect to displacement. The potential energy profile of
the charge-transfer reaction is thus appropriately described by two
intersecting parabolas as shown in Fig. 8.5.1. The zero-point energy
of the initial state oscillator is displaced -Q from the zero-point energy
(taken as zero for convenience) of the final state oscillator. The abscissa
q represents, in one dimension, the multidimensional displacements of
water molecules near the ion.
The reorganizational energy Es is the energy which must be
supplied to the solvated ion in the initial state so that it may have the
configuration of the ground state of the product ion. The energy
Es is marked in Fig. 8.5.1. The activation energy E* is the energy
which must be supplied to the reacting ion before electron transfer
obeying the Frank-Condon principle can occur, and is shown in
Fig. 8.5.1.
Writing UR and Up as the energies of the initial- and final-state
harmonic oscillators, we have, by inspection of Fig. 8.5.1,
Up = k(q - d)2 (8.5.2)
UR = kq2 - Q (8.5.3)

Fig. 8.5.1. Parabolic potential energy curves for initial- and

final-state harmonic oscillators.
116 CHAPTER 8: BASIC ELECTRODE KINETICS

At the intersection point of the two curves,

(8.5.4)
and
q = q' (8.5.5)

Substituting (8.5.4) and (8.5.5) in (8.5.2) and (8.5.3) and subtracting

(8.5.3) from (8.5.2) yields

kd 2 - 2kq'd +Q= 0 (8.5.6)

and rearranging yields:

q' = (Q + kd2)/2kd (8.5.7)

From Fig. 8.5.1, it is clear that

Es = kd 2 (8.5.8)

and substituting (8.5.8) in (8.5.7) yields

(8.5.9)

But from Fig. 8.5.1 it is clear that

E* = k(q')2 (8.5.l0)

and substituting (8.5.10) into (8.5.9) yields

E* = (Q + Es)2/4Es (8.5.1)

It is apparent from the above that (8.5.1) does not depend on any
particular model of charge transfer since the details of the actual
charge-transfer process were neglected in the derivation of (8.5.1).
Further, since (8.5.2) and (8.5.3) are the equations to parabolas, it is
apparent that (8.5.1) is only valid when the initial and final states of
the system approximate harmonic oscillators.

The image interaction energy between a univalent ion and a

metallic conductor is given by

(8.7.l)
 ANSWERS 117

where E is the dielectric constant and, is the distance between the metal
and the ion situated in the OHP. Assuming the position of the ion can
fluctuate by as much as I A, we have for the fluctuation of the image
energy
(8.7.2)
taking the metal-to-OHP distance as 4 A. Substituting numerical values
for eo and E into (8.7.2), we have
LI Vimage = -4.8 X 10-10 x 300/4 x 6 x 10-8 x 12
-0.05 eV
-0.05 x 23.06 kcal mole-1
-1.2 kcal mole-1
Consequently, the magnitude of the image energy fluctuations is only
about 10 % of typical activation energies for charge-transfer processes
(10-15 kcal mole-I).
In an elementary charge-transfer reaction in the gas phase, an
electron leaves a metal electrode and is captured by an ion situated
I cm away from the metal. Calculation of the image energy of the
electron as it leaves the metal and calculation of the Coulombic energy
of the electron as it approaches the ion provide ways of calculating
the energy barrier through which the electron must tunnel to reach
the ion.
The image interaction energy between a metal and the electron
is given by
(8.7.3)
where 'M is the distance between the metal and the electrode. Since
Vimage becomes very small when 'M is large, the zero of the energy scale
is the electron at rest at infinity. Taking the dielectric constant as 1,
this interaction energy at a separation of 1 A is given by
Vimage = 4.8 X 10-10 x 300/1 X 10-8 x 4
= -3.60eV
The energy of interaction between an electron and the ion is
Coulombic and given by
(8.7.4)
where '1 is the electron-ion separation. Again the electron at rest at
infinity is the zero of the energy scale.
118 CHAPTER 8: BASIC ELECTRODE KINETICS

-2·0
......
>CIl

::::J
-4·0

EF

-6·0

0 2·0 4·0 6·0 a·o

Jt M (,a)

Fig. 8.7.1. The barrier through which an electron must tunnel in a

charge-transfer process.

Figure 8.7.1 shows a plot of the electron energy as a function of

the electron~metal separation and also as a function of the metal~ion
separation. The distance between the metal and the ion is taken as
8 A. The energy of the electron in the metal is the Fermi energy,
which, on the scale of the electron at rest at infinity, is the negative
of the electron work function of the metal. For a nonradiative transfer
of an electron from the metal to the ion to take place, there must
be vacant electron levels in the ion at the Fermi energy of the metal.
The energies of the electron in the metal and of the ion are marked in
Fig.8.7.!.
The intersecting image force curve and Coulombic force curve
form a barrier through which the electron must tunnel to the ion. Very
close to the metal or the ion, (8.7.3) or (8.7.4) no longer gives a true
picture of the energy of the electron since both Uimage and Ucoul
become very large and negative when the electron is close. The true
situation close to the metal or the ion is better represented by the dotted
lines in Fig. 8.7.1, which take into consideration the fact that at the
metal or the ion, the electron energy is that of the Fermi energy of the
metal.
 ANSWERS 119

It is clear from Fig. 8.7.1 that the electron has to tunnel through
a barrier of width I equal to 4.4 A and height V2 - E equal to 2.8 eV.
The probability of tunneling P r through a barrier of width I and height
V2 at a level E is given (roughly) by the Gamow equation as
Pr = exp{ -(47Tljh) [2m(V2 - E)]1/2} (8.7.5)

where m is the mass of the particle. For the units in the exponential
to cancel, V2 - E must be in ergs when m is in grams and I in centi-
meters.
Substituting numerical values in (8.7.5), we have

47T x 4.4 X 10-8

Pr = exp [ - 6.624 x 10-27

x (2 x 9.1 x 10-28 x 2.8 x 1.60 x 10-12)1/2]

= 5.6 X 10-4

where the factor of 1.60 x 10-12 converts electron-volts to ergs.

It should be noted that (8.7.5) applies strictly only to a rectangular
barrier and for this reason, the above value of P r is only approximate.
A more accurate value of P r could be obtained by considering an
Eckhardt barrier.

o Consider the homogeneous exchange reaction

Fe3 + + Fe + ->- Fe2+ + Fe3+
2 (8.9.2)

proceeding along the path

Fe~~c + Fe~tc ->- [Fe~tsc + Fe~tsc]

H (8.9.4)
Fe~tc + Fe~~c +-- [Fe~tsc + Fe~tsc]
where the subscripts esc and nesc refer to the equilibrium solvent
configuration and the nonequilibrium solvent configuration sur-
rounding ions in their normal and activated states respectively. The
free energy of activation is the free energy difference between the
activated and initial states and is consequently

(8.9.5)
120 CHAPTER 8: BASIC ELECTRODE KINETICS

The corresponding heterogeneous reaction

Fe 3 + + e(M) _ Fe2+ (8.9.3)
proceeds along the path

Fe~tc + e(M) - [Fe~tsc + e(M)]

H (8.9.6)
Fe~~c +--- [Fe~tsc]
The electrochemical free energies of activation, L1G~et and L1G~et for
the forward and reverse reactions respectively are given by

(8.9.7)
and
(8.9.8)

Assuming that activation of the homogeneous and heterogeneous

processes takes place by the same unspecified mechanism, (8.9.7) and
(8.9.8) can be substituted into (8.9.5), leading to the result

(8.9.9)
However, at the reversible potential, the rates of the forward and
reverse heterogeneous reactions are the same, and for equal concentra-
tions of reactants and products,
-* = L1 -(--*
L1 Ghet Ghet (8.9.10)
and under these conditions (8.9.9) becomes
L1G~om = 2L1G~et (8.9.11)
The rate constants for the homogeneous and heterogeneous
reactions can be written
khom = Zhom exp(-L1G~om/RT) (8.9.12)
khet = Zhet exp(-L1G~et/RT) (8.9.13)
from which, together with (8.9.11), it follows that
(khom/Zhom)I/2 = (khet/Zhet) (8.9.1)
Since in this derivation of (8.9.1) no assumptions have been made
about the mechanism of activation, the experimental verification of
(8.9.1) does not constitute evidence in favor of the electrostatic model
of electrode processes.
 CHAPTER 9

MORE BASIC ELECTRODE

KINETICS

'1' (a) Calculate the diffusion limiting current for the oxidation
~ of an organic compound at an electrode in a quiescent solution.
Assume six electrons are involved in the reaction. Corganic = 10-2 mole
liter-I; DOrganiC = 2 X 10-5 cm 2 sec-i.
(b) Calculate the transition time for the same system at a current
density of 0.30 A cm- 2 •

For the cathodic reaction

A + 2e- -->- 2B (9.2.1)

io = 1.10 X 10-3 A cm- 2 and (8YJI8i)n->O = -28.3 ohms cm 2 at 25°C.

What is the stoichiometric number of the reaction and what is its
mechanistic significance in this case?

8 (a) Figure 9.3. I shows galvanostatic transients of current

L:J density 35 rnA cm- 2 for a Pt electrode in two solutions. One
contains 1.3 fLmoles liter- i benzene and the other is a blank. The blank
curve corresponds to the formation of an oxide film on the Pt, the
other to the oxidation of adsorbed benzene plus the formation of the
oxide film. Calculate the charge required to oxidize the adsorbed
benzene.
(b) Plots of equilibrium benzene coverage versus potential
121
122 CHAPTER 9: MORE BASIC ELECTRODE KINETICS

1.8 II ;'
w
:I:
z
:> 1.6
-I
« 1.4
i=
z
W
I- 1.2
~
w
0
0
1.0
a::
l-
e.>
w 0.8
-I
w
T IoI
0.6

0 2 4 10
I (msec)

Fig. 9.3.1. Galvanostatic transients of a platinum

electrode in a 1.3 fLmoles Iiter- 1 benzene solution and
for a blank solution .

obtained by radiotracer and galvanostatic transient methods are given

in Figs. 9.3.2 and 9.3.3, respectively. Assuming that benzene is adsorbed
intact and that the oxidation takes place according to
(9.3.1)

..'e
u
..
'2 0 .7 0.65 _ ____
w 0.63
. 10- 10
~
a:: 0 .6 Mole
11: 10 - 10
em· .! Molt
w em-z

8w 0.5 0.30 V 0.65V

z
w
~ 0.4
w
<am
Q 0.3 0 .1 0 .2 0 .3 0.4 0.5 0.6 0 .7 0 .8
ELECTRODE POTEN TI AL (V, NHE)
Fig. 9.3.2. Benzene coverage estimated by radiotracer
techniques on a p al t inum electrode.
 PROBLEMS 123

N
'eu
.0
§~ 1.2
.§
UJ
~ 1.0
a:
UJ
8> 0.8
UJ
z
~ 0.6
z
UJ
!D
"'Q 0.4 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
ELECTRODE POTENTIAL (V,NHEI
Fig. 9.3.3. Benzene coverage estimated by galvanostatic
transient techniques on platinum.

compare numerically the results of these two methods at 0.30 and

0.65 V NHE. What conclusions can be drawn about the principal
species present on the surface at the two potentials?

Q (a) The adsorption of propane from 85 % HaP04 has been

~ studied by Brummer (1965). At elevated temperatures, the
initial adsorption obeys the diffusion equation

(9.4.1)

where Q is the anodic charge per square centimeter of geometric area

needed in a galvanostatic transient to oxidize the adsorbed material
to CO 2 and H-l. The process involves n electrons, and t is the time. If
adsorption occurs without partial oxidation, calculate the initial slope
of the Q versus (1/2 curve using Cpropane = 1.6 X 10-4 mole liter-1 and
Dpropane = 1.07 X 10- 6 cm 2 sec-I. What would be the slope per real
square centimeter of surface if the roughness factor is three?
(b) The fractional surface area free of adsorbed propane can be
measured by cathodic charging with hydrogen:
H-l + e- -+ Hads (9.4.1)
where each adsorbed hydrogen atom occupies one platinum atom on
the surface. At a chosen potential, propane is allowed to adsorb on
124 CHAPTER 9: MORE BASIC ELECTRODE KINETICS

the electrode for a certain time t and the unoccupied surface is then
charged with hydrogen by applying a cathodic pulse. From the number
of coulombs passed, the fractional coverage by hydrogen OH , which
corresponds to the fraction of surface sites not covered by adsorbed
propane, is obtained. The experiment is repeated for different times of
adsorption of propane. Figure 9.4.1 shows plots of OH against [1/2 for
adsorption of propane at two potentials.
Taking unit coverage by adsorbed hydrogen as 1.3 X 1015
atoms cm- 2 of platinum, calculate the number of surface sites occupied
by one propane molecule at 0.2 V and also at 0.4 V.

(a) By considering the first step of the hydrogen evolution

reaction as a quasi equilibrium

1.0

0.9

0.8
0.2 VOLT
0.7

0.6

0.5
8H
04
\
0.4 VOLT
\
0.3
\
\
0.2
\
\
0.1
\
\
0 2 4 6 8 10 12 14
1 1
t'2 (sec'2 )
Fig. 9.4.1. Fractional coverage by hydrogen obtained
by cathodic charging experiments as a function of t ' /2 ,
where t is the time of adsorption of propane in phosphoric
acid.
 PROBLEMS 125

H+ + e- ? Had (9.5.1)
derive the relation
dB/dE = -FB(l - B)/RT (9.5.2)

where B is the fractional coverage by Had at potential E.

(b) During galvanostatic transients, the equilibrium (9.5.1) can
give rise to large pseudocapacitances C given by
C = -Fr dB/dE (9.5.3)
where ris the unit coverage by Had (1.7 X 10-9 atom cm- 2). The
experimental value of the pseudocapacitance on an iron electrode at the
potential, -0.45 V NHE, is 230 fLF cm- 2 • Calculate the corresponding
fractional coverage by adsorbed hydrogen.

r-;l Table 9.6.1 lists the standard potentials for the electrode
~ processes that may occur on iron electrodes in aqueous solution.
Construct a potential-pH diagram (Pourbaix diagram) for iron. From
the diagram, deduce the lower pH limits of stability of iron in (a) oxygen-
saturated solution containing 10-2 mole liter-1 ferric ions, (b) oxygen-
free solution containing 10-2 mole liter-1 ferrous ions. Assume that
oxide films provide complete protection when they are thermodyna-
mically stable.

TABLE 9.6.1. Standard Potentials

for the Electrochemical Processes Occurring at Iron

Reaction Standard potential, V

Fe2+ + 2e- :;:::': Fe -0.44

Fe3+ + 3e- :;:::': Fe -0.04
+ 8H+ + 2e- :;:::': 3Fe + + 4H 20
Fe 3 0. 2 0.98
+ 8W + 8e- :;:::': 3Fe + 4H 20
Fe 3 0. 0.08
3Fe20 + 2H+ + 2e- :;:::': 2Fe 0. + H 20
3 3 0.22
2H+ + 2e- :;:::': H2 0.00
O 2 + 4H++ 4e-:;:::': 2H 02 1.23
Fe 20 + 6H+ + 2e-:;:::': 2Fe2+ + 3H 20
3 0.73
2Fe3+ + 3H 20:;:::,: Fe 20 + 6H+
3 *
* pKs = 0.72, where K, is the solubility product.
126 CHAPTER 9: MORE BASIC ELECTRODE KINETICS

r:;--] Using the following data, construct an E-log i diagram

L2J (Evans-Hoar diagram) for the corrosion of iron in hydrogen-
saturated oxygen-free solution of pH = 3.1 and aFe2+ = 0.02 (molar
scale).
For iron dissolution, io = 9 X 10-7 A cm-2 ; anodic d1J/d(log i) =
0.04 V; the corrosion potential of iron is -0.215V(RHE)and Epe2+/Fe =
-0.44 V. For the hydrogen evolution reaction (h.e.r.), the cathodic
d1J/d(log i) = -0.12 V.
Calculate (a) the corrosion rate, (b) the exchange current density
for the hydrogen evolution reaction on iron, (c) the cathodic protection
current required to reduce the corrosion rate to zero.

Iol One preparation method for aniline depends on the reduction

~ of nitrobenzene with hydrogen using colloidal platinum as a
catalyst. Examine the kinetics of the process assuming that it occurs by
an electron-transfer mechanism at the platinum. Neglect the polarization
of the hydrogen evolution reaction and take the following parameters
for the reduction of nitrobenzene: io = 10-8 A cm- 2 ; d1J/d(log i) =
-0.12 V; and the reversible potential is 0.S7 V (RHE).
Calculate the rate of reduction in moles per liter per second if
there are 1010 particles of Pt per liter of average diameter 2 x 10-4 cm.

f9l Potential sweeps can yield transient data at high sweep rates,
~ and steady-state data at low sweep rates.
(a) Estimate the minimum sweep rate that can be used to measure
the coverage of a Pt electrode by benzene at 50°C before readsorption
contributes an error of 10 %. Assume adsorption follows the equation
~ _ -.2: Beq (~)1/2 (9.9.1)
dt 1 / 2 - Krmax 7T

where K is the equilibrium constant for adsorption and equals 6 x lOS

cm3 mole-I; r max is the coverage corresponding to 0 = 1 and equals
2.5 x 10-10 mole cm- 2 ; Oeq is the equilibrium coverage and may be
taken as 0.3; D = 1.0 X 10-5 cm 2 sec-I.
(b) The oxidation of ethylene in sulfuric acid solution has been
investigated by the potential sweep technique. The apparent d1J/d(log i)
on platinum at SO°C increases from the steady-state value of 0.145 V
 PROBLEMS 127

to about 0.20 V as the sweep rate goes from 10-5 V sec- I to 10-3 V sec-I
and thereafter remains constant. The Tafel region extends over a span
of 0.35 V. Show that the higher slope is consistent with an irreversible
diffusion-controlled reaction for which (Delahay, 1953)

(9.9.2)

where A = (rxF/RT)v; n is the number of electrons transferred in the

overall reaction; rx is the transfer coefficient; v is the sweep rate in volts
second-I; X(At) is a couple function of (At) and is plotted in Fig. 9.9.1.

r:jQl For the ethylene oxidation reaction on platinum at 80°C, the

~ following parameters are found over the pH range 0.5-12.5
and the potential range 0-1 V (RHE);

JH ~ 20 kcal mole-I; ECO

2
= 100 %; [oE/o(ln i)]j) = 2RT/F
(Oi/OP)E < 0; diL/dp ~ 1

ECO
2
is the Coulombic efficiency for oxidation of ethylene to CO2 , P is

6·0

-
.......
,c:.
4·0

2·0

-3,0 -2,0 -1-0

log x (M)
Fig. 9.9.1. Plot of the function (At) against the logarithm of the couple
function x(At). (Gileadi et al., 1966.)
128 CHAPTER 9: MORE BASIC ELECTRODE KINETICS

the partial pressure of ethylene, and iL is the diffusion-limited current.

If the same mechanism applies in both acid and alkaline solution, what
is the rate-determining step prior to diffusion control? Consider only
(OH) and (C 2 H4) as possible adsorbed species.

For the oxalic acid oxidation reaction on platinum at 80°C, the

following empirical relationship was found in acid solution:

(9.11.1)

Assume that the reacting species is the undissociated acid; the only
adsorbed species that need to be considered are OH, HC 20 4 , and
HC0 2 ; charge-transfer steps are always single-electron steps. Then
write two possible reaction paths for each of the following possibilities:
(a) OH radicals are not involved; (b) the oxalic acid molecule undergoes
a chemical reaction step:

(9.11.2)

and thereafter OH radicals are not involved; (c) two OH radicals are
involved per molecule of oxalic acid. Assuming Temkin adsorption
conditions, establish the r.d.s. (if any) for each path that is consistent
with the observed potential dependence of (9.11.1), and then show
which mechanisms can be eliminated on the basis of the other data.

Q The kinetic parameters in Table 9.12.1 were observed for the

~ anodic oxidation of each of four saturated hydrocarbons on
platinum in 85 % phosphoric acid at potentials below 0.5 V RHE.
Write down about a dozen partial mechanism hypotheses
(sequences up to and including the rate-determining step) and tabulate
the values for the three parameters above predicted by each mechanism

TABLE 9.12.1. Kinetic Parameters for the Anodic Oxidation

of Saturated Hydrocarbons on Platinum

(0£/0 In i)p.T (0 In i/o In P)aH 0

2

89-102 ± 5 RT/F o
 ANSWERS 129

assummg Langmuir adsorption conditions and low coverages by

adsorbed species. By comparing the parameters with experimental
values, reduce the number of possible mechanisms to two. Estimate
which of the two would be energetically unfavorable from the following
data. Bond strengths are: C-C = 93.1 kcal mole-I; C-H = 106 kcal
mole-I; Pt-C- 44 kcal moie- 1 ; Pt-H = 62 kcal mole-I; also,
LlHad(H 2 0) = 20 kcal mole-I; the ionization energy of H, I H - H + =
313 kcal mole-I; LlH(H g >- H:
01n ) = -263 kcal mole-I; the work
function of platinum rpPt = 123 kcal mole-I. Write down the most
probable mechanism for the oxidation of saturated hydrocarbons up
to the rate-determining step and give one possibility for the rest of the
reaction.

ANSWERS

(a) Under steady-state conditions, the flux J of the organic

molecules arriving at the electrode is given by Fick's first law:

J -D dCjdx moles cm- 2 sec- l (9.l.l)

where the concentration is expressed in moles per cubic centimeter

and J and dCjdx are vectors directed toward and away from the
electrode, respectively. For an oxidation involving six electrons,
J = i/6F (9.1.2)
where i is the current density. Assuming a linear concentration gradient,

dCjdx = (Crlectrode - Cbu1k)/O (9.1.3)

where 0 is the diffusion layer thickness and may be taken as 0.05 cm in

unstirred solution. The gradient is a maximum when Celectrode = 0
and the maximum, or limiting current, is given from (9.l.l)-(9.1.3) as

(9.1.4)

which becomes, substituting given values,

iL = 6 .c< 96,500 x 2 x 10-5 x 10-5/5 X 10- 2

= 2.3 mA cm- 2
130 CHAPTER 9: MORE BASIC ELECTRODE KINETICS

(b) As shown by Sand (1900), the transition time T is given by

T = 7TD(zFCj2i)2 (9.1.5)

where the concentration C is again expressed in moles cm-3 • Substituting

the given values, we find

T = 7T X 2 X 10-5 [6 X 96,500 X 10-5 /2 X 3 X 10-1 ]2

= 3x 10- sec 3

I3l (a) Charge passed during a galvanostatic transient of current

L.:J density i is iT (coulombs cm- 2), where T is the time to reach
steady state. Therefore, the charge associated with Pt oxidation is
iTox and the charge Q associated with benzene oxidation (see Fig. 9.3.1)
IS

Q= i(Ttot - Tox) (9.3.2)

Therefore,

Q = 35 X 10-3 (8.0 X 10-3 - 5.0 X 10-3)

= 1.05 X 10-4 coulomb cm- 2

(b) From radio tracer measurements, the coverage at 0.30 V,

FO• 30 , is (Fig. 9.3.2)
F O. 30 = 0.65 X 10-10 mole cm- 2

Assuming that benzene is adsorbed without chemical reaction, the

oxidation during a galvanostatic transient will produce 18 electrons per
molecule:
(9.3.3)

The charge required to oxidize benzene adsorbed at 0.30 V is

QO.3 = 18 X 96,500 X 0.65 X 10-10 coulomb cm- 2

=c 1.13 x 10-4 coulomb cm- 2

This value is in close agreement with the value measured by galvano-

static transients (1.14 X 10-4 coulomb cm- 2). The benzene molecule jo,
 ANSWERS 131

therefore adsorbed intact at this potential. Similarly, the charge required

to oxidize benzene at a coverage of 0.63 mole cm- 2 would be
0.85 x 10-4 coulomb cm- 2 measured by galvanostatic transients. If
one could attribute the discrepancy to partial oxidation of the benzene
during adsorption, say to some species C6 H n O", , then the number of
electrons produced per molecule would be given by the equation

The number of electrons (n + 12 - m) required to account for the

discrepancy is (0.86/1.09) x 18 or about 14 in place of 18 for benzene.

r;J (a) For the equilibrium, Eq. (9.5.1), the rate in the forward
~ direction equals that in the reverse direction, and, making use
of the Butler-Volmer equation, we have

kC H +(1 - B) exp( -f3EF/RT) = tB exp[(1 - (3) EF/RT] (9.5.4)

the equilibrium constant is

-> f-

K = k/k (9.5.5)

and substituting (9.5.4) into (9.5.5) yields

K = [B/C H +(1 - B)] exp(EF/RT) (9.5.6)

Differentiating with respect to E,

-(KCH +/B2) dB/dE = (F/RT) exp(EF/RT) (9.5.7)

and substituting for K yields (9.5.2):

dB/dE = -FB(l - B)/RT

(b) Substituting (9.5.3) into (9.5.2) yields

C = £2rB(1 - B)/RT (9.5.8)

and rearranging and substituting numerical values gives

B(1 _ B) = 230 X 10-6 x 8.314 x 298

(96,500)2 x 1.7 X 10-9
B= 0.04
132 CHAPTER 9: MORE BASIC ELECTRODE KINETICS

r:;l Standard potentials are on the scale of the standard hydrogen

L2J electrode. For this calculation, it is convenient to have all
potentials on the reversible hydrogen electrode scale (RHE), i.e., the
potential of the electrode measured against a hydrogen electrode in the
same solution:

EFe 2+/Fe(RHE) ~ E~e2+/Fe + (RT/2F) In aFe + (RTjF)pH

2+ (9.7.1)

-0.44 - 0.050 + 0.183

-0.307 V

The Evans-Hoar diagram is shown in Fig. 9.7.1. The iron dissolution

line can be drawn since its slope (0.040 Vjdecade of current density) and
one point on the line [io, E Fe 2+/Fe(RHE)] are known. The corrosion
potential is a mixed potential with iron dissolution and hydrogen
evolution proceeding at equal and opposite rates. Consequently, the
hydrogen evolution line can be drawn with a slope of 0.12 V jdecade of
current density through the point where the iron dissolution line crosses
the corrosion potential, -0.215 V.

-0,'

-0,2
EcorrosIon
.

10- 6 10-5 10-4

(A cm- 2 )
Fig. 9.7.1. The Evans-Hoar diagram for the corrosion of iron.
 ANSWERS 133

From Fig. 9.7.1, the corrosion rate of iron in this solution can be
seen to be 1.8 x 10- 4 A cm- 2 and io for the h.e.r. is 3.0 x 10- 6 A cm- 2 •
The dissolution of the metal ceases when its potential is lowered to
its reversible potential, a cathodic current density of 1.2 x 10- 3 A cm- 2,
called the cathodic protection current density, being required to
maintain this potential.

In'] (a) To cause an error of 10 %, the sweep must be of sufficient

L!J duration for (-}eq/lO to readsorb. A relation between (-) and t can
be obtained by integrating (9.9.1). The constant of integration is zero
because (-) is zero at t = 0:

(9.9.3)

Substituting (-}eqjlO in (9.9.3) for (-) yields

(9.9.4)

and rearranging gives

(9.9.5)

Substituting given values yields

= (6·, 108 x 2.5 X 1O-1 0j20)2 7T /I0- 5

17.6 sec

A reasonable amplitude for a voltage sweep is 1.5 V, thus the minimum

sweep rate would be ,,-,0.1 V sec-I.
(b) The Tafel slope aVjo(ln i) can be written

oV oV o[ln X('\t)]
----- (9.9.6)
cOn i) o[ln X(,\t)] oOn i)

But from Eq. (9.9.2), c[ln X(At)]jc(ln i) = 1, and (9.9.6) becomes

cV cV
(9.9.7)
cOn i) c[ln X(,\t)]
I.'(At) cV
(9.9.8)
o[ln X(,\t)] o(,\t)
134 CHAPTER 9: MORE BASIC ELECTRODE KINETICS

Now, the voltage at any time t during a voltage sweep of rate v is given
by
v= Vo + vt (9.9.9)

where Vo is the initial voltage. From (9.9.8) and the definition of A,

we have
At = (rxF/RT)(V - Vo) (9.9.10)
Therefore,
oV/o(At) = RT/rxF (9.9.11)

Now, the steady-state drJld(log i) of 0.145 V at 80°C is consistent

with rx = 0.5. At may be estimated from (9.9.10). Substituting numerical
values we get
At = 96,500 X 0.35/2 X 8.314 X 353
= 5.75
At At = 5.75, the slope of the plot in Fig. 9.9.1 is 3.1 and hence

o(At)/o[ln X(At)] = 3.1/2.303 (9.9.12)

Substituting (9.9.11) and (9.9.12) into (9.9.8) yields

oV/o(ln i) = 3.1RT/2.303rxF

and the anodic slope, 8 Vl8(log i), is given by

oV/o(log i) = 3.1 X 8.314 X 353/0.5 X 96,500

= 0.19 V
Consequently, the apparent Tafel slope at high sweep rates is consistent
with an irreversible diffusion-controlled process.
 CHAPTER 10

SOME ELECTRODIC REACTIONS

OF INTEREST

r71 From the data of Table 10.1.1 for the hydrogen evolution
L.!.J reaction, calculate the relative electrocatalytic activity of an
iron electrode with respect to that of a nickel electrode.

TABLE 10.1.1. Kinetic Parameters for Hydrogen Evolution

Overpotential at a d'Y//d(log i),

Electrode fixed current density, V V

Fe -0.427 -0.120
Ni -0.350 -0.112

r:;--, Galvanostatic transient experiments are performed on the

L2.J deposition of silver from aqueous solution near the reversible
potential. The mechanism is

(lO.2.1a)
Agad ~ Aglattice (10.2.1 b)

where step (b) is the diffusion of the adion Agad to a growth site where
it is rapidly incorporated into the lattice. If step (b) is the rate-deter-
mining step (r.d.s.), the overpotential 'Y/D will be given by
-'Y/D = 'TsDiRTj£2Co (10.2.2)
where 'TSD is rise time (measured after the double-layer charging is
135
136 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

complete) of the galvanostatic transient of current density i; Co is the

equilibrium concentration of adions and the listed values in Table 10.2.1
are constant within experimental error. Use an average value of Co to
calculate 'TJD at each current density. Calculate the overpotential for
step (a) rate-determining on the basis of an exchange current density of
0.30 A cm- 2 . Which step is the r.d.s. ?

TABLE 10.2.1. Equilibrium Adion Concentration Co

in the Electrodeposition of Silver
at Near-Equilibrium Conditions at 25°C

Equilibrium adion
Current density i, concentration Co , Rise time
rnA cm- 2 mole cm- 2 TSD , JLsec

o 3.3 X 10-11
6.3 2.1 X 10- 11 187
8.9 2.6 X 10-11 156
15.0 2.3 X 10-11 96
22.6 3.2 X 10-11 75
44.7 5.8 X 10-11 53

Consider the hydrogen evolution reaction

H+ + e- ~ Had (l0.3.1a)
(l0.3.1b)

where, on the medium-overvoltage metals, step (b) is the r.d.s., show that,
for low 8 (fractional coverage by Had),

(dO/dry) = -(FK/RT) exp( -F'TJ/RT) (10.3.2)

and that at high coverage by Had, i.e., 0 --+ 1,

dO/d7) --+ 0

~ The study of certain electrode reactions requires solutions of

l2J extreme purity. Estimate the electrode area required to purge
190 ml of solution by anodic preelectrolysis of an impurity initially
present at 5 x 10-6 M within 24 hr. Take the maximum acceptable
 PROBLEMS 137

impurity content as 10-11 M, the diffusion coefficient of the impurity as

10-5 cm 2 sec-I, and the diffusion layer thickness as 0.05 cm. What
time would be required if the solution was stirred or circulated over the
electrode?

Tafel slopes at low overpotentials for the oxygen reac-

tion
(10.5.1)

on iridium are 2RT/3F (anodic) and -2RT/F (cathodic). At higher

anodic overpotentials, the Tafel slope becomes 2RTjF. (a) What is the
stoichiometric number at low overpotential? (b) Show that the low
overpotential data are consistent with the "electrochemical oxide"
mechanism with the second step as the r.d.s.:

K
H 20 +:!: OHad + H+ + e- (8 0H ~ 1) (l0.5.2a)
"i
OHad +:!: Oad + H+ + e- (r.d.s.) (lO.5.2b)
"
r
K'
20ad +:!: O2 (8 0 ~ 1) (l0.5.2c)

(c) Which step as the r.d.s. is consistent with the high anodic
overpotential data?

~ Draw two intersecting Morse-type curves for the species M-H

L:J and H+-(H 20), respectively, to represent the initial and final
states of the rate-determining step

where M-H represents a hydrogen atom chemisorbed on the metal

electrode and H+-(H 20) a hydrated proton in the OHP.
(a) How will the heat of activation vary with the heat of adsorption
of hydrogen? If the intersecting parts of the curves can be approximated
by straight lines, show geometrically that

tan y/(tan ,.\ + tan y) = fJ (10.6.2)

138 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

where f3 is the symmetry factor and y is the angle between the M-H line
and the vertical; A is the angel between the H+-(H 20) line and the
vertical.
(b) What will the corresponding relationship be if the r.d.s. is
the next step?
(10.6.3)
(c) Express the results of (a) and (b) in a qualitative statement
about the respective effects of adsorption energy of hydrogen on the
metal on the exchange current density of the reaction.

r::;-) Table 10.7.1 lists the kinetic parameters of processes that can
L2J occur during the deposition of iron from acidified 0.05 M
ferrous solution. Assuming activation control, calculate the pH at
which the iron deposition current is ten times that of water discharge.
Find graphically the potential which gives rise to the maximum efficiency
for iron deposition at this pH.

TABLE 10.7.1. Kinetic Parameters for Iron Deposition

in Acidified 0.5 M Fe2+ Solution

Reacting EO,
species VNHE log io d1} i d(log i) logh

Fe'" -0.44 pH - 9 0.12 -2

HsO+ 0.0 -(0.5 pH + 4) 0.12 -pH
H 2O 0.0 -8.0 0.12 (very large)

(a) From a consideration of the electrode-kinetic condition for

equilibrium in the simple charge-transfer redox reaction
(10.8.1)
show that the Fermi level in the electrode at the reversible potential
is the independent of the electrode material. Note that the Fermi level
is equivalent to the electrochemical potential of electrons in the electrode
material.
(b) I n spite of the fact that the free energy of activation of a simple
charge-transfer redox reaction is independent of the electrode material,
 PROBLEMS 139

a roughly linear relationship has been observed between log io and <P
(the electron work function) for the reaction
(10.S.2)
for a series of metals. What sign would you expect for the slope
d(log io)jd<P if the relationship arose from the effect of the electrokinetic
potential upon ionic concentrations in the Helmholtz plane?

(a) By integrating the steady-state diffusion equation expressed

in spherical coordinates

(10.9.1 )

between '0 and '0 + 0, where '0 is the radius of a spherical electrode,
othe diffusion layer thickness, and C the concentration of the diffusing
species at " show that the maximum rate of diffusion in terms of the
limiting current i L is given by
(10.9.2)
where Coo is the concentration of the diffusing species in the bulk of
the solution.
(b) The overpotential at the tip of a metal dendrite growing by
electrodeposition has three components: activation 'YJA , diffusion YJ D ,
and Kelvin overpotential TJK . The Kelvin contribution arises because
the surface energy of the sharp tip alters the reversible poteQtial of the
electrode reaction:
7)K = 2yVjzFr o (10.9.3)
where y is the surface energy and V the molar volume of the deposit.
For a fixed total overpotential YJ, derive the following expression for the
optimum radius of curvature of the tip of a dendrite for the most rapid
growth rate:
I ± [I + DCooz2p2YJj2y VioJ1/ 2
(10.9.4)
zFYJj2yV
Assume
(l0.9.5)

where YJA is the activation overpotential.

140 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

(c) Using the data given below for the growth of zinc dendrites,
calculate the radius of the tip of the dendrite.
(d) Calculate the ratio of the current density on the dendrite tip
to that on the planar part of the electrode.
YJ. 0.02 V: D = 7 X 10- 6 cm 2 sec-1
C£ == 10 1 M: y = 2000 ergs cm- 2
V = 9.2 ml mole- 1 ; T =35°C
is - 5 X 10- cm;
2 io = 0.10 A cm- 2

~ The rate of ethylene oxidation at a given potential on a series

~ of metals and alloys is shown in Fig. 10.10.1. The mechanism
of the reaction is
C 2 H4 ~ (C 2 H 4)ad (lO.lO.la)
H 20 ~ OHad + H+ + e- (IO.IO.lb)
(C 2 H 4 hd + (OHhd ~ (C 5 HtOhd (lO.lO.lc)
and the species (C 2 H 50)alj further reacts to produce CO 2 and water.
Step (b) is the r.d.s. on platinum, but on all other substrates, it is step (c).
The abscissa in Fig. 10.10.1 is the latent heat of sublimation of the
metal, a parameter related to the metal-metal bond strength.
Rationalize the volcano shape of the curve in terms of Temkin
adsorption of OH radicals and competition between OH radicals and
C 2 H 4 for adsorption sites.

r.j1l By a consideration of the work of Damjanovic, Dey, Genshaw,

~ Hoare, Huq, Krasilchikov, Rao, Rosenthal, Schumilowa, and
Veselovski, discuss:
(a) The dependence of the rate of oxygen reduction upon the
degree to which the surface is covered with oxide.
(b) The dependence of exchange current density upon the M-O
bond strength.
(c) The likely rate-determining steps for O 2 dissolution on
platinum.
What directions appear to be the most promising for fundamental
research on the catalysis of oxygen reduction?
 PROBLEMS 141

Ao
o Pd

Co- Au Rh
025%-75·/0 0

o.
60 80 100 120 140 180
L (Kcol mole-I)

Fig . 10.10.1. Rate of the electrochemical

oxidation of ethylene on metal and alloy elec-
trodes as a function of the latent heat of sub-
limation of the electrode material. (Data from
Kuhn et at., 1967.)

G2l Investigate the mechanism of the electrochemical oxidation of

~ hydrocarbons, with special reference to the work of Bagotsky,
Breiter, Brummer, Grubb, Johnson, Kuhn, Niedrach, Piersma, and
Wroblowa.
Give the best available replies to the folIowing questions:
(a) In view of the projected rapid development of atomic power,
what is the economic reason for interest and research in the electro-
chemical oxidation of hydrocarbons? Is there a sociological advantage
to be looked for in such research?
(b) Is adsorption of organic materials from solution onto electrodes
usualIy with or without dissociation? How may adsorption with
dissociation be detected?
(c) In steady-state measurements, various parameters of the
reaction, e.g., stoichiometric number, cathodic and anodic Tafel
142 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

slopes, and reaction orders, can often be determined and compared

with values predicted for various model mechanisms. Transient and
potential sweep measurements often lead to information on the nature
and surface concentration of intermediates. Discuss the complimentary
roles of the two types of measurements for hydrocarbon oxidation.
(d) What generalization seems to be possible with respect to the
rate-determining step of organic oxidation mechanisms?
(e) The rational evaluation of nonnoble catalysts for hydro-
carbon oxidation depends upon elucidation of the rate-determining
steps in hydrocarbon oxidation, and this, in turn, would be greatly
helped by knowledge of organic radicals on surfaces. Discuss available
and potential approaches to surface radical determination.
(f) What research, then, would you advocate at the present time
for furthering our understanding of the mechanism and catalysis of the
electrochemical oxidation of hydrocarbons to CO 2 ?

ANSWERS

I.jl The relative electrocatalytic activity is given by the ratio of the

~ exchange current densities iO(Fe)/io(Nil . The high-field approxima-
tion of the Butler-Volmer equation is, for a cathodic process,
i= io exp( -f37JF/RT) (10.1.1)
from which it is apparent that
d7J/d(log i) = -2.303RT/f3F (10.1.2)
and substituting (10.1.2) into (10.1.1) yields

i= io exp[2.3037J d(log i)/d7J] (10.1.3)

Since the overpotentials given in Table 10.1.1 refer to a fixed

current density, we have, by substituting in (10.1.3) and dividing,
iO(Fe) exp(2.303 x 0.350/0.112)
iO(NI) exp(2.303 x 0.427/0.120)
=0.37
 ANSWERS 143

With step (1O.3.1b) as the r.d.s., then step (1O.3.1a) may be

considered to be in quasiequilibrium:

and the rates of the forward and reverse reactions are equal. From the
Butler-Volmer equation, we have for the equilibrium (10.3.1)

k,CH+(l - 0) exp( -f3TJF(RT) = krO exp[(1 - (3) TJF(RT] (10.3.2)

which rearranges to

0(1 - 0)-1 = KCH+ exp( -TJF(RT) (10.3.3)

where
(10.3.4)

Differentiating (10.3.3) with respect to TJ,

[(1- 0)-1 + 0(1- 0)-2] dO(dTJ = -(FKCH+(RT) exp( -TJF(RT) (10.3.5)

and rearranging yields

[1 + 0(1 - 0)-1] dO(dTJ = -(1 - O)(FKCH+(RT) exp( -TJFJRT) (10.3.6)

At low coverage, (1 - 0) --+ 1, and (10.3.6) becomes

(dO(dTJ) = -(FK(RT) exp( -TJF(RT) (10.3.7)

At high coverages, 0 --+ 1, and (10.3.6) becomes

dO(dTJ = 0

The stoichiometric number v is given by Parsons (1951)

(cf. Bockris and Reddy, 1970) as

(10.5.3)

where n is the number of electrons transferred in the overall reaction,

144 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

and oYJ/o In ia and oYJ/o In ic are the anodic and cathodic Tafel slopes,
respectively. Substituting the given values in (10.5.3) yields

v = (4F/RT)[(3F/2RT) + (F/2RT)]-1
=2
(b) For step (10.5.2b) as the r.d.s., the anodic current ia is given by

ia = 4Fk/JOH exp[(l - (3)YJF/RT] (10.5.4)

Since step (lO.5.2a) may now be regarded as being in quasiequilibrium,

we have for this step

(10.5.5)

Since BOH ~ I, then (l - B OH ) ~ 1, and substituting (10.5.5) into

(10.5.4) yields
(10.5.6)
Therefore,
In ia + In CH + = In 4Fkf K + (2 - (3) YJF/RT (10.5.7)
and differentiating at constant C H + yields

(10.5.8)

which is as observed.
For the cathodic reaction,
(10.5.9)
and since step (l0.5.2c) may be considered to be in quasiequilibrium:

(10.5.10)

where Po 2 is the partial pressure of oxygen. Hence,

(10.5.11)

from which it follows that

(10.5.12)

which is the observed cathodic Tafel slope.

 ANSWERS 145

(c) For step (10.5.2a) rate-determining, the anodic current is

given by
i" = 4Fk exp[(l - ~) YJF/RT] (10.5.13)

from which it follows that

which is consistent with the high-overpotential data.

From the Butler-Volmer equation, we have that the current

for iron deposition is given by

log i . ~ log io - [( V - VR )/0.12] (10.7.1)

where V is the electrode potential and VR is the reversible potential

for iron deposition, on the N HE scale, and

(10.7.2)

since the activity of metallic iron is unity. Substituting numerical values

in (l 0.7.2) yields

VR = -0.44 + (8.314 x 298/2 x 96,500) In(0.05)

, -0.48 V

Substituting for V R and io in (\0.7.1) yields

log iFeH = pH - 9.0 - [(V + 0.48)/0.120] (10.7.3)

Similarly, for water discharge, we have

log i H20 = -8.0 - [( V + 0.060pH)/0.120] (10.7.4)

since for water discharge

VR = 0.0 + (RT/2F) In(CH +)2 (10.7.5)

When
(10.7.6)
146 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

and from (10.7.3), (10.7.4), and (10.7.6), we have

pH
- 90 -
.
V + 0.48 = 1.
0.120
0 _ 80 _ V
.
+ 0.12
0.060pH
(10.7.7)

and condition (10.7.6) is satisfied, independent of potential, when

pH = 4.0.
Under mixed activation and diffusion control, the current density
for iron deposition is related to the potential by

V = VR - dO d-:
og IFe2+
) [log iFe2+ + (9.0 - pH)]

+ RT In (1 _ .iFe2+ ) (10.7.8)
2F lL_Fe2+

and substituting numerical values into (10.7.8) yields

V = -0.48 - 0.120(log i Fe 2+ + 5.0) + 0.030 log[l - (iFe2+/1O-2)]

(10.7.9)
and the potential corresponding to various current densities for iron
deposition can be found by substituting values for i Fe 2+ into (10.7.9).
The results are plotted in Fig. 10.7.1.
At a potential of 0.50 V NHE, the overpotential for discharge of
H30+ is given by

Y) = V - VR
= -0.50 - 0.060pH
= -0.260 V
and substituting for Y) in the Butler-Volmer equation, we have

log iH 0+
3
= -(0.5pH + 4.0) + (0.260/0.120)
= -3.80

However, since the diffusion-limited current for H30+ discharge is

10- 4 A cm- 2 (Table 10.7.1), discharge of hydroxonium ions will
proceed at a current density of 10-4 A cm- 2 at all potentials more
negative than -0.50 V. This is shown in Fig. 10.7.1.
Water discharge obeys Eq. (10.7.4) and its potential-current
density relation is also plotted in Fig. 10.7.1.
 ANSWERS 147

'"' -0.90
ILl
z
Z
;>
~

...J
«
~
z
ILl
t-
O
0.

10- 5 10-4 10- 3 10- 2

CURRENT DENSITY (A cm- 2 )

Fig. 10.7.1. Current densities for iron deposition, water discharge, and hydro-
xonium ion discharge as a function of electrode potential.

The current density efficiency for iron deposition is given by

(10.7.10)
Interpolating the values of i Fe 2+, i H20 , and i H • O + from Fig. 10.7.1, the
curve of efficiency against potential (Fig. 10.7.2) can be constructed.
Under the stated conditions, the maximum efficiency is seen to be 88 %
at -0.80 V NHE.

(a) A general integration of (10.9.1) yields

r 2 dC/dr = k (10.9.6)

where k is a constant and integration of (10.9.6) between limits ro and

ri = ro + S yields

CTt - CTO = ITi kr2 dr (10.9.7)

TO

(10.9.8)
148 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

90
z
0
i=
V5
~
w
a
~
a::
fr
r
u
z
w
U
lL..
lL.. 40
W
t-
Z
w 30
a::
a::
=>
u 20
0~ -0.90 -0.80 -0.70 -0.60
POTENTIAL (V, NHE)
Fig. 10.7.2. Current efficiency for iron deposition as a function
of electrode potential.

The flux at the electrode ir o is given by Fick's first law as

J ro = - D(dCjdr )(r~'o) (10.9.9)

and substituting (10.9.6) and (10.9.8) into (10.9.9) gives

(10.9.10)

Now
i = zFiro (10.9.11)

and since the maximum flux across the diffusion layer is obtained when
C r o is zero and C, , = COX) the diffusion limiting current iL is given by
(10.9.2)
iL = zFDC,:/ro

(b) The total overpotential 1) is given by

1) = 1) A + 1) + 1)K
D (10.9.12)
 ANSWERS 149

where 1JA is the activational overpotential given by the Tafel equation

and the diffusional overpotential is given by

(10.9.13)

Substituting (10.9.3) and (10.9.13) into (10.9.12) yields

YJ = iRT + iRT + 2y V (10.9.14)

iozF iLZF zFro

and differentiating with respect to ro yields

O = ~ RT +~ RT + iRT
-F==V-=C'--
_ 2y V
(10.9.15)
dro iozF dro iLzF zF z 00
D 2
Zrro

The current density will be a maximum when

diJdr o = 0 (10.9.16)

and substituting (10.9.16) and (10.9.2) into (10.9.15) yields

(10.9.17)

Substituting (10.9.17) and (10.9.2) into (10.9.14) yields

YJr 2 _ 2y Vr o _ 2yVDCoo _ 2y Vr o = 0 (10.9.18)

o zF io zF

which is quadratic in ro and has the solution (10.9.4):

(c) By balancing the units in (10.9.4), it can be seen that Coo is

in moles per milliliter and y in joules per square centimeter and V in
milliliters per mole. Substituting the given values for zinc yields

ro = 3.7 x 10-5 cm

(d) Substituting the value of ro and other given values into (10.9.17)
gives the current density at the tip of the dendrite:
hiP = 2.1 X 10-1 A cm- 2
This value is seen to be considerably less than the diffusion limiting
150 CHAPTER 10: SOME ELECTRODIC REACTIONS OF INTEREST

current obtained by substituting numerical values into (10.9.2), i.e.,

iL = 2.7 A cm- 2 • Consequently, the growth of dendrites under the
present conditions is only partly diffusion-limited.
On the planar part of the surface, the diffusion-limited current it *
is given by
(10.9.19)

and substituting numerical values in (10.9.19) yields

If the current on the planar part of the surface was activation controlled,
it would be given by (10.9.5),

i = (ioF/RT) I7JA I
which yields, on substituting numerical values,

i = 7.8 X 10-2 A cm- 2

Consequently, on the planar part of the surface, the current is diffusion

controlled with a value of 2.7 x 10-3 A cm- 2 , and

itip/ipl ane = 2.1 x 10-1/2.7 x 10-3

= 78
It should be noted that when the current on the dendrite tip is
purely diffusion controlled, we have, from (10.9.2) and (10.9.19),

itip/ipl ane = 8/ro (10.9.20)

and this ratio can become very large. However, ro is no longer given
by (10.9.4), since in the derivation of (10.9.4), it was assumed that

7JD = iRT/iLZF (10.9.21)

and (10.9.21) does not hold when i = iL .

 CHAPTER 11

SOME ASPECTS OF
ELECTROCHEMICAL
TECHNOLOGY

r:;( What principle of thermodynamics determines whether or not

L...!J a material is stable in contact with a solution? Write explicit
equations to show whether or not a metal in contact with a solution is
thermodynamically stable if (a) hydrogen ions are available to partici-
pate in a cathodic reaction; (b) oxygen is available to participate in a
cathodic reaction. According to these criteria, determine whether iron
is stable in aqueous solution of pH = 3, and whether tin is stable in
aqueous solution of pH = 7 in contact with air. What essential
consideration has been omitted from this treatment of metal stability?

E;e2+jFe = -0.44 V: E~n2+jsn = -0.136 V; E~+.02jH20 = 1.23 V.

~ Answer the following elementary questions concerning electro-

L2J chemical energy producers:
(a) What is the sign of the free energy change for overall chemical
reaction in a working fuel cell ?
(b) Write an equation which represents the ideal, maximum
efficiency of a fuel cell in terms of the heat of reaction and the cell
potential.
(c) In which range do practical fuel cell efficiencies lie? In what
range do the efficiencies of internal combustion engines lie?
(d) What is the main negative feature of the fuel cell-electric
motor combination as a source of mechanical power?
151
152 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

rJl The corrosion potential of iron in a certain solution is

L.:J -0.21 V RHE. Assuming that hydrogen enters microvoids in
the metal which are approximately spherical and of diameter 10-5 cm,
determine whether or not the stated conditions will lead the embrittle-
ment of the iron. Take Young's modulus of iron as 1.2 x 1012 dynes cm- 2
and its surface energy as 1000 ergs cm- 2 •

r:l The emission of unsaturated hydrocarbons from automobile

~ exhausts has been recognized as the principal cause of smog
since the later 1960's.

(a) What oil- (or gasoline-) burning systems are alternatives to the
internal combustion engine as power sources for automobiles?
(b) Apart from the low level of research on batteries, what is the
principal obstacle to the development of a battery-powered automobile?
(c) What other electrochemical power sources can be considered
possible replacements for the automobile internal combustion engine?
(d) What is the greenhouse effect? What long-term environmental
damage is expected to result from the emissions of the automobile
internal combustion engine and other fossil-fuel-consuming power
sources?

'5l Increasing energy consumption, limited reserves of fossil fuels,

L.:.J and the greenhouse effect will result in the increasing use of
nuclear energy. Nuclear generators are only efficient when sufficiently
large and heat dissipation necessitates that they be situated on the
shore or on platforms in the ocean. Thus, the site of power consumption
will frequently be several hundred miles from the nuclear generator.
In a concept known as the "hydrogen economy," the electricity
will be used to electrolyse water and the hydrogen produced then
pushed through pipes to homes and plants where, in some applications,
it may be burned directly to produce heat, and in others converted to
electricity at on-site fuel cells.
Use the following information to calculate the cost of producing
electrolytic hydrogen in cents/lOOO SCP (SCP = 1 ft3 at 1 atm and
60 0 P). Neglect in your calculation revenue from sale of oxygen
 PROBLEMS 153

~
I-
ID 70
'"Q
"-
~
l-
(/)
0
u
z
0
1i5
(/)

~
en
Z
<[
a::
I-
>-
(!)
a::
w
z
w

DISTANCE FROM POWER STATION (mi.)

Fig. 11 .5 .1. Cost of energy transmission plotted

against distance from the power station for transmission
as electricity and as hydrogen gas.

produced and sale of heavy water produced. * The information below

is based on an Allis-Chalmers basic electrolysis cell producing
40 x 106 SCF of hydrogen per day (Gregory ef al., 1971).

Cell voltage 1.78 Vat 800 A/ft2 at 250°F

Electrical energy O.25¢/kWhr at reactor
Maintenance and operation 4.2¢/1000 SCF hydrogen
Labor and overhead on labor O.8¢/1000 SCF hydrogen
Desalinated feed water O.4¢/ IOOO SCF hydrogen
Depreciation, local tax, and Insurance 9 % on capital investment of $11,800,000

Use Fig. 11.5.1 to calculate the cost of supplying heat energy to

a consumer 300 miles from the nuclear generator in the form of
hydrogen and in the form of electricity transmitted at 345 kV. The

* Heavy water production is estimated at O.0041bjl OOO SCF of hydrogen . The

1972 market value of heavy water is $28/lb. The market value of both oxygen and
heavy water could be expected to fall considerably.
154 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

calorific value of hydrogen is 325 Btu/SCF. Local distribution costs are

$1.94/106 Btu for electricity and 17¢/106 Btu for natural gas. For
hydrogen, the local distribution cost will be somewhat higher, but
comparable to that of natural gas.
Compare the cost to this consumer of electricity produced by an
on-site hydrogen-air fuel cell of typical efficiency to that directly
transmitted. Add 25 % to the cost of the fuel-cell-produced power to
cover the amortization of equipment.
How much fresh water per day would be produced by a hydrogen-
air fuel cell providing a household with 50 kWhr per day?

r;r From a survey of recent publications, compare and contrast

~ the pros and cons of the following electrochemical energy
storage devices from the viewpoint of application to automobiles:

(a) Hydrogen-air fuel cell.

(b) Alkali metal-halogen battery.
(c) Alkali metal-sulfur battery.
(d) Zinc-air battery.
(e) Nickel-zinc battery.
(f) AI-Cu cell using propylene carbonate.

f7l Consider the feasibility of running a commuter automobile on

~ solar energy. Assume that, at the earth's surface, the power
density of solar radiation is about 50 mW cm-2 over a period of about
12 h. Estimate the amount of energy per day you could expect to
collect and store in batteries from clean solar cells covering the roof,
trunk, and hood of a compact automobile. The efficiency of good
photocells is 10-12 %.
Experience with compact electric vehicles weighing about 2000 lb
shows that commuter vehicles of moderate speed, less than 40 mph,
require about I kWhr per five miles. Assuming that the solar cell
output can be electrochemically stored with 80 %efficiency, determine
the feasibility of commuting in such a vehicle. Take 30 miles as the
average daily distance traveled by commuter automobiles.
 PROBLEMS 155

The direct electrochemical oxidation of cyanide in alkaline

solution proceeds according to

CN- ---+ CN + e- (l1.8.la)

(l1.8.1b)

(l1.8.lc)

where the reaction of cyanogen with hydroxide ions is a homogeneous

process taking place in the bulk of the solution. It is desired to use this
process to reduce a 50,000 ppm concentration of cyanide in a factory
effluent to 0.5 ppm. From free-energy calculations and other considera-
tions, determine whether it would be practical to carry out the oxidation
of cyanide in the fuel cell mode or whether the cell would have to be
driven. Take the standard free energies of formation of CN-, C2N 2 ,
and CNO- in aqueous solution as 40, 70, and -24 kcal mole-I,
respectively.
Sketch out a fluidized-bed type of electrode system appropriate for
the purification of the effluent, and discuss the optimization of the
operating conditions; current, flow rate, etc. If you conclude that the
cell has to be driven, calculate the cost of electricity for processing
1000 gallons of effluent per day.

r;-J Consider a situation in which a country of 100 x 106 people

~ derives its entire supply of electric power by burning coal of
sulfur content I ;;;. The electric power consumption is 10 kWhr per
person per day.
What is the efficiency of a modern coal-burning power plant and
how many tons of coal, of calorific value 13,000 Btu/lb, would be
required per year? How many tons of S02 would be produced per year
and how many tons of H2S04 could be produced from it? Compare
this yearly per capita production of sulfuric acid with that of a highly
industrialized nation. (The yearly per capita production of sulfuric
acid in the United States is about 300 lb.)
Use the data below to decide whether or not it is possible to carry
out the S02 to H2S04 conversion in the fuel cell mode. Consider the
relative merits of obtaining H2 S04 in this way compared to those of
156 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

producing H2S0 4 from elemental sulfur, the stress being placed on the
environmental and economic aspects:
.dG/(H 2S04 )(aq.) = -177 kcal mole-1
.dG/(H 2S03)(aq.) = -129 kcal mole-1

Q In one proposal for dealing with junked cars, the automobile

~ body is shredded in a hammer mill and the scrap compressed
into disc-shaped billets 30 cm in diameter and 5 cm thick. The billets
are dropped into guides in an electrochemical cell which locate them a
few centimeters from stainless steel cathodes. The metals are recovered
by anodically dissolving the billet and depositing the metals at the
cathodes. Between the anodes and the main cathodes on which iron
is deposited are a series of grids held at intermediate potentials to
collect the heavy metal components of the junked car. The grids are
designed so that they can be replaced periodically and the heavy metals
collected. Zinc would be plated out beyond the main cathodes on an
electrode held at a more negative potential. Iron would be collected
in the form of powder beneath the main cathodes.
Evaluate the technical feasibility of the proposal from the following
considerations:
(a) What electrolyte solution would be the most suitable?
(b) What determines the maximum current density for dissolution
of the billet? Estimate the numerical value of this current density.
(c) At what potentials should the various cathodes be maintained
in order to effect the recovery of all the metallic components of the
billet, except aluminum (see Table 1l.l0.1)?
(d) How can a build-up of ferrous ions in solution due to the
codeposition of hydrogen on some of the cathodes be avoided without
promoting the reaction
(11.10.1)
on auxiliary anodes?
(e) Make rough estimates of the amount of electricity required to
process one car.
(f) How long would it take, at the maximum current density,
to process one car? Estimate roughly the size of the cell needed to
process 1.6 x 104 junked cars per year.
 PROBLEMS 157

TABLE 11.10.1. Breakdown of Metallic

Components of a Typical Car

Metal Amount,Ib

Iron 3000
Zinc 54
Aluminum 51
Copper 32
Lead 20
Nickel 5
Chromium 5

Discuss factors which would affect your choice of direct energy

conversion systems for the following situations:
(a) Powering a torpedo.
(b) Supplying a village in a prImitIve community with electric
power (1 kWhr per person per day).
(c) Powering a car.
(d) Powering a truck.
(e) Powering a large boat.
(f) Providing auxiliary power in space vehicles.
(g) Powering a city.

In each case, give:

(a) A qualitative discussion.
(b) Quantitative calculations in support of your choice taking into
consideration economy, power efficiency, and power per unit weight.
For example, economy is not important for (a) or (f), but is important
for all the others.
(c) A discussion of the effects on the environment, including the
greenhouse effect, in the years after 2000 A.D.

A relation exists between the cost of power produced at an

atomic reactor power station and its size. Investigate this relation
158 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

in the literature. How big would the average plant have to be to provide
electricity at half the present cost? Suppose the entire supply of U. S.
electricity were atomic, at one-fifth the present cost in present dollars,
what would be the number of plants which would have to be built and
approximately what would the total cost of construction be in present
dollars? How long would it take to amortize this investment if the
electricity cost is reduced to one-fifth its present cost?
If there were, in this way, a large excess of cheap electricity, to
what profitable use could it be put? Consider environmental improve-
ments that could be brought about by that use.

ANSWERS

When a metal M is placed in contact with a solution, it will be

unstable so long as the process

(11.1.1)

where the electrons are retained by the metal, proceeds spontaneously.

In the absence of species in the solution capable of accepting the
electrons from the metal, (11.1.1) will soon cease as the potential
difference built up between the metal and the solution opposes the
further transfer of metal ions into the solution. However, if species are
available in the solution to accept the electrons, for example, if one
of the processes

H++e--+tH 2 (11.1.2)
H+ + t02 + e- -+ tH 0 2 (11.1.3)

or some general cathodic reaction

(11.1.4)

can occur, it will depolarize the metal and the reaction (11.1.1) can
continue to take place. Reactions (11.1.2) and (11.1.3) are the usual
 ANSWERS 159

depolarizing reactions. Hence, the metal will be unstable if one of the

overall processes

(11.1.5)

01.1.6)

proceeds spontaneously, i.e., if it proceeds with a negative change of

free energy.
Since reactions (l1.1.5) and (11.1. 6) are the overall cell reactions
for the cells

M i MZ+(aq.) : H+(aq.) I H2 (11.1.7)

M I Mz+(aq.) • H+(aq.)02(g) I H 20 01.1.8)

the free-energy changes can be computed from the cell potentials as

-
LlG
zF =
[0 RT
E H +/H2 + 2F In
(CH +)2]
PH 2 -
[0
EM'+/M +
RT
zF In CM'+
]

(11.1.9)

- ~~ = [E~+,02/H20 + :} In(C H +)4 Po.] - [E~H/M + ~: In C MH]

(11.l.l0)
where the activity coefficients have been taken as one.
Consider, now, iron in contact with a solution of pH = 3. Initially,
the concentration of ferrous ions in the solution will be extremely small
and the second term in brackets on the right-hand side of (11.1.9)
will be large and negative. Consequently, LlG will be large and negative
and iron dissolution will proceed spontaneously until a finite concentra-
tion of ferrous ions is built up in the solution near the electrode. In
practice, this concentration can be taken as 10-6 M. By a similar
argument, the partial pressure of hydrogen PH 2 ,can be taken as 10-6
atm. Substituting numerical values into (11.1.9) yields

-LiG/zF = [0.0 + 0.0] - [-0.44 - 0.18]

= 0.62 V

Consequently, LlG is negative and the process proceeds spontaneously.

Iron is thermodynamically unstable in this solution.
160 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

Substituting the numerical values for tin in a solution of pH = 7

into (11.1.9) yields

-iJG/zF = [0.0 - 0.240] - [-0.136 - 0.18]

= +0.076 V
Consequently, tin is thermodynamically unstable in this solution due to
hydrogen depolarization, although the driving force, -3.5 kcal, is very
small. Substituting numerical values into (11.1.10) and taking Po as
0.20 yields
•

-iJG/zF = [1.23 - 0.43] - [-0.136 - 0.18]

= 1.11 V
This shows that tin is thermodynamically unstable in this solution due
to oxygen depolarization, (11.1.6), as well.
The essential consideration neglected in this treatment is the
consideration of the rates of the processes. Thus, tin is found to not
corrode perceptively in solution of pH = 7 in contact with air; the
depolarization reactions 01.1.5) and (11.1.6) are extremely slow on tin.

The corrosion potential of iron is a mixed potential composed

of iron dissolution,

Fe ~ Fe2+ + 2e- (11.3.1)

and hydrogen evolution,

01.3.2)

each proceeding at the corrosion potential. For hydrogen evolution,

the corrosion potential on the RHE scale, Ecor is the overpotential 'Y}H
of the process,
Ecor = 'Y}H (11.3.3)

The pressure at which hydrogen is formed on the surface of the

electrode may be conceived to be that which thermodynamically causes
a decrease in the reversible potential by the amount 'Y}H • This pressure
can be calculated from the Nernst equation:
'Y}H = -(RT/2F) In PH. (11.3.4)
 ANSWERS 161

assuming that the h.e.r. takes place by means of proton discharge

followed by combination of H atoms.
Rearranging (11.3.4), we have

(11.3.5)

and substituting numerical values yields

PH. = exp(2 x 0.21 X 96,500/8.314 X 298)

= 101 atm

Equilibrium is established when the fugacity of hydrogen on the

surface of the metal and the fugacity of hydrogen in microvoids inside
the metal become equal.
It has been shown that cracks will develop at a microvoid if the
pressure inside the microvoid exceeds a certain critical pressure Pcrit
(Beck et al., 1966). Further,

Pcrit = (16yY/3d)1/2 (11.3.6)

where y is the surface energy of the metal, Y is Young's modulus, and d

is the diameter of the microvoid. Substituting the given values for iron,
we have
Pcrit = (16 X 103 X 1.2 X 1012/3 X 10-5)1/2
= 2.5 X 1010 dynes cm- 2
= 2.5 X 104 atm

Consequently, when the fugacities of the hydrogen on the surface

and in the microvoids are equal, the pressure of hydrogen in the micro-
voids would be considerably in excess of Perit and the stated conditions
will lead to the propagation of cracks and the embrittlement of the iron.

~ To determine the cost of hydrogen production, it is first

~ necessary to calculate the number of moles of hydrogen in
1 SCF and the amount of electrical power required to produce I SCF
of hydrogen:
I SCF = 273 X 28.3/288.4 liters (NTP)
= 273 x 28.3/288.4 X 22.4 moles
162 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

The evolution of I SCF of hydrogen therefore requires:

273 x 28.3 x 2 x 96,500/288.4 x 22.4 = 23.2 x 104 coulombs

Since the cell voltage is 1.78 V, the electrical energy required is

23.2 X 104 x 1.78/1000 x 3600 = 0.115 kWhr/SCF

and the cost of electrical power to produce 1000 SCF of hydrogen is

0.115 X 10 3 x 0.25 = 28.7¢/1000 SCF

The total cost of producing hydrogen can be arrived at by summing

the cost of electrical power, the cost of services, and the depreciation.
The latter, per 1000 SCF of hydrogen, is given (see Table 11.10.1) by
11.8 x 106 x 100 x 0.09/40 X 103 x 365 = 7.3¢/1000 SCF
Therefore, the total cost of hydrogen production is

28.7 + 7.3 + 4.2 + 0.8 + 0.4 = 41.4¢/1000 SCF

= 41.4/0.325¢/106 Btu
= 127¢/106 Btu

The total cost of supplying heat energy to a consumer in the form of

hydrogen is the sum of the production, transmission (see Fig. 11.5.1),
and local distribution costs. Taking the latter at 20¢/106 Btu, we have
that the total cost is
127 + 15 + 20 = 162¢/106 Btu
For heat energy in the form of electricity, w~ have that 106 Btu
require
0.293 x 106 ;1000 kWhr

and are produced at a cost, at the reactor, of

0.293 X 106 x 0.25/1000 = 73.3¢/106 Btu
The total cost of heat energy to a consumer 300 miles from the
reactor is obtained by summing the cost of electrical power at the
reactor, cost of transmission at 345 kV (Fig. 11.5.1), and the cost of
local distribution:
73.3 + 57 + 194 = 324¢/l06 Btu
 ANSWERS 163

Consequently, for the supply of heat energy, hydrogen enjoys a 2:1 cost
advantage over directly transmitted electrical energy.
For the supply of electrical energy, the cost of directly transmitted
electric power will be the same at 324¢/106 Btu. The cost of producing
electrical energy by burning hydrogen in a hydrogen-air fuel cell will
include the cost of hydrogen at the site, the efficiency of the fuel cell,
and the depreciation of the fuel cell. Since hydrogen-air fuel cells are
typically 75 %efficient, we have the cost of electrical energy per 106 Btu
as
162 x 1.25/0.75 = 270¢/106 Btu

where the factor of 1.25 covers the amortization of the fuel cell.
Thus, electrical energy produced from hydrogen enjoys a 1.2:1
cost advantage over directly transmitted electrical energy for a consumer
300 miles from the reactor, and this advantage will increase with
distance.
The pure water produced by an on-site hydrogen-air fuel cell will
contribute significantly to the total fresh-water requirements of the
average household as shown by the following calculation. Burning one
mole of hydrogen produces one mole of water and 2 x 96,500 coulombs.
Since the fuel cell is 75 %efficient, the cell voltage will be 0.75 X 1.23 V.
Therefore, burning one mole of hydrogen produces

0.75 x 1.23 x 2 x 96,500/1000 x 3600 = 0.049 kWhr/mole

Hence, the production of 50 kWhr produces 50/0.049 moles of water
as a by-product, i.e., 18.2 liters per day.

Gl The overall length and width of a typical compact automobile

~ are about 410 and 160 cm, respectively. Assuming that skillful
design permits all of this area to be covered by solar cells, a total
collection area of 6.56 x 104 cm 2 could be anticipated.
The total energy collected over a 12 h period of sunshine of average
power densi ty, 50 m W cm -2, using photocells of 12 %efficiency would be

6.56 X 104 x 0.050 x 12 x 0.12/1000 = 4.73 kWhr

Taking into account the efficiency of the storage batteries, the total
energy available to drive the vehicle in a 24 hr period would be

4.73 x 0.80 = 3.78 kWhr

164 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

Consequently the vehicle could be expected to have a daily range of

about 19 miles. While this range falls somewhat short of the average
daily distance traveled by commuter automobiles, it is apparent that
such vehicles could be of considerable value in many smog-endangered
communities. At the present time, however, the cost of solar cells is
prohibitively high owing to lack of research and development in the
field.

~ The efficiency of a modern coal-burning power plant is about

~ 34 %. Since 1 kWhr is equivalent to 3410 Btu, the total electrical
power requirement of the country of 100 x 106 people each consuming
3650 kWhr per year is
100 x 106 x 3650 x 3410 Btu/year

Since the efficiency of the power plant is 34 %, the amount of coal

required by the country is
100 x 106 x 3650 x 3410/13,000 x 2000 x 0.34 = 1.41 x 108 tons/year

Therefore, the amount of sulfur burned in the process of electrical

power generation is 1.41 x 106 tons/year, which produces
1.41 x 106 x 64/32 = 2.82 x 106 tons S02/year
and the amount of sulfuric acid that could be produced as a by-product
of electrical power generation is:
l.41 x 106 x 98/32 = 4.32 x 106 tons H 2S04 /year

This production rate of sulfuric acid corresponds to a yearly per capita

production of 86 Ib, or about 29 % of the requirements of a highly
industrialized country.
For the oxidation of sulfur dioxide (H 2SOa in aqueous solution)
to be carried out in the fuel cell mode, it must be coupled with a suitable
cathodic reaction which takes place at the other electrode. Oxygen
reduction is clearly the best choice in this case, as it leads to the forma-
tion of no undesirable by-products. Further, the use of oxygen in the
air means there are no additional fuel costs, and the high oxygen rever-
sible potential 0.23 V, RHE) results in a high cell voltage and power
density. The oxygen electrode, however, does suffer from the drawback
that its exchange current density is low and considerable polarization
loss usually occurs at this cathode.
 ANSWERS 165

The oxidation of H 2SOa at the anode proceeds according to

(11.9.1)

while the reduction of oxygen at the cathode proceeds according to

(11.9.2)

By summing (11.9.1) and (11.9.2), we obtain the overall cell reaction

(11.9.3)

Since oxygen is in its standard state [L1C,o(02) = 0], the standard

free-energy change of (11.9.3) is given by

-177 - (-129) = -48 kcal mole-1

where the negative sign indicates that the process proceeds spon-
taneously with a cell reversible voltage given by

4800 X 4.18/2 X 96,500 = 1.04 V

Further, since hydrogen ions do not appear in Eq. (11.9.3),

it means that the overall cell voltage is independent of pH; the potentials
of both the anode and the cathode vary with pH in the same way.
The calculation assumes the partial pressure of oxygen is one and the
concentrations of H 2S04 and H2SO a are equal.
The cell voltage of 1.04 V is the reversible cell voltage and is not
attainable under practical conditions, due to polarization at each
electrode and resistance losses inside the cell. The polarization loss at
an electrode depends on the current density and electrocatalytic activity
of the electrode material for the reaction taking place. Polarization at
the sulfurous acid electrode, under practical conditions, is an unknown
quantity, but it is probably the same magnitude as that at a working
oxygen electrode. This being the case, we could anticipate a working
cell voltage of about 500 mY. The generation of electrical power, as
well as the sale of the sulfuric acid produced, would seem to make the
removal of S02 from coal-burning power-plant flue gases economically
attractive, for it would produce electricity as a by-product.
Sulfur dioxide in the atmosphere is known to constitute a con-
siderable health hazard and has been shown to promote respiratory
ailments in man. Its effect on the environment is hardly less alarming.
166 CHAPTER 11: ELECTROCHEMICAL TECHNOLOGY

Sulfur dioxide stunts and blights natural vegetation and is responsible

for extensive property damage by hastening the corrosion of metals,
the dissolution of marble, and the weathering of protective and
decorative coatings. Of course, not all the sulfur dioxide in the atmos-
phere comes from coal-burning power plants, but it is difficult to
understand why sulfur dioxide is not scrubbed from flue gases and
converted into commercially valuable sulfuric acid.
 CHAPTER 12

MISCELLANEOUS PROBLEMS

r:-l The symmetry factor f3 is a central one in electrode kinetics.

L..!J Deduce its relation for a general multistep reaction to the
transfer coefficient ex and to the stoichiometric number v.
Draw diagrams to illustrate the circumstances under which you
would expect to attain "barrierless discharge," i.e., a rate at which
there is no longer any effect of overpotential on the reaction rate.

Gl Benzoic acid (which can be considered relatively undissociated,

~ and hence, for the purposes of the problem, a nonelectrolyte)
is in an aqueous solution at saturation. The solution is now made
1 Min NaCI. On the assumption that the appropriate model is that of
ions solvated by water molecules and that water molecules so restricted
cannot aid the solution of benzoic acid, calculate the change of solu-
bility of the nonelectrolyte.
The NaCI is replaced by N(Pr)4CI. What qualitative change would
be expected in the solubility of the benzoic acid and why?

r;-] In ellipsometry, it is assumed that the refractive index of the

L.2J metal is independent of the potential of the electrode. On this
basis, changes in optical properties of the surface are regarded as
occurring entirely due to the formation of films thereon. However, it
has been suggested that when the electrode potential changes, it is
tantamount to a change of electron concentration in the surface of
the electrode and therefore to a change in the refractive index.
Review the recent literature on this topic and on this basis estimate
167
168 CHAPTER 12: MISCELLANEOUS PROBLEMS

numerically the extent to which electromodulation at the surface

interferes with the ellipsometric determination of adsorption.

I:l To pump blood around, the body needs a power source of

~ about 20 W. Carry out calculations to see whether an internal
fuel-ceIl-powered electrical pump could be a replacement for the human
heart, the fuel being provided by the body. What would be the probable
anodic and cathodic reactions? What would be the orders of magnitude
of the electrode areas involved? (Calculate this from reasonably
assumed values of exchange current, etc.) Where could the electrodes
be placed in the body? What competing reactions may interfere and
could they be electrochemically controlled? What would be the effects
of local pH changes arising from the reduction of O 2 in the blood
(neutral pH) and from the oxidation of some substances?

Consider the following well-known reactions:

S02 + ~02 ---+ S03 (12.5.la)

S03 + H 0 ---+ H 2S04
2 (12.5.lb)
NO + i02 ---+ N0 2 (12.5.2a)
2N0 2 + l02 ---+ N 20 5 (12.5.2b)
N 0 + H 0 ---+ 2HN03
2 5 2 (12.5.2c)
N2 + 3H -+ 2NH3
2 (12.5.3)
N2 + O ---+ 2NO
2 (12.5.4a)
2NO + 4H2 ---+ N2H4 + 2H 0 2 (12.5.4b)

Suggest fuel cell possibilities for each of these reactions by writing

appropriate anodic and cathodic reactions. Estimate standard free-
energy changes for the overall processes occurring in appropriate media
at 25°C. For processes that are thermodynamically unfavorable at
this temperature, repeat the estimations for higher temperatures within
ranges which you think are practical.
Use solubility and diffusion coefficient data to estimate limiting
current densities at each electrode and, by analogy to similar processes,
estimate the activation overpotential at the limiting current density.
 PROBLEMS 169

By examination of the literature, estimate a common multiplying factor

for the current density in going from planar to porous electrodes. In
this way, very roughly estimate the power density of actual fuel cells
based on the above reactions. Also calculate the rate of product
formation per square meter of electrode and discuss the practicality of
any spontaneous electrolytic syntheses which seem to be suggested
by your calculations.

One mechanism for the electrocatalytic reduction of O 2 is said

to involve the rate-determining step

O2 + H+ + e- ---+ OHads + Oads (12.6.1)

Consider from a quantum-mechanical viewpoint the factors which
affect the velocity of this reaction at an electrode-solution interface.
By what criterion can one know if the process is likely to be adiabatic
or nonadiabatic? Draw qualitative potential energy profile diagrams
for the reaction pathway orthogonal to the electrode surface. How will
the reaction rate depend on the Fermi level in the metal? On the bonding
to the substrate? What will be the state of O 2 binding to the electrode?

r::l The analysis of transient data, after the double-layer charging

L2J is complete, is used to determine the concentration of radicals
on the surface and the rate constants of non-rate-determining steps.
The variation of the observed current or potential generally lasts
in the millisecond region. However, for the oxidation of a number of
organic compounds, the time variation of the current of a potentiostatic
transient lasts up to tens of minutes. The interpretation of the early
part of the transient has been made in terms of diffusion to the surface.
Discuss the time dependence of the oxidation of unsaturated hydro-
carbons from this viewpoint assuming that the discharge of a water
molecule is the rate-determining step and that the coverage by organic
species is high. Show that at constant potential
d[ln( -di/dt)v1ldt = -D/K8 (12.7.1)
where D is the diffusion coefficient, 8 the diffusion layer thickness, and K
the distribution constant of the organic between the electrode and the
solution. How would you obtain K and thus examine the applicability
of this equation?
170 CHAPTER 12: MISCELLANEOUS PROBLEMS

r:l In the early 1970's, several types of electrochemical power

L!J sources seem to hold promise of developing into nonpolluting
and non-C0 2 emitting power sources suitable for automobiles. They
may be broadly classified as follows:
(a) High energy-density power storers.
(b) Fuel cells burning nonhydrocarbon synthetic fuels, e.g.,
hydrazine.
(c) Fuel cells burning hydrogen.
Of course, there are also some negative aspects associated with each
type of power source. Review these possibilities as quantitatively as
possible in the light of the latest information available in the literature
and decide which seems to be the optimal system to concentrate
developmental work on at the present time.

~ A problem of antipollution technology is the removal of

L2.J foreign matter from lakes. It has been suggested that much of
this material is colloidal and it may be possible to flocculate it electrically
(Kuhn, 1971) by evolving oxygen bubbles below it, and floating the
coagulated solids to the surface for burning, sweeping away, etc.
Perform appropriate calculations to examine this concept. What
concentration of colloidal matter is typical of polluted lakes such as
Michigan? What area and type of electrode structure could be envisaged
for the achievement of Kuhn's concept? What sort of field and current
would be needed? Could significant quantities of a Great Lake be
revived in this fashion in ten years? What order of magnitude of
expenditure would be involved in capital investment and in operating
costs?

f10l What is the Esin-Markov effect and how is it related to contact

~ adsorption? Show that the isotherm derived from the BDM
theory of the double layer is quantitatively consistent with the Esin-
Markov effect for smaller contact-adsorbed ions, e.g., CI-, Br-. Why
do inconsistencies appear for larger anions?

A solution consists of 0.1 M Cu++ and 0.3 M Ni++. It is desired

to recover 99.9 % of the copper in its most desirable form, i.e.,
 PROBLEMS 171

as bars or strips of metal. Carry out the necessary calculations to act

as a basis for an engineering design of a continuous copper extractor.
Consider the transport properties and the deposition potentials of the
ions in this solution and estimate the exchange current densities for
the deposition of each. Decide on the potential, type of electrode, flow
rate, and type of flow (laminar or turbulent) of the solution that would
lead to optimal recovery of pure copper. The design of the electrode
should be compatible with minimal cell volume.
Derive an equation for the fraction of copper removed in one pass
of the solution through the extractor and hence for the extraction time
as a function of flow rate. Calculations ofthe optimal electrode potential
and flow rate should be performed and the cell voltage and efficiency
estimated. What other considerations are important?

f121 Review the physical and chemical properties of the sodium

L.::J tungsten bronzes and the recent literature pertaining to their
use as electrodes. Discuss these aspects of bronze electrodes:
(a) Evidence which may lead to the sodium tungsten bronzes
being considered as semiconductor electrodes.
(b) The promotion by Pt of the catalytic activity of the bronzes for
oxygen reduction. Is it the special properties of traces of Pt on the
surface of the bronzes or the properties of bronzes doped by Pt which
is important?
(c) The value of studies on the bronzes to the future development
of oxygen electrodes.
(d) The difficulties associated with the preparation of porous
electrodes from the bronzes and how such difficulties might be overcome.
(e) The economic aspects of sodium tungsten bronze electrodes
containing a few hundred ppm of platinum for oxygen reduction and
evolution.
 APPENDIX 1

VALVES OF SOME IMPORTANT

PHYSICAL CONSTANTS

Avogadro's number, NA = 6.023 X 1023 mole-1

Boltzmann constant, k = l.381 X 10-16 ergs deg-1
Electron charge, eo = 4.803 x 10-10 esu
Electron rest mass, n1, = 0.101 X 10- 28 g
Faraday constant, F = 96,490 coulombs mole-1
Gas constant, R = 8.314 J deg- 1 mole- 1
= 1.987 cal deg- 1 mole- 1
Mechanical equivalent of heat, J = 4.184 J cal-1
Planck constant, h = 6.626 X 10- 27 erg sec
Triple point of water, To = 273.16°K
Velocity of light, c = 2.998 X 1010 cm sec- 1
Volume of perfect gas (NTP) Vo = 22.414 liters mole- 1

173
 APPENDIX 2

SOME USEFUL CONVERSION

FACTORS

1 ft3 = 28.32 liters

1 atm = 1.013 X 106 dynes cm- 2
1 Btu = 1054 J
= 0.2929 Whr
1 kWhr = 36 x 105 J
1 eV = 23.06 kcal mole-1
I coulomb = 3 x 109 esu (statcoulombs)
1A = 3 x 109 statamperes
1V = 0.333 X 10- 2 statvolt
I ohm = 0.111 x 10-11 statohm
1F = 9 X 1011 em
1 esu em- 1 = 1 statvolt

174
 BIBLIOGRAPHY
Beck, W., J. O'M. Bockris, J. McBreen, and L. Nanis, Proc. Roy. Soc.,
A290:220 (1966).
Bockris, J. O'M., The Electrochemistry of Cleaner Environments,
Plenum Press, New York (1971).
Bockris, J. O'M., and D. Drazic, Electrochemical Science, Taylor and
Francis, London (1972).
Bockris, J. O'M., and Z. Nagy, Ecological Electrochemistry (1972)
[projected date of publication, 1973].
Bockris, J. O'M., and A. K. N. Reddy, Modern Electrochemistry,
Plenum Press, New York (1970).
Bockris, J. O'M., and P. P. S. Saluja, J. Phys. Chem., 76 (15):2140
( 1972).
Bockris, J. O'M., and S. Srinivasan, Fuel Cells-Their Electrochemistry,
McGraw-Hill, New York (1969).
Bockris, J. O'M., N. Bonciocat, and F. Gutmann, A Primer in Electro-
chemical Science, Wykeham Press, London (1972) [pubI.1973].
Booth, F., J. Chem. Phys., 19:391 (1951).
Brummer, S. B., J. I. Ford, and M. J. Turner, J. Phys. Chem., 69:3424
(1965).
Conway, B. E., Electrochemical Data, Elsevier, New York (1952).
Conway, B. E., Proc. Roy. Soc., A247:400 (1958).
Conway, B. E., in Modern Aspects of Electrochemistry, Vol. 3, Ed.
by J. O'M. Bockris and B. E. Conway, Plenum Press, New York
(1964).
Conway, B. E., Electrode Processes, Ronald Press Co., New York
(1965).
Conway, B. E., J. O'M. Bockris, and H. Linton, J. Chem. Phys. 24:834
(1956).
Criss, C. M., and J. W. Cobble, J. Amer. Chem. Soc., 83:3223 (1961).
175
176 BmLIOGRAPHY

Delahay, P., J. Amer. Chem. Soc., 75:1190 (1953).

Delahay, P., New Instrumental Methods in Electrochemistry, Inter-
science, New York (1954).
Delahay, P., The Double Layer and Electrochemical Kinetics, Inter-
science, New York (1965).
Eigen, M., and L. deMaeyer, Proc. Roy Soc. (London), A274:505
(1958).
Emi, T., and J. O'M. Bockris, J. Phys. Chem., 74:159 (1970).
Eyring, H., D. Henderson, and W. Jost, Physical Chemistry-An
Advanced Treatise, Vols. 9A and 9B, Academic Press, New York
(1970).
Falkenhagen, H., and W. Ebeling, in Ionic Interactions, Ed. by S.
Petrucci, Academic Press, New York (1971).
Frank, H. S., in Chemical Physics of Ionic Solutions, Ed. by B. E.
Conway and R. G. Barradas, John Wiley and Sons, New York
(1966).
Frank, H. S., and P. T. Thompson, in The Structure of Electrolyte
Solutions, Ed. by W. Hamer, John Wiley and Sons, New York
(1960).
Fricke, H., Phys. Rev., 24:575 (1924).
Friedman, H. L., in Modern Aspects of Electrochemistry, Vol. 6,
Ed. by J. O'M. Bockris and B. E. Conway, Plenum Press, New
York (1971).
Frumkin, A. N., Z. Phys., 35:792 (1926).
Frumkin, A. N., and B. B. Damaskin, in Modern Aspects of Electro-
chemistry, Vol. 3, Ed. by J. O'M. Bockris and B. E. Conway,
Plenum Press, New York (1964).
Fuoss, R. M., and F. Accasina, Electrolytic Conductance, Interscience,
New York (1959).
Gileadi, E., G. Stoner, and J. O'M. Bockris, J. Electrochem. Soc.,
113:585 (1966).
Goffredi, M., and T. Shedlovsky, J. Phys. Chem., 71:2176 (1967).
Gregory, D. P., p. Y. L. Ng, and G. M. Long, in The Electrochemistry
of Cleaner Environments, Ed. by J. O'M. Bockris, Plenum Press,
New York (1971).
Halliwell, H. F., and S. C. Nyburg, Trans. Faraday Soc., 58:1126
(1963).
Harned, H. S., and B. B. Owen, The Physical Chemistry of Electrolyte
Solutions, Reinhold, New York (1958).
Hemmes, P., and S. Petrucci, J. Phys. Chem., 72:3986 (1968).
 BffiLIOGRAPHY 177

Henderson, P., Z. Physik. Chern., 59:118 (1907).

Henderson, P., Z. Physik. Chern., 63:325 (1908).
Horiuti, J., and M. Polanyi, Acta Physicochern. URSS, 2:505 (1935).
Kovac, Z., Dissertation (University of Pennsylvania) (1964).
Kuhn, A. T., H. Wroblowa, and J. O'M. Bockris, Trans. Faraday Soc.,
63: 1458 (1967).
Latimer, W. M., Oxidation Potentials, Prentice-Hall, Englewood
Cliffs, New Jersey (1952).
Levich, V. G., in Physical Chemistry-An Advanced Treatise, Vol. 9B,
Ed. by H. Eyring, D. Henderson, and W. Jost, Academic Press,
New York (1970).
Marcus, R. A., J. Phys. Chem., 67:853, 2889 (1963).
Nanis, L., and J. O'M. Bockris, J. Phys. Chern., 67:2865 (1963).
Nagarajan, M. K., and J. O'M. Bockris, J. Phys. Chern., 70:1854
(1966).
Nightingale, Jr., E. R., J. Phys. Chern., 63:1381 (1959).
O'Ferrall, R. A. More, G. W. Koeppl, and A. J. Kresge, J. Arner.
Chem. Soc., 93:1 (1971).
Parsons, R., Trans. Faraday Soc., 147:1332 (1951).
Parsons, R., Handbook of Electrochemical Constants, Butterworths,
London (1959).
Passynski, A., Acta Physicochem. URSS, 8:385 (1938).
Planck, M., Ann. Physik., 39:161 (I 890a).
Planck, M., Ann. Physik., 40:561 (I 890b).
Pourbaix, M., Atlas D' Equilibres Electrochimiques, Gauthier-Villars,
Paris (1963).
Pourbaix, M., Lectures on Electrochernical Corrosion, Plenum Press,
New York (1973).
Rampolla, R. W., R. C. Miller, and C. P. Smyth, J. Chern. Phys.,
30:566 (1959).
Sand, H. J. S., Phil. Mag., 1:45 (1900).
Solomons, c., J. H. R. Clarke, and J. O'M. Bockris, J. Phys. Chern.,
49:455 (1968).
Vetter, K., Electrochernical Kinetics, Academic Press, New York (1967).
Zachariasen, W., J. Arner. Chem. Soc., 54:3841 (1932).
Zana, R., and E. Yeager, J. Phys. Chern., 71:521 (1967).