Electrochemical Cells

Introduction

Electrochemical cell either coverts chemical energy into electrical energy or

converts electrical energy into chemical energy. When a cell converts chemical energy
into electrical energy, it is called a Galvanic cell and in the reverse operation, it is called
an electrolytic cell. Galvanic cell acts as a power source. Many times a single cell can act
as both the Galvanic as well as the electrolytic cell e.g. car battery.

A redox reaction occurs in an electrochemical cell. Each of the two half reactions
occurs at a different electrode. Electrochemical cell therefore consists essentially of a
cathode and anode placed in a solution containing one or more redox species. To sustain
the reaction, electron transferred to the anode during the oxidation must travel to the
cathode where it is picked up by the oxidant. Electrons flow from the anode to the
cathode. Electrodes must therefore be conducting bodies. They are usually made of
metals, but materials such as graphite, conducting polymers or semiconductors are also
used. The electrical conductivity of metals varies from 102 to 104 S.cm-1.

A ⎛1⎞A
R= or σ = ⎜ ⎟
Aσ ⎝R⎠A

Electrodes are connected to each other externally using an external load or using a
power source. Electrons flow from the anode to the cathode in the external circuit. Hence
electric current flows from the cathode to the anode in the external circuit. A simple
example of an electrochemical cell is the Daniell cell, shown below. It can be operated
both in the Galvanic and the electrolytic modes.
Galvanic Mode
In this mode, the following reactions occur at the zinc and the copper electrode
respectively
Zn ↔ Zn 2 + + 2e − (anode)
Cu 2 + + 2e − ↔ Cu (cathode)

1
 Zinc electrode acts as anode and copper as cathode. The overall reaction is
Cu 2 + + Zn ↔ Cu + Zn 2 +
This reaction is spontaneous in the forwards direction i.e. deposition of copper and
dissolution of zinc.

Direction of current flow

Cathode
Anode +

Zn Cu

ZnSO4 CuSO4

Anion Exchange
Membrane
In order to prevent mixing between the CuSO4 and ZnSO4, it is necessary to
separate the two half cells by a partition. Thus cell has an anodic (left) and cathodic (right)
compartments. However, this partition cannot be impervious; otherwise Zn2+ ions will
accumulate in the anodic compartment without the balancing SO42- ions and SO42- ions
will be excess in the cathodic compartment. Anodic compartment will be positively
charged and the cathodic compartment will be negatively charged. Positively charged
zinc ions will adsorb on the anode and increase its potential. On the other hand sulfate
ions will adsorb on copper electrode and will reduce its potential. Potential of the anodic
compartment will increase and that of the cathode compartment will decrease. The
potential difference between the two electrodes will rapidly decrease and cell will stop
working. A simple way to avoid this is to use a anion exchange membrane, which will
allow only SO42- ions to move from the right to the left compartment. Because the

2
membrane surface is positively charged, it will exclude the cations from its interior. Thus
both ZnSO4 and CuSO4 solutions will remain electroneutral.
Reversible Operation of the cell
If we connect a voltmeter across the cell, practically no current will flow through
the cell. The cell will attain equilibrium. We then say that the cell is operating reversibly.
Each of the two half reactions will also attain equilibrium. Hence
μ zn ( electrode ) = μZn 2+ (solution ) + 2μe − ( electrode ) (anode)

μCu 2+ (solution ) + 2μe − ( electrode ) = μ Cu ( electrode ) (cathode)

The potentials with overbar represent electrochemical potentials, which may be viewed as
sum of the chemical and the electrical potentials.
We can expand the potential terms as follows
μ Zn ( electrode ) = μ 0 zn (s ) + RT ln a Zn ( s )

μ Cu ( electrode ) = μ 0 Cu ( s ) + RT ln a Cu ( s )

μZn 2+ (solution ) = μ 0Zn 2+ + RT ln a Zn 2+ + z zn 2+ Fφsolution

μCu 2+ (solution ) = μ 0Cu 2+ + RT ln a Cu 2+ + z Cu 2+ Fφsolution

μe − ( electrode ) = μ 0e− + RT ln a e − + z e − Fφelectrode

The following simplifications can be made : We use standard state of metals as their pure
state and since metal remains pure during the reaction we can write a Zn (s ) = a Cn (s ) = 1 .

Further, we assume that electrons do not interact with each other in metals and hence
a e − = 1 . We also note that z zn 2+ = z cu 2+ = 2 and z e − = −1

The equations for the half reactions reduce to

(
μ 0Zn (s ) = μ 0Zn 2+ + RT ln a Zn 2+ + 2Fφanolyte + 2 μ 0e− − Fφanode )
( )
μ 0Cu 2+ + RT ln a Cu 2+ + 2Fφcatholyte + 2 μ 0e− − Fφcathode = μ 0Cu (s )

Rearrangement gives

φanode − φanolyte =
1 0
(
μ 2+ + RT ln a Zn 2+ + 2μ 0e − μ 0Zn (s )
2F Zn
)

3
 φcathode − φcatholyte =
1 0
2F
(
μ Cu 2+ + RT ln a Cu 2+ + 2μ 0e− − μ 0Cu (s ) )
We define the standard free energy of the reduction reactions
Zn 2 + + 2e − ↔ Zn
and Cu 2 + + 2e − ↔ Cu
as
( )
ΔG 0 Zn ↔ Zn 2 + + 2e − = μ 0Zn ( s ) − μ 0Zn 2+ − 2μ 0e

ΔG (Cu ↔ Cu
0 2+
+ 2e ) = μ
− 0
Cu ( s ) − μ 0Cu 2+ − 2μ 0e

We can then write

Δφanode =
1
2F
[ (
− ΔG 0 Zn ↔ Zn 2 + + 2e − + RT ln a Zn 2+ ) ]
Δφcathode =
1
2F
[ (
− ΔG 0 Cu ↔ Cu 2 + + 2e − + RT ln a Cu 2+ ) ]
Δφ represents the potential difference between the electrode and the electrolyte. If we
take the catholyte and the anolyte with unit activities, then the potential differences would
be

Δφ0anode = −
1
2F
(
ΔG 0 Zn 2 + + 2e − ↔ Zn )
Δφ0cathode = −
1
2F
(
ΔG 0 Cu 2+ + 2e − ↔ Cu )
In general
1
Δφ0 = − ΔG 0
nF
Here n is the number of electrons transferred during the reaction per atom of species
undergoing reduction.
It is important to realize that it is not possible to measure ΔG 0 on an electron
transfer reaction except through measurement of potential difference between the
electrode and the solution. However, to do this we have to measure the potential of the
solution. This can only be done by introducing another electrode into the anolyte (or
catholyte) and measuring the potential difference across the anode (or cathode) and this
new electrode. However, no electrode can be considered as an inert electrode. An

4
introduction of an electrode always results in setting up of reaction equilibrium at the
electrode surface. For example, if you introduce zinc as electrode in the cathodic half cell,
then reaction equilibrium of zinc dissolution will be established and the potential
difference obtained will not be correct. Instead, if we introduce copper electrode, then
copper deposition equilibrium will be established and we will get not potential difference
been the new electrode and the cathode. If we introduce any other metal electrode, it will
either dissolve or copper will deposit on it. It is thus impossible to measure the potential
difference between the electrode and the solution.
Reference Electrode
A way out of this problem is to use a reference electrode, whose potential remains
constant. We introduce reference electrode into the solution and measure the potential
difference between the working electrode and the reference electrode. One of the
reference electrodes is the standard hydrogen electrode. The potential of the standard
hydrogen electrode is chosen as zero by a common agreement. Such an arbitrary choice is
permitted since potential is not an absolute quantity and hence one can use any suitable
potential as a datum. The standard hydrogen electrode is schematically shown below

Pt H2
aH2 = 1

HCl
aH+=1

Hydrogen gas is bubbled into aqueous solution of HCl. The reaction occurring at the
standard hydrogen electrode is
2H + + 2e − ⇔ H 2

5
The electrode is chosen is platinum. Platinum is an inert electrode and hence we are sure
that it does not dissolve in the acidic solution. This electrode is referred to either as
standard hydrogen electrode (SHE), or normal hydrogen electrode (NHE). It is also called
reversible hydrogen electrode (RHE).
At the equilibrium
(
μ 0H 2 ( g ) + RT ln a H 2 = 2μ 0H + + 2RT ln a H + + 2Fφsolution + 2 μ 0e− − FφSHE )
The activities of H+ ion and that of H2 are maintained at unit values. The gas is hydrogen
at 1 bar partial pressure (Usually Hydrogen gas at 1 atmosphere is used. It is saturated
with water hence its partial pressure is nearly equal to 1 bar). To achieve the
condition a H + = 1 , a high concentration of the acid is required. Activity coefficient of HCl

as a function of molality is shown below. We can see that at the molality of HCl of 1.192
m, the activity of HCl is unity.
m 0.7 0.8 0.9 1.0 1.2 1.4
γ HCl 25 0 C 0.772 0.783 0.795 0.809 0.840 0.876

At the unit hydrogen ion activity, pH of the solution is zero.

pH = − log10 a H = − log10 (1) = 0
The equation representing the equilibrium reduces to

φSHE − φsolution =
1
2F
( ) 1
(
2μ 0H + + 2μ 0e− − μ H 2 = − ΔG 0 2H + + 2e ↔ H 2
2F
)
The reference electrode must be introduced in the solution. It is, however,
separated from the solution by a membrane (or porous glass partition). Suppose we
introduce SHE in the catholyte and connect it to the cathode through a voltmeter so that
no current flows through the circuit. The reaction equilibrium will soon be established in
which the potential of the solution inside the reference electrode half cell will be the same
as that of the catholyte. Copper being more noble, will act as cathode and reference
electrode as anode. Thus hydrogen will form H+ ions. Some of them have to diffuse into
the anolyte. At the equilibrium,

φSHE
0
− φ0catholyte = −
1
2F
(
ΔG 0 2H + + 2e ↔ H 2 )

6
 φ0cathode − φ0catholyte = −
1
2F
(
ΔG 0 Cu 2 + + 2e − ↔ Cu )
Subtraction gives

φ0cathode − φSHE
0
=−
1
2F
[ (
ΔG 0 Cu 2 + + 2e − ↔ Cu − ΔG 0 2H + + 2e ↔ H 2 ) ( )]
Standard electrode potential
The expression above is called the standard electrode potential for the reaction
Cu 2 + + 2e − ↔ Cu and is denoted by E 0 Cu 2 + + 2e − ↔ Cu ( )
In general

E 0R = −
1
nF
(
ΔG 0R − ΔG SHE
0
)
where ΔG SHE
0 0
is the standard free energy of hydrogen reduction reaction. Since E SHE =0
by convention, we can write

φ0cathode = −
1
2F
[ ( )
ΔG 0 Cu 2+ + 2e − ↔ Cu − ΔG 0 2H + + 2e ↔ H 2 ( )]
E 0 for the reaction
Cu 2 + + 2e − ↔ Cu
is 0.34 V and that for the reaction
Zn 2 + + 2e − ↔ Zn
it is -0.76 V. Now

φ0cathode = −
1
2F
[ ( )
ΔG 0 Cu 2+ + 2e − ↔ Cu − ΔG 0 2H + + 2e ↔ H 2 = 0.34V ( )]

And φ0anode = −
1
2F
[ ( )
ΔG 0 Zn 2+ + 2e − ↔ Zn − ΔG 0 2H + + 2e ↔ H 2 = −0.76V ( )]
We call these differences as standard half cell potentials. From these, we can obtain the
full cell potential as
1
[ ( ) (
ΔG 0 Cu 2 + + 2e − ↔ Cu − ΔG 0 Zn 2 + + 2e ↔ Zn = 0.34 − (−0.76) = 1.1V
φ0cathode − φ0anode = −
2F
)]
It is important to note that this difference is not affected by the choice of the reference
electrode since the corresponding terms cancel. Now
ΔG 0 (Cu 2 + + 2e − ↔ Cu ) − ΔG 0 (Zn 2 + + 2e ↔ Zn ) = ΔG 0 (Cu 2 + + Zn ↔ Zn 2 + + Cu )

7
From this, we get

φ0cathode − φ0anode = −
1
2F
( )
ΔG 0 Cu 2+ + Zn ↔ Zn 2+ + Cu = 1.1V

We call this difference the standard cell potential. In general,

1
φ0cathode − φ0anode = − ΔG 0
nF
where ΔG 0 is the standard free energy of the cell reaction.
ΔG 0 (Cu 2+ + Zn ↔ Zn 2+ + Cu ) = −2x96487 x1.1 = −212.3 kJ.mol −1
The overall reaction is spontaneous since the free energy is negative.
Nernst Equation
Actual potentials of the electrodes are related to the standard potentials by the
following equations
RT
φanode − φ0anode = ln a Zn 2+
nF
RT
φcathode − φ0cathode = ln a Cu 2+
nF
In this case all electrode potentials are referred to SHE. In general,

RT ⎛⎜ a [Oxi ] ⎞⎟
φ − φ0 = ln
nF ⎜⎝ a [Re d ] ⎟⎠

In the present case a [Re d ] represents the activity of the metal and is taken as unity. The

above equation is called the Nernst equation.

The potential difference is given by

RT ⎛⎜ a Cu 2+ ⎞
⎟
Δφ = Δφ0 + ln
nF ⎜⎝ a Zn 2+ ⎟
⎠
The actual potential difference will depend on the concentrations of copper and the zinc
ions in the respective solutions. The potential difference increases as the activity of the
copper ions increases relative to that of zinc ions. Note that
RT 8.314 x 298 0.02568
= = = 0.01284 V (12.84 mV)
nF 2 x 96487 2
Thus to obtain the equilibrium potential difference of 1.2 V, we need

8
 ⎛ a 2+ ⎞ 100 a 2+
ln⎜ Cu ⎟= = 7.79 or Cu = 2416
⎜ a 2+ ⎟ 12.84 a Zn 2+
⎝ Zn ⎠
We see that to obtain a small increase in the potential difference, we need a very large
difference between the activity of zinc and copper ions in the two compartments. This
also means that the equilibrium potential of the cell will not change significantly during
its use. During the use the copper concentration will decrease and zinc concentration will
increase. This will reduce the cell potential Δφ with time, albeit slowly. A significant
reduction will only occur when cupric ions are almost completely depleted.
Thermodynamics of Electrochemical Reactions
How much energy the cell will generate if operated reversibly? As a good
approximation we assume that the cell potential remains constant at 1.1 V. Then
Energy = Δφ0q
Here q is the total charge flowing through the cell. It can be obtained from Faraday’s law
q = 2VF[CuSO 4 ]i
Thus if the cell has a volume of 1 liter and initially contains 1 M copper sulfate, then
q = 2 x1x 96487 x1 = 192974 C
And energy obtained
Energy = 1.1x192974 = 212.3 kJ
We see that under reversible operation the maximum work available is (− ΔG ) times the
number moles of the metal available.
We can obtain different thermodynamic characteristics of the reaction
(Cu 2+
+ Zn ↔ Zn 2+ + Cu ) using the cell potential as follows
ΔG = − nFΔφ
From Gibbs-Helmholtz equation, we can obtain the entropy of the reaction
⎛ ∂ΔG ⎞ ⎛ ∂Δφ ⎞
ΔS = −⎜ ⎟ = nF⎜ ⎟
⎝ ∂T ⎠ p ⎝ ∂T ⎠ p
From the above two quantities, we can derive the expression for the enthalpy of the
reaction as

9
 ⎡ ⎛ ∂Δφ ⎞ ⎤
ΔH = ΔG + TΔS = −nF⎢Δφ − T⎜ ⎟ ⎥
⎢⎣ ⎝ ∂T ⎠ p ⎥⎦

Electric potential in the solution

Now consider the circuit taken anticlockwise, which joins the cathode and the
anode externally by the actual connection and joins them inside the solution by an
imaginary path. By the Kirchoff’s first law of electric circuits, all the potential
differences across the circuit should add to zero. This implies that the potential must rise
from the anode to the cathode and this rise must exactly balance the drop in the external
part of the circuit. No drop can occur in the bulk solution since there is no current flowing
in the circuit. A small potential drop may exist across the anion exchange membrane, it is
due to exclusion of cations. This exclusion is called the Donnan exclusion. However for
sufficiently thin membranes, it is of the order of a few mV. So where does the rest of the
potential drop occur?

− Zn + Cu

Double layers

The figure above shows the electric potential profile in a Daniell cell under no
current condition. In this case, nearly all potential drop occurs across the electric double
layer regions. Zinc electrode, being negatively charged, attracts zinc ions. Hence the
potential increases in the double layer associated with zinc electrode. On the other hand,
copper electrode is positively charged and it will attract sulfate ions around it. As a result,
potential decreases in the double layer surrounding this electrode. In the bulk of the
solution, potential will be uniform.

10
 In general, the potentials of the catholyte and the anolyte are different. The
difference is called the liquid-junction potential. We can compute the liquid junction
potential as follows. Since sulfate ions communicate with each other across the anion
exchange membrane, at equilibrium, the electrochemical potentials of the ions in the two
half cells must be equal and hence
zSO 2− Fφanolyte + RT ln a SO 2− ( anolyte ) = zSO 2− Fφcatholyte + RT ln a SO 2− ( catholyte )
4 4 4 4

RT a SO 24− ( catholyte )
Or φcatholyte − φanolyte = ln
2F a SO 2− ( anolyte )
4

If the concentration of catholyte (copper sulfate) is higher than that of anolyte (zinc
sulfate) the liquid junction potential is positive.
Irreversible operation of the cell
Now consider the case when the voltmeter is replaced by a resistance of finite
magnitude. The actual potential difference across the cell will be less than 1.1 V. The
reason is the several potential drops that occur in the cell.
Δφ = (Δφ)rev + (ηsurface + ηconcentration )c − (ηsurface + ηconcentration )a − ηo

Among the terms to the right, (Δφ)rev is the potential difference, when the cell is

operating under reversible condition ( (Δφ)rev = 1.1V) , ηsurface is the surface overpotential,

ηconcentration is the concentration overpotential and ηo is the Ohmic potential drop across the
cell .
Note that Δφ is also given by
Δφ = IR ext
The internal current must be equal to the external current at steady state since no charge
can accumulate in the circuit. The current in the circuit must adjust itself in such way that
the last two equations are simultaneously satisfied. i.e.
IR ext = (Δφ)rev + (ηsurface + ηconcentration )c − (ηsurface + ηconcentration )a − ηo

From this expression, it is clear that as R ext decreases, the current must rise. However,
there is a limit to this current. This happens because all the overpotentials are functions of
current. Their magnitudes increase with increase in the current. Since these overpotential
terms are positive for the anode and negative for the cathode, they make the negative

11
contribution to the right hand side. As a result the right hand side decreases as the current
increases. A limiting current will be reached when any decrease in the resistance does not
increase the current any further.
Surface Overpotential
The surface overpotential, ηsurface , provides a driving force for the reaction. At the
anode, the electrons are withdrawn from Zn to form the zinc cation. For this reaction to
be driven in positive direction, free energy of electrons in the anode must be lower than
that at the equilibrium. This requires that the potential of the anode be higher than that at
the equilibrium. On the other hand, at the cathode, the electrons are added to cupric ions
to form the copper metal. For this reaction to be driven in positive direction, free energy
of electrons in the cathode must be lower than that at the equilibrium. This requires that
the potential of the cathode be lower than that at the equilibrium. We define the surface
overpotential as
ηsurface = φelectrode − φ electrode,eq

Based on this discussion we see that cathodic overpotential is negative and anodic
overpotential is positive. Since cathodic overpotential is added and anodic overpotential
is subtracted, both these reduce the electrode potential. The surface overpotential is
estimated by the Butler-Volmer equation.
Concentration overpotential
Now consider concentration overpotential. Zinc ions are generated at the anode
and cupric ions are consumed at the cathode. The zinc ions, generated at the interface, are
transported into the bulk solution. This transport occurs by diffusion and requires a
driving force and hence the concentration of the zinc ions just outside the double layer is
greater than that in the bulk. Consequently, the electrochemical potential of the zinc ions
just outside the double layer will be greater than that in the bulk. Since, the
electrochemical potential of zinc ions in the double layer is uniform electrode potential is
higher than at the equilibrium, where the concentration of zinc ions outside the double
layer is the bulk concentration. This phenomenon is called the concentration polarization
and the deviation in the potential from the equilibrium situation is called concentration
overpotential. The concentration potential is positive at the anode and negative at the

12
cathode. In general, polarization is the term used with reference to electrode and means
deviation from equilibrium.
The concentration overpotential can be estimated as follows. We use Nernst film
model for diffusion. We assume a stagnant film of thickness δ at the solid-liquid interface.
Transport of ions occur through this film by the process of diffusion (and by migration, if
electric field exists there). The flux of zinc ions jZn 2+ (mol.m-2.s-1) crossing this film is

jZn 2+ = D Zn 2+
(c *
Zn 2+
− c Zn 2+ )
δ
The current crossing the anode is
c*Zn 2+ − c Zn 2+
I = jZn 2+ z Zn 2+ FA = D Zn 2+ z Zn 2+ FA anode
δ
Here A anode is area of cross section of the anode.
Rearrangement gives
Iδ
c*Zn 2+ = c Zn 2+ + (1)
D Zn 2+ z Zn 2+ FA anode

Now at equilibrium
RT
φeq
anode − φ anode =
0
ln a Zn 2+
nF
Under the condition when the current is being drawn
RT
φanode − φ0anode = ln a *Z2+
nF n

Subtraction yields the concentration overpotential at the anode as

RT ⎛⎜ a Z 2n+ ⎞ RT ⎛ c*Z2+ γ *Zn 2+ ⎞

*

ηconcentration ,a = φanode − φ eq
= ln ⎟= ln⎜ n ⎟ (2)
nF ⎜ a Zn 2+ ⎟ nF ⎜ c 2+ γ 2+ ⎟
anode
⎝ ⎠ ⎝ Zn Zn ⎠
Elimination of c*Z2+ between equations 1 and 2, we obtain the final expression for the
n

concentration overpotential at the anode as

RT ⎡⎛⎜ γ Zn 2+ ⎞⎛ ⎞⎤
*
⎟⎜1 + Iδ ⎟⎥
ηconcentration ,a = ln ⎢
nF ⎢⎣⎜⎝ γ Zn 2+ ⎟⎜ D 2+ z 2+ FA anode c 2+
⎠⎝ Zn Zn Zn
⎟⎥
⎠⎦
The overpotential increases as current density increases, film thickness increases,
diffusion coefficient decreases and the bulk concentration decreases.

13
 In a similar way, we can obtain the expression for the cathodic concentration
overpotential.
The current crossing the cathode is
c Cu 2+ − c*Cu 2+
I = D Cu 2+ z Cu 2+ FA cathode
δ

RT ⎛⎜ a C 2u+ ⎞ RT ⎛ c*C 2+ γ *Cu 2+ ⎞

*

ηconcentration ,c = φcathode − φ eq
= ln ⎟= ln⎜ u ⎟
nF ⎜ a Cu 2+ ⎟ nF ⎜ c 2+ γ 2+ ⎟
cathode
⎝ ⎠ ⎝ Cu Cu ⎠
From these two equations, we obtain

RT ⎡⎛⎜ γ Cu 2+ ⎞⎛ ⎞⎤
*
⎟⎜1 − Iδ ⎟⎥
ηconcentration ,c = ln ⎢
nF ⎢⎣⎜⎝ γ Cu 2+ ⎟⎜ D 2+ z 2+ FA cathode c 2+
⎠⎝ Cu Cu Cu
⎟⎥
⎠⎦
Note that the anodic overpotential is negative.
Now consider ohmic potential IR int . It is the potential difference needed to carry
current from the anode to the cathode inside the cell. In the anodic compartment, the
current is carried by zinc ions which move away from the electrode and sulfate ions
which move towards the electrode. In the cathodic compartment, the current is carried by
cupric ions which move away from the electrode and sulfate ions which move towards
the electrode. All these motions require energy since the solvent produces drag on the
ions. Part of the available free energy is lost in motion against the drag. This is the ohmic
drop of potential and depends on how much current is drawn from the cell. Larger the
current, greater is the ohmic drop. The current density is related to the field by the
following equation
i = κE
Where i is the current density (A.m ), E is the electric field ( V.m-1) and κ is the
-2

electrical conductivity of the solution ( S.m-1). From this we get

η0 , a η0 , c
I = A a κa = A c κc
Aa Ac

Here A a and A c are the area of cross section of anodic and cathodic compartments,

respectively and κ a , κ c are the conductivities of the respective solutions

Rearrangement gives

14
 IA a IA c
ηo ,a = and ηo ,c =
A a κa A c κc
Hence the total Ohmic potential drop across the cell is
⎛ A A ⎞
ηo = ηo,a + ηo,c = I⎜⎜ a + c ⎟⎟
⎝ A a κa A c κc ⎠
Conductivity depends on concentration and mobility of ions. Mobility of an ion depends
on its size, its interaction with the solvent (degree of hydration) and on the viscosity of
the solvent.
Ohmic overpotential is the most important potential drop in many electrochemical
cells and one of the important design considerations is how to reduce this drop. Ohmic
overpotential can be reduced by reducing the distance between the electrodes, by
reducing the current density in the cell and by increasing the conductivity. It is not
necessary that current carriers are the same ions as those participating in the reaction at
the electrode. Hence to increase the conductivity, indifferent electrolytes are added.
These are the electrolytes which do not take part in the charge transfer reactions, but act
only as charge carriers.
Now consider a case where the resistance in the external circuit is replaced by a DC
power source. Assume that the positive end of the power source is connected the positive
end (cathode) of the Daniell cell. If the equilibrium potential difference of the power
source is less than 1.1 V, cell will charge the power source. The current flow will stop
when EMF of the power source is exactly 1.1 V. Note that in this case, the electrodes of
the Daniell cell will be at the same potential as when the voltmeter was connected to it.
This can be checked by using reference electrode.

Dc Power Source

Zn Cu

ZnSO4 CuSO4

15
 Cell connected to DC source is as shown in the figure and the potential profile under
no current condition is shown in the next figure

− Zn + Cu

Double layers

When we increase the EMF of the source is increased beyond 1.1 V, the current will
begin to flow through the cell in the direction opposite to that when the cell was
discharging. The cell operated in this manner is called the electrolytic cell.
Here, the copper electrode will act as anode and zinc electrode as cathode. Note that
anode is now at a higher potential than the cathode. It is now positive and the cathode
(zinc electrode) is negative. Current will still flow from the cathode to the anode in the
external circuit and yet we call the cathode as the negative electrode because it is
connected to the negative end of the power source. Internally, the current will flow from
copper electrode to zinc electrode. The potential of copper electrode will be higher than
the equilibrium value and that of the zinc electrode will be lower than the equilibrium
value. This overpotential drives the current. Various contributions to the potential
difference are listed below
Δφ = (Δφ)rev + (ηsurface + ηconcentration )a − (ηsurface + ηconcentration )c − ηo

Note that the anodic potential differences are positive and are added to (Δφ)rev , whereas

the cathodic potential differences are negative and are subtracted from (Δφ)rev . Hence

Δφ is greater than (Δφ)rev . This difference increases and consequently the current
increases as the EMF of the DC source increases.

16
2.4 Classification of cells based on Design
Cells can also be classified based their design. The following are the classes
(1) Cells with single common electrolyte (2) Electrolyte concentration cells (3) Electrode
concentration cells. (4) Cells with different electrolytes
Cells with single common electrolyte: As the name suggests, there is a single electrolyte,
which is common to both the electrodes. The cell is normally an electrolytic cell.
Example is the following cell
Pt H 2 (g ) HCl(aq) AgCl(s) Ag

In the above equation, a vertical line represents the interface between two adjacent phases.
The left hand side electrode is always anode and the right hand side is the Cathode. The
liquid junction (not present in the above cell) is shown by two parallel vertical lines.
In the above cell, aqueous HCl is the electrolyte. The anode is hydrogen electrode at
which the following reaction occurs
1
H 2 → H + + e−
2
The cathode is Ag/AgCl electrode at which the following reaction occurs
AgCl + e − → Ag + Cl −
The standard potential of the cell is 0.22 V. In the electrolytic mode, the cell can be used
for production of hydrogen on a laboratory scale.
(1) Electrolyte concentration cells: Here the cell is divided into two compartments
separated by a semipermeable membrane. The two compartments are identical except
the concentration of the electrolyte is different. The following cell is an example
Pt H 2 (g ) HCl(aq, m1 ) HCl(aq, m 2 ) H 2 (g) Pt , m 2 > m1

The zero current potential of the cell is given by

RT ⎛ a HCl (m 2 ) ⎞
E= ln⎜ ⎟
F ⎜⎝ a HCl (m1 ) ⎟⎠

Note that the standard potential of this cell is zero

(2) Electrode concentration cells: In these cells, electrodes are of different composition.
The gas electrodes operating at different gas pressures and electrode consisting of

17
 amalgams of different concentrations are the examples of this class. A hydrogen
electrode of this type can be represented by
Pt H 2 (g, p1 ) HCl(aq) H 2 (g, p 2 ) Pt ( p1 > p 2 )

At the anode (left electrode) of this Galvanic cell, the reaction is

1
H 2 → H + + e−
2
At the cathode exactly opposite reaction occurs
1
H + + e− → H2
2
The potential of the cell depends upon the difference in the pressures of hydrogen in the
two compartments.
The zero current potential of the cell is

RT ⎛ p1 ⎞
E= ln⎜ ⎟
F ⎜⎝ p 2 ⎟⎠

(4) Cells with different electrolytes: The example of this cell is Daniell cell.
Example 1: Find the zero-current potential for the following cell
Cu / Zn / Zn 2 + // Cl − / AgCl / Ag / Cu '
Solution: The half-cell reactions are
Zn ⇔ Zn 2+ + 2e(Cu ) (anode)

2AgCl + 2e(Cu ' ) ⇔ 2Ag + 2Cl − (cathode)

The cell reaction can be written as
Zn + 2AgCl + 2e(Cu ' ) ⇔ Zn 2 + + 2Ag + 2Cl − + 2e(Cu )
At equilibrium
μ Zn + 2μ AgCl + 2μ Cu
e
'
= μ Zn 2+ + 2μ Ag + 2μ Cl − + 2μ Cu
e

Rearrangement gives
2(μ Cu
e − μ e ) = μ Zn 2+ + 2μ Ag + 2μ Cl − − μ Zn − 2μ AgCl
' Cu

(
2 μ Cu
e − μe
' Cu
) (
= −2F φCu ' − φCu = −2FE )
Hence − 2FE = μ 0Zn 2+ + RT ln a Zn 2+ + 2μ 0Ag + 2μ 0Cl − + 2RT ln a Cl − − μ 0Zn − 2μ 0AgCl

Noting that

18
 − 2FE 0 = μ 0Zn 2+ + 2μ 0Ag + 2μ 0Cl − − μ 0Zn − 2μ 0AgCl

we obtain
− 2FE = −2FE 0 + RT ln a Zn 2+ + 2RT ln a Cl −

or

E = E0 −
RT
2F
[
ln (a Zn 2+ )(a Cl − )
2
]
Examples of Galvanic Cells
Examples of a variety of Galvanic cells are presented in this section
(a) The dry cell battery
It is the Lachanche cell. It is used in flashlights and transistor radios. The anode
consists of a zinc container and the cathode is a carbon rod placed at the center of the cell.
The zinc container is filled with manganese dioxide which is soaked in an aqueous starch
paste containing ammonium chloride, zinc chloride as electrolytes. The following
reactions occur
Anode: Zn (s) Æ Zn2+ (aq) + 2e-
Cathode: 2NH4+(aq) +2MnO2 (s) + 2e- Æ Mn2O3(s) + 2NH3 (g) + H2O (l)
The overall reaction is
Zn(s) + 2NH4+(aq) +2MnO2 (s) Æ Zn2+ (aq)+Mn2O3(s) + 2NH3 (g) + H2O (l)
The voltage produced by the dry cell is about 1.5 V. This battery is not rechargeable.
(b) The Lead Storage Battery
This battery is used in automobiles. It consists of six identical cells connected in
series. Each cell produces two volts and hence the battery produces 12 volts. This is
enough to power the ignition circuit of an automobile. The cathode of each cell is spongy
lead and anode is lead dioxide supported on lead frame. Aqueous solution of sulfuric acid
acts as the electrolyte. The cell reactions are
Anode: Pb (s) + SO42- (aq) Æ PbSO4 (s) + 2e-
Cathode: PbO2(s) +4H+(aq) + SO42- (aq)+ 2e- Æ PbSO4(s) + 2H2O (l)
The overall reaction is
Pb(s) + 4H+(aq) +PbO2 (s)+2 SO42- (aq) Æ 2PbSO4(s) + 2H2O (l)

19
The exchange current density in this battery is very high and hence one can draw a large
current (required for ignition) from this battery. Another advantage of this battery is that
it is rechargeable. The following reactions occur during recharging cycle.
Anode: PbSO4 (s) + 2e- Æ Pb (s) + SO42- (aq)
Cathode: PbSO4(s) + 2H2O (l) Æ PbO2(s) +4H+(aq) + SO42- (aq)+ 2e-
The overall reaction is
2PbSO4(s) + 2H2O (l) Æ Pb(s) + 4H+(aq) +PbO2 (s)+2 SO42- (aq)
The reactions during recharging are exactly opposite to those during discharging.
During discharging process, sulfuric acid is consumed and its concentration, and
hence its density, progressively decreases. The heath of the cell can be checked using a
hydrometer. The density should be greater than 1.2 g.(mL)-1. In cold climates, lead
battery can go dead due to increase in the viscosity of sulfuric acid. In such cases Ohmic
drop across the cell reduces the overvoltage and hence the current rating of the cell. Thus
the battery appears to go dead. Warming of the battery regains its normal operation.
(c) Solid State Lithium Battery
These are the batteries with no fluid in it. This battery uses a solid polyelectrolyte
(anionic) as a medium between two electrodes. The anode is lithium and the cathode is
made up of TiS2 or V6O13 (insertion compound). The reactions are
Anode: Li Æ Li+ + e-
Cathode: TiS2 + e- Æ TiS2-
Lithium ion migrates from anode to cathode and pairs with TiS2-. Use of lithium has lot
of advantages. It is easy to oxidize. The standard reduction potential is -3.05 V. the
voltage of the battery is as high as 3 V. Secondly, its molecular mass is low (6.941 g.mol-
1
) and hence small mass of lithium can pack large amount of power. It is also
rechargeable. Much work is still needed to make these batteries more reliable and long
lasting, but they are being viewed as the batteries of the future.
(d) Fuel Cells
They convert chemical energy in fuels directly to electrical energy instead of going
through thermal cycle as is conventionally done. The advantage is the high efficiency of
conversion.

20
 Lecture-3
Butler-Volmer Equation
Significance of Butler-Volmer Equation:
The most significant advantage of electrochemical reactions is that the rate of
these reactions can be varied over a wide range by varying the electrode potential.
Every reaction requires energy for its activation. In thermochemical reactions, it is
supplied as thermal energy, the intensity of which increases with increase in
temperature. In electrochemical reactions, this energy is supplied in the form of
electrical energy. For those reactions which have very high energy of activation, the
rate of reaction in the thermochemical mode becomes significant only at high
temperatures. Some of these reactions could be conducted electrochemically at room
temperature by providing adequate overpotential (difference between the actual
electrode potential and the Nernst potential). Butler-Volmer equation is one of the
important equations in Electrochemistry. It relates the rate of electron transfer reaction
to the overpotential.
Specific Rate of Reaction
Consider the following electron transfer reaction occurring at an electrode.
O + ne ⇔ R (1)
It is assumed that the reaction is conducted under sufficiently high rate of agitation so
that the transport of the reactant to and the product from the electrode does not affect
the rate of the reaction.
Since the reaction is reversible, the net specific rate (rate per unit area of the
electrode) of the reaction can be written as the difference between the specific rates of
the forward and the backward reactions.
G H
r=r−r (2)
G H
In this equation, r is the specific rate of conversion of O + ne to R; r is the specific
rate of conversion of R to O + ne . It is important to note that since charge transfer
occurs on the electrode surface, the total rate of the reaction is proportional to
electrode surface area. The specific rate (defined per unit area of the electrode) is
therefore independent of the area of the electrode.
According to the activated complex theory (also called the transition state
theory), every reaction occurs in two steps. In the first step, the reactant, adsorbed on
the electrode, forms an activated complex in which the electrons are partially or

1
wholly involved. This complex, A* (see Figure-1 below), has a higher free energy
than both O*+ne and R* (O* and R* are the adsorbed forms of O and R).
G L
ΔG and ΔG are the free energy of activation of the forward and the backward reaction
respectively. Thus for the forward reaction, we have

A*

G
ΔG G
ΔG

H
ΔG

O∗ + ne

R*

Figure-1: Energy of the reactants/product as a function of reaction coordinates.

G
(
ΔG = μA∗ − μO∗ + n μe ) (3)
G
Thus ΔG is the difference between electrochemical potential of the activated complex
and the sum of the electrochemical potentials of the oxidant and the free electrons
which take part in the reaction.
In a similar manner we can write for the backward reaction
H
ΔG = μA∗ − μR∗ (4)
The following condition is valid even under nonquilibrium state.
μO∗ = μO and μ R∗ = μ R (5)

Here, μO and μ R are the electrochemical potentials of O and R in the bulk solution. In
writing eq 5, we recognize the fact that whenever the potential of the electrode
changes, the composition of the double layer changes extremely fast (reason is,
double layer is very thin and hence very short time is required for the motion of ions
through the double layer) so that electrostatic potential of any species attain equality
at all points in the double layer, practically instantly.
Using eq 5, we can rewrite eq 3 and 4 as
G
ΔG = μA∗ − (μO + n μe ) (6)
H
and ΔG = μA∗ − μ R (7)
Note that the activated complex A* has existence only on the electrode surface and
hence has no counterpart in the bulk solution.

2
 The specific rates of the reaction are proportional to the probability that the
reactants/products will surmount the activation barrier, which is given by the
Boltzmann factor. Thus
G G − ΔGG RT H H H
r = Be and r = Be − ΔG RT (8)
G H
where B and B are constants.
Substituting, the rates from eq 8 into Eq 2, we obtain
G G H H
r = Be − ΔG RT − Be − ΔG RT (9)
Cathodic Current Density
Noting that per mole of R going to O, nNav electrons are transferred from the
electrode to the solution and therefore nF Coulombs of electric charge is transferred,
we can relate the cathodic current density i with the rate of reaction
i cath = nFr (10)

i cath represents the current entering (electrons leaving) the electrode per unit surface
area. It is subscripted as cathodic current density since it corresponds to net reduction.
If oxidation (opposite of reaction 1) is actually occurring at the electrode then i cath will
be negative. Combining eq 9 and 10, we obtain
G G H H
i cath = nFBe − ΔG RT − nFBe − ΔG RT (11)
G
We can write terms to the right as the forward current density i and the backward
H
current density i , respectively. Thus
G H
i cath = i − i (12)
For a give composition of the solution, we can adjust the electrode potential in such a
way that no net current flows through the electrode and hence i cath = 0 . Eq 11, under
this condition, reduces to
G Ge H He
nFBe − ΔG RT
− nFBe − ΔG RT
=0 (13)
Here the subscript e corresponds to the zero net current (equilibrium) state. We can
also rewrite eq 13 as
Ge He
i =i (14)
Ge G Ge H H He
where i = nFBe − ΔG RT
and i e = nFBe − ΔG RT
(15)
In other words, at the equilibrium, forward current exactly matches with the backward
current.

3
 We can write the activation energies under equilibrium condition, using eq 6
and 7, thus
G
ΔG e = μA∗e − μO + μee ( ) (16)
H
and ΔG e = μA∗e − μR (17)

In the above equation, μ∗Ae is the electrochemical potential of the activation complex

μ ee that of electron, both at equilibrium. Note that μ O and μ R in the above equations

are not changed since we are keeping the bulk composition constant and only
changing the electrode potential.
Subtracting eq 17 from eq 16, we obtain
G H
ΔG e − ΔG e = μR − μO + n μee ( ) (18)
However at the equilibrium,
μO + n μee = μ R (19)
Comparison of eq 18 with eq 19 gives us the following equality
G H
ΔG e = ΔG e (20)
This is easy to understand this. At the equilibrium, the driving force for the forward
and the backward reactions should be equal.
Substitution of eq 20 into eq 13 yields
G H
B=B (21)
This allows us to transform eq 11 to the following form

( )
G H
i cath = nFB e − ΔG RT − e − ΔG RT (22)
G H
where B is the common symbol for B or B .
We could write eq 22 in a modified form as follows

( G
i cath = nFBe− ΔG RT e − (ΔG − ΔG ) RT − e − (ΔG − ΔG ) RT
e e
(23)
H e
)
G H
Note that, since ΔG e = ΔG e , we have used a common symbol ΔG e to represent both.
Comparing the term outside the bracket in eq 23 with that in eq 15, we see that
G H
it equals i e (a common symbol for both i e and i e ). It is called the exchange current
density and is commonly denoted by symbol i 0 . We shall use this symbol and rewrite
eq 23 as

(
i cath = i 0 e − (ΔG − ΔG
G e
) RT − e− (ΔGH − ΔG ) RT e
) (24)

i 0 = nFBe− ΔG
e
RT
where (25)

4
Thus the exchange current density is the current density of the half-step (either
forward or backward step) at the equilibrium.
The Butler-Volmer Equation
When the electrode potential differs from the Nernst potential, the current
density of one of the half steps is expected to rise above i 0 , whereas that of the other

half step is expected to fall below i 0 .

Subtracting eq 16 from and 6, we obtain
G
( ) (
ΔG − ΔG e = μA∗ − μA∗e − n μe − μee ) (26)

But μe = μ 0e − Fφ and μee = μ 0e − Fφe (27)

where φ is the electrode potential under nonequilibrium condition and φe is that under
the equilibrium condition (Nernst Potential).
Eq 26 can now be simplified to the following form
G
( )
ΔG − ΔG e = μA∗ − μA∗e + nF(φ − φe ) (28)
We define the overpotential as
η = φ − φe (29)
It represents the deviation from the Nernst potential. Using this symbol, we can
rewrite eq 28 as
G
ΔG − ΔG e = (μA∗ − μA∗e ) + nFη (30)
Also, combining eq 7 and 17, we obtain
H
ΔG − ΔG e = μ A∗ − μA∗e (31)

The change in the transition state of A, μA∗ − μA∗e , is caused by the change in the
potential of the electrons alone. Electrochemical potentials of both the oxidant O and
the reductant R are kept constant. The change in the chemical potential of electrons
is nFη . Entire change may not be transmitted to the activation complex since electrons
taking part in the reaction may not be fully associated with the transition complex. We
expect that
− nFη < μA∗ − μA∗e < 0 (32)
We can therefore write
μA∗ − μA∗e = −αnFη 0 < α <1 (33)

5
The actual change in μA∗ − μA∗e will depend on the extents to which electrons are
associated with the activation complex.
The term α is commonly known as the symmetry factor. It is also called the
transfer coefficient. If the structure of the activated complex is exactly half way
between O* +ne and R*, then we expect α to be 0.5. It is customary to use α = 0.5 as
an approximation. The justification is that for majority of reactions, it is
approximately 0.5. However, there is an appreciable number of reactions, where
α differs significantly from 0.5.
Combining eqs 30 and 33, and also 31 and 33, we obtain
G
ΔG − ΔG e = (1 − α)nFeη (34)
H
ΔG − ΔG e = −αnFeη (35)
Substituting these quantities in eq 26, we obtain
(
i cath = i 0 e − (1− α ) nFη RT − eαnFη RT ) (36)
The above equation is called the Butler-Volmer equation. We see from this equation
that the cathodic current density is positive only when the overpotential η is negative.
This is easy to understand. When, the overpotential is negative, free energy of
electrons in the electrode is higher than that at the Nernst potential. Hence the
electrode acts as a electron donor (i.e. cathode). On the other hand if η is positive, the
potential energy of electrons is lowered and hence electrons from the reductant are
extracted by the electrode. Electrode in this situation acts as anode. The current i cath is
negative, indicating it is the anodic current.
One important point about the Butler-Volmer equation is that one must
use proper stoichiometry of the reaction in order to get the mechanism right. For
example in the hydrogen ion reduction reaction one electron is involved in the
stoichiometry (n=1). Writing the equation as

2H + + 2e − ⇔ H 2
(n = 2) is not correct since it does not represent the correct mechanism of the reaction.
In actual mechanism, hydrogen ion is reduced to form hydrogen atom. Hydrogen
atoms on the adjacent sites combine to form hydrogen molecule. In most cases, the
rate controlling step involves transfer of a single electron. Hence, n is taken as 1, in
general.

6
Exchange Current Density
The exchange current density defined by eq 25

i 0 = nFBe− ΔG
e
RT
(37)
represents the current density in any one direction(forward or backward) at the
electrode when it is at the Nernst potential.
Exchange current density of some reactions along with the transfer
coefficients for some reactions are shown below
Table-1 Exchange current density of reactions
Reaction Electrode i0, A.m-2 α

H + + e − ⇔ (1 2)H 2 Pt 7.9 -

H + + e− ⇔ 1 2 H 2 Cu 0.01 -

H + + e− ⇔ 1 2 H 2 Ni 0.063 0.58

H + + e− ⇔ 1 2 H 2 Pb 5x10-8 -

H + + e− ⇔ 1 2 H 2 Hg 7.9x10-9 0.5
Fe3+ + e − ⇔ Fe 2 + Pt 25 0.58
4+ − 3+
Ce + e ⇔ Ce Pt 0.40 0.75

We see from this table that for the same reaction, different electrodes have different
values of the exchange current densities. Thus for hydrogen ion reduction reaction,
platinum electrode will produce a large current density for small values of the
overpotential. On the other hand, Mercury will require a large value of the
overpotential to produce noticeable values of the current density. This leads us to
define ideal polarizable electrodes, for which the exchange current density is zero.
Mercury nearly acts as ideal polarizable electrode. On the other hand, an ideal
nonpolarizable electrode is the one which produces an infinite current density for a
small value of the overpotential. Platinum electrode acts as nearly ideal polarizable
electrode for many reactions. Following point, related to electrode polarizability is
worth noting. Polarizability of an electrode not only depends on the nature of the
electrode but also on the nature of the reaction. For example mercury is polarizable
for reduction of cations, but is nonpolarizable for the oxidation of anions. Similarly,
the exchange current density for N2/N3- couple on platinum is 10-72 A.m-2 which
extremely small compared to H+/H2 couple.

7
Dependence of the exchange current density on concentration:
We can write the Butler-Volmer equation in an alternative form as follows
i cath K ⎡ (1 − α )nF ⎤ H ⎡ αnF ⎤
r= = k [O]exp ⎢− Δφ⎥ − k[R ]exp ⎢ Δφ ⎥ (38)
nF ⎣ RT ⎦ ⎣ RT ⎦
where Δφ is the difference between the electrode φ and the solution potential φs .

Δφ = φ − φs (39)
K L
k and k are the (forward and backward) rate constants of the reactions and [O] and
[R] are the concentrations of the oxidant and the reductant, respectively, in the
solution. This is a standard way we write the rate expression for a reversible reaction,
except that an additional driving force supplied by the imposed potential is included.
The transfer coefficient 1 − α represents the fraction of the total imposed potential
which promotes the cathodic reaction.
Let at certain potential difference Δφ = Δφe , the rate of the reaction is zero. In
that case, eq 38 reduces to
K ⎡ (1 − α )nF e ⎤ H ⎡ αnF e ⎤
k[O]exp ⎢− Δφ ⎥ = k[R ]exp ⎢ Δφ ⎥ (40)
⎣ RT ⎦ ⎣ RT ⎦
On rearrangement, the above equation gives

RT ⎛ k[R ] ⎞
Δφ e = − ln⎜ K ⎟ (41)
nF ⎜⎝ k[O] ⎟⎠

Writing the overpotential as

RT ⎛ k[R ] ⎞
η = Δφ − Δφe = Δφ + ln⎜ K ⎟ (42)
nF ⎜⎝ k[O] ⎟⎠

and substituting into eq 38, we obtain

i cath K ⎡ (1 − α )nF ⎛ RT ⎛ k[R ] ⎞ ⎞⎤

= k[O]exp ⎢− ⎜η − ln⎜ K ⎟ ⎟⎥
nF ⎣⎢ RT ⎝ ⎜ nF ⎜⎝ k[O] ⎟⎠ ⎟⎠⎦⎥
(43)
H ⎡ αnF ⎛ RT ⎛ k[R ] ⎞ ⎞⎤
− k[R ]exp ⎢ ⎜η− ln⎜ K ⎟ ⎟⎥
⎜
⎣⎢ RT ⎝ nF ⎜⎝ k[O] ⎟⎠ ⎟⎠⎦⎥

Rearrangement of the above equation yields

(1−α )
i cath K ⎛ k[R ] ⎞ ⎡ (1 − α )nF ⎤ H ⎛ kK [R ] ⎞
−α
⎡ αnF ⎤
= k[O]⎜⎜ K ⎟⎟ exp ⎢− η⎥ − k[R ]⎜⎜ ⎟⎟ exp ⎢− η (44)
nF ⎝ k[O] ⎠ ⎣ RT ⎦ ⎝ k[O] ⎠ ⎣ RT ⎥⎦
Comparison with eq 36 gives

8
 K α L
( )(
i 0 = nF k[O ] k[R ] )
1− α
(45)
Thus we see that the exchange current density is a function of the reactant
concentrations. The transfer coefficient plays a role in this dependence.
Activation energy of electrochemical reaction:
Various processes contribute to the activation energy of an electrochemical
reaction. They are listed below
1. Desolvation of the adsorbed reactants
2. Structural changes in the reactants to form activated complex. This may involve
bending/stretching or even breaking of some bonds.
3. Movement of the electrons across the surface of the electrode.
Redox reactions involving no bond breaking has a high exchange current density.
This is evident from the last two entries of Table-1.
By coating the electrode surface with a catalyst, it is possible to reduce the
activation energy of the reaction and thereby increase the exchange current density.
Polarization Curves

1 i
4
2
5

Anodic Range 3
3 η
Cathodic Range 0

2
5
1
4

Figure-1: Overpotential-Current Density Diagram for different electrodes

1. normal electrode 2. polarizable electrode 3. nonpolarizable electrode 4.
ideally polarizable electrode 5. ideally nonpolarizable electrode

9
 We have seen from Table-1 that for the same reaction, different electrodes have
different exchange current densities. Thus platinum electrode will produce a large
current density for small values of the overpotential. On the other hand, Mercury will
require a large value of the overpotential to produce noticeable values of the current
density.
We can study the characteristics of electrodes using the overpotential-current
density diagram as shown in Figure-1. The plot is a curve and is called the
polarization curve. The plot has two regions. The anodic region corresponds to
positive values of the overpotential whereas the cathodic region corresponds to
negative overpotential. The current density has a positive value (as per our
convention) in the cathodic region and negative value in the anodic region. The
diagrams have the following characteristic features
At small values of the overpotential (< 10 mV) the plot is linear. This is called the
linear region of the Butler-Volmer equation. In this region, it is possible to linearized
Butler-Volmer equation using the approximation e x = 1 + x . Thus
i cath = i 0 (e − (1− α ) nFη RT − e αnFη RT ) ≈ i 0 (− (1 − α)nFη RT − αnFη RT ) = −i 0 nFη RT (46)

The slope of the plot is − i0 nF RT . The slope is directly proportional to the exchange
current density. The slope of the plot at the origin can therefore be used to compute
the exchange current density.
Example-1: What is the effective resistance of at 250C of an electrode interface when
the overpotential is small? Evaluate it for 1 cm2 of (a) Pt, H2/H+, (b) Pb,H2/H+
electrodes.
Solution: At small values of the overpotential (< 10 mV), it is possible to linearized
Butler-Volmer equation to give
i cath = −i 0 nFη RT
The effective resistance of the electrode interface can now be defined as
η RT
R int = − = (47)
i cath A i 0 nFA

(0.0257 V )
(a) For Pt,H2/H+ , i 0 = 7.9A ⋅ m −2 , R int = = 32.53 Ohm
( )
7.9A ⋅ m − 2 × 1 × 1 × 10 − 4

(b) For Pb,H2/H+, i 0 = 5 × 10 −8 A ⋅ m −2 ,

(0.0257 V )
R int = = 51.4 × 108 Ohm
( −8 −2
)
5 × 10 A ⋅ m × 1 × 1 × 10 −4

10
Tafel Plot
At high vales of the overpotential, another asymptote is reached. In this case one of
the exponential term becomes very large and the other very small. The small term can
be dropped. Thus in the cathodic region (negative values of the overpotential), we
have
i cath = i 0e − (1− α ) nFη RT (48)
Applying natural logarithm to both sides of the above equation we obtain
ln i cath = ln i0 − (1 − α ) nηF RT (49)

The plot of ln i cath versus (− η) is a straight line with slope equal to (1 − α ) nF RT and

the intercept on y-axis equal to ln i0 . Such a plot is called a Tafel Plot.

The Tafel plot can also be drawn with positive overpotential where following relation
holds
ln (− i cath ) = ln i 0 + α nηF RT (50)
The experimental setup for obtaining the polarization curve is shown below.

Using both the plots together, it is possible to obtain all the parameters of the Butler-
Volmer equation ( α , n and i 0 ).

11
The electrode of interest is called the working electrode. Potential difference is
applied across the working electrode and another electrode called the
counterelectrode. The current flowing through this circuit is measured. If the working
electrode has area A, and the current flowing across it is I, then the current density is
I/A. The potential difference between the working and the counter electrode is not
useful. It is necessary to measure the potential of the working electrode directly. This
is done using a third electrode that is the reference electrode. The reference electrode
is connected to the working electrode through a high impedance voltmeter. Practically
no current flows through the circuit formed by the working and the reference
electrode. When the current is varied in the circuit formed by the working and the
counterelectrode, the potential of the working electrode varies. This potential is
measured with respect to the reference electrode using the voltmeter in the reference
electrode –working electrode circuit. The Nernst potential is that potential when no
current flows in the working-counterelectrode circuit.
A point to be taken care of is that the reference electrode and the working
electrode are near each other so that there is negligible potential difference between
the two points in the solution, with respect which the potentials are measured. The
reference electrode is provided with a Luggin capillary as shown below.

12
 Luggin capillary consists of a bent tube with a large enough opening, at one
end, to accommodate a reference electrode and a usually much smaller opening at the
other end, which is only large enough to insure diffusional movement of the
electrolyte. The device minimizes any ohmic drop in the electrolyte associated with
the passage of current in an electrochemical cell.

Overpotential-Current relation for a cell

When a cell is at equilibrium, its potential is given by the zero current
potential or the Nernst potential of the cell.
⎡ RT ⎛ a O ⎞ ⎤ ⎡ 0 RT ⎛ a O ⎞ ⎤
Δφe = φec − φea = ⎢ E 0c − ln⎜ ⎟ ⎥ − ⎢E a − ln⎜ ⎟ ⎥ (51)
⎣⎢ n c F ⎜⎝ a R ⎟⎠ c ⎦⎥ ⎣⎢ n a F ⎜⎝ a R ⎟⎠ a ⎦⎥

Let in actual operation, the overpotential of the cathode be ηc and that of the anode be

ηa . In that case the cell potential is

( ) (
Δφ = φec + ηc − φ ea + ηa = Δφe + (ηc − ηa ) ) (52)
We define the cell overpotential as
Δη = ηc − ηa (53)
Comparing eq 52 and 53, we obtain
Δη = Δφ − Δφe (54)
Applying Butler-Volmer equation for both cathode and anode, we obtain the
corresponding current densities as
(
i c = i0 c e − (1− α c ) n c Fη c RT
− eα c n c Fη c RT
) (55)

i a = i 0a (e − (1− α a ) n a Fη a RT
− eα a n a Fη a RT
) (56)

In practice, there are two situations (1) Here the potential φ of the cell may be
specified and one is interested in the current drawn from the cell or fed into the cell.
(2) Another problem is when the external load (resistance) is given and one needs to
find the current drawn from the cell. In both situations, the values of ηc and ηa are
unknown. However, we know that the cathodic and anodic currents must be equal but
opposite in sign. Hence we can write
(
i c A c + i a A a = i 0 c A c e − (1− α c ) n c Fη c RT
− eα c n c Fη c RT
)+ i 0a (
A a e − (1− α a ) n a Fη a RT
− e α a n a Fη a RT
)= 0
(57)

13
If we know Δφ , then using eq 54, we obtain Δη . Eq 53 and 57, can then be
simultaneously solved to compute the values of ηc and ηa .
If the resistance in the external circuit is specified, and its value is R, then we
can write
Δφ = i c A c R (58)
Eq 53,54,57 and 58 are simultaneously solved to give the needed values.
Note that in all this analysis we have neglected the potential drop due to
concentration polarization, ohmic drop and the drop due to the junction.
Corrosion
Consider, as a typical case, corrosion of iron surface. The corrosion reactions are
Fe 2 + (aq) + 2e − ⇔ Fe(s) E 0 = −0.44V (59)
1
and H + (aq) + e − ⇔ H2 E 0 = 0V (60)
2
4H + (aq) + O 2 (g ) + 4e − ⇔ 2H 2O(l) E 0 = 1.23V (61)
In absence of air, reaction 61 does not occur and hence we only consider reactions
(59) and (60).
In the corrosion process, there are cathodic and anodic regions on the iron
surface as shown in the figure below.
Fe2+
H2 H+ H+ H2
Fe

e- e-
The electrons are released into the metal at the anode (central region) and are
consumed at the cathode. The metal attains a potential at which the cathodic and the
anodic currents are balanced (corrosion potential). This corrosion potential lies at a
value which is lower than the Nernst potential of the cathodic reaction (hydrogen ion
reduction) and higher than the Nernst potential of the anodic reaction (iron oxidation
to Fe2+. The cathodic (or anodic) current is called the corrosion current and is given
by
Icorrosion = i cath A cath = (− i ano )A ano = i cath (− i ano )A cath A ano (62)

If we assume that overpotentials are high enough so as to fall in the Tafel region, then

14
 i cath = i 0cath e −( φ corr − φ cath ) F 2 RT and i ano = −i 0ano e ( φ corr − φ ano ) F 2 RT
e e
(63)
Here we consider only single electron transfer mechanism and the transfer coefficient
of 0.5
Substituting eq 63 into 62, we obtain

Icorrosion = i 0cath e − ( φ corr − φ cath ) F 2 RTi 0ano e( φ corr − φ ano ) F 2 RT A cath A ano
e e

(64)
= i0 Ae (φ cath − φ ano )F / 4 RT = i0 Ae EF 4 RT
e e

where E is the Nernst potential of the cell formed by couple of the reaction.
The cathodic Nernst potential (for hydrogen evolution) depends on pH. Thus

RT ⎛⎜ 1 ⎞
⎟ = E 0 − RT 2.303pH = 0 − 0.059pH V.
E cath = E 0 − ln
F ⎜⎝ a H + ⎟
⎠ F

Hence the Nernst potential of the cell is

E = -0.059 pH + 0.44 V.
It is seen that the value of E is strongly dependent on pH and decreases with increase
in pH.
Example-1: A typical exchange current density of H+ discharge at platinum is 7.9
A.m-2 at 250C. What is the current density at the electrode when its overpotential is (a)
10 mV (b) 100 mV (c) -0.5 V (d) -5 V? Take α = 0.5 .
Solution: We use Butler-Volmer equation.
(
i cath = i 0 e − (1− α ) nFη RT − eαnFη RT )
The cathodic current should will be negative for positive overpotential (electrode will
behave as anode) and positive for negative values of the overpotential.

(a)
( )
i cath = 7.9 e −0.5×1×(0.01V ) (0.0257 V ) − e0.5×1×( 0.01V ) (0.0257V ) = 7.9(0.8441 − 1.215)
−2
= −3.1 A ⋅ m

(b)
( )
i cath = 7.9 e −0.5×1×(0.1V ) (0.0257 V ) − e 0.5×1×(0.1V ) (0.0257 V ) = 7.9(0.1429 − 6.997)
= −54.1A ⋅ m − 2

(c)
( )
i cath = 7.9 e −0.5×1× ( −0.5 V ) (0.0257 V ) − e 0.5×1× ( −0.5 V ) (0.0257V ) = 7.9(16774 − 0)
−2
= 1.33 × 10 A ⋅ m 5

(d)
( )
i cath = 7.9 e −0.5×1× ( −5 V ) (0.0257V ) − e0.5×1×( −5 V ) (0.0257 V ) = 7.9(1.764 × 10 42 )
−2
= 1.39 × 10 A ⋅ m 43

In actual practice, we will never be able to obtain such a high values of the current
density since transport processes will limit the current to approximately 105 A.m-2

15
Example2: To first approximation, significant evolution or deposition occurs in
electrolysis if the overpotential exceeds about 0.6 V. To illustrate the criterion
determine the effect that increase in the overpotential from 0.4 to 0.6 has on the
current density in the electrolysis of 0.1 m NaOH(aq), which is 1.0 A.m-2 at 0.4 V and
250C. Take a =0.5
Solution : (
i cath = i 0 e − (1− α ) nFη RT − e αnFη RT )
Hence
(i )2 =
(e − (1− α ) nFη RT
− eαnFη RT
=
) (
e −0.5× 0.6 0.0257 − e 0.5× 0.6 0.0257 )
(i )1 (e − (1− α ) nFη RT
− eαnFη RT ) (
e − 0.5× 0.4 0.0257 − e 0.5×0.4 0.0257 )
e 0.5× 0.6 0.0257
= 0.5× 0.4 0.0257
= e0.5× 0.2 0.0257 = 48.96
e
(i )2 = 48.96 × 1.0 = 48.96 A.m 2

16
 Lecture-4
Thermodynamics of Electrolytes
4.1 Activity and activity coefficient
It is customary to express chemical potential of a species in solution by the
following equation
μ i = μ i0 + RT ln (a i ) (1)

where μ i0 = chemical potential of i in the arbitrarily defined standard state

and a i = activity of species i.
This manner of expressing the chemical potential has its origin in the chemical
potential for component of an ideal gas mixture (derived on the basis of statistical
mechanics)
μ i = μ i0 + RT ln (x i ) (2)

Here μ i0 = chemical potential of the gas phase species i in the pure state and xi is the
mole fraction of i in the actual mixture.
In eq 2, − R ln(x i ) represents the translational entropy (number of ways the
molecules of i can be arranged in the different translational energy states). Smaller the
value of xi, greater is this entropy and lesser is the value of μ i . And since a component
has a tendency to reduce its chemical potential, it will diffuse from the higher to lower
concentration. This is also consistent with the entropy maximum principle, which
state that for a fixed total energy, the process will occur in the direction in which the
system entropy is maximized.
In eq 1, a i is an analog of concentration of the species i. In the case of nonideal
gas, the term quantifying the nonideality should also be included in the activity. Thus,
we can write
ai = γixi (3)
γ i is called the activity coefficient, which quantifies the nonideality. It is a positive
quantity. γ i < 1 implies that, there is an attractive interaction among the molecules.
This reduces the effective mole fraction and hence the chemical potential.
From eq 1, we see that the standard state is the state in which the activity
equals unity. For this to happen, we must have γ i x i = 1 in the standard state. If we
define the standard state as the pure component state, we have a problem if
nonideality is still present in the pure component state. To avoid this problem, we
define the standard state as the imaginary pure component state in which all the
nonidealities are extinguished. This is not a problem since, in practice we are not
required to produce this state. We are only required to account for these nonidealities
in computation of γ i .
4.2 Activity of solution of electrolytes
Use of pure component as the standard state for a species in a liquid mixture is
convenient only when the component, in its pure state, exists as liquid. For
electrolytes (say NaCl), pure state is solid state. Hence, to compute the chemical
potential of an electrolyte solution with pure electrolyte as the reference state, one
must include the work of dissolution of solid electrolyte in the solvent. This
complicates the situation. To avoid this and alternative reference state is needed.
For electrolytic solution it is customary to use molarity or molality, instead of
mole fraction, to represent concentration. Thus we write
μi = μi0 + RT ln(ci γ i ) (4)

Or μi = μi0 + RT ln (mi γ i ) (5)

In the above equation
ci = Molar concentration of i, mol.L-1 of the solution
m i = Molal concentration of i, mol.Kg-1 of pure solvent.

ci0 and mi0 are the corresponding values in the standard state. γ i and γ i0 are the activity
coefficients in actual and the standard state respectively.
Among the two, molality is the more popular unit. Thus in a solution of
electrolyte activity a i can be written as
a i = mi γ i (6)
It is customary to use the reference state as for i as 1 molal solution of i in the solvent
in which the solute-solute interactions are extinguished ( m0i = 1 and γ i0 = 1 ).
Hence using this standard state, we can rewrite eq 17 as
μi = μi0 + RT ln (mi γ i ) (7)
Again, the standard solution is the solution of the unit activity.
Activity coefficient of an ion in solution depends on its interaction with other
ions as well as with the solvent. Interaction among ions is mainly electrostatic
(Coulombic) in nature. Other interactions (van der Waals) are much weaker and play
a minor role, especially in dilute solutions. Interaction of ion with solvent (water)
cannot be neglected. This interaction, called solvation interaction, mainly involves
ion-dipole interaction. However, the need to obtain this interaction is avoided by
defining the standard state as 1 molal solution in which ion-ion interactions are
absent, but ion-solvent interactions are present to an extent as if each ion is
completely solvated. This definition of the standard solution has the following
implication.
γi → 1 as m i → 0 (8)
The reason is as mi → 0 the ions are far apart and the electrostatic interaction among
them is absent and ions are completely solvated.
4.3 Mean activity coefficient
Consider an electrolyte represented by the formula MpXq (we call this
electrolyte as p, q electrolyte. Thus CuSO4 is a 1:1 electrolyte and BaCl2 is a 1:2
electrolyte). It dissociates in the aqueous solution according to the following equation
M ν+ X ν− → ν + M z+ + ν − X z − (9)

Assume that it is a strong electrolyte so that reaction 9 goes to completion. In the

above equation z + and z − are the valencies of the cations and anions respectively.
Electroneutrality dictates that
ν+z+ + ν−z− = 0 (10)
The chemical potential of the electrolyte can be written as the sum of the potential of
the cations and the ions it produces on dissolution. Thus
μ M ν+ X ν− = ν + μ + + ν − μ − (11)

where μ + and μ − are the chemical potentials of the cations and anions respectively
and are given by
μ + = μ0+ + RT ln γ + m + (12)

and μ − = μ 0− + RT ln γ − m − (13)
Substituting eq 12 and 13 into eq 11, we obtain
( ) (
μ M ν+ X ν− = ν + μ 0+ + RT ln γ + m + + ν − μ 0− + RT ln γ − m − ) (14)

Rearrangement of the above equation yields

( ) ( ) (
μ M ν+ X ν− = ν + μ 0+ + RT ln m + + ν + μ 0− + RT ln m − + RT ln γ ν++ γ ν−− ) (15)

We now define a quantity γ ± as

 (
γ ± = γ ν++ γ ν−− )(
1 ν + +ν − )
(16)
and rewrite eq 15 as
( ) ( )
μ M p X q = ν + μ 0+ + RT ln m + γ ± + ν − μ 0− + RT ln m − γ ± = ν + μ + + ν −μ − (17)

where we redefine the ionic chemical potentials as

μ + = μ0+ + RT ln m + γ ± (18)

and μ − = μ 0− + RT ln m − γ ± (19)
This new definitions of the chemical potentials of the ions is needed since we can
measure only one activity coefficient for an electrolyte in the solution, that is γ ± . Also,
we do not need γ + and γ − individually in practice and the knowledge of γ ± is
sufficient.
4.4 Debye-Huckel theory
In dilute solutions, ions are completely solvated. Hence, there is no deviation
from the standard state a far as the ion-solvent interaction is concerned. As for the
ion-ion interactions, solvent effectively screens the short range ion-ion interactions.
Only interaction which dominates is the electrostatic interaction. The model proposed
by Debye-Huckel accounts quantitatively for this electrostatic interaction. The details
of the model are explained in the Appendix. The final expression for the activity
coefficient for ion i is

z i2α I
ln γ i = − (20)
1 + Ba I
where I represents the ionic strength of the solution and is defined as
1
I= ∑ m jz2j
2 j
(21)

Here m j is the molality of ion i and z j is the corresponding valancy.

F2e 2ρs
α= (22)
8π(εε 0 RT )
32

F2 2ρs
and B= (23)
(εε0RT )1 2
The values of the parameters α and B for aqueous solutions are given in the table
below
T, 0C 0 25 50 75
 (
α, kg.mol −1 )
12 1.1324 1.1762 1.2300 1.2949

(
B, kg.mol −1 ) ⋅ (nm )
12 −1 3.248 3.287 3.326 3.368

The term ‘a’ represents sum of the solvated ionic radii of the positive and the
negative ions. Usually, it is used as a fitted parameter.
From eq 20, we can obtain γ ± as follows. From eq 16, we have
1
ln γ ± = [ν + ln(γ + ) + ν − ln(γ − )] (24)
ν+ + ν−
From eq 20, we get

z 2+ α I
ln γ + = − (25)
1 + Ba I

z 2−α I
and ln γ − = − (26)
1 + Ba I
Substituting the expression from eq 25 and 26 into eq 24, we obtain

ln γ ± = −
( )
1 ⎡ ν + z 2+ + ν − z 2− α I ⎤
⎢ ⎥ (26)
ν+ + ν− ⎣ 1 + Ba I ⎦
Using eq 10, we write ν + z + = −ν = z − in the first term and ν − z − = −ν + z + in the second
term of the numerator of the term in the bracket. This gives

z+z− α I
ln γ ± = − (27)
1 + Ba I
Note that Debye-Huckel theory always predicts negative value of ln γ ± and
hence γ ± < 1 , indicating attractive interaction among ions.
At very low concentration (ionic strength less than 0.01m) eq 27 can be
simplified by neglecting the term in the denominator. We then get the Debye-Huckel
limiting law
ln γ ± = − z + z − α I (28)

Many times the above expression can also be written as

log γ ± = − z + z − A I (29)

where the logarithm is to the base 10. The value of A at 250C is 0.509.
Example-1: Estimate the man activity coefficient of CaCl2 at 250C, in an aqueous
solution that is 0.01 m CaCl2 and 0.03 m NaF. Use Debye-Huckel limiting law.
Solution: We denote various ions as follows
Ca2+=1, Na+=2, Cl- = 3 and F- =4
The molality and valancy of each of these are given below
m1 = 0.01 and z1 = 2; m2 = 0.03 and z2 = 1, m3 = 0.02 and z3 = -1; m4 = 0.03 and z4 =
-1
The ionic strength is given by

I=
1
2
[0.01 × (2 ) + 0.03 × (1) + 0.02 × (− 1) + 0.03 × (− 1) = 0.06
2 2 2 2
]
log γ ± = − z + z − A I A = 0.509

For CaCl2, z + = 2 , z − = −1

log γ ± = − (2)(−1) × 0.509 0.06 = −0.24935 and hence γ ± = 0.5632

Example-2: (a) The Coulomb potential has a very long range. Consider a spherical
shell of electrons around a single proton, there being one electron per every square
centimeter of the surface of the shell. What force the proton exerts on the entire shell
of electrons, when the latter’s radius is (i) 10 cm (2) 1 m (3) 106 km.
(b) The shielded Coulomb potential has much shorter range. Repeat the calculation of
part (a) using the following potential with ε = 1 and λ = 10 cm .
zie − r λ
φ= e
4πεε 0 r
Solution: (a) The Coulombic potential is given by
zie
φ=
4πεε 0 r

In this case z i e is the charge on the proton (z i = 1) and r is the radius of the shell on
the surface of which the potential computed. The field is calculated from the potential
as follows
K K dφ e
E = −∇φ = − r̂ = r̂
dr 4πεε0 r 2
K
In the above equation r̂ is the unit vector in radial direction. E represents the force per
unit charge. The total charge on the shell is 4πr 2e × 10 4 . The factor 104 is added to
convert m2 of area of the shell into cm2. Hence the total force is
K e 2 × 10 4
F=
e
4πεε 0 r 2
(
r̂ − 4 πr 2
e × 10 4
)
= −
εε 0
r̂
Note that the force is independent of the radius of the shell. The reason is, although
with increase in the radius of the shell field decreases (inversely as r-square), the
number of electrons increase (directly as r-square). These opposing factors exactly
compensate each other. Hence (i) to (iii) have the same answer namely

e 2 × 104 (1.602 x10−19 ) × 104

2

F= = −12
= 2.89 × 10 − 23 N (attractive)
εε0 1 × 8.85 × 10
(b)
K dφ d⎛ e ⎞ e d ⎛ e− r / λ ⎞ ee − r / λ ⎡ r⎤
E = − r̂ = − ⎜⎜ e − r / λ ⎟⎟r̂ = − ⎜⎜ ⎟⎟r̂ = ⎢⎣1 + λ ⎥⎦ r̂
dr dr ⎝ 4πεε 0 r ⎠ 4πεε0 dr ⎝ r ⎠ 4πεε 0 r 2

The force on the shell of radius r is given by

K ee − r / λ ⎡ r⎤ e 2 × 104 ⎡ r ⎤ −r / λ
F=
4πεε 0 r 2
1 +
⎢⎣ λ ⎥⎦ 4 πr (
2
e × 10 4
r̂ =
εε 0
) ⎢⎣1 + λ ⎥⎦ e r̂

Note that now the force depends on the radius of the shell. Let us compute it for the
three cases
(1.602 × 10−19 ) 2 × 104 ⎡ 0.1⎤ − 0.1 / 0.1
(i) r = 10 cm , F = ⎢⎣1 + 0.1⎥⎦ e = 2.13 × 10− 23 N(attractive)
1 × 8.85 × 10−12

(1.602 × 10−19 ) 2 × 10 4 ⎡ 1 ⎤ −1 / 0.1

(ii) r = 1 m , F = ⎢⎣1 + 0.1⎥⎦ e = 1.45x10-26 N (attractive)
1 × 8.85 × 10−12

(1.602 × 10−19 ) 2 × 104 ⎡ 109 ⎤ −109 / 0.1

(iii) r = 106 km , F = ⎢1 + 0.1 ⎥ e =0
1 × 8.85 × 10−12 ⎣ ⎦
4.5 Activity of Solvent
Activity of solvent could be computed, once the activity of the electrolyte is
estimated. We use Gibbs-Duhem equation to achieve this. At a constant temperature
and pressure, this equation reduces to

∑ n dμ
j
j j =0 (30)

If we assign j =s to the solvent, we can rewrite the above equation as

1 Ms
− dμ s =
ns
∑ n dμ
j≠ s
j j =
Ws
∑ n dμ
j≠ s
j j = M s ∑ m jdμ j
j≠ s
(31)

In the above equation, M s is the molecular weight of the solvent ( kg.mol-1) and Ws is
its weight (in Kg). Differentiation of eq 5 gives
dμi = RTd ln(mi γ i ) (32)
Substitution into eq 31, give
 ⎡ ⎤
− dμs = RTM s ∑ m jd ln (m j γ j ) = RTM s ⎢∑ (dm j + m jd ln γ i )⎥ (33)
j≠ 0 ⎣ j≠s ⎦
Substitution of the expression for ln γ i from eq 20 into eq 33 and subsequent
differentiation gives
⎡ 2αI ⎤
− dμ s = RTM s ⎢∑ (dm j − d I )⎥ (34)
⎢⎣ j≠s 1 + Ba I
2
( ⎥⎦ )
Integration of the above equation between m j = 0 (pure solvent) to m j = m j yields

(see appendix-II for details)

−
μ s − μ 0s
RT
2
= − ln(a s ) = M s ∑ m j − αM s I 3 2σ Ba I
3
( ) (35)
j

where σ(x ) is defined as

3 ⎡ 1 ⎤
σ (x ) = ⎢ x − 2 ln (1 + x ) − 1 + x + 1⎥ (36)
x3 ⎣ ⎦
We obtain osmotic coefficient as

− ln a s
2 32
αI σ Ba I ( )
φs = = 1− 3 (37)
Ms ∑ m j ∑mj
j≠ s j≠ 0

Note that σ(x ) approaches 1 as x → 0 and hence 1 − φs is proportional to square root of

the ionic strength in dilute solutions of electrolytes.
4.6 Improvements over Debye-Huckel theory
Debye-Huckel limiting law is valid only at very low electrolyte concentrations
as is evident from the following table, which compares experimental value of γ ± of
aqueous NaCl at 250C with those computed using the limiting law
m, molal 0.001 0.002 0.005 0.01 0.02
− ln γ ± (Exp) 0.0155 0.0214 0.0327 0.0446 0.0599

− ln γ ± (Limiting D/H law) 0.0162 0.0229 0.0361 0.0510 0.0722

% error 4.52 7.01 10.4 14.3 20.5

A accurate method is to use eq 27 with appropriate value of a. Values of ‘a’ for a few
electrolytes are given below
 electrolyte HCl HBr LiCl NaCl KCl
a(nm) 0.45 0.52 0.43 0.40 0.36

The comparison of eq 27 with experiments, for NaCl ( Ba = 1.32) , is shown below.

m, molal 0.001 0.002 0.005 0.01 0.02
− ln γ ± (Exp) 0.0155 0.0214 0.0327 0.0446 0.0599

− ln γ ± (Eq 27) 0.0156 0.0216 0.0330 0.0451 0.0609

% error 0.65 0.93 0.92 1.12 1.66

Debye-Huckel theory predicts that the mean activity coefficient of an

electrolyte decreases with increasing concentration. This is true for dilute solutions.
Theory begins to deviate qualitatively from the experiments above the concentrations
of the order of 1 m. Beyond this concentration, activity coefficient begins to increase
with increase in the electrolyte concentration. For nonelectrolytes such sucrose,
activity coefficient increases with concentration right from very low concentration,
implying that the origin of this increase lies in solute-solvent interaction rather than
electrostatic (ion-ion) interaction
A plausible explanation is that hydration of ion partitions solvent into free and
bound categories. The bound water is a part of ion and hence the effective
concentration of ion in the solution should be based on the free solvent. As the
concentration of the electrolyte increases, concentration of free solvent decreases and
hence the effective concentration of electrolyte increases significantly above its
formal concentration (as expressed by molality). This implies higher activity and
hence a higher value of γ ± . This is estimates as follows ( Robinson and Stokes)
Let n be the number of moles of an electrolyte and n s be the moles of solvent.

Let n h + be the hydration number of the cation of the electrolyte (defined as the
number of molecules of solvent associated with one cation of electrolyte) and let
n h − be the hydration number of the anion of the electrolyte.
We considered two ways of computing the Gibbs free energy of the solution.
The first is the formal way in which we assume that all solvent molecules are present
as free molecules. Hence
G = n (ν + μ′+ + ν − μ′− ) + n sμ s (38)
Here primes are added to denote that these are formal chemical potentials. In the
second case, we consider the hydrated solute molecules and write
G = n (ν + μ + + ν −μ − ) + (n s − n h + ν + n − n h − ν − n )μ s (39)
express the chemical potentials in the form of activity and activity coefficients.
Thus
n (ν + ln γ + + ν − ln γ − ) + n ln x − (n h + ν + n + n h − ν − n ) ln a s = n (ν + ln γ′+ + ν − ln γ ′− ) + n ln x ′
(40)
In the above equation, we express the activity of electrolyte in terms of its mole
fractions, x and x’, because it is more accurate to do so.
We define mean hydration number as
n h +ν + + n h−ν −
nh = (41)
ν+ + ν−
and modify eq 41 to the following form
n s + n − nn h
ln γ′± = ln γ ± − n h ln a s − ln (42)
ns + n
The last term can be written in terms of the molality to obtain
1 + (m / M s )
ln γ′± = ln γ ± − n h ln a s + ln (43)
1 − (m / M s )(n h − 1)

Since ln γ ± can be obtained by DH theory, we write

z+ z− α I 1 + (m / M s )
ln γ ± = − − n h ln a s + ln (44)
1 + Ba I 1 − (m / M s )(n h − 1)

The last two terms are additive and tend to increase the value of ln γ ± . It appears to fit
the activity data well over a wide range of concentrations of electrolytes
4.7 Measurement of activity coefficient.
A variety of techniques are used to measure the activity coefficient of
electrolyte. Most of them measure activity solvent in the form of osmotic coefficient
defined by the following equation ( for electrolyte MpXq).
− ln a s − ln a s
φs = = (45)
M s ∑ m j M s m(p + q )
j≠ s

The activity coefficient of the electrolyte is then obtained using the following
equation ( See Appendix-III for the proof)
 ⎛ 1 − φs ⎞
m
− ln γ ± = 1 − φs + ∫ ⎜ ⎟dm (46)
0⎝
m ⎠

It is important to note that to obtain the mean activity coefficient of an electrolyte at a

given concentration, m, we need to measure the osmotic coefficient between 0 and m
in order to evaluate the integral in the above equation. It is preferable to split the
integral into two parts and write
⎡ γ± ⎤ m
⎛ 1 − φs ⎞
− ln ⎢ ⎥ = φs (m 0 ) − φs + ∫ ⎜ ⎟dm (47)
⎣ γ ± (m 0 ) ⎦ m 0⎝
m ⎠

Here m 0 is a sufficiently low concentration so that γ ± (m 0 ) could be obtained using eq

27.
Methods which are used to measure the activity of the solvent are: elevation of
boiling point, depression of freezing point, osmometry and isopiestic technique. The
principles of these methods are described here.
(i) Elevation of boiling point: When electrolyte is added to water, boiling point of
water increases. At the boiling point, the vapor phase is in equilibrium with the
liquid phase. Hence we can write
μ s (v ) =μ s (l ) (48)

Since the vapor phase is pure solvent, μ s (v ) = μs0 (v ) , where superscript 0 corresponds
to pure state and so
μ s0 (v ) = μ 0s (l) + RTb ln a s (49)

Here Tb is the boiling point of the solution of the electrolyte.

Rearrangement of the equation above gives
μ s0 (v ) − μ 0s (l ) Δ vap H − TΔ vapS Δ vap H Δ vapS
ln a s = = = − (50)
RTb RTb RTb R

Δ vap H and Δ vapS are respectively the enthalpy and the entropy of vaporization. For

pure water, a s = 1 and Tb = Tb* and eq 50 reduces to

Δ vap H Δ vapS
0= *
− (51)
RT b R

We assume that the values of Δ vap H , Δ vapS do not change between Tb and Tb* . This is

reasonable assumption since the difference between these two temperatures is small.
Subtraction of eq from eq yields
 Δ vap H ⎛ 1 1 ⎞ Δ H ⎛ T − T* ⎞ Δ HΔT
ln a s = ⎜⎜ − * ⎟⎟ = − vap ⎜⎜ b * b ⎟⎟ = − vap 2 b (52)
R ⎝ Tb Tb ⎠ R ⎝ Tb Tb ⎠ R Tb* ( )
In the last step we replaced Tb by Tb* in the denominator because difference between
these two temperatures is small.
The above equation allows us to obtain the activity coefficient of the solvent
in the solution of the electrolyte by measuring the elevation in its boiling point.
(ii) Depression of freezing point: When electrolyte is added to water, freezing
point of water decreases. At the freezing point, the solid phase is in equilibrium with
the liquid phase. Hence we can write
μ s (s ) =μs (l) (53)

Since the solid phase is pure solvent μ s (s ) = μ s0 (s ) , where superscript 0 corresponds to

pure state and so
μ s0 (s ) = μ s0 (l) + RTf ln a s (54)

Here Tf is the freezing point of the solution of the electrolyte.

Rearrangement of the equation above gives
μ s0 (l ) − μ 0s (s ) Δ H − TΔ fusS Δ H Δ S
ln a s = − = − fus = − fus + fus (55)
RTf RTb RTb R

Δ fus H and Δ fusS are respectively the enthalpy and the entropy of fusion. For pure

water, a s = 1 and Tf = Tf* and the equation above reduces to

Δ fus H Δ fusS
0=− + (56)
RTf* R

We assume that the values of Δ fus H , Δ fusS do not change between Tf and Tf* . This is
reasonable assumption since the difference between these two temperatures is small.
Subtraction of eq 56 from eq 55 yields

Δ fus H ⎛ 1 1 ⎞ Δ H ⎛ T∗ − T ⎞ Δ HΔT
ln a s = − ⎜⎜ − * ⎟⎟ = − fus ⎜⎜ f * f ⎟⎟ = − vap 2 f (57)
R ⎝ Tf Tf ⎠ R ⎝ Tf Tf ⎠ ( )
R Tf*

Again, in the last step we replaced Tf by Tf* in the denominator because difference

between these two temperatures is small. ΔTf = Tf∗ − Tf is positive ( since ln a s is

negative.) and hence it is called depression of the freezing point.
(iii) Osmometry: In the membrane version of this technique, two compartments of a
reservoir, one containing the test electrolyte solution and the other containing pure
solvent, are connected by a semipermeable partition. This partition allows only the
solvent molecules to pass through and not the ions of the electrolyte. Since the solvent
in the pure solvent compartment has a higher chemical potential than that in the other
compartment, it tends to permeate into the other compartment. This is prevented by
the application of pressure to the compartment containing the electrolyte. The excess
pressure needed to prevent the solvent permeation is called the osmotic pressure and
is denoted by π . It is related to the activity of the electrolyte as shown below
Let the pressure of the pure solvent compartment be p. Then the pressure of the
solution compartment would be p + π . Since the solvent in the two compartments is at
equilibrium, we can write
μ s (p + π) = μ s0 (p ) (58)
The term on the left of the equation can be expanded as
μ s (p + π) = μ 0s (p + π) + RT ln a s (59)
Combining these two equations, we obtain
μ s0 ( p) = μ 0s (p + π) + RT ln a s (60)
For pure solvent, chemical potential can be expressed as a function of temperature and
pressure by the following equation
dμ 0s = Vm dp + Sm dT (61)

where Vm is the molar volume of the solvent and Sm is the molar entropy. Since the
temperature is constant , the second term on the right hand side of the above equation.
Moreover, Vm does not depend on pressure (liquid is practically incompressible).
Hence, we can write
μ s0 (p + π) = μ s0 (p ) + Vm π (62)
Combining Eq 60 and 62 , we obtain
Vm π
ln a s = − (63)
RT
The advantage of the osmometry lies in its high sensitivity to the small concentrations
of electrolytes. For water at 250C, the value of Vm RT = 7.265x10 −9 Pa −1 . This
implies that with a high sensitivity pressure measuring device, with accuracy of say 1
Pa, we can measure a value of (− ln a s ) of the order of 10-8 (roughly corresponding to
10-8 m solution of electrolyte).
(iv)Isopiestic technique: This is a simple technique in which two electrolyte solutions
have a common vapor space. Solvent activity of one of the two electrolyte solutions
(say solution-1) is known as a function of the electrolyte concentration, whereas the
other solution (solution-2) is that of the test electrolyte. The two solutions are initially
weighed and the allowed to equilibrate. The solutions are weighed again after they
reach the equilibrium. Since solvent in the two solutions are in equilibrium, we can
write
a s (1) = a s (2) (64)

Since a s (1) is known a s (2) can be obtained. By simultaneously coupling a number

of vessels by a common vapor space, we can obtain the activity data for a range of
concentrations of the test electrolyte in one experiment. The method is not accurate
for electrolyte concentration lower than about 0.1 molal.
(iv) Electrochemical method: In this method, the activity of an electrolyte is directly
measured, in contrast to the techniques described above. It uses concentration cell in
which the two compartments have different concentrations of the same electrolyte. It
is preferable to use a cell without liquid junction (also called cell without
transport/transference). For example consider the following cell
Ag, AgCl HCl(m′) H 2 (Pt ) − (Pt )H 2 HCl(m ) AgCl, Ag (65)

(Pt)-(Pt) represents two faces of the same platinum electrode. The same hydrogen
pressure is maintained in both the cells
We consider this cell as a combination of two cells. The reactions in the right
cell are
AgCl + e − → Ag + Cl − (m)
1
H + (m ) + e − → H2
2
The overall reaction is
1
AgCl + H 2 → Ag + Cl − (m) + H + (m )
2
And the cell potential is

RT ⎛⎜ a Cl− ( m ) a H + ( m ) ⎞⎟
ΔE(R ) = ΔE 0 − ln (66)
F ⎜ fH2 ⎟
⎝ ⎠
Where ΔE 0 is the standard cell potential for the above cell. The reactions in the left
cell are
1
H + (m′) + e − → H 2
2
AgCl + e − → Ag + Cl − (m′)
The overall reaction is
1
Ag + Cl − (m′) + H + (m′) → AgCl + H 2
2
And the cell potential is

RT ⎛⎜ f H2 ⎞
⎟
ΔE(L) = − ΔE 0 − ln (67)
F ⎜ a Cl− ( m′) a H + ( m′) ⎟
⎝ ⎠
The overall cell potential is obtained by adding the potentials of the left and the tight
cell.

RT ⎛⎜ a Cl− ( m ) a H + ( m ) ⎞⎟ RT ⎛⎜ f H2 ⎞
⎟
ΔE = ΔE (R ) + ΔE (L) = ΔE 0 − ln − ΔE 0 − ln
F ⎜ f ⎟ F ⎜ a a ⎟
⎝ H2 ⎠ ⎝ Cl ( m′) H ( m′) ⎠
− +

RT ⎛⎜ a Cl− ( m ) a H + ( m ) ⎞⎟ RT ⎛⎜ mγ − (m )mγ + (m ) ⎞⎟
=− ln =− ln (68)
F ⎜ a Cl− ( m′) a H + ( m′) ⎟ F ⎜⎝ m′γ − (m′)m′γ + (m′) ⎟⎠
⎝ ⎠
2RT ⎛ mγ ± (m ) ⎞
=− ln⎜⎜ ⎟⎟
F ⎝ m′γ ± (m′) ⎠
This allows comparison of the mean activity coefficients of HCl in the two
compartments. If we use a very dilute solution in one of the two compartments then
we can use DH theory to find the mean activity coefficient in that compartment and
use the cell potential to obtain the activity coefficient of HCl in the other
compartment.
In practice, it is easier to use only one compartment of the above cell, i.e.
(Pt )H 2 HCl(m) AgCl, Ag

The cell potential is

RT ⎛⎜ a Cl− ( m ) a H + ( m ) ⎞⎟ 2RT ⎛⎜ m HCl( m ) γ ± ⎞⎟

ΔE = ΔE 0 − ln = ΔE 0 − ln (69)
F ⎜ fH2 ⎟ F ⎜ f H 2 ⎟⎠
⎝ ⎠ ⎝
Rearrangement gives
 2RT ⎛⎜ m HCl( m ) ⎞⎟ 2RT
ΔE + ln = ΔE 0 − ln (γ ± ) (70)
F ⎜ f ⎟ F
⎝ H 2 ⎠
Thus a plot of LHS versus m yields a curve ( which becomes a straight line at low
concentrations) with intercept on y axis of ΔE 0 ( 0.268 V for this cell). The curve
directly provides the relation between ln (γ ± ) and m.
Appendix I: Derivation of the Debye-Huckel Equation for the ionic activity
coefficient
Consider a solution containing ions. Molar concentration of ion i is denoted by c j∞

(the molal concentration by m j∞ ) and the charge on the ion is denoted by zj.

Electroneutrality of the solution requires that

∑z c
j
j j∞ =0 (A1)

Locally, however, ionic distribution will not be uniform. Each ion will attract
oppositely charged ions around it. Consider an ion having valancy z i .

c j∞ , z i
Central ion of valancy zi

Ionic cloud

The charge on the central ion is z ie . Due to this charge, an electric field will be
generated around it. The field will attract oppositely charged ions towards it. Ionic
cloud will be formed around the central ion. The distribution of the concentration of
ions in the cloud is the Boltzmann distribution
⎛ z eφ ⎞ ⎛ z Fφ ⎞
c j = c j∞ exp⎜⎜ − j ⎟⎟ = c j∞ exp⎜⎜ − j ⎟⎟ (A2)
⎝ kT ⎠ ⎝ RT ⎠
c j∞ , thus represents concentration of ion j, far away from the central ion.

The decay of the potential is governed by Poisson equation.

ρe
∇ 2φ = − (A3)
εε0

where ρe is the electric charge density and is given by

ρe = F∑ z jc j (A4)
j

ε0 is the permittivity of the vacuum ( = 1.602x10-19 C2.J-1.m-1) and ε is the dielectric

constant of the solution ( for dilute solutions, dielectric constant of pure water can be
used).
 Since the cloud has spherical symmetry, we can have only the radial
component of the Laplacian in Eq A3. using this simplification, we get the following
equation after combining eq A2, A3 and A4.
1 d ⎛ 2 dφ ⎞ F ⎛ z Fφ ⎞
⎜r
r dr ⎝ dr ⎠
2 ⎟=− ∑
εε 0 j
z jc j∞ exp⎜⎜ − j ⎟⎟ a≤r<∞ (A5)
⎝ RT ⎠
In the above equation a represents the average value of the sum of the radii of the pair
of hydrated ions (For a solution containing only one kind of cation and anion, sum of
the radii of the cation and the anion dominates the average since the electrostatic
repulsion prevents like charged ions to approach very close to each other.).
The boundary conditions, required to solve the above second order ordinary
differential equation are:
dφ zie
=− (A6)
dr 4πεε 0a 2
The above equation is the statement of Gauss law applied on the sphere, with center
as that of the central ion, and radius equal to a. The other boundary condition is
φ → 0 as r → ∞ (A7)
In order to solve eq A5 analytically, Debye-Huckel linearized the equation using the
approximation e x = 1 + x for small x. Thus
1 d ⎛ 2 dφ ⎞ F z Fφ
⎜r
r dr ⎝ dr ⎠
2 ⎟=− ∑
εε 0 j
z jc j∞ (1 − j )
RT
(A8)

The first term in the summation in the above equation can be neglected on the basis of
eq A1. The simplified form of eq A8 is
1 d ⎛ 2 dφ ⎞ φ
⎜r ⎟= (A9)
r 2 dr ⎝ dr ⎠ λ2
12
⎛ ⎞
⎜ εε 0 RT ⎟
where λ=⎜ 2 ⎟ (A10)
⎜ F ∑ z j c j∞ ⎟
2

⎝ j ⎠
The term λ has units of length and is called the Debye length.
The solution of eq A9, subject to boundary conditions A6 and A7 is
z i e e − (r − a ) λ
φ= (A11)
4πεε 0 r 1 + (a λ )

Note that the potential due to central ion alone is φc

 zie
φc = (A12)
4πεε0 r
The second factor in Eq A11 is less than 1 and it decays more than exponentially.
This sharp decay is caused by the oppositely charged ions in the cloud. The cloud thus
acts as a screen. The Debye length λ is an important parameter, which is indicative of
the rate of decay of potential with r. Smaller the value of λ greater is the decay. The
Debye length is therefore also called the screening length.
Eq A10 can be recast in a slightly different form. We define ionic strength of a
solution as
1
I= ∑ m j∞ z 2j
2 j
(A13)

and rewrite eq A10 as

12
⎛ εε RT ⎞
λ = ⎜⎜ 02 ⎟⎟ (A14)
⎝ 2F Iρs ⎠
where, ρs represents kg of solvent present in Liter of the solution. For dilute solution
it is nearly equal to the density of the solvent in Kg.L-1. For 0.1 m aqueous solution of
1,1 electrolyte at 250 C, the value of λ is 0.96 nm. It is inversely proportional to
square root of ionic strength.
The nonideal contribution to μi , i.e. RT ln γ i equals the Nav times the reversible
electrostatic work required to be done to form the cloud of ions around a single
central ion of species i. The contribution to the potential due to cloud is

z i e ⎛ e − (r − a ) λ ⎞
φ − φc = ⎜⎜ − 1⎟⎟ (A15)
4πεε 0 r ⎝ 1 + (a λ ) ⎠

The value of this potential at r = a is

zie ⎛ 1 ⎞
(φ − φc ) r = a = − ⎜⎜ ⎟ (A16)
4πεε0λ ⎝ 1 + (a λ ) ⎟⎠

To obtain the reversible work, we charge the central ion i in infinitesimal steps. We
bring charge dq from infinity to r = a (we assume the central ion to be of radius a.
Once we bring the charge at the surface of the ion it is as good as bringing it to its
center.). Suppose we have charged the central ion to a charge q. The extra work done
to increase its charge to q+dq equals (φ − φc ) r = a , q dq
 q ⎛ 1 ⎞
dw i = − ⎜⎜ ⎟dq (A17)
4πεε0λ ⎝ 1 + (a λ ) ⎟⎠

(note that at standard state we need to do work equal to φc r = a dq ). The total work to

charge the central ion is

zie

wi = − ∫
q ⎛
⎜⎜
1 ⎞
⎟⎟dq = −
(zi e ) ⎛
2
⎜⎜
1 ⎞
⎟ (A18)
0
4πεε0λ ⎝ 1 + (a λ ) ⎠ 8πεε0λ ⎝ 1 + (a λ ) ⎟⎠

Hence

N av (z i e ) ⎛ ⎞ Fezi2 ⎛ ⎞
2
1 1
RT ln γ i = − ⎜⎜ ⎟⎟ = − ⎜⎜ ⎟ (A19)
8πεε0λ ⎝ 1 + (a λ ) ⎠ 8πεε 0λ ⎝ 1 + (a λ ) ⎟⎠

Fezi2 ⎛ 1 ⎞
or ln γ i = − ⎜⎜ ⎟ (A20)
8πεε 0 RTλ ⎝ 1 + (a λ ) ⎟⎠

The above equation can be recast in a more convenient form after substituting the
expression for λ from A10. The resulting form is

z i2α I
ln γ i = − (A21)
1 + Ba I
where

F2e 2ρs
α= (A22)
8π(εε 0 RT )
32

F2 2ρs
and B= (A23)
(εε0RT )1 2

Appendix II : Derivation of Equation 35

We begin with eq 33
⎡ ⎤
− dμs = RTM s ⎢∑ (dm j + m jd ln γ i )⎥ (A24)
⎣ j≠s ⎦
From eq 20 we obtain

z i2 α I
ln γ i = − (A25)
1 + Ba I
Substitution of A25 into A24 yields
⎡ ⎛ I ⎞⎟ ⎤
− dμs = RTM s ⎢∑ (dm j − m jz 2j αd⎜⎜ ⎟) ⎥ (A26)
⎢⎣ j≠ s ⎝ 1 + Ba I ⎠ ⎥⎦
 1
We can rewrite equation A24 by recalling that I = ∑ m jz 2j and get
2 j≠s

⎡ 2αI ⎤
− dμ s = RTM s ⎢∑ (dm j ) − d I ⎥ (A27)
⎢⎣ j≠s 1 + Ba I
2
( ⎥⎦ )
We integrate the equation between I = 0 (pure solvent) to I = I and m j = m j

μs ⎡ mj I
2αI ⎤
− ∫ dμ 0 = RTM s ⎢∑ ∫ dm j − ∫ d I ⎥ (A28)
μ s0 ⎢⎣ j≠ s 0 0 1 + Ba I
2
(⎥⎦ )
or

μ s − μ s0
I
I
− = M s ∑ m j − 2αM s ∫ d I (A29)
RT j≠ 0 0 1 + Ba I
2
( )
Consider the last term, we write 1 + Ba I = x . From this we get
x −1 dx
I= or d I =
Ba Ba
Ba I +1 Ba I +1
( x − 1) 2
I
I 1 1 2 1
∫ (1 + Ba I )
0
2
d I=
(Ba )3 ∫ 1
x 2
dx =
(Ba )3 ∫
1
1− + dx
x x2
Ba I +1

=
1
(Ba )3
⎡ 1⎤
⎢⎣ x − 2 ln x − x ⎥⎦
⎡
=
1 1 ⎤
(
⎢ Ba I − 2 ln Ba I + 1 − Ba I + 1 + 1⎥ (A30)
(Ba )3
)
1⎣ ⎦
⎡
( ⎤ I
) ( )
32 32
3I 1
= Ba I − 2 ln Ba I + 1 − + 1⎥ = σ Ba I
(
3 Ba I ⎣
3 ⎢
) Ba I + 1 ⎦ 3

Where
3 ⎡ 1 ⎤
σ (x ) = x − 2 ln (1 + x ) − +1 (A31)
x 3 ⎢⎣ 1 + x ⎥⎦
Combining eq A30 and A29, we obtain

−
μ s − μ s0
RT
2
= M s ∑ m j − αM s I3 2 σ Ba I
3
( ) (A32)
j≠ s

3 ⎡ 1 ⎤
σ (x ) = 3 ⎢
x − 2 ln (1 + x ) − + 1⎥ (A33)
x ⎣ 1+ x ⎦
Appendix III : Derivation of Equation
We begin with eq 33
⎡ ⎤
− dμs = RTM s ∑ m jd ln (m j γ j ) = RTM s ⎢∑ (dm j + m jd ln γ i )⎥ (A34)
j≠ 0 ⎣ j≠s ⎦
For a p:q electrolyte the above equation reduces to
 − dμs = RTM s [(p + q )dm + m(pd ln γ + + q ln γ − )]
(A34)
= RTM s (p + q )[dm + md ln γ ± ]

Also dμ s = RTd ln a s (A35)

Combining these two equations, we obtain
− d ln a s = M s (p + q )[dm + md ln γ ± ] (A36)

− ln a s
Now φs = (A37)
M s m(p + q )

or − d ln a s = M s (p + q )d (mφs ) (A38)
Combining A-36 and A-38, we obtain
dm + md ln γ ± = d (mφs )

1 1 ⎛ 1 − φs ⎞
or − d ln γ ± = − d (mφs ) + dm = −dφs + ⎜ ⎟dm (A39)
m m ⎝ m ⎠
Integrating the equation from m = 0 to m = m and noting that ln γ ± = 0 and φs = 1 at

m = 0 , we get

⎛ 1 − φs ⎞
m
− ln γ ± = 1 − φs + ∫ ⎜ ⎟dm (A40)
0⎝
m ⎠
 Lecture-5
Ion Transport in Solution-I
1 Ionic mobility
We can treat electrochemical potential of an ion in the manner similar to any other
potential and assert that electrochemical force acting on the ion ( f i , N.ion-1) equals
negative of the gradient of electrochemical potential μi (divided by Avogadro
number, because the electrochemical potential is based on one mole of the ion)
1
fi = − ∇ μi (1)
N av
The electrochemical potential can be written as the sum of the chemical and the
electrical potentials.
μi = μi + z i Fφ (2)
Here, z i is the valancy of the ion (sign being included) and φ is the electric potential.
This motion of the ion will be opposed by the fluid drag. This drag is directly
proportional to velocity of the ion and is given by Stokes equation
f di = 6πηa i vi (3)

In the above equation η represents the fluid viscosity (Pa.s) and ai is the radius of the
hydrated ion.
At certain velocity, called terminal velocity v ti , the electrochemical force and
the drag force become equal.
K 1 K
6πηa i v ti = − ∇ μi (4)
N av
It takes very short time for an ion to attain the terminal velocity even when its initial
velocity is very different. It is, therefore, reasonable to assume that ion moves at all
times at its terminal velocity.
We define the mobility of an ion by the following equation
K K
v ti = − u i∇ μi (5)
Thus we see that mobility of an ion represents the velocity it will attain per unit
electrochemical potential gradient. Units of mobility are: m2.mol.s-1.J-1.
Comparison of eq 4 and 5 gives us the following relation for the mobility.
1
ui = (6)
6πηa i N av

1
It is seen from the above equation that the mobility is inversely proportional to fluid
viscosity and radius of the ion. However, it does not depend on the charge of the ion.
However, for a given field strength, the terminal velocity does depend on the charge
of the ion.
We could rewrite eq 5 as
K K K K
v ti = − u i (∇μi + z i F∇φ) = −u i (RT∇ ln a i + z i F∇φ)
K
K K (7)
= −u i (RT∇ ln (c i γ i ) + z i F∇φ)

K ⎛ d ln γ i ⎞ K
Writing ∇ ln γ i = ⎜⎜ ⎟⎟∇ ln c i (8)
⎝ d ln c i ⎠
We can rewrite eq 7 as

K ⎡ ⎛ d ln γ i ⎞ K K ⎤
v ti = − ⎢RTu i ⎜⎜1 + ⎟⎟∇ ln c i + z i u i F∇φ⎥ (9)
⎣ ⎝ d ln ci ⎠ ⎦
We define the Diffusion coefficient as
⎛ d ln γ i ⎞
D i = RTu i ⎜⎜1 + ⎟⎟ (10)
⎝ d ln c i ⎠
and rewrite eq 9 as
K K
v ti = −(Di∇ ln ci + z i u i F∇φ)
K
(11)

Note that D i depends on the concentration of the electrolyte, although the

mobility may not depend on it. At very low concentration of the electrolyte, γ i → 1 ,
and eq 10 reduces to the Nernst-Einstein equation.
Di
ui = (12)
RT
Combining eq 6 and 12, we obtain the following relation for D i (Stokes-Einstein
relation)
kT
Di = (13)
6πηa i
where k is the Boltzmann constant.
The above set of equations is valid only for very dilute solutions. For the
concentrated solutions, motion of the given ion not only depends on its own
electrochemical potential gradient but also the electrochemical potential gradients of
the rest of the ions in its vicinity, because the motion of the neighboring ions produces
motion in the solvent which in turn affects the motion of the given ion.

2
Example-1: Determine the terminal velocity of sodium ion at 250C in an aqueous
solution having uniform composition, subjected to an electric field of magnitude 100
V.m-1. Also compute the diffusion coefficient of sodium ion assuming the solution to
be very dilute. Mobility of sodium ion in solution at 250C has the value of 5.384x10-13
m2.mol.s-1.J-1.
Solution: We use eq 5 to compute the terminal velocity. Here
G K K K
∇ μi = ∇μ i + z i F∇φ = z i F∇φ
K
The term ∇μ i vanishes because the composition is uniform.
The terminal velocity is now obtained from eq 5
K K K
v ti = u i ∇ μi = u i z i F ∇φ

Substituting the values

K
v ti =5.384x10-13 x1x 96488x100 = 5.19x10-6 m.s-1 = 5.19 μm.s-1.

Diffusion coefficient = u i RT = 5.384x10-13x 8.314x 298 = 1.334x10-9 m2.s-1.

Mobilities of various ions, along with their diffusion coefficients, in very dilute
solutions, are listed in Table-1 below.
Table-1 Mobilities and Diffusion Coefficients of Ions in water at 250C
Ion Dix109 uix1013 Ion Dix109 uix1013
m2.s-1 m2.mol.s-1.J-1 m2.s-1 m2.mol.s-1.J-1
H+ 9.312 37.59 OH- 5.260 21.23
Li+ 1.030 4.157 Cl- 2.032 8.201
Na+ 1.334 5.384 Br- 2.084 8.411
K+ 1.957 7.899 I- 2.044 8.250
NH4+ 1.954 7.887 NO3- 1.902 7.677
Ag+ 1.648 6.652 HCO3- 1.105 4.460
Mg2+ 0.7063 2.851 CH3CO2- 1.089 4.395
Ca2+ 0.7920 3.713 SO42- 1.065 4.299
Sr2+ 0.7914 3.194 Fe(CN)63- 0.896 3.62
Ba2+ 0.8471 3.419 Fe(CN)64- 0.739 2.98
Cu2+ 0.72 2.91 ClO4- 1.792 7.233
Zn2+ 0.71 2.87 BrO3- 1.485 5.994
La3+ 0.617 2.49 HSO4- 1.33 5.37

3
2 Nernst-Planck equation
We obtain the flux of ion (mol.m-2.s-1) relative to the mean fluid motion as
K K
J i = v ti c i (14)
Using eq 11, we can rewrite the above equation as
K K K K K
J i = −c i (D i ∇ ln ci + z i u i F∇φ) = −D i ∇c i − Fz i u i ci ∇φ (15)
The above equation is called the Nernst-Planck equation.
For dilute solutions, the Nernst-Planck equation is simplified using Nernst-Einstein
equation to
K ⎛K Fz c K ⎞
J i = − D i ⎜ ∇c i + i i ∇φ ⎟ (16)
⎝ RT ⎠
The first term on the right of eq 15 is called the diffusion flux (flux
contributed by the concentration gradient) and the second is called the migration flux
(that contributed by the electric field).
K
In the Nernst-Planck equation, Ji is the flux of ion i in the frame which moves
K
with the fluid. In the stationary frame, the flux of i is given by N i and is related to
K
J i as follows
K K K
Ji = N i − c i v (17)
K
In the above equation, c i v represents the flux of ion i due to bulk transport
(convective flux). Subtraction of this bulk flux from the total flux yields the diffusion
flux.
K
Substituting Ji from eq 15 into eq 17 and rearranging the resulting equation, we
obtain
K K K K
N i = −Di∇ci − Fz i u i ci∇φ + ci v (18)
Thus the total flux in stationary frame consists of diffusion flux, migration flux and
the flux due to convection.
3 Equation of Continuity
Continuity equation is a material balance for ion i at any point in the fluid and
can be written as
∂ci K K
= −∇ ⋅ N i + R i (19)
∂t
Each term has units of mol of i per unit volume of the liquid per unit time. The term
on the left is the rate of accumulation of i. This accumulation occurs because of the

4
net input of the flux of i at the point (first term on the right)and rate of generation of i
due to homogeneous chemical reaction (second term). R i excludes generation by
charge transfer reactions. These reactions do not occur in the bulk, but only at the
electrode surface. Their effect is incorporated through boundary conditions.
K
Substituting N i from eq 18 into eq 19, we obtain
∂ci K K K
= ∇ ⋅ (Di∇ci + Fz i u ici∇φ − ci v ) + R i
K
(20)
∂t
In dilute solution, we assume the diffusion coefficient of an ion is independent of
composition. Hence
∂ci G K G K
= Di∇ 2ci + Fz i u i∇ ⋅ (ci∇φ) − ∇ ⋅ (ci v ) + R i (21)
∂t
We expand the last-but-one term as
G K K K K K
∇ ⋅ (ci v ) = ci∇ ⋅ v + v ⋅ ∇ci (22)
We can consider the fluid to be incompressible and drop the first term on the right of
the above equation. Combining the resulting equation with eq 21, we obtain
∂ci G K K G
= Di∇ 2ci + Fzi u i∇ ⋅ (ci∇φ) − v ⋅ ∇ci + R i (23)
∂t
This is more convenient form of the equation of continuity.
4 Electroneutrality
Another equation which is obeyed at every point in the electric field is the
Poisson equation
ρc
∇ 2φ = − (24)
εε 0
The charge density is related to the concentration of ions by the following equation
ρc = F∑ z ici (25)
i

Combining eq 24 and 25, we obtain

F
∇ 2φ = − ∑ z i ci
εε 0 i
(26)

Conversely, we can write

εε0 2
∑z c
i
i i =−
F
∇φ (27)

5
The value of the constant εε0 F has a value of 7.18x10-15 mol.V-1.m-1. So unless the
potential variation is such that its Laplacian has a value of the order of 1014 V.m-2, we
can replace the right hand side by zero i.e.

∑z c
i
i i =0 (28)

Eq 28 is called the electroneutrality condition. This condition is obeyed in most

situations. The exceptions are the double layer regions where the order magnitude of
the Laplacian can easily exceed 1014 V.m-2. In electrochemistry, we shall accept this
as a valid condition except when we treat double layers.
In accepting electroneutrality, what we are saying is that, although in
microscopic sense there is always a deviation from electroneutrality its effect dies out
on macroscopic scale.
Once, we accept electroneutrality condition, we discard Poisson equation. It is
wrong to say that Poisson equation reduces to Laplace equation due to
electroneutrality (right hand side of eq 26 is zero). We must use some other equation
to determine the potential (which may reduce to the Laplace equation under certain
conditions).
5 Electrical Conduction
When ion moves, it carries electric current due to its charge. The contribution
K
of ion i to the current density is given by Fz i N i . The total electric current density is,
therefore given by
K K
i = F∑ z i N i (29)
i
K
Substituting the expression for N i from eq 18 into eq 29, we get
K K K K
i = −F∑ z i Di∇ci − F2∇φ∑ z i2 u i c i + Fv∑ z i c i (30)
i i i

The last term is zero due to electroneutrality. Hence we obtain

K K K
i = −F∑ z i Di∇ci − F2∇φ∑ z i2 u i ci (31)
i i

We define the conductivity of the solution κ as

κ = F 2 ∑ z i2 u i c i (32)
i

and rewrite eq 31 as
K K K
i = −F∑ z i Di∇ci − κ∇φ (33)
i

6
Conductivity has units of S.m-1.
If the concentration is uniform, the first term on the right vanishes and we obtain
K K
i = − κ∇φ (34)
Eq 34 is a representation of the Ohm’s law for solution of electrolytes at uniform
concentration. When the concentration gradients exist, eq 34 gives the contribution to
current by migration alone.
In absence of concentration gradients,
K ⎛ ⎞K
i = −⎜ F2 ∑ z i2 u i c i ⎟∇φ (35)
⎝ i ⎠
From the above equation we can obtain the current carried by species j as
K K
ij = − F2 z 2j u jc j∇φ (36)

we define the transport number (transference number) of species j, t j , as the fraction

of the total current carried by the ion j, i.e.

z 2j u jc j
tj = (37)
∑ zi2u ici
i

If we sum eq 37 over j we get

∑z u c 2
j j j

∑t = z u c j
=1 (38)
∑
j 2
j i i i
i

6 Diffusion Potential
Note that when the concentration gradients exist in the solution, Ohm’s law is
no longer valid. Even the direction of the current density would not match that of the
electric field. In conductivity measurements, an alternating current is used in order to
prevent buildup of concentration gradients in the solution.
It is interesting to rearrange eq 35 as
K
K i F K
∇φ = − − ∑ z i D i ∇c i (39)
κ κ i
If there is no current in the solution, eq 39 reduces to
K F K
∇φ K
i =0
=− ∑
κ i
z i D i ∇c i (40)

7
The above equation indicates that the even in the absence of current, electric potential
gradient may exist in the solution. This potential gradient is called the diffusion
potential. If the diffusion coefficients of all ions are equal then
K F K F K⎛ ⎞
∇φ Ki =0 = − D∑ z i∇ci = D∇⎜ ∑ z i ci ⎟ = 0 (41)
κ i κ ⎝ i ⎠
where we have used electroneutrality condition in the last step. Diffusion potential is,
therefore, absent when the diffusion coefficients of the ions are equal.
The reason for the development of diffusion potential is that in absence of
current, there should be no flow of charge. When two ions diffuse at different speed,
there would be current generation. Development of electric field accelerates ions of
one charge and retards the ions of the opposite charge. A proper magnitude should
make them both move with the same speed so that the current is zero.
We can obtain an alternative expression for the diffusion potential as follows.
We replace D i by u i RT in eq 39, on the basis of the Nernst-Einstein equation, and get
K
K i FRT K
∇φ = − −
κ κ i
∑ z i u i ∇c i (42)

We further modify the second term of the above equation by replacing κ in terms of
mobilities using eq 32
K K
K i RT z i2 u i ci∇ ln ci
∇φ = − − ∑ z
κ F∑ z 2j u jc j i
(43)
i
j

Using the definition of the transference number (eq 37), we can rewrite the above
equation as
K K
K i RT t i∇ ln ci
∇φ = − −
κ
∑ z
F i
(44)
i

This gives us an alternative form of diffusion potential as

K
K RT t i∇ ln c i
∇φ K = −
i =0 ∑ z
F i
(45)
i

Example-2 Calculate the diffusion potential in 1 mol.m-3 HCl solution when the
concentration gradient of 1 kmol.m-4 is imposed.
Solution: We first calculate the transport numbers for hydrogen and chloride ions
using eq 37

8
 z 2j u jc j
tj =
∑ zi2u ici
i

The mobilities are obtained from Table-1 as u H + = 37.59 m2.mol.s-1.J-1 and

u Cl − = 8.201 m2.mol.s-1.J-1. Hence

(1) 2 (37.59x10 −13 )(1)

tH+ = = 0.821
(1)2 (37.59 x10 −13 )(1) + (−1)2 (8.201x10−13 )(1)
t Cl − = 1 − 0.821 = 0.179

The diffusion potential is the given by eq 46 as

K
K RT t i∇ ln ci ⎛ 0.821 0.179 ⎞ 1000
∇φ Ki = 0 = − ∑ = −0.0257⎜⎜ + ⎟⎟ ×
(−1) ⎠
= −16.49 V.m −1
F i zi ⎝ 1 1
7 Laplace equation:
We multiply the continuity eq 19 by zi and carry out summation over all ions
to obtain
∂ ∑ z ici K K
i
= −∇ ⋅ ∑ z i N i + ∑ z i R i (46)
∂t i i

The left hand side term is zero in the light of the electroneutrality condition 28. The
second term on the right is also zero since no charge is generated during reaction. It is
only distributed among the reactants. Eq 46, therefore, reduces to
K K
∇⋅i = 0 (47)
We have used eq 29 for the current density to arrive at eq 47.
Eq 47 is the charge conservation equation and states that current moving in at
a point should be equal to that moving out.
Substitution of eq 33 for current density into eq 47, we obtain
K
− F∑ z i∇ ⋅ (Di∇ci ) − κ∇ 2φ = 0 (48)
i

When the concentrations in the solution are uniform, the first term is zero and we
obtain the Laplace equation.
∇ 2φ = 0 (49)
Laplace equation, therefore, describes potential distribution in a system in which the
concentration gradients are absent. It is also used as an approximation where, electric
filed is strong enough so that diffusion contribution to current can be neglected in
comparison to migration contribution.

9
We apply the concepts developed in the previous sections to some specific situations.
8. Solution of a Single electrolyte
Consider a strong electrolyte M ν + X ν − . Let c, c + and c − be the concentration of
salt, cations and anions, respectively, in the solution. We then have
c+ c−
c= = (50)
ν+ ν−
Electroneutrality of the salt in the solid state gives
ν+ z+ + ν−z− = 0 (51)
where z + and z − are the valencies of the cation and the anion respectively. These
equations automatically satisfy the electroneutrality condition.
c+ z + + c− z − = 0 (52)
We now use the continuity equation
∂c i G K K G
= D i∇ 2 ci + Fz i u i∇ ⋅ (ci ∇φ ) − v ⋅ ∇(c i ) + R i (53)
∂t
We assume that no reaction occurs in the bulk. We write the equation for both the
cations and the anions
∂c + G K K G
= D +∇ 2c + + Fz + u +∇ ⋅ (c + ∇φ) − v ⋅ ∇c + (54)
∂t
∂c − G K K G
= D −∇ 2c − + Fz − u −∇ ⋅ (c −∇φ) − v ⋅ ∇c − (55)
∂t
Using eq 1, we replace c + and c − in terms of c, the concentration of the salt and get
∂c G K K G
= D +∇ 2c + Fz + u + ∇ ⋅ (c∇φ ) − v ⋅ ∇c (56)
∂t
∂c G K K G
= D −∇ 2c + Fz − u −∇ ⋅ (c∇φ ) − v ⋅ ∇c (57)
∂t
Subtraction of eq 57 from eq 56 gives
G K
(D + − D− )∇ 2c + F(z + u + − z − u − )∇ ⋅ (c∇φ) = 0 (58)
G K
We use this equation to eliminate ∇ ⋅ (c∇φ ) from eq 56 and obtain
∂c z u (D − D − ) 2 K G
= D +∇ 2 c + + + + ∇ c − v ⋅ ∇c (59)
∂t (z + u + − z − u − )
Simplification of eq 59 gives
∂c K G
= D∇ 2c − v ⋅ ∇c (60)
∂t

10
where D is effective diffusion coefficient of the electrolyte and is given by
z + u +D− − z −u −D+
D= (61)
(z + u + − z − u − )
For very dilute solutions, we use the Nernst-Einstein relationship, to simplify eq 61 to
yield
D + D − (z + − z − )
D= (62)
(z + D + − z − D − )
We see from eq 60 that the electrolyte behaves as if it is a neutral species. The
effective diffusion coefficient lies somewhere between that of the cation and the
anion. Note that the equation is valid even when current is flowing through the
solution.
Using the definition of transport number
z 2j u jc j
tj = (63)
∑ zi2u ici
i

And the Nernst-Einstein equation we can write (for very dilute solutions)
z 2+ D + ν + c z 2− D −ν − c
t+ = t− = (64)
z 2+ D + ν + c + z 2− D −ν −c z 2+ D + ν + c + z 2− D − ν − c
Eq 64 can be simplified further using eq 52 to obtain
z + D+ z−D−
t+ = t− = (65)
z+ D+ − z −D− z + D + − z − D−
Thus for very dilute solutions of a single electrolyte, the transport numbers are
independent of the electrolyte concentration.
Example-3: Determine the effective diffusion coefficient of sodium sulfate in its very
dilute aqueous solution at 250C. Also determine the transport numbers for the ions .
The diffusion coefficient of sodium ion 1.334x10-9 m2.s-1 and that of sulfate ion is
1.065x10-9 m2.s-1
Solution: D+ = 1.334x10-9 m2.s-1, z+ =1, D-=1.065x10-9 m2.s-1, z- = -2
D + D − (z + − z − ) 1.334 × 10−9 × 1.065 × 10−9 (1 + 2)
D= = = 1.230 × 10− 9 m2.s-1
(z + D + − z − D − ) (1)(1.334 × 10 ) − (−2)(1.065 × 10 )
−9 −9

z +D+ 1x1.334 x10−9

t+ = = = 0.385
z + D + − z − D − 1x1.334 x10− 9 − (−2) x1.065x10− 9

t − = 1 − t + = 1 − 0.385 = 0.615

11
Note that sulfate ions carry more share of current compared to sodium ions in spite of
having lower diffusion coefficient. The reason is their charge is double that of
sodium.
Once the concentration distribution is known from the solution of eq 60, the
potential distribution can be obtained using the eq 31 for the current density.
K K K
i = −F∑ z i Di∇ci − F2∇φ∑ z i2 u i ci (66)
i i

For a single salt solution, the above equation can be simplified with the help of eq 50
to yield
K
i K K
− = (D + − D − )∇c + (z + u + − z − u − )Fc∇φ (67)
z +ν+F
This equation allows us to determine the potential distribution, provided the current
density distribution is known.
In the absence of current, eq 67 reduces to
K D+ − D− K
∇φ = − ∇ ln c (69)
F(z + u + − z − u − )
This is the diffusion potential.
Example-4 A quiescent aqueous solution of a salt MX is situated between two flat
parallel plates of metal M. The initial concentration of the salt is c0. A constant direct
current is passed between the plates under such a condition that the only electrode
reactions are dissolution of the anode and deposition of M on the cathode. Estimate,
the potential profile and the concentration profile of MX in the solution at steady state
and also the maximum possible current density. Assume the solution to be dilute and
ignore changes of temperature and physical properties.
Solution: The figure below shows the electrolytic cell.
The reactions are
M + + e = ⇔ M (cathode) (70)
M ⇔ M + + e = (anode) (71)
We begin with the continuity equation for ion i (eq 19)
∂ci K K
= −∇ ⋅ Ni + R i (72)
∂t
Since we consider steady state, and also no homogeneous reaction, the first and the
third terms are zero and we are left with
K K
∇ ⋅ Ni = 0 (73)

12
 M+

Anode Cathode

x
L

We neglect the end effects and assume that the gradient of the flux exists only in the
x direction. This allows us to write equation 23 as
dNix
=0 (74)
dx
where N ix is the x component of the flux of ion i. We can write the above equation
for cation M+ and anion X-. Thus
d(N + )x
=0 (75)
dx
d(N − )x
=0 (76)
dx
Integration of the above equations yield
(N + )x = C1 (N − )x = C2 (77)
where C1 and C2 are constants.
Since anions X- do not undergo any electron transfer at the anode, its flux at
anode must be zero.
(N − )x = 0 = 0 (78)

This gives us C 2 = 0 and hence (N − )x = 0 at all points in the cell. We can also see
that since cation takes part in electron transfer reaction, we have
(N + )x = 0 = ( N + )x = L = C1 (79)
Let this constant value be denoted by N+.
The current density is given by eq 29

13
 K K
i = F∑ z i N i (80)
i

This can be expanded to give

i = FN + (81)
K
where i is the x component of the current density, i .
Next, we write the Nernst-Planck equation (eq 18)
K K K K
N i = −Di∇ci − Fz i u i ci∇φ + ci v (82)
Since there is zero fluid velocity, we drop the last term. We also write the x-
components of fluxes for both the cation and the anion.
dc + dφ
N + = −D + − Fu + c + (83)
dx dx
dc − dφ
N − = 0 = −D − + Fu − c − (84)
dx dx
Eq 84 can be simplified to yield
dφ D − dc −
= (85)
dx Fu − c − dx

dφ
We eliminate from eq 83 using eq 85 and obtain
dx
dc + D − u + c + dc +
N + = −D+ − (86)
dx u − c − dx
Using Nernst-Einstein equation we replace mobility ratio by the corresponding ratio
of the diffusion coefficients. In addition, we use electroneutrality to get: c − = c + .
Using these in eq 86, we obtain
dc + dc dc
N + = −D + − D + + = −2D + + (87)
dx dx dx
Since N+ is constant, we can integrate the above equation to obtain the concentration
profile of the cations as
N+
c+ = − x + C3 (88)
2D +
Since the total quantity of the metal ions must remain constant, we can write
L L
1 N+ N L
L ∫0 2D + L ∫0
c0 = c + dx = − xdx + C3 = − + + C3 (89)
4D +

This gives us

14
 N+L
C3 = c 0 + (90)
4D +
Substituting C3 from eq 90 into eq 88, we obtain
N + (L − 2x )
c + = c0 + (91)
4D +
The concentration profile is thus a straight line with the concentration at x = L / 2
equal to c0 .
Electroneutrality dictates that the concentration profile of the anion, that is
identical to that of the cation and hence is also given by eq 91, with c + replaced by c − .
Thus we have a situation where the anion develops a concentration profile in the
absence of flux. Such a situation never arises in nonionic systems.
The potential distribution can be obtained from eq 85, which can be modified
as
dφ RT d ln c +
= (92)
dx F dx
Integration yields
RT
φ= ln c + + C4 (93)
F
The potential difference across the cell is obtained from the above equation as

RT c + x = 0
Δφ = φ x = 0 − φ x = L = ln (94)
F c+ x = L

Note that this is a concentration cell and hence the potential is related to activity of the
potential determining ion (cation in this case).
Using eq 91, we can rewrite eq 94 as

RT ⎛ c0 + N + L 4D + ⎞
Δφ = φ x = 0 − φ x = L = ln⎜ ⎟ (95)
F ⎜⎝ c 0 − N + L 4D + ⎟⎠

It is seen from eq 91 that the lowest value of c + occurs at x = L . A limiting situation

is reached if the concentration c + x=L

becomes zero. Eq 91, then reduces to

N+ L 4c0 D +
0 = c0 − or N+ = (96)
4D + L
This the highest value of the flux attainable in the cell. It is called the limiting flux.
The highest value of the current density that can be drawn from the cell is called the
limiting current density. It can be obtained by combining eq 81 with eq 96

15
 4Fc0 D +
i lim = (97)
L
It is not possible to increase the current density beyond the limiting value by
increasing the cell potential. This limiting current density is cause by the
concentration gradient. Whenever a concentration gradient exists in a cell, we say
concentration polarization has occurred.
9 Effect of agitation on transport ions
Example-5 In the cell described in the last example, agitation is provided at the
intensity which is sufficient to reduce the diffusion film thickness to δ = L / 1000 .
Calculate the limiting current density. Assume the solution to be dilute and ignore
changes of temperature and physical properties.
Solution:

C x =0

C0 Diffusion film
Diffusion film
M+ C x=L

Anode Cathode

In this case, the agitation allows the concentration to remain uniform, equal to c0 in
the bulk of the fluid. The diffusion resistance is confined to a thin film of thickness δ
Eq 83 and 84 holds in the films
dc + dφ
N + = −D + − Fu + c + (98)
dx dx
dc − dφ
N − = −D− + Fu − c − =0 (99)
dx dx
dφ
Eliminating between eq 1 and 2, we obtain
dx

16
 dc +
N + = −2 D + (100)
dx
This equation on integration gives
N+
c+ = − x + C3 (101)
2D +

The boundary condition for the film around the anode is c + = c0 at x = δ . This gives

N+
C3 = c 0 + δ (102)
2D +
From eq 101 and 102, we obtain
N+
c+ = c0 + (δ − x ) (103)
2D +
This is the concentration profile in the film around the anode. The concentration of
the cations in the film is greater than that in the bulk. This is understandable since the
movement of the cations is from the electrode to the bulk.
Eq 101 is also valid in the diffusion film at the cathode. The boundary
condition is c + = c 0 at x = L − δ

N+
C3 = c0 + (L − δ ) (104)
2D +
Using this in eq 101, we get
N+
c + = c0 + (L − x − δ ) (105)
2D +
The potential gradient in the film is given by eq 92
dφ RT d ln c +
= (106)
dx F dx
In the film around the anode, the potential drop is
⎛ ⎛ ⎞⎞
⎜ c 0 + ⎜⎜ N + δ ⎟⎟ ⎟
RT ⎛⎜ c + x =0 ⎞⎟ RT ⎜ ⎝ 2D + ⎠ ⎟ = RT ln⎛⎜1 + N + δ ⎞⎟
Δφ1 = ln = ln⎜ ⎟ F ⎜ 2Dc ⎟ (107)
⎜ ⎟
F ⎝ c + x =δ ⎠ F ⎜ c0 ⎝ 0 ⎠
⎟
⎜ ⎟
⎝ ⎠
This equation can also be written in terms of the current density using eq 81
i = FN + (108)
The resulting equation is

17
 RT ⎛ iδ ⎞
Δφ1 = ⎜⎜1 + ⎟ (109)
F ⎝ 2FDc0 ⎟⎠

Similarly the potential drop in the film around the cathode is

RT ⎛⎜ c 0 ⎞⎟ RT ⎛ iδ ⎞
Δφ 2 = ln =− ln⎜⎜1 − ⎟ (110)
F ⎜⎝ c + x = L ⎟⎠ F ⎝ 2FD + c0 ⎟⎠

In the bulk, the concentration is uniform and hence the charge transport occurs only
by migration. The current density can be obtained as
dφ
i = −κ (111)
dx
where
( )
κ = F2 ∑ z i2 u i ci = F2 z 2+ u + c + + z 2− u − c − = F 2c 0 (u + + u − ) (112)
i

The total potential drop in the bulk is obtained by integration of eq 111

iL
Δφ3 = (113)
κ
Thus, we see that in the bulk, both the cations and anions carry current.
The total potential drop will be the sum of the three contributions

RT ⎛ 1 + iδ 2FD + c 0 ⎞ iL
Δφ = Δφ1 + Δφ 2 + Δφ3 = ln⎜ ⎟+ (114)
F ⎜⎝ 1 − iδ 2FD + c 0 ⎟⎠ κ

The limiting situation occurs when the concentration of the cation drops to zero at the
cathode. c + = 0 at x = L , in that case eq 105 gives
N +δ 2D + c 0
c0 − = 0 or N + = (115)
2D + δ
The limiting current is then given by
2FD + c 0
i lim = FN + = (116)
δ
We compare this to the value in the absence of agitation example -4, i.e.
2Fc0 D +
i lim = (117)
(L 2)
Comparison between eq 116 and 117 shows that due to agitation, the limiting current
has increased by the factor of L 2δ , (500 fold in this case) .
When L is large, the ohmic potential drop iL κ dominates the overall drop of
the potential across the cell. This situation can arise in industrial electrochemical

18
reactors where the electrodes are large and hence placing them near each other is
difficult. Note that it is necessary to provide agitation in the gap between the
electrodes, otherwise δ will increase (which will not only increase Δφ1 + Δφ 2 , but will
also reduce the limiting current). A special attention should therefore be given to
electrode geometry and the method of providing agitation. In addition to this, it may
be necessary to increase the conductivity of the solution. This is accomplished by
using a supporting electrolyte at high concentration.
10 Effect of Indifferent (supporting) electrolyte
Example-6: Consider the same cell as in example-5, except that the agitation is
absent. The cell contains AgNO3 at a concentration of 0.1 mol.m-3 and KNO3 at100
mol.m-3. A voltage, just sufficient to cause the concentration of the silver ions at the
cathode to drop to zero, is impressed across the cell. Assume again that the only
reaction is dissolution and deposition of silver. (a) Calculate the concentration
gradient of three ionic species present (b) Show that the effect of the potential on the
movement of silver ions is negligible
Solution: We begin with the continuity equation and obtain
dN ix
=0 (118)
dx
This gives on integration, that fluxes of all components are constant. These are
denoted by N + , N′+ and N − as fluxes of Ag+, K+ and NO3- respectively. Since K+ and
NO3- do not undergo charge transfer
N′+ = 0 and N − = 0 (119)
writing the expression for the fluxes, we obtain
dc + dφ
N + = −D+ − Fu + c + (120)
dx dx
dc′+ dφ
N′+ = − D′+ − Fu′+ c′+ =0 (121)
dx dx
dc − dφ
N − = −D− + Fu − c − =0 (122)
dx dx
We rearrange these equations after applying Nernst-Einstein equation
dc + Fc + dφ N
+ =− + (123)
dx RT dx D+

dc′+ Fc′+ dφ
+ =0 (124)
dx RT dx

19
 dc − Fc − dφ
− =0 (125)
dx RT dx
The electroneutrality dictates that
c + + c′+ = c − (126)
Subtraction of eq 125 from eq 124 yields
d(c′+ − c − ) F(c′+ + c − ) dφ
+ =0 (127)
dx RT dx
Using eq 126we modify eq 127 to obtain
dφ RT dc +
= (128)
dx F(c′+ + c − ) dx

Since the concentration of KNO3 is very large compared to AgNO3, c′+ + c − >> c + and
hence the potential gradient is reduced to a very small value. Thus using a high
concentration of indifferent electrolyte, we can almost completely eliminate the
potential gradient in the cell. Eq 123 then transforms to
dc + N
=− + (129)
dx D+
Integration yields
N+x
c+ = − +C (130)
D+
where C is a constant and can be obtained from the initial concentration as
L L
1 N N L
c0 = ∫ c + dx = − + ∫ xdx + C = − + + C (131)
L0 D+ L 0 2D +

Hence
N+L
C = c0 + (132)
2D +

N + (L − 2x )
c + = c0 + (133)
2D +

When c + = 0 at x = L, we obtain
2c 0 D +
N + lim = (134)
L
The corresponding limiting current is
2Fc0 D +
i lim = (135)
L

20
We see that this current is half that in the absence of the indifferent electrolyte. The
reason is that the elimination of the potential gradient eliminates the transport of silver
ions by migration.
11. Both ions taking part in the charge transfer reaction
In the previous case, we considered a situation where only one ion took part in
the charge transfer reaction. Here we consider another situation where both cations
and anions of a species participate in the reaction.
Example-7:
Let us consider another case where HCl is being electrolyzed. The reactions are
1
H − + e− ⇔ H 2 (cathode) (136)
2
1
Cl − ⇔ Cl 2 + e − (anode) (137)
2

Cl2 H2

HCl

Anode Cathode

The equations to be solved are

d(N + )x
=0 (138)
dx
d(N − )x
=0 (139)
dx
This on integration gives
(N + )x = C1 (N − )x = C2 (140)
Let these constant values be denoted by N+ and N- respectively.
Next, we write fluxes in terms of gradients

21
 dc + FD + c + dφ
N + = −D+ − (141)
dx RT dx
dc − FD − c − dφ
N − = −D− + (142)
dx RT dx
Electroneutrality requires that
c+ = c− (143)
Using this, we can transform eq 142 to the following form
dc + FD − c + dφ
N − = −D− + (144)
dx RT dx
We must also have
N+ = −N− (145)
at all points. This condition is required to balance the current in the outer circuit.
Adding eq 141 and 144, we obtain
dc + F(D + − D − )c + dφ
− (D + + D − ) − =0 (146)
dx RT dx
We use above equation to obtain the potential gradient as
dφ RT (D + + D − ) dc +
=− (147)
dx Fc + (D + − D − ) dx
Substitution the above expression into eq 141 gives
⎛ (D + D − ) ⎞ dc +
N + = − D + ⎜⎜1 − + ⎟⎟ (148)
⎝ (D + − D − ) ⎠ dx
dc +
or N + = 2D +eff (149)
dx
D+D−
D + eff = (150)
(D + − D− )
The concentration profile is a straight line.
N + (L − 2 x )
c + = c0 − (151)
4D + eff

H+ c0

0 L 22
x
It is seen that the concentration profile of H+ ions is increasing from anode to cathode.
This is counterintuitive, since hydrogen ions are consumed at the cathode and hence
we expect the H+ ion concentration to decrease from the anode to cathode. Note
however, that driving force is the potential gradient and not the concentration
gradient. The concentration gradient is the result of the potential gradient. Since the
mobility of hydrogen ion is much large than that of chloride ion, chloride ion needs a
higher value of the driving force compared to hydrogen. The reverse concentration
gradient of hydrogen slows down the movement of H ions and enhances the
movement of Cl ions. In other words, the direction of the concentration gradient and
the direction of flux will match only for the slow moving component.
The potential profile is obtained by integration of eq 147
RT (D + + D − )
φ=− ln c + + C (152)
F (D + − D − )
Hence the potential difference at the two ends of the cell is

RT (D + + D − ) ⎛⎜ c + x = L ⎞⎟
Δφ = φ x = 0 − φx = L = ln (153)
F (D + − D − ) ⎜⎝ c + x = 0 ⎟⎠

The limiting current density is when c + x = 0 = 0 . The value is

4Fc0 D + eff
i lim = (154)
L

23
 Lecture-6
Ion Transport in Solution-II
6.1 Equivalent Ionic conductance
Conductivity κ is dependent both on the nature of the electrolyte as well as its
concentration. To compare different electrolytes for their ability to conduct, we must
have a quantity which is independent of concentration. To achieve this, we define two
additional quantities based on the conductivity for a single salt solution. First is the
molar conductivity Λ m , defined as
κ
Λm = (1)
c
The second quantity is the equivalent conductivity
κ κ
Λ= = (2)
cν + z + cν − (− z − )
Table-2 compares these two quantities for some salts
Table-2 Molar and Equivalent conductivities of electrolytes at 250C and 0.01 M
Electrolyte Molar Conductivityx102 Equivalent
S.m2.mol-1 conductivityx102
S.m2.eq-1
KCl 1.413 1.413
NaCl 1.185 1.185
MgCl2 2.292 1.146
Na2SO4 2.248 1.128

This table shows that equivalent conductivity of the electrolytes is a better basis for
comparison since it is insensitive to valencies of the ions present.
In the literature, instead of ionic mobilities, values of the equivalent ionic
conductance are reported. They are related to ionic mobilities by the following
equation
λ i = z i F2 u i (3)

In very dilute solutions, we can also obtain the diffusion coefficient from the ionic
conductance using the Nernst-Einstein relation
RTλ i
Di = (4)
z i F2

1
 We can relate equivalent conductivity of a salt with the ionic conductance of
the constituting ions as follows. Using the definition of conductivity
κ = F2 ∑ z i2 u i ci (5)
i

We can rewrite eq 2 as
F2 ∑ z i2 u i ci
Λ= i
=
(
F2 z +2 u + c + + z −2 u − c − ) ⎛ z2 u c z2 u c ⎞
= F2 ⎜⎜ + + + + − − − ⎟⎟ (6)
cν + z + cν + z + ⎝ cν + z + cν − (− z − ) ⎠
The last part of the above equation is obtained using the following relation.
ν+z+ + ν−z− = 0 (7)
Eq 6 can be simplified to
⎛ z2 u z2 u ⎞
Λ = F2 ⎜⎜ + + + − − ⎟⎟ = F2 ( z + u + + z − u − ) = λ + + λ − (8)
⎝ z+ (− z − ) ⎠
The transport numbers are related to ionic conductance as follows. From the
definition,
z 2j u jc j
tj = (9)
∑z u c
i
2
i i i

we find that
z +2 u + c + z +2 u + cν +
t+ = = (10)
z +2 u + c + + z −2 u − c − z +2 u + cν + + z 2− u −cν −

Using electroneutrality of crystalline state ( ν + z + + ν − z − = 0 ), we can modify eq 10 to

z+u+ λ+
t+ = = (11)
z−u + − z−u − λ + + λ −

λ−
Similarly t− = (12)
λ+ + λ−

6.2 Kohlrausch law

At low concentrations of electrolytes, the equivalent conductivity of
electrolytes is found to vary according to the following empirical relationship
Λ = Λ0 − Am1 2 (13)
where m is the molal concentration of the electrolyte. This relationship is called the
Kohlrausch law. Onsager has provided a theoretical justification for the above
equation based on Debye-Huckel theory and the resulting form of the equation is
called the Debye-Huckel-Onsager equation. The basis of the derivation of the

2
equation is that every ion is surrounded by a cloud of the oppositely charged ions.
When the central ion moves under the application of electric field, cloud does not
immediately responds due to its inertia. It lags behind the central ion. As a result there
is a charge separation between the central ion and the cloud. The cloud pulls-back the
central ion and hence reduces its velocity. More spread is the cloud, lesser is its force
on the central ion. Therefore, the reduction in the velocity of the central ion due to the
cloud is inversely proportional to the Debye length. This theory yields the following
expressions for the mobilities.

F ⎛ z+ Fω ⎞
u + = u 0+ − ⎜⎜ + u 0+ ⎟⎟ (14)
λN av ⎝ 6πη 6εε0 RT ⎠

F ⎛ z− Fω ⎞
u − = u 0− − ⎜⎜ + u 0− ⎟⎟ (15)
λN av ⎝ 6πη 6εε0 RT ⎠

Where u 0+ and u 0− are the values of the mobilities of cation and anion respectively, at
infinite dilution, η is the fluid viscosity, λ is the Debye length given by
12
⎛ εε RT ⎞
λ = ⎜⎜ 02 ⎟⎟ (16)
⎝ 2F Iρs ⎠

⎛ 2q ⎞
and ω = z + z − ⎜⎜ ⎟
⎟ (17)
⎝ 1 + q ⎠

⎛ z z ⎞⎛ λ + + λ − ⎞
q = ⎜⎜ + − ⎟⎟⎜⎜ ⎟⎟ (18)
⎝ z + + z − ⎠⎝ z + λ + + z − λ − ⎠
Using eq 14 and 15, we can obtain the equivalent ionic conductances as

F ⎛ z + F2 Fω z + 0 ⎞
2
λ + = λ0+ − ⎜ + λ+ ⎟ (19)
λN av ⎝ 6πη 6εε0 RT ⎟⎠
⎜

F ⎛ z − F2 Fω z − 0 ⎞
2
λ − = λ0− − ⎜ + λ− ⎟ (20)
λN av ⎝ 6πη 6εε0 RT ⎟⎠
⎜

Adding the two, we get the equivalent conductivity of the electrolyte as

Λ = Λ0 − ⎜
(
F ⎛ z + + z − F2
2 2
+
)
Fω
(
z + λ0+ + z − λ0− )⎞⎟⎟ (21)
⎜
λN av ⎝ 6πη 6εε 0 RT ⎠
Since the Debye length is inversely proportional to the square root of the ionic
strength, and the ionic strength in turn is directly proportional to molality, we obtain
eq 13.

3
 For a symmetric electrolyte, we can simplify eq 21 as

Λ = Λ0 − ⎜
(
F ⎛ z + + z − F2
2 2
+
) Fωz + 0 ⎞
Λ⎟ (22)
⎜
λN av ⎝ 6πη 6εε 0 RT ⎟⎠

which could be written in an abbreviated form as

Λ = Λ0 − (a + bΛ0 ) c (23)
where c is the concentration in mol.L-1 .
Example 1: Determine equivalent ionic conductance of sodium ion and sulfate ion
from their mobilities and hence estimate the equivalent conductivity of sodium sulfate
in very dilute solutions.
Solution: Mobilities of sodium and sulfate ions are, 5.384 x 10-13 m2.mol.s-1.J-1 and
4.299x10-13 m2.mol.s-1.J-1 respectively. We use eq 3 to calculate the equivalent ionic
conductance of the ions as
λ + = z + F2 u + = 1x (96488) 2 x 5.384 x10 −13 = 5.012 x10 −3 S.m 2 .equv.−1

λ − = z − F2 u − = 2 x (96488) 2 x 4.299 x10 −13 = 8.005x10 −3 S.m 2 .equv.−1

The equivalent conductivity of sodium sulfate can now be calculated as

Λ = λ + + λ − = 5.012 x10 −3 + 8.005x10 −3 = 0.01302 S.m 2 .equv.−1

The calculated value is greater than that listed in the table (0.01128 S.m 2 .equv.−1 ). The
reason is that the tabulated value is for 0.01 M solution, whereas the computed value
is at infinite dilution. This shows a strong dependence of the electrolyte concentration
on the equivalent conductance.
6.3 Measurement of transport number
The most commonly used method for the measurement of transport number is
Hittorf’s method. The apparatus consists of two electrode compartments joined by a
tube. As an example, consider that the electrodes are silver and the cell is initially
filled with the test aqueous solution of AgNO3. A DC current is then passed through
the cell. After time tb, the concentrations of AgNO3 in the two electrode
compartments are measured and are used for estimating the transport numbers of
silver and nitrate ions.
We note that silver dissolves from the electrode in the anodic compartment
and equal amount of silver is deposited on the electrode in the cathodic compartment.
Part of the silver ions generated in the anodic compartment travels to the cathodic
compartment via the central tube. Result is that the concentration of the silver ions in

4
the anodic compartment increases, but not at the same rate as that at which it is
generated. To maintain the electroneutrality, rate of flow of nitrate ions into the
anodic compartment from the cathodic compartment must match the net rate of
increase in the silver ions in the anodic compartment. Since the central tube acts as a
carrier, we expect that the concentration of silver nitrate in the central compartment to
nearly remain constant during the experiment at the initial concentration c0 .
If the current passed through the circuit is fixed at I, the rate of dissolution of
silver at the anode will be I F and this is also equal to the rate of deposition of silver
at the cathode. Let Δt be the time over which the current is passed into the solution.
Then the amount of silver generated in the anodic compartment, or the amount
depleted from the cathodic compartment would be IΔt F .

c0
ca (t ) cc (t)

The flow of current through the central compartment will also be I and is
caused by motion of silver ions from the left to the right and that of nitrate ions from
the right to the left. The fraction of the total current carried by the silver ions will be
It + , where t + is the transport number of the silver ion. This results in the transport of
It + Δt F ions from the left to the right compartment. Hence the net amount (in moles)
of the silver ions accumulated in the anodic compartment is
Δn = IΔt F − It + Δt F = I(1 − t + )Δt F = It − Δt F (24)

5
 Rearrangement gives
Δn
t− = (25)
IΔt F

We can also obtain the expression for t + as

IΔt F − Δn
t+ = 1− t− = (26)
IΔt F
This quantity Δn represents the number of moles of silver accumulated in the anodic
compartment or the number of moles of silver depleted from the cathodic
compartment. If both the cathodic and the anodic compartments have equal vaolume
V, then we can write
V(c a − c c )
Δn = (27)
2
Here c a and c c are the concentrations of silver ion in the anodic and the cathodic
compartment, respectively.
The main assumption in this treatment is that there is no change in the
concentration of the electrolyte in the central compartment. In reality, the
concentration of the electrolyte in the central compartment should increase with time
as a result of diffusion/migration of silver ions from the anodic compartment and
nitrate ions from the cathodic compartment. This effect will be least when the
concentrations in the three compartments are identical. Hence to obtain a high
accuracy, we must restrict to very small values of Δn . This necessitates a high
accuracy of measurement of Δn . The methods such as measurement of the difference
in the conductivities of the solutions in the anodic and the cathodic compartments can
be used to find small differences in the concentrations of these compartments.
6.4 Equation for diffusion in moderately concentrated solutions
When the solution is not very dilute, some changes are required in the relevant
equations. The Nernst-Plank equation i.e.
K K K K
N i = −Di∇ci − Fz i u i ci∇φ + ci v (28)
is still valid, but the ionic diffusion coefficient is not related to mobility by the Nernst-
Einstein equation, but by the following equation (eq 10 of lecture-5).
⎛ d ln γ i ⎞
D i = RTu i ⎜⎜1 + ⎟⎟ (29)
⎝ d ln c i ⎠

6
The mobility itself is a function of concentration and can be obtained using Debye-
Huckel-Onsager equation. Rest of the conditions such as electroneutrality in the bulk
solution, Kirchoff’s law, i.e.
K K
∇⋅ i = 0 (30)
and the Laplace equation ( in the absence of concentration gradient) are still valid.
6.5 Equations for diffusion in concentrated solutions
In the dilute solutions, following rule governs the motion of an ion in the
solution.
K K
v ti = − u i∇ μi (31)

In the above equation μi represents the electrochemical potential, u i represents

K
mobility of I and v ti represents the steady velocity, attained by the ion, relative to the
solvent. We can also write the above equation in terms of the absolute velocities as
shown below
K K K
vi − v 0 = −u i∇ μi (32)
K K
In the above equation, vi is the absolute velocity of ion i and v0 is that of the solvent.
The above equation can be rearranged as
K K K
vi − v 0
− ∇ μi = (33)
ui

The reciprocal of u i can be viewed as the coefficient of frication between component

K
i and the solvent and − ∇ μi may be viewed as the friction force
In concentrated solutions, we modify the equation to the following form
n x (v − v )
K K
K
− ∇ μi = ∑
j i j
(34)
j=1 u ij

The summation extends over all the components in the solution. This equation implies
that the force of friction depends on relative velocity of species i with respect each of
the remaining species. The term x j u ij is the friction coefficient between components

i and j. Mole fraction x j of component j is introduced in order to account for the

concentration dependence of the friction coefficient. Thus for the same relative
velocity between i and j, friction increases with increase in mole fraction of j (due to
greater number of collisions). We expect u ij to be concentration independent.

7
 Writing x j = c j c T , where c T is the molar density of the solution, we can

recast eq 34 in the following form as

n c c (v − v )
K K
K
− c i ∇ μi = ∑
i j i j
(36)
j=1 c T u ij

The above equation can also be written as

K
− c i∇ μi = ∑ K ij (v i − v j )
n
K K
(37)
j=1

ci c j
K ij = (38)
c T u ij

And relating the mobilities to pair-diffusion coefficients by relation

D ij
u ij = (39)
RT
we obtain
RTc i c j
K ij = (40)
c T D ij

We sum eq 37 with respect to i

K
∑ i i ∑∑ K ij (vi − v j )
n n n
K K
c ∇ μ = − (41)
i =1 i =1 j=1

But the left hand side is zero by Gibbs-Duhem equation and hence, we get

∑∑ K (v − vj)= 0
n
K
n
K
ij i (42)
i =1 j=1

The above equation can be rearranged to give

∑∑ (K − K ji )v i = 0
n n
K
ij (43)
i =1 j=1

Since this is valid for any arbitrary set of velocities, we require that
K ij = K ji (44)

implying symmetry of the matrix of friction coefficients. Comparison between eq

40and 44 show that the pair diffusion coefficient is also symmetric, i.e.
D ji = D ij (45)

Noting that
K K
N i = ci vi (46)

8
We can rewrite eq (36) as
K K
K RT n c j N i − c i N j
− c i ∇ μi = ∑ D
c T j=1
(47)
ij

Expanding the electrochemical potential in terms of the electrical and the chemical
potential, we obtain
K K
c K zcF K 1 n c j Ni − ci N j
− i ∇μ i − i i ∇φ =
RT RT cT
∑
j=1 D ij
(48)

This is the standard form of the Maxwell-Stefan equations. These are total of n
equations one for each species.
6.6 Concentrated solution containing a single electrolyte
Let us simplify the Maxwell-Stefan equation for a solution of a single
electrolyte. There are three species i.e. cation, anion and the solvent. Since only two
equations are independent, we write eq 48 only for cations and anions as follows
K K K K
c+ K z +c+ F K 1 ⎡ c − N + − c + N − c0 N + − c + N 0 ⎤
− ∇μ + − ∇φ = ⎢ + ⎥ (49)
RT RT cT ⎣ D +− D +0 ⎦

K K K K
c− K z −c− F K 1 ⎡ c + N − − c − N + c0 N − − c − N 0 ⎤
− ∇μ − − ∇φ = ⎢ + ⎥ (50)
RT RT cT ⎣ D +− D −0 ⎦
K
We add the two equations. The term containing ∇φ vanishes due to electroneutrality.
The resulting equation is
K K K K
c K c ⎡ N + − c+ v0 N − − c− v0 ⎤
− ∇μ e = 0 ⎢ + ⎥ (51)
RT cT ⎣ D +0 D −0 ⎦

where c is the concentration of the electrolyte and μ e is its chemical potential.

μ e = ν + μ + + ν −μ − (52)
K K
In eq 51, N 0 is replaced by c 0 v 0 . Rearrangement of eq 51 gives
K K K
ccT K ⎡ D −0 N + + D + 0 N − − cv 0 (ν + D −0 + ν − D + 0 ) ⎤
− ∇μ e = ⎢ ⎥ (53)
c 0 RT ⎣ D + 0 D −0 ⎦
Writing the current density as
K
i K K
= z+ N+ + z−N− (54)
F
K
we eliminate N − from eq 53 using eq 54 and obtain

9
 K K
−
z − ccT K
∇μ e = ⎢
( ) K
⎡ (D −0 z − − z + D +0 )N + + D +0 i F − cv 0 ν + (z − D −0 − z + D +0 ) ⎤
⎥ (55)
c 0 RT ⎣ D + 0 D −0 ⎦
We define the diffusion coefficient of the electrolyte as
D 0+ D 0− (z − − z + )
D= (56)
z + D 0 + − z − D 0−

We also define the transport number t + as

z + D +0
t 0+ = (57)
z + D 0+ − z − D0−
The above two equations are same as those obtained for dilute solution of a single
electrolyte, in the previous lecture. Using these definitions, and writing − (z + ν + ν − )
in place of z − on the left hand side of eq 55, we get
K
K Dν + ccT K t 0+ i K
N+ = − ∇μ e + + c + v0 (58)
c 0 (ν + + ν − )RT z+F
K K
If we eliminate, N + instead of N − between equations 51 and 52, we obtain, the
K
equation for N − as
K
K Dν − cc T K t 0− i K
N− = − ∇μ e + + c − v0 (59)
c 0 (ν + + ν − )RT z−F
K K K K K ⎛ d ln γ ± ⎞
Writing, ∇μ e = RT∇ ln a e = RT (∇ ln m + ∇ ln γ ± ) = RT∇ ln m⎜1 + ⎟ and noting
⎝ dm ⎠
c
that m = , M w being the molecular weight of water, we can modify a part of
c0 M w
first term on the RHS of eqs 58 and 59 as
DccT K DccT ⎛ dγ ± ⎞ K DccT ⎛ dγ ± ⎞ K
∇μ e = ⎜1 + ⎟∇ ln m = ⎜1 + ⎟∇(ln c − ln c 0 )
c 0 RT c 0 ⎝ dm ⎠ c 0 ⎝ dm ⎠
DccT ⎛ dγ ± ⎞⎛ d ln c 0 ⎞ K ⎛ d ln c 0 ⎞ G
= ⎜1 + ⎟⎜1 − ⎟∇ ln c = (ν + + ν − )D c ⎜1 − ⎟∇c
c 0 ⎝ dm ⎠⎝ d ln c ⎠ ⎝ d ln c ⎠
(60)
where D c is the concentration based diffusion coefficient of the electrolyte and is
defined as
Dc T ⎛ dγ ± ⎞
Dc = ⎜1 + ⎟ (61)
c 0 (ν + + ν − ) ⎝ dm ⎠
We can now rewrite eq 58 and 59 as

10
 K
K ⎛ d ln c 0 ⎞ K t 0+ i K
N + = −D c ν + ⎜1 − ⎟∇c + + c+ v0 (62)
⎝ d ln c ⎠ z+F
K
K ⎛ d ln c 0 ⎞ K t 0− i K
N − = − D c ν − ⎜1 − ⎟∇c + + c− v0 (63)
⎝ d ln c ⎠ z −F
We use one of the two expressions in the continuity equation ( in the absence of
reaction)
∂c + K K
= −∇ ⋅ N + (64)
∂t
and get
K K
∂c K K K ⎡ ⎛ d ln c 0 ⎞ K ⎤ i ⋅ ∇t 0+
+ ∇ ⋅ (cv 0 ) = ∇ ⋅ ⎢D c ⎜1 − ⎟∇c −
d ln c ⎠ ⎥⎦ z + ν + F
(65)
∂t ⎣ ⎝
6.7 Alternative form of the Diffusion equation
Equation 37 can also be cast in an alternative form. We can write it as
K
ci ∇ μi = ∑ K ij (v j − v 0 ) − (v i − v 0 )∑ K ij
n
K K K K n
j=1 j=1
(66)
= ∑ M ij (v j − v 0 )
n
K K
j=1

Where
M ij = K ij , i≠ j
= K ij − ∑ K ik
(67)
i= j
k

M ij is a symmetric matrix. Also note that in the set of eq 49 are only n-1 equations are
K
independent since ∑ c i∇ μi = 0 . Hence we can select any n-1 equations from this set.
i

We omit the equation corresponding to solvent 0. In the matrix form we can write
these n-1 equations as
[c ∇K μ ] = [M ][vK
i i
0
ij j
K
− v0 ] (68)

[ ]
Here the new matrix M 0ij is n-1 order square matrix obtained from M ij by deleting [ ]
row and column corresponding to the reference component.
We can invert eq 51 to the following form
K K
[
K K
] [ ][
−1
v j − v 0 = M ij0 ci∇μ i = − L0ji c i∇μi ] [ ][ ] (69)

where [L ] = −[M ]
0
ji
0 −1
ij (70)

Equation 69 can be written in the expanded form as

11
 K K K
v j − v 0 = −∑ L0jk c k ∇μ k (71)
k≠0

The diffusion coefficients L0jk are symmetric with respect to indices i and k (Onsager

reciprocity relationship)
L0jk = L0kj (72)

6.8 Conductivity and transference numbers

The expression for the current density is
K
i = F∑ z jc j v j = F∑ z jc j (v j − v 0 )
K K K
(73)
j j

K
The second sum in eq 73 is identical to the first sum since v 0 ∑ z jc j = 0 due to
j

electroneutrality.
Combining eq 71 with eq 73, we find
K K
i = −F∑ z jc j ∑ L0jk c k ∇ μk (74)
j≠ 0 k ≠0

If the solution has uniform composition, the potential gradient reduces to

K K
∇μ k = z k F∇φ (75)
Substitution of the above expression into eq 74 gives
K K
i = −F2∇φ∑ z ici ∑ L0ik z k c k (76)
i≠0 k≠0

Noting the definition of conductivity as

K K
i = − κ∇φ (77)
We obtain the following expression for conductivity on the basis of eq76
κ = F2 ∑∑ L0ik z i ci z k c k (78)
i ≠0 k ≠ 0

We can express the current density as

K K
i = ∑ ik (79)
k

Comparing eq 73 and 79, we obtain

K
ij = Fz jc j (v j − v 0 )
K K
(80)

Noting that eq 71 reduces, under uniform composition, to

K K K
v j − v 0 = −F∇φ∑ L0jk z k c k (81)
k≠0

we obtain (from eqs 80 and 81)

12
 K K K
ij = t 0j i = −F2∇φz jc j ∑ L0jk z k c k (82)
k ≠0

Where, t 0j is a transference number. From eq 77 and 82, we obtain

K K
− t 0j κ∇φ = − F2∇φz jc j ∑ L0jk z k c k (83)
k ≠0

F 2 z jc j
or t 0j =
κ
∑L
k ≠0
0
jk z k ck (84)
K
Note that t 00 = 0 , since i0 = 0 as seen from eq 80.
Summing eq 84 over j, we obtain
F2
∑tj
0
j =
κ
∑∑ L
j≠ 0 k ≠ 0
0
jk z j z k c jc k = 1 (85)

where we have used eq 78. The transference number is dependent on the species
chosen as the reference, because the magnitudes of L0jk depend upon it (see eq 71). In

the present case, we have chosen solvent (0) as the reference. However, any other
species could be chosen as the reference.
6.9 Fundamental equation for diffusion in electric field
When the composition of the solution is nonuniform, we have to use eq 74 for
the current density
K K
i = −F∑ z ici ∑ L0ik c k ∇ μk (86)
i≠0 k≠0

Rearrangement yields
K K
i = −F∑ c k ∇ μk ∑ L0ki z ici (87)
k ≠0 i≠0

In writing the above equation, we have made use of the symmetry of matrix L0ik .
Using the expression 84 for the transference number
F2 z k c k
t 0k =
κ
∑L
i≠0
0
zc
ki i i (88)

We can rewrite eq 87 as
FK t0 K
i = − ∑ k ∇ μk (89)
κ k ≠0 z k

It is important to note that t 0k z k is not zero even for uncharged species (see eq 88).
Hence uncharged species also contribute to the current. This is easy to understand

13
because the motion of an uncharged species will induce motion in rest of the species,
including the charged species and this indirect motion will contribute to the current.
Eq 89 is the fundamental equation for diffusion in concentrated solutions. Its
use in computing the junction potential will be illustrated later. It can be solved for the
electric potential profile provided, the concentration profiles of all the species are
known. These profiles are obtained by solving the continuity equation
∂ci K K
= −∇ ⋅ N i + R i (90)
∂t
K
where, Ni is obtained from eq 71 as follows
K K K K
N i = c i vi =ci v 0 − ∑ L0ik c k ∇ μk (91)
k ≠0

6.10 The Henderson equation

We shall use the fundamental equation to study the junction potentials. Eq 89,
is rewritten below
K
Fi t0 K
= −∑ k ∇ μ k (92)
κ k ≠0 zk

Expanding the electrochemical potential in the above equation into the chemical and
the electrical potentials and rearranging the resulting equation, we get
K FK t0 K
F∇φ = − i − ∑ i ∇μi (93)
κ i ≠ 0 zi

We can also recast eq 93 in terms of activities as shown below

K FK t0 K
F∇φ = − i − RT ∑ i ∇ ln a i (94)
κ i ≠0 z i

In the absence of the current, eq 94 reduces to

K RT t 0i K
− ∇φ = ∑ ∇ ln a i
F i ≠ 0 zi
(95)

The above equation could also be written in differential form as follows

RT t 0i
− dφ = ∑
F i≠0 z i
d ln a i (96)

Transition
ci = ci(0) region ci = ci(L)

x 14
x=0 x=L
Consider a transition region of length L between the two solutions of uniform
concentrations ci (0) and ci (L) . We wish to find the difference φ(0) − φ(L )
Integration of eq 96 between 0 and L gives
L L
RT t i0 d ln a i
∫ dφ = − ∑
F i≠0 ∫0 z i dx
dx (97)
0

To obtain the potential difference across the transition region one must know the
concentration profiles of the species in the transition region and also dependence of
the activity coefficients and transport numbers on concentrations. However, we can
obtain the solution of eq 97, provided certain assumptions are made. We replace
activities by concentrations. We also assume linear gradient of species i.e.
⎛x⎞
ci ( x ) = ci (0) + [ci (L) − ci (0)]⎜ ⎟ (98)
⎝L⎠
We can rewrite eq 97 as
L
RT t i0 dci
φ(L ) − φ(0) = − ∑
F i ≠ 0 ∫0 z i cidx
dx (99)

We also use the definition of the transference number, which is applicable to dilute
solutions. i.e.
z i2 u i ci
t 0i = (100)
∑ z2ju jc j
j

Hence

RT
L ∑ u z [c (L) − c (0)]dx
i i i i
φ(L ) − φ(0 ) = −
FL ∫0
i≠0
(101)
⎧ ⎛ x ⎞⎫
∑ u z ⎨ci (0) + [ci (L) − ci (0)]⎜ ⎟⎬
2
i i
i≠0 ⎩ ⎝ L ⎠⎭
Let
β(x ) = ∑ u i z i2ci ( x ) H = ∑ u i z i [ci (L) − ci (0)] (102)
i≠0 i≠0

Eq 101 then reduces to

L
RTH dx
φ(L ) − φ(0 ) = − ∫ (103)
F 0 β(0)L + [β(L ) − β(0)]x

Integration gives

15
 RTH ⎛ β(0 ) ⎞
φ(L ) − φ(0 ) = ln⎜⎜ ⎟ (104)
F[β(L ) − β(0)] ⎝ β(L ) ⎟⎠
The above equation is called the Henderson equation.
For a single symmetric electrolyte (− z − = z + = z ) eq 104 simplifies to

RT(u + − u − ) ⎛ c(0 ) ⎞
φ(L ) − φ(0 ) = ln⎜ ⎟ (105)
zF(u + − u − ) ⎜⎝ c(L ) ⎟⎠
And for 1:1 electrolyte, the equation reduces to

φ(L ) − φ(0 ) =
RT
(t + − t − ) ln⎛⎜⎜ c(0) ⎞⎟⎟ (106)
zF ⎝ c(L ) ⎠
6.11 Liquid Junction potentials
Electrochemical cells consist of two half-cells, each containing one electrode.
The two cells, in general, have different composition. Since the current has to be
carried from one electrode to the other, the half cells must communicate with each
other. This communication of the current occurs though a liquid junction. The liquid
junction is a transition region between the two half-cells. The two half cell can have
different constitutions. It is therefore necessary to minimize the contamination of the
liquid of one half cells by that from the other half cell. The diffusion process can be
slowed down by either using a porous separator or by placing the solutions in two
separate compartments and joining them by a capillary. The capillary, which is filled
with a salt surrounded by its saturated solution, is called the salt bridge.
The liquid junction potential can be computed by the Henderson equation.
For example, consider a liquid junction of two 1:1 electrolytes (denote by superscripts
I and II) with a common anion. Also assume that both the solutions are at the same
concentration. For this case
β(0 ) = u I+ (1) 2 c I+ (0) + u − (−1) 2 c −I (0) = u +I c + u − c (107)

β(L ) = u II+ (1) 2 c +II (L) + u − (−1) 2 c −II (L) = u +IIc + u −c (108)

and H = u I+ [0 − c] + u II+ [c − 0] = c(u II+ − u I+ ) (109)

φ(L ) − φ(0 ) =
(
RTc u II+ − u I+ ) ⎛ u I+ c + u − c ⎞ RT ⎛ u I+ + u − ⎞
⎜ ⎟= ln⎜ ⎟
[( ) ( ln
)]
F u II+ c + u −c − u I+ c + u −c ⎜⎝ u II+ c + u −c ⎟⎠ F ⎜⎝ u II+ + u − ⎟⎠
(110)

Noting the relation between the equivalent conductance and mobilities,

Λ = λ + + λ − = z + F2 u + + z − F2 u −

We see that

16
 (
ΛI = (1) F2 u +I + (−1) F2 u − = F2 u +I + u − ) (112)

and ΛII = (1) F2 u II+ + (−1) F2 u − = F (u

2 II
+ + u− ) (110)

In the above expressions ΛI and ΛII represent the equivalent conductance of

electrolytes I and II respectively.
Substitution into eq 104 gives

RT ⎛ ΛI ⎞
Δφ L = φ(L ) − φ(0) = ln⎜ ⎟ (113)
F ⎜⎝ ΛII ⎟⎠

Where we have used symbol Δφ L to denote the liquid junction potential. The above
equation is called the Lewis and Sargent equation. It is valid only when the
concentrations in both the cell compartments are equal.
Table below compare the liquid junction potentials for pairs of electrolytes
with chloride as common anion. The calculations are based on the Lewis and Sargent
equation. The concentrations of all solutions are 0.1 M.

Junction Δφ L (calculated), mV Δφ L (experimental), mV

HCl/KCl 28.52 26.78
HCl/LiCl 36.14 34.86
KCl /LiCl 7.62 8.79
KCl/NaCl 4.86 6.42
NaCl/NH4Cl -4.81 -4.21

Note that except for H+ ion which has much higher mobility compared to
other ions, all other cations have nearly same mobility values. Hence the junction
potentials are low for the pairs not involving H+ ion as is evident from the table. The
calculated values are also in close agreement with the experiments.
Salt bridge is a special liquid junction joining two different electrolytes. It is
tube, joining the two electrode compartments, which is filled with a concentrated
solution of an equitransferent electrolyte, one having close values of the cation and
the anion transference numbers (e.g. KCl, in which the transference number of K is
0.49 and that of Cl is 0.51). Considerable reduction in the junction potential could be
achieved. We can see this from eq 106 that

17
 RT 0 0 ⎛ c(0 ) ⎞
Δφ L = (
t + − t − ln⎜⎜ ) ⎟⎟ (114)
zF ⎝ c(L ) ⎠
By using an equitransferent electrolyte, we reduce the difference t 0+ − t 0− . For example
using KCl bridge it is possible to reduce the junction potential of HCl/KCl junction
from 29 mV to 3 mV.
Example: Estimate the liquid junction potentials for the following junctions
(a) 0.001 N H2SO4 / 0.01 N. H2SO4
(b) 0.001 N KCl / 0.01 N. KCl
Solution: Since both these are symmetric 1:1 electrolytes we use eq 114 to compute
the liquid junction potential, which is valid for dilute solutions
RT 0 0 ⎛ c(0 ) ⎞
Δφ L = (t + − t − )ln⎜⎜ c(L) ⎟⎟ (114)
zF ⎝ ⎠
The transference numbers are obtained from the mobilities of the ions using eq 100.
i.e.
z i2 u i ci
t 0i = (115)
∑ z 2ju jc j
j

(a) The mobilities are u+ (of H+)= 37.59x10-13 m2.mol.s-1.J-1, u-(HSO4-)= 5.37x10-13
m2.mol.s-1.J-1, z+ = 1 and z- = -1 also c+ = n and c- = n ( n = normality)
(1) 2 u + n u+ 37.59
t = 2
0
= = = 0.875
(1) u + n + (−1) u − (n ) u + + u − 37.59 + 5.37
+ 2

and hence t 0− = 1 − t 0+ = 1 − 0.875 = 0.125 . Substituting these in eq 114 we obtain

⎛ 0.001 ⎞
Δφ L = 0.0257[0.875 − 0.125]ln⎜ ⎟ = −0.0414 V
⎝ 0.01 ⎠
(b) In this case The mobilities are u+ = 7.899 x10-13 m2.mol.s-1.J-1 and u- = 8.201x10-13
m2.mol.s-1.J-1, z+ = 1 and z- = -1 also c+ = n and c- = n ( n = normality)
(1) 2 u + n u+ 7.899
t 0+ = = = = 0.491
(1) u + n + (−1) u − n u + + u − 7.899 + 8.201
2 2

and hence t 0− = 1 − t 0+ = 1 − 0.491 = 0.509 . Substituting these in eq 114 we obtain

⎛ 0.001 ⎞
Δφ L = 0.0257[0.491 − 0.501]ln⎜ −4
⎟ = 5.92 x10 V
⎝ 0.01 ⎠

18
 Lecture-7
Double Layer
7.1 Introduction
Consider a charged electrode surface which is placed in a solution containing
strong electrolyte species. Assume that the surface charge density of the electrode is σ .
Due to the surface charge, an electric potential is generated in the solution. The potential
at the locations which are far removed from the surface is constant. At these locations,
the solution is electrically neutral. We call this region, the bulk solution. The condition of
electroneutrality in the bulk can be expressed as
ρ b = ∑ cib z i = 0 (1)
i

Because of the charge on the surface, a charge distribution exists in its vicinity. If the
surface is positively charged, it will attract anions and repel cations. This leads to
anisotropy in the charge distribution near the interface. The region in which this
anisotropy exists is called the double layer. The idea of the double layer was first
proposed by Helmholtz, who imagined it to be consisting of two layers of charges, the
surface charge layer and the adjacent layer of oppositely charged ions which completely
neutralize the surface charge. Hence he named it as the double layer. The modern theory
of double layer was proposed by Guoy and Chapman and is called Guoy-Chapman
theory, which we describe here.
It is important to note that, it is entropically unfavorable to have an extreme
distribution where all anions and no cations are located in the double layer.
Thermodynamically, the electrochemical potential of any species, at equilibrium, should
be same at all locations in the solution. If not, the potential gradient will make the species
to diffuse till the gradient vanishes. Thus
μi = μib (2)
Here, μi is the electrochemical potential of species i at any location in the double layer

and μib is the chemical potential of the same species in the bulk. The electrochemical
potential of an ionic species i in as electric field is given by
μ i = μ i0 + RT ln ci γ i + z i Fφ (3)

1
where, μi0 is the chemical potential of i at standard state, ci is the concentration of i , γ i ,
the activity coefficient, z i the valancy of the ion, e , the charge on the primary particle, φ ,
the potential of the field.
In the bulk,
μib = μi0 + RT ln cib γ ib (4)
Applying the condition of equality of the potentials of i , we obtain
μ i0 + RT ln ci γ i + z i Fφ = μ i0 + RT ln cib γ ib (5)
The above equation, on rearrangement yields.
ci γ i ⎛ z Fφ ⎞
= exp⎜ − i ⎟ (6)
ci γ i
b b
⎝ RT ⎠
The above equation provides us with the distribution of concentration of ion i in the
double layer as a function of the electric potential. If the solution is dilute, γ i = γ ib = γ i∞ ,
where, the last term is the activity coefficient of the ion at infinite dilution. In such a case,
we simplify the above equation to
⎛ z Fφ ⎞
ci = cib exp⎜ − i ⎟ (7)
⎝ RT ⎠
The distribution of concentration, expressed by the above equation is called the
Boltzmann distribution.
7.2 Poisson-Boltzmann Equation:
The charge density at any location in the double layer is given by the following
equation
ρ = ∑ ci z i F (8)
i

Substituting for ci from the Boltzmann distribution, we obtain

⎛ z Fφ ⎞
ρ = ∑ Fz i cib exp⎜ − i ⎟ (9)
i ⎝ RT ⎠
If we substitute the above expression for the charge density into the Poisson equation, we
obtain
F ⎛ z Fφ ⎞
∇ 2φ = −
εε 0
∑z c
i
b
i i exp⎜ − i ⎟
⎝ RT ⎠
(10)

2
The above equation is called the Poisson-Boltzmann equation. It is the basic equation of
the double layer theory. The solution of the equation tells how the potential varies with
distance in the double layer. The boundary conditions for the above equation can be
obtained as follows. We measure the distance in the double layer from the surface of the
electrode, in the direction normal to the surface and denote it by x . In the bulk, which is
far removed from the surface, electroneutrality prevails and hence the potential is zero.
φ = 0 as x → ∞ (11)
To obtain the boundary condition corresponding to the surface, we consider a Gaussian
pillbox, whose one of the two flat faces is just behind the surface and is inside the solid
the other face is parallel to the surface at distance δr from it.

E x =δx Solution

E=0 Electrode

We apply the Gauss theorem to the volume enclosed by the pillbox. If the flat faces are of
unit area, the charge enclosed by the pillbox is
q = σ + ρδx (12)
where ρ is the charge density in the solution enclosed by the pillbox. If we apply the
Gauss law on this box, we obtain
K K σ + ρδx
∫S ⋅ n ds = εε 0
E (13)

Since the field is in the direction normal to the surface, the contribution to the integral on
LHS by the curved surface of the pillbox is zero. Hence we are left only with the two flat
faces.
K K K K σ + ρδx
E⋅n + E⋅n = (14)
x =0 − x = δx εε 0
Since electrode is a conductor the field inside the solid is zero and we can drop the first
term on the LHS of the above equation. Writing the field in terms of the gradient of
potential, we obtain

3
 ∂φ   σ + ρδx
− x ⋅n = (15)
∂x r = δr εε 0
Note that at x = δx , the direction of the normal is x̂ . Hence
∂φ σ + ρδx
− = (16)
∂x x =δx εε 0
Taking the limit as δx → 0 , we obtain
∂φ σ
− = (17)
∂x x =0 εε 0
The above equation provides the required boundary condition. We take some special
cases of the above equation.
7.3 Guoy-Chapman Model
This theory is applicable for solution of a single symmetric electrolyte
(e.g. NaCl, ZnSO 4 ). In this case, both the cation and the anion of the electrolyte have the
same valency z (z + = − z − = z) . The Poisson-Boltzmann equation can be written for this
case as

∇ φ=−
2 F
εε 0
b
i i
⎛ z i Fφ ⎞
∑i z c exp⎜⎝ − RT ⎟⎠ = − εε ze [
Fc b −Fzφ RT
− ze Fzφ RT
= ]
2Fzc b
εε 0
⎛ Fzφ ⎞
sinh⎜ ⎟ (18)
0 ⎝ RT ⎠
For a plane surface, the above equation reduces to
d 2φ 2Fzcb ⎛ Fzφ ⎞
= sinh ⎜ ⎟ (19)
dx 2
εε 0 ⎝ RT ⎠
In the above equation x represents the distance measured from the surface. We write the
dimensionless potential as
Fzφ
y= (20)
RT
The Poisson-Boltzmann equation then becomes
d 2 y 2F 2 c b z 2
= sinh y (21)
dx 2 εε 0 RT
The inverse Debye length is defined as
12
⎛ 2F 2 I ⎞
κ = ⎜⎜ ⎟⎟ (22)
⎝ εε 0 RT ⎠

4
For a symmetric electrolyte
1
I= ∑
2 i
cib z i2 = c b z 2 (23)

Hence
12
⎛ 2F2 c b z 2 ⎞
κ = ⎜⎜ ⎟⎟ (24)
⎝ εε 0 RT ⎠
Substituting the above in the Poisson-Boltzmann equation, we obtain
d2y
= κ 2 sinh y (25)
dx 2
The above equation has the following boundary conditions
dy
y = 0 and = 0 as y → ∞ (26)
dx
and
dφ σ dy Fzσ
− = or − = (27)
dx x =0 εε 0 dx x = 0 εε 0 RT
The equation can be solved as follows. Substituting
dy d 2 y dp dp
= p and 2
= =p (28)
dx dx dx dy
We can convert the Poisson-Boltzmann equation to
1 d p2( )
= κ 2 sinh y (29)
2 dy
The above equation can be solved to give
p 2 = 2κ 2 cosh y + C (30)

As x → ∞ both p → 0 and y → 0 . Substitution in the above equation gives C = −2 κ 2 and

2
⎛ dy ⎞ ⎛ y⎞
p = ⎜ ⎟ = 2κ 2 (cosh y − 1) = 4 κ 2 sinh 2 ⎜ ⎟
2
(31)
⎝ dx ⎠ ⎝2⎠
Taking square root, we obtain
dy ⎛y⎞
= −2 κ sinh ⎜ ⎟ (32)
dx ⎝2⎠
Negative sign is retained for the following reason. If the surface charge is positive, the
potential in the double layer has positive sign but it decreases with increase in x and

5
 ⎛ y⎞
hence dφ dx and consequently dy dx is negative. On the other hand sinh ⎜ ⎟ is positive
⎝2⎠
and hence negative sign is required to match the signs on both sides. If the surface
charge is negative, then the potential is negative, and increases with x, hence dy dx is

⎛ y⎞
positive. But sinh ⎜ ⎟ is negative and hence we need to retain negative sign.
⎝2⎠
At the surface (i.e. at x = 0 ),

dy ⎛ ys ⎞
= −2κ sinh ⎜⎜ ⎟⎟ (33)
dx x =0 ⎝ 2⎠
We also have
dy Fzσ
− = (34)
dx x = 0 εε 0 RT
Combining the two equations and the subsequent rearrangement gives
⎛ Fzσ ⎞
ys = 2 sinh −1 ⎜⎜ ⎟⎟ (35)
⎝ 2εε 0 RT κ ⎠
Putting back the expressions
12
⎛ 2F2 c b z 2 ⎞ Fzφs
κ = ⎜⎜ ⎟⎟ and y s = (36)
⎝ εε 0 RT ⎠ RT

into the eq 35, we obtain

⎛ ⎞
⎜ ⎟
Fzφs ⎜ Fzσ ⎟
= 2 sinh −1 ⎜ 12 ⎟
(37)
RT ⎜ ⎛ 2F 2 c b z 2 ⎞ ⎟
⎜ 2εε 0 RT⎜⎜ εε RT ⎟⎟ ⎟
⎝ ⎝ 0 ⎠ ⎠
Simplifying the above expression we get
⎛ σ ⎞
sinh −1 ⎜ ⎟
2RT
φs = (38)
Fz ⎜ 8RTεε c b ⎟
⎝ 0 ⎠
The above equation is the relation between the surface charge and the surface potential.
For small values of the surface charge density, eq 35 can be simplified by noting that for
small y, sinh y = y . Hence

6
 2Fzσ Fzφs 2Fzσ
ys = or = (39)
εε 0 RTκ RT εε 0 RTκ
The above equation can be simplified to give
2σ
φs = (40)
εε 0 κ
Thus for small values of the surface charge density, the surface potential is proportional
to the surface charge density. This fact is made use of in obtaining the surface charge
density from zeta potential (a form of the surface potential, obtained using electrokinetic
experiments)
We can integrate the equation (32) ( dy dx = −2κ sinh (y 2) ) between the surface
and any point in the double layer to give
y x
dy
∫s sinh (y 2) = −2κ∫0 x = −2κx (41)
y

Here ys is the value of dimensionless surface potential. We can integrate the expression
on the left by the substitution q = exp(y 2) , dq = (1 2) exp(y 2)dy = (q 2)dy . Then
(y 2)
e( y 2 )
⎛ e ( y 2 ) − 1 e (y 2 ) + 1 ⎞
e
⎛ q −1 ⎞
s
4dq
− 2κx = ∫ = 2 ln⎜⎜ ⎟⎟ = 2 ln ( y 2 ) ⋅ (ys 2 ) ⎟
⎜ (42)
⎛ 1 ⎞ ⎝ q + 1 ⎠ e⎝ ⎠
⎛⎜ ys 2 ⎞⎟ ⎜e +1 e − 1 ⎟⎠
⎛⎜ ys 2 ⎞⎟
q ⎜ q − ⎟ ⎝
e ⎝ ⎠
⎜ q ⎟⎠
⎝
The above equation on rearrangement gives

(y 2) (e(
= (
ys 2
) ( s
)
) + 1 + e (y 2 ) − 1 e − κx 1 + tanh (ys 4)e − κx
=
e
(e ys 2)
) ( s
)
+ 1 − e (y 2 ) − 1 e − κx 1 − tanh (y 4)e
s − κx
(43)

For small ys , the above equation can be simplified by substituting exp(y ) = 1 + y and
tanh y = y

⎛ ys ⎞ − κx
1 + ⎜⎜ ⎟⎟e
y
1+ = ⎝4⎠ or y = ys e − κx (44)
2 ⎛ ys ⎞ − κx
1 − ⎜⎜ ⎟⎟e
⎝4⎠
In other words
y φ
= e − κx or s = e − κx (45)
y s
φ

7
This means for small surface potentials, the potential in the double layer varies
exponentially with the distance. At a distance of κ −1 , the potential will fall to 1 e times its

value at the surface. Hence κ −1 is called the effective thickness of the diffuse double
layer.
For large values of the surface potentials ys >> 1 , and for x >> κ −1 , we can write
tanh (ys 4) = 1 and hence

1 + e − κx
e( y 2 ) = (46)
1 − e − κx
Taking logarithm of both sides and simplifying the expression for x >> κ , we get
( ) ( )
y = 2 ln 1 + e − κx − 2 ln 1 − e − κx = 4e − κx
Or
4RT − κx
φ= e (47)
zF
Thus even in this case, the potential at some distance away from the surface is
4RT
exponential but with a apparent surface potential equal to , which is independent of
zF
the actual value of the surface potential. For a monovalent counterion at room
temperature, the apparent surface potential is
4RT 4 × 8.314 × 298 × 1000
= mV ≈ 100mV
zF 1 × 96500
This implies that when the surface charge is very high, its influence is only over a short
distance from the surface. This is clear from the surface boundary condition
∂φ σ
− = , which implies that when σ is high, the potential decays very rapidly.
∂x x =0 εε 0

∂φ σ
(Incidentally, the boundary condition − = can also be derived using the charge
∂x x =0 εε 0
∞ ∞ ∞
d 2φ dφ dφ
balance, i.e. σ = − ∫ ρdx = εε 0 ∫ 2 dx = εε 0 = −εε 0 )
0 0
dx dx 0 dx x = 0

We have already obtained the equation relating the surface charge density and the
surface potential, i.e.

8
 ⎛ σ ⎞
sinh −1 ⎜ ⎟
2RT
φs = (48)
Fz ⎜ 8RTεε c b ⎟
⎝ 0 ⎠
It is seen that the surface potential increases with increase in the surface charge density.
At the temperature of 298 K,

2 × 8.314 × 298 × 1000 ⎛ σ ⎞

φs = sinh −1 ⎜⎜ ⎟ mV
⎟
−12
96500z ⎝ 8 × 80 × 8.314x 298x8.854 ×10 × c ⎠
b

51.3 ⎛ 266.9σ ⎞
= sinh −1 ⎜⎜ ⎟
⎟
z ⎝ ⎠
b
c

Thus for a 1:1 electrolyte (z = 1) , with bulk concentration of 0.1 M and σ = 0.16 C ⋅ m −2 ,
we get
⎛ 266.9 × 0.16 ⎞
φs = 51.3 sinh −1 ⎜⎜ ⎟ mV = 51.3 × 2.158 = 111mV
3 ⎟
⎝ 0.1× 10 ⎠
Note that for a given charge density, the gradient of the potential at the surface is
fixed. The surface potential will therefore depend upon the thickness of the double layer.
Lesser the thickness lower is the surface potential. Thus, higher the bulk concentration of
the electrolyte or higher the valency, lesser is the surface potential.
7.4 Stern Model
The Guoy-Chapman treatment fails when the surface potential φs is high. By
Boltzmann equation,
csi
= e − Fz i φ
s
RT
b
ci

If φs is 300 mV, z i = −1 , T = 298 , then

csi ⎛ 96500 × 1 × 0.3 ⎞

= exp⎜ ⎟ = exp(11.68) = 118756
⎝ 8.314 × 298 ⎠
b
ci

If cib = 1mol ⋅ m −3 , then cs = 118756 mol ⋅ m −3 = 118 M . This value is physically

impossible. This means that the surface potential cannot attain a value beyond a certain
limit. Stern argued that there must be a relatively immobile layer of adsorbed molecules
at the surface. A substantial drop of potential must occur in this layer. He called this layer
as Stern layer. Stern model is, thus, a combination of the Helmholtz and Guoy-Chapman

9
models. It postulates partial adsorption of molecules and ions forming the Helmholtz
layer (also called the Stern layer), followed by the diffused double layer. Potential drop is
linear in the Stern layer, while it follows Poisson-Boltzmann equation in the diffused
double layer. The following are the details of the Stern model.
Let S0 be the total moles of available sites on the surface. Each site has

valancy z s . Then the total surface charge is z s FS0 . Let Ss be the moles of site which are

occupied by the adsorbed molecules. The moles of unoccupied sites are S0 − Ss . The

surface potential is ψ 0 and that at the outer boundary of the stern layer is ψ δ . The
equilibrium between the bulk solution and the outer plane of the Stern layer is given by
the Boltzmann equation

ciδ ⎛ z i Fφδ + u ic ⎞
= exp⎜⎜ − ⎟⎟ (49)
cib ⎝ RT ⎠
Here, z i is the valancy of the adsorbed ion i and u ic is the extra potential due to van der
Waals forces as well as other forces such as those due to hydrogen bond and chemical
bond.
The solvent adsorbs on the rest of the sites according to
csδ ⎛ u ⎞
= exp⎜ − s ⎟ (50)
⎝ RT ⎠
b
cs

Note that csδ ∝ S0 − Ss and ∑c

i
δ
i ∝ Ss . Also

∑c δ
i
cib (
⎛ z i Fφδ + u ic − u s ) ⎞⎟
i
b
=∑ exp⎜⎜ − ⎟ (51)
c s i csb ⎝ RT ⎠
or

Ss σ − σs
= 0
cb ⎛ z Fφδ + u ic − u s
= ∑ ib exp⎜⎜ − i
( ) ⎞⎟ (52)
S0 − Ss σs ⎟
i cs ⎝ RT ⎠
Here σ0 and σs are the surface charge densities of the bare and the occupied surface
respectively. This implies that adsorbed counterions have the same valancy as the surface
groups.

10
 cib
When the bulk solution is dilute, the ratio b
can be replaced by x ib , the mole
cs
fraction of i in the bulk. Hence

σ 0 − σs
= ∑ x ib exp⎜⎜ − i
(
⎛ z Fφδ + u ic − u s ) ⎞⎟ (53)
σs ⎟
i ⎝ RT ⎠
Or
σs 1
= (54)
σ0 (
⎛ z Fφδ + u ic − u s
1 + ∑ x ib exp⎜⎜ − i
) ⎞⎟
⎟
i ⎝ RT ⎠
Here σs is the effective charge density of the Stern layer. Since the diffuse double layer

( )
exists outside the Stern layer x δ < x < ∞ , we can apply the Poisson-Boltzmann equation
in that region as.
d 2φ
2
= − κ 2φ (55)
dx
The boundary conditions for the above equation are
dφ σ
− = s , φ → 0 as x → ∞ (56)
dx x = x δ εε 0
The last three equations can be solved simultaneously to yield the potential profile and
hence the value of φδ . The surface potential can then be obtained by assuming that the
potential drop in the Stern layer is linear.

dφ φ0 − φ δ σ
− = = s (57)
dx x =0 δ ε′ε 0
Here, δ is the thickness of the Stern layer and ε′ is the dielectric constant of the stern
layer. This dielectric constant is likely to be substantially lower than that in the bulk.
The value of δ ε′ can be estimated by using electrocapillary effect and the related
studies.
Consider σs = 0.05 C.m −2 , ε′ = 30 , ε 0 = 8.854 x10 −12

φ0 − φδ 0.05
= −12
= 1.88x108 V.m −1
δ 30x8.854x10

11
Thus for δ = 0.5nm , φ0 − φδ = 0.094V . This implies that a substantial potential drop
occurs in the Stern Layer.

Solvent molecule Solvated cation

Negatively charged Electrode

Adsorbed anion

1 2 3
Fig: Double Layer Diagram
1.Inner Helmoholtz plane 2. Outer
Helmoholtz plane 3. Diffuse Layer
The stern model postulates that double layer consists of two regions, a Stern layer,
consisting of the adsorbed molecules/ions and the diffuse layer, consisting of mobile
molecules/ions. The molecules in the Stern layer consist of two types, those which are
held by pure electrostatic forces and those which are adsorbed through other forces. The
later are called specifically adsorbed molecules. Most cations are adsorbed by purely
electrostatic attraction to negatively charged surfaces. These cations can not completely
shed their water of hydration. Hence the surface of a cathode consists of a layer of water
molecules followed by a layer of cations. The plane passing through the centre of these
cations is called the outer Helmholtz plane. On the other hand, most anions or other
neutral molecules (say anionic and non-ionic surfactants) may adsorb specifically, i.e.
through forces other than electrostatic. These molecules can completely shade their water
of hydration and hence can approach much closer to the surface. A plane through the

12
centre of these specifically adsorbed ions/molecules is called the inner Helmholtz plane.
These two planes divide the Stern layer into two regions, the region between the surface
and the inner Helmholtz plane and that between the inner and the outer Helmholtz plane
(see the figure below).
The dielectric constant of these two regions can be widely different and hence a
more appropriate model is two Stern layers model. However, it is not possible to measure
the dielectric constants of the two layers separately and hence certain degree of
empiricism sets in, in these models, which reduce their efficacy.
7.7 Capacitance of double layer
Capacitor is a device to store electric charge. It consists of two metal plates
having a potential difference imposed on them. The region between the plates is filled
with a medium with low dielectric constant. Due to the field generated by the potential
difference, the plates are polarized. The plate at high potential is positively charged and
the other one is negatively charged. The charge on the plates are equal, but opposite in
sign. The dipoles of the medium get orientated so as to nullify (at least partially) the
applied field. This lowers the potential difference between the plates. The integral
capacity of the capacitor (per unit area) is defined as
σ
C= (58)
Δφ
where σ is absolute charge density ( C.m-2) on each plate and Δφ is the imposed potential
difference. The SI unit of C is Farad.m-2. (A capacitor is nonlinear if σ does not vary
linearly with Δφ . In such cases, C is not constant, but varies with Δφ . In such cases, it is
advantageous to define differential capacity per unit area, Cd as follows

⎛ dσ ⎞
C d = ⎜⎜ ⎟⎟ (59)
⎝ dφ ⎠
For a capacitor consisting of two infinite parallel plates placed at distance d
between them and filled with a medium with dielectric constant ε , we can relate the
surface charge density with the potential difference as follows. We write the Poisson
equation between the plates as
d 2φ
=0 (60)
dx 2

13
The right hand side is zero since there is no charge between the plates. The Gauss law
applies on the left plate ( kept at x=0) gives
dφ σ
− = (61)
dx x = 0 εε 0
We also know the potential difference Δφ between the plates. Integration of eq 60 and
application of eq 61 gives
dφ σ
=− (62)
dx εε 0
Further integration yields
Δφ σ
= (63)
d εε 0
This produces the following expression for the capacity
σ εε 0
C= = (64)
Δφ d
The capacitance (also called capacity) is directly proportional to the dielectric constant of
the medium and inversely proportional to the distance between the plates. Note that this
is a linear capacitor.
We can view a double layer as a capacitor. One of the plates is the electrode. The
other plate is considered to be imaginary plane in the bulk solution. As we apply a
potential difference between the electrode and the bulk solution, the electrode gets
charged. The double layer can therefore be viewed as a capacitor.
If we assume that the double layer consists of only Helmholtz layer of thickness
d, then it can be treated as a parallel plate capacitor for which
ε h ε0
C = Cd = (65)
dh

Here d h is the thickness of the Helmholtz layer and ε h is its dielectric constant. This
implies that the differential capacity of the double layer is independent of the solution
concentration as well as the electrode potential.
On the other hand, if we assume that Guoy-Chapman model is valid i.e. only the
diffuse double layer is present then the charge density on the electrode is related to the
electrode potential by eq 38

14
 ⎛ σ ⎞
sinh −1 ⎜ ⎟
2RT
φs = (66)
Fz ⎜ 8RTεε c b ⎟
⎝ 0 ⎠
Noting that this relation in nonlinear, we obtain the differential capacitance iof the double
layer as
⎛ ∂σ ⎞ ⎛ Fzφs ⎞
C d = ⎜⎜ s ⎟⎟ = εε 0 κ cosh⎜⎜ ⎟⎟ (67)
⎝ ∂φ ⎠ c b ⎝ 2RT ⎠
Since the inverse Debye length κ is directly proportional to the square root of the ionic
strength of solution, the differential capacity of the Guoy-Chapman layer is proportional
to the square root of the ionic strength. It also depends on the hyperbolic cosine of the
electrode potential.
It is possible to view a double layer as a composite capacitor. It consists of a Stern
layer and a diffuse layer. The Stern layer acts as a parallel plate capacitor (Helmholtz
capacitor) and its differential capacity is denoted by C H and that of the diffuse layer by
CG. We can split the total potential drop across the double layer as
Δφ = ΔφS + Δφd (68)
Differentiation by σ yields
⎛ dΔφ ⎞ ⎛ dΔφS ⎞ ⎛ dΔφd ⎞
⎜ ⎟=⎜ ⎟+⎜ ⎟ (69)
⎝ dσ ⎠ ⎝ dσ ⎠ ⎝ dσ ⎠

1 1 1
Or = + (70)
Cd CH CG

C H can be assumed as constant and C G is obtained from eq 67 as

⎛ Fzφs ⎞
CG = εε 0 κ cosh⎜⎜ ⎟⎟ (71)
⎝ 2RT ⎠

⎛ Fzφs ⎞
For low values of the surface charge, φs is small and hence cosh⎜⎜ ⎟⎟ → 1 . C G in this
⎝ 2RT ⎠
case is proportional to κ , which increases with increase in the concentration of the
electrolyte. At high electrolyte concentrations, C G >> C H , and from eq 70, we see that

C d = C H . Thus at sufficiently high concentration of the electrolyte, the double layer is

15
effectively a Helmholtz capacitor. On the other hand, at sufficiently low electrolyte
concentrations, C H >> C G and the double layer behaves as a Guoy-Chapman capacitor

having capacity C G .

7.8 Electrocapillary effect: Lippmann equation

It is neither possible to measure the concentration profile, nor the potential profile
in the double layer. How does one discern the structure of the double layer? It was
observed by Lippmann that interfacial tension of mercury electrode-electrolyte interface
changes with change in the potential of the electrode. This effect is known as
electrocapillary effect. Lippmann showed that change in the interfacial tension with
electrode potential is related to the surface charge density of the electrode by the
following equation, which is known as Lippmann equation
⎛ ∂γ ⎞
σ = −⎜⎜ ⎟⎟ (72)
⎝ ∂Δφ ⎠μ ,T
In the above equation, γ is the interfacial tension (J.m-2), Δφ is the potential difference
between the electrode and the solution (V) and σ is the surface charge density (C.m-2).
Constancy of μ implies that the composition of the solution is kept constant.
Lippmann equation can be derived as follows. We first prove the Gibbs
adsorption equation. We consider an interfacial region between two phases α and
β having area A. In the interfacial region, concentrations of the components vary with
distance. The Gibbs interface is a plane arbitrarily located in the interfacial region. The
bulk concentrations are extended to the interface. The quantity A(a iα − a βi ) is not
accounted in this procedure and is associated with the interface. This is called the excess
mass of i in the interface.
The excess mass varies with the location of the Gibbs interface. It can be positive
or negative. It is customary to choose the Gibbs plane so that the excess mass of the
solvent is zero.
We write the fundamental equation of thermodynamics for the Gibbs
interface
dG = −SdT + Vdp + γdA + ∑ μi dn i (73)
i

16
 z= L2

β cβi
Interfacial
Zone. Width
Exaggerated a βi
z=0 Gibbs Interface
a αi ci(z)

α
c αi
z = -L1

In the above equation, μi is the electrochemical potential of species i in the bulk phase
and A is the interfacial area. γ is the interfacial tension and can be considered as the
excess free energy of the interface per unit area. At constant temperature and pressure,
the above equation simplifies to
dG = γdA + ∑ μi dn i (74)
i

Since the variables in the differential are extensive, we can integrate the equation over the
system size to obtain
G = γA + ∑ μi n i (75)
i

Total differentiation of the above equation gives

dG = γdA + dγA + ∑ μi dn i + ∑ dμi n i (76)
i i

Comparison between eq 70 and 72 gives, the Gibbs-Duhem equation for the interface
Adγ + ∑ n i dμi = 0 (77)
i

Note that μi are the electrochemical potentials of species in the bulk phases. Dividing eq
77 by A , we obtain
dγ + ∑ Γi dμi = 0 (78)
i

ni
where Γi = (79)
A

17
It is called the interfacial excess of species i .
Consider now a system consisting of a mercury electrode and an electrolyte say
NaCl. The interface in this case is consists of Na+ and Cl- ions, water, mercury atoms and
electrons. Suppose we vary the electric potential of mercury and measure it with a normal
Hg-Hg2Cl2 electrode which is introduced in the electrolyte using a Luggin capillary. At
the reference electrode, the following reaction occurs
1
Hg 2 Cl 2 + e − ↔ Hg + Cl − Δφ0 = 0.2676V (80)
2
Within a certain range of potentials, mercury acts as an ideally polarizable electrode, that
is, it is polarized in order to balance the excess charge in the solution. For example, if Cl-
ions are generated at the Hg-Hg2Cl2 electrode, by forward reaction of eq80, then those
ions will accumulate in the double layer of Hg electrode. This excess negative charge is
exactly balanced by the positive charge generated inside mercury phase (this charge
resides at the interface.). This positive charge is generated by the consumption of
electrons at Hg-Hg2Cl2 electrode. In this case Hg electrode is at positive potential with
respect to Hg-Hg2Cl2 electrode. In the opposite situation, i.e. when Hg electrode is at
negative potential, sodium ions balance the negative charge on the mercury electrode and
there is a net consumption of Cl- ions at Hg-Hg2Cl2 electrode.
At the equilibrium, we have
1
μHg 2Cl 2 (β ) + μe − (β ) = μHg (β ) + μCl− (α ) (81)
2
Here, the reference electrode phase represented by β , the solution phase by α and the
mercury phase by ε . We differentiate the equation and note that d μHg 2 Cl2 (β ) = 0 and

d μHg (β ) = 0 because both are uncharged species and their composition does not change.

Hence we obtain
d μe − (β ) = d μCl− (α ) (82)

We expand eq 78 as
dγ + ΓNa + d μ Na + (α ) + ΓCl − dμCl− (α ) + ΓH 2O dμ H 2O (α ) + ΓH g dμ Hg (ε ) + Γe − d μe − (ε ) = 0 (83)

18
Since Gibbs plane is chosen so as to obtain ΓH 2O = 0 , we can drop the corresponding term

fro eq 83. Also dμ Hg (ε ) is zero because Hg is pure. This reduces eq 83 to the following

form.
K
dγ + ΓNa + d μ Na + (α ) + ΓCl − dμ Cl− (α ) + Γe − d μe − (ε ) = 0 (84)

Next we apply charge balance equation

ΓNa + − ΓCl − − Γe − = 0 (85)

We use eq 85 to eliminate ΓCl − fro eq 84 and obtain

dγ + ΓNa + d μ Na + (α ) + (ΓNa + − Γe − )d μCl − (α ) + Γe − dμ e − (ε ) = 0 (86)

Noting that
μ Na + (α ) + μCl− (α ) = μ NaCl (α ) (87)

we modify eq 86 to
K
dγ + ΓNa + dμ NaCl (α ) − Γe − d μCl− (α ) + Γe − dμ e − (ε ) = 0 (88)

Writing surface charge density σ as

σ = − FΓe − (89)

and also using eq 82, we obtain

[ K
]
dγ + ΓNa + dμ NaCl (α ) − σ d μe − (β ) − dμ e − (ε ) = 0 (90)

But
K
[
Δφ = F μ e − (ε ) − μe − (β ) ] (91)

Hence
dγ + ΓNa + dμ NaCl (α ) + σdΔφ = 0 (92)

This is the Lippmann equation. When the composition of the electrolyte is kept constant
and only its potential is varied, then it reduces to
dγ + σdΔφ = 0 (93)
which on rearrangement yields eq 72.
The differential capacitance of the double layer is given by eq 59, which on
combining with eq 72 yields
⎛ ∂2γ ⎞
Cd = −⎜⎜ ⎟
2 ⎟
(94)
⎝ ∂Δφ ⎠μ ,T

19
7.9 Electrocapillary curves
Electrocapillary studies are conducted using the dropping mercury electrode
instrument. It is shown in the figure below. It consists of a capillary of 50 μm internal
diameter. It is fed from a reservoir of mercury so that the available head of mercury is
200-1000 mm. Mercury issues from the capillary in the form of nearly spherical drops.

Dropping Mercury Electrode

Potentiaostat

electrolyte

The Drop grows until its weight is no longer supported by the capillary. The mass
of the drop, M, is therefore related to the interfacial tension by the following relation
2πrγ cos θ
M= (95)
(ρm − ρ)g
Here r represents radius of the capillary, ρm is the density of mercury, ρ is the
density of the electrolyte solution and θ is contact angle between mercury and the
capillary.

20
 The mass of the drop can be obtained by counting the number of drops falling per
unit time. Thus, if m is the mass flow rate of mercury and n is the number of drops falling
from the capillary per second, then
m
M= (96)
n
Using these equations, it is possible to measure the interfacial tension between
mercury and the electrolyte.
The drop disengaged from the capillary falls in the pool of mercury at a short
distance below. The capillary is immersed in the electrolyte solution under study. The
mercury drop is made the working electrode and the mercury pool the counterelectrode.
The most important aspect which distinguishes the dropping mercury electrode from the
rest of the electrodes is that when drop falls through the electrolyte at its terminal
velocity, it agitates the liquid and helps to attain uniformity. If the potential is kept fixed,
then each drop behaves exactly as any other drop. The experiments are, therefore, highly
reproducible. Since the charge accumulated on the drop flows as current in the circuit, it
is possible to measure the surface charge density of the drop by equation
I
σ= (97)
4πrd2 n
The following figure shows plots of interfacial tension of mercury electrolyte
interface as a function of electrode potential. The potential is relative to normal calomel
electrode and shifted by 0.48 V in order that the surface tension maximum of KOH is 0
V. Potential so obtained is called the rational potential.

21
 420
KOH

380
Ca(NO3)2

340 NaCl
γ, mN.m −1

300
NaBr KCNS
KI

260

-1.4 -1.0 -0.6 -0.2 0.2 0.6

Potential, V

Following observations can be made about these curves

1. It is seen that the electrocapillary curve is approximately a parabola. It shows a
maximum. From Lippmann equation we see that the interfacial tension reaches an
extremum when the surface charge of the electrode is zero. This point is called the point
of zero charge. Let the potential difference at this point be denoted by Δφ0 . The equation
of the curve can be described by the following equation
γ = a + bΔφ − cΔφ2 (98)

We also have γ max = a + bΔφ0 − cΔφ02 and b − 2cΔφ0 = 0 . These two equations can be
combined with eq 86 to give
( )
γ max − γ = 2cΔφ0 (Δφ0 − Δφ) − c Δφ02 − Δφ 2 = c(Δφ0 − Δφ)
2

This on rearrangement gives

γ = γ max − c(Δφ0 − Δφ)
2
(99)
Note that c is a positive constant.
This gives the following expressions for the surface charge density and the double
layer capacitance respectively.

22
 ⎛ ∂γ ⎞
σ = −⎜⎜ ⎟⎟ = 2c(Δφ − Δφ0 ) (100)
⎝ ∂Δφ ⎠μ ,T

⎛ ∂2γ ⎞
Cd = −⎜⎜ ⎟ = 2c
2 ⎟
(101)
⎝ ∂Δφ ⎠μ ,T
Thus we see that capacity of the double layer is constant, provided the curve is a perfect
parabola. We can the write the equation for the electrocapillary curve as
Cd
γ = γ max − (Δφ0 − Δφ)2 (102)
2
Constant capacity of double layer is indicates that the double layer behaves as
Helmholtz capacitor. Actual electrocapillary curves, however, differ from perfect
parabola.
2. It is seen that the large part of the negative branch of the electrocapillary curve is
independent of the type of the electrolyte. This implies that the potential and charge
relationship for this branch is the same for all electrolytes. This is possible only if there is
no specific adsorption of ions on the electrode surface. The surface potential of the
electrode depend on the structure of the double layer and on the ionic strength of the
solution. It is expected that all the curves are at the same ionic strength. Hence, only the
structure of the double layer matters. If there is a specific adsorption, then different ions
will adsorb to a different extent. The structure of the double layer will then be different.
From this we conclude that the specific adsorption does not occur in the negative branch.
Since we expect cations to adsorb in this branch (electrode surface is negative), we
conclude that cations are not specifically adsorbed i.e. they adsorb under the influence of
the electrostatic forces alone. The reason that cations are not specifically adsorbed is that
they are strongly hydrated (being small) and hence they come in contact with the
electrode through water layer. On the other hand the positive branch splits into different
curves for different electrolytes. We therefore expect anions to be specifically adsorbed.
A greater slope of the curve implies greater charge on the electrode and hence greater
charge screening, which implies a greater specific adsorption. From the figure, we see
that the specific adsorption follows the trend: I − > Br − > Cl − > OH − .
3. Since OH- ions are strongly hydrated, it is reasonable to assume that they are not
specifically adsorbed on the electrode. Hence the point of zero charge in KOH

23
corresponds to mercury-water interface. The potential of the electrode corresponding to
this point is -0.48 V with respect to normal calomel electrode. The scale in the figure is
shifted by 0.48 V so that this point corresponds to zero. This new scale is called the
rational potential scale. If anion is specifically adsorbed, the effective surface charge
becomes negative even when the electrode has no intrinsic charge. Hence the rational
potential is negative at the point of zero charge. More negative the potential is stronger is
the adsorption.
The table below lists the potentials of zero charge for various electrolytes with
respect to normal calomel electrode in KCl.
Electrolyte Concentration, Potential Electrolyte Concentration, Potential
M (V) M (V)
LiCl 0.1 -0.5592 HCl 0.1 -0.558
NaCl 0.1 -0.5591 NH4Cl 0.1 -0.5587
KCl 1 -0.5555 CaCl2 0.1 -0.5586
KCl 0.7 -0.5535 BaCl2 0.1 -0.5587
KCl 0.1 -0.5585 CoCl2 0.1 -0.5585
KCl 0.01 -0.5936 AlCl3 0.1 -0.5585
KCl 0.001 -0.640 KNO3 0.1 -0.5166
NaF 0.1 -0.474 KBr 0.1 -0.5741
KHCO3 0.1 -0.4728 KCNS 0.1 -0.626
KOH 0.1 -0.4767 KI 0.1 -0.723

4. It is seen that it is the anion and not the cation which governs the potential of zero
charge.
5. As the concentration of the electrolyte decreases, the surface potential becomes more
negative. This is consistent with eq 66 (rewritten below) which shows that φs increases as

c b decreases for the same charge density

⎛ σ ⎞
sinh −1 ⎜ ⎟
2RT
φs = (103)
Fz ⎜ 8RTεε c b ⎟
⎝ 0 ⎠

24
The effective charge density should in principle decrease with decrease in the electrolyte
concentration, but since surface concentration of the counterions is much higher than the
bulk, adsorption plateau may be reached even at very low bulk concentrations and hence
it is reasonable to assume that the effective surface charge density does not vary with
concentration of the electrolyte.
The strong dependence of φs on the concentration of the electrolyte at low concentrations
indicates that diffused double layer theory works well at low concentration. It is also
found that the electrocapillary curve deviates from the parabolic form as the electrolyte
concentration decreases.

25
 Lecture-8
Instrumental Techniques in Electrochemistry-I
8.1 Classification of the techniques
Kinetics of electrode processes could be dynamically studied using three-
electrode system (working-counter-reference electrode). The working electrode is a
microelectrode (of the size of the order of few microns) so that fluid around the
electrode is not affected by convective currents in the fluid. Hence it can be assumed
that transport to and from the electrode occurs as if it is placed in a stagnant fluid of
infinite dimension.
The available techniques can be broadly classified into two classes. They are:
1. Controlled potential techniques: In this technique potential difference between the
working and the reference electrode is varied with time in a controlled manner and the
current in the primary circuit (working and the counterelectrode) is measured as a
function of time. This class of technique is also called chronoamperometry. The
voltage is controlled using an instrument called potentiostat.
2. Controlled current techniques: In this technique, the current in the primary circuit is
varied with time using an instrument called Galvanostat and the potential difference
between the working and the reference electrode is measured as a function of time.
This class of techniques is also called chronopotentiometry.
8.2 Controlled Potential techniques
A typical setup is shown below

I(t) Counter
electrode
Function Reference electrode
Generator Potentiostat
Working
electrode
E-controlled
I(t)

Ammeter

Function generator provides a program to enable the potentiostat to vary the potential
difference between the working electrode and the reference electrode in a controlled
manner. Using an ammeter, the current in the primary circuit is measured.

1
 Depending on the manner in which the potential is varied, we have a variety of
potential control techniques. These are listed below
1. Single step, double step and pulse chronoamperometry: In this technique, the
potential function is a single step or a combination of a positive step followed by the
equal size of negative step or a pulse. The current is measured as a function of time.
2. Polarography: Instead of a solid metal working electrode, the technique uses a
dropping mercury electrode. The drops of mercury are formed at regular intervals at
the tip of a glass capillary. Variation of the current in the primary circuit during the
growth of the drop is measured. Successive drops may be subjected to same or
varying potential.
3. Chronocoulometry: In this technique, instead of measuring current, charge passed
through the circuit in measured as a function of time.
4. Cyclic voltammetry: In this case, potential-time function is saw-toothed i.e. a linear
increase followed by a linear decrease at the same rate.
8.3 Controlled current techniques (chronopotentiometry):
A setup for chronopotentiometry is shown below. A controlled current source
varies current with time in a programmed manner. The potential is measured between
the working and the reference electrode.

Ammeter

Counter electrode
Programmed current source
Potential recorder
Working electrode

The technique can be further classified as follows

1. Constant current chronopotentiometry
2. Chronopotentiometry with linearly increasing current.
3. Current reversal chronopotentiometry
4. cyclic chronopotentiometry

2
8.4 Cyclic Voltammetry
This is the most popular technique to characterize electrode kinetics. This
technique is also known as linear potential sweep chronoamperometry or linear sweep
voltammetry. The potential sweep is saw-toothed as shown in Figure-1

E (-)

Em

E0

Ei t
0 tm 2tm

Figure-1: Potential sweep in cyclic voltammetry

Potential begins from some initial value Ei, which is above the Nernst potential,
reduces linearly with time goes to minimum value Em at time tm and then increases
back again at the same time rate till the initial value is reached. It is customary to
draw this diagram by plotting decreasing E along positive x-axis.

I
A + e− → A −

0 E (-)
Ei E0 Em

A − → A + e−

Figure 2: A typical cyclic voltammogram

3
 The part of the sweep between E0 and Em is the cathodic sweep, whereas that
between Ei and E0 is anodic. Different sweep rates are used to get complete
information about the system. A typical current versus time plot is shown in Figure-2.
It is explained qualitatively with an example of a redox reaction shown in the figure.
Initially the system contains only A. Since the potential is well above the
Nernst potential, only nonfaradaic current flows through the system. When the
potential approaches the Nernst potential, the reduction begins and the cathodic
current begins to rise. As the potential grows more negative, the concentration of A
near the electrode begins to drop and that of A- begins to rise. The rate of reduction
does not rise as fast as that observed in the beginning. Finally, current reaches a peak
and begins to fall due to lack of availability of the oxidant A. The fall continues till
voltage reaches a plateau. This is the limiting current where the concentration at the
electrode is zero and the rate of the reaction is controlled of diffusion of reactant and
the product to and from the electrode.
When a reverse sweep is applied, current remains at the plateau up to some
value of the potential and then begins to fall. This is because the reduction kinetics
becomes controlling. When potential approaches the Nernst potential, current fall is
much rapid since anodic reaction is now beginning to occur and there is a high
concentration of A- in the vicinity of the electrode. This continues till A- begins to
limit the oxidation, then current rise is reduced and finally a minimum value is
attained. With further increase in the potential, the current profile turns up and finally
current becomes zero when potential is sufficiently above the Nernst Potential.

I
A + e− → A −

0 E (-)
Ei E0 Em

Fig 3: Cyclic voltammogram of irreversible reduction process

The overall shape of the curve allows us to obtain nature of the electrodes
processes. Details of the kinetics could be obtained by studying the set of the curves
4
obtained over a range of the sweep rates. For example, if anodic peak does not appear
during the return sweep, see figure -3, it implies that the reduction is irreversible i.e.,
once A- is formed, it cannot be oxidized back to A.
It is also possible that some processes are slow and therefore do not occur if
the sweep rate is too high. Thus by comparison of cyclic voltammogram at different
sweep rates one can get idea about relative speeds of different steps.
Example-1
A cyclic voltammogram of ClC6H4CN (p-chlorocyanobenzene) in acid
solution is as shown in the figure below. Explain the nature of the diagram based on
the following mechanism.
ClC6 H 4CN + e − ⇔ ClC6 H 4CN − (1)

ClC6 H 4CN − + H + + e − ⇔ C6 H 5CN + Cl − (2)

C6 H 5CN + e − ⇔ C6 H 5CN − + Cl − (3)

0 E (-)

Figure-4 Cyclic Voltammogram forExample-1

Solution: During the cathodic sweep, step-1 takes place first i.e. conversion of p-
chlorocyanobenzene to the corresponding anion radical. Step-2, may occur
simultaneously along with step1. Step-3 has a higher negative value of the Nernst
potential compared to step-1 and hence, does not occur at low cathodic potential. The
peak appears when the concentration of p-chlorocyanobenzene at the electrode
surface reduces to a low value. The current begins to decrease and continues to
decrease until the potential becomes sufficiently cathodic to promote the third step.
The current then begins to increase until the concentration of the benzyl cyanide at the
electrode surface also becomes very low. The second peak appears and current begins

5
to decrease. During the reverse sweep, current slowly deceases till step-3 begins to
reverse. Thus the current becomes anodic. Meanwhile, p-chlorocyanobenzene is also
accumulating at the electrode surface. Hence it begins to reduce and a cathodic peak
is observed. Beyond this potential is low enough so that no reaction occurs and we are
back to the base line.
8.5 Theory
Let us consider a reaction O + ne ⇔ R . Let us consider that the electrode is a
infinite plane and the transport occurs in semi-infinite domain in a solution which
initially contains O alone at concentration c∗O . It is assumed that a sufficiently high
concentration of a supporting electrolyte is used so that the motion of O and R in the
solution is controlled by the diffusion and not the migration. We take different cases.
Case-1: Plane Electrode, Equilibrium Reaction, step change in the electrode
potential- We assume that the reaction is so fast that the equilibrium prevails at all
times at the electrode, as dictated by Nernst equation
RT c R (0.t )
φ( t ) = φ0′ − ln (4)
nF c O (0, t )
It is convenient to write the equation above as
c O (0.t ) ⎡ nF ⎤
θ= = exp ⎢ (
φ( t ) − φ0′ ⎥) (5)
c R (0, t ) ⎣ RT ⎦
The continuity equations for O and R can be written as
∂c O ( x , t ) ∂ 2c O ( x , t )
= DO (6)
∂t ∂x 2
∂c R ( x , t ) ∂ 2c R ( x, t )
= DR (7)
∂t ∂x 2
The initial conditions are
c O (x ,0 ) = c∗O , c R (x ,0 ) = 0 (8)
The boundary conditions corresponding to the bulk of the fluid are
c O (∞, t ) = c∗O , c R (∞, t ) = 0 (9)
The flux balance equation is
⎛ ∂c ( x, t ) ⎞ ⎛ ∂c ( x, t ) ⎞
DO ⎜ O ⎟ + DR ⎜ R ⎟ =0 (10)
⎝ ∂x ⎠ x =0 ⎝ ∂x ⎠ x =0
Before we consider the linear sweep voltammetry, let us consider a simpler
problem of step voltammetry. Here at time t < 0 , the electrode is at a potential at

6
which no current flows. At t = 0 , we change the potential suddenly to a negative value
(on the reduction wave). In this case θ is a constant.
We solve the above set of equations using Laplace transform. The Laplace
transform is defined by the following equation
∞
cO (x, s ) = ∫ cO (x, t )e − st dt (11)
0

Here, the overbar on cO (x , s ) indicates the Laplace transform of the original quantity, s
being the Laplace transform parameter. We use the initial condition given by eq 8, to
obtain
c∗O
c O (x , s ) = + A(s )e
−x s DO
(12)
s

cR (x , s ) = B(s )e
− x s DR
(13)
These equations automatically satisfy the boundary conditions in the bulk.
Transformation of eq 10 in the Laplace domain yields
⎛ ∂ c ( x , s) ⎞ ⎛ ∂ c ( x , s) ⎞
DO ⎜ O ⎟ + DR ⎜ R ⎟ =0 (14)
⎝ ∂x ⎠ x =0 ⎝ ∂x ⎠ x =0
Substituting eq 12 and 13 into eq 14, we obtain
− D O A(s ) s D O − D R B(s ) s D R = 0 or B(s ) = − A(s )ξ (15)

ξ = (D O D R )
12
where (16)
We can use eq 15 to transform eq 13 to the following form

cR (x , s ) = −A (s )ξe
− x s DR
(17)
From eq 5, we have
cO (0, s ) = θcR (0, s ) (18)
From eq 12, 17 and 18, we get
c∗O c∗O
+ A(s ) = −θA(s )ξ or A(s ) = − (19)
s s(1 + ξθ)
Eq 12 and 17 can now be transformed to
c∗O c∗O
cO (x, s ) =
−x
−
s DO
e (20)
s s(1 + ξθ )

ξc∗O
c R (x , s ) = −
−x s DO
and e (21)
s(1 + ξθ)
The inverse transforms of eq 20 and 21 yield the following concentration profiles

7
 c∗O ⎛ x ⎞
c O (x, t ) = c∗O − erfc⎜ ⎟ (20)
1 + ξθ ⎜2 D t ⎟
⎝ O ⎠

ξc∗O ⎛ x ⎞
and c R (x , t ) = erfc⎜ ⎟ (21)
1 + ξθ ⎜2 D t ⎟
⎝ R ⎠

The current is given by the product of the interfacial flux and electrode area, A
⎛ ∂c ( x, t ) ⎞
I(t ) = nFADO ⎜ O ⎟ (22)
⎝ ∂x ⎠ x = 0
Eq 22 could be written in Laplace domain as
⎛ ∂c ( x , s) ⎞
I(s ) = nFADO ⎜ O ⎟ (23)
⎝ ∂x ⎠ x =0
We note from equation 20that

⎛ ∂cO (x, s ) ⎞ c∗O 1

⎜ ⎟ = (24)
⎝ ∂x ⎠ x =0 (1 + ξθ) sD O
Hence

nFAc∗O DO
I(s ) = (25)
(1 + ξθ) s
Laplace inversion gives

nFAc∗O D O
I (t ) = (26)
(1 + ξθ ) πt
The reaction becomes diffusion controlled, if c O (0.t ) = 0 , in which case θ = 0 and eq
26 reduces to the following form

DO
I d (t ) = nFAc∗O (27)
πt
This is called the Cottrell equation. Comparing 26 and 27, we obtain
I d (t )
I (t ) = (28)
(1 + ξθ)
This implies that the shape of the current versus time curve is independent of
the reversible couple used, but its magnitude is scaled by factor 1 (1 + ξθ ) . I(t ) = 0 for
very positive potential and I d (t ) for very negative potential. Thus it always lie

between 0 and I d (t ) .

8
 Suppose one does a series of experiments in which, several potential steps are
used, each of different magnitude. If one measures current at the end of a fixed time τ
and then plots i(τ ) as a function of the potential step, then
I(τ) 1
= (29)
I d (τ) (1 + ξθ)

⎡ nF
Noting that θ = exp⎢ (φ − φ0′ )⎤⎥ , we can transform eq 29 as
⎣ RT ⎦
I(τ) 1
= (30)
I d (τ) ⎡ nF ⎤
1 + ξ exp ⎢ (φ − φ0′ )⎥
⎣ RT ⎦
For a large positive potential, exponential term tends to infinity and the ratio
I(τ) I d (τ) tends to zero. In the other extreme of large negative potential, exponential

term tends to zero and the ratio I(τ) I d (τ) tends to one. Thus I(τ) Id (τ ) varies from
zero to one. The potential at which the ratio is 0.5 is called the half wave potential and
is denoted by φ1 2 . It is easy to see from eq 31 that at the half wave potential

⎡ nF ⎤
ξ exp ⎢ (φ − φ0′ )⎥ = 1 , hence we get
⎣ RT ⎦

RT ⎛ D R ⎞
φ1 2 = φ0 ' + ln⎜ ⎟ (31)
2nF ⎜⎝ D O ⎟⎠

Since the second term on the left is usually smaller than the first, φ1 2 ≈ φ0' .

Case-2: Spherical Electrode, Equilibrium Reaction, step change in the electrode

potential-The above treatment can be extended to spherical electrode of radius r0 . The
diffusion equation changes to
∂c O (r, t ) D O ∂ ⎛ 2 ∂c O (r, t ) ⎞
= 2 ⎜r ⎟ (32)
∂t r ∂r ⎝ ∂r ⎠
∂c R (r, t ) D R ∂ ⎛ 2 ∂c R (r, t ) ⎞
= 2 ⎜r ⎟ (33)
∂t r ∂r ⎝ ∂r ⎠
The initial conditions are
c O (r,0) = c∗O , c R (r,0) = 0 r > r0 (34)
The boundary conditions corresponding to the bulk of the fluid are
c O (∞, t ) = c∗O , c R (∞, t ) = 0 (35)
The flux balance equation is

9
 ⎛ ∂c (r, t ) ⎞ ⎛ ∂c (r, t ) ⎞
DO ⎜ O ⎟ + DR ⎜ R ⎟ =0 (36)
⎝ ∂r ⎠ r = r0 ⎝ ∂r ⎠ r = r0

The equations could be cast into that for plane interface by transformation ψ O = c O r

and ψ R = c R r .
Eq 32 and 33 transform to
∂ψ O (r, t ) ∂ 2ψ O (x, t )
= DO (37)
∂t ∂r 2
∂ψ R (r, t ) ∂ 2ψ R (x, t )
= DR (38)
∂t ∂r 2
The solution in the Laplace domain is
c∗O A(s ) − r
cO (r, s ) = +
s DO
e (39)
s r
B(s ) − r
cR (r, s ) =
s DR
e (40)
r
The flux balance equation 36 allows us to modify eq 40 to
A(s )ξ 2 γ (s ) − r0
cR (r, s ) = −
s DO − ( r − r0 ) s D R
e e (41)
r
Here

1 + r0 s D O
γ (s ) = (42)
1 + r0 s D R

Noting that

cO (0.t ) ⎡ nF ⎤
= θ = exp ⎢ φ − φ 0′ ⎥( ) (43)
cR (0, t ) ⎣ RT ⎦
We obtain from eq 39 and 41,
⎛ 1 ⎞ r0 c*O r0
A (s ) = −⎜⎜ ⎟⎟ s DO
e (44)
⎝ 1 + ξ γ (s )θ ⎠ s
2

The profiles in the transformed domain are

c∗O ⎛ 1 ⎞ r0 c*O −(r − r0 )
cO (r, s ) = − ⎜⎜ ⎟ s DO
e (45)
s ⎝ 1 + ξ 2 γ (s )θ ⎟⎠ rs

⎛ ξ 2 γ (s ) ⎞ r0c*O −( r − r0 )
cR (r, s ) = ⎜⎜ ⎟⎟ s DR
e (46)
⎝ 1 + ξ γ (s )θ ⎠ rs
2

The current is given by

10
 ⎛ ∂c (r, t ) ⎞
I(t ) = nFADO ⎜ O ⎟ (47)
⎝ ∂r ⎠ r = r0
In the Laplace domain, this equation transforms to
⎛ ∂c ( r, s) ⎞
I(s ) = nFADO ⎜ O ⎟ (48)
⎝ ∂r ⎠ r = r0
Substitution of eq 45 yields

nFAc∗O D O ⎛⎜ 1 + r0 s D O ⎞⎟
I(s ) = (49)
1 + ξ 2 γ (s )θ ⎜⎝ r0s ⎟
⎠

1 + r0 s D O
The presence of γ (s ) = does not allow us to invert the transform into the
1 + r0 s D R

time domain in an analytical form. We can however obtain the following limiting
cases. Note that s in the above expression plays role of inverse of time. Thus large
values of s correspond to short time and small values correspond to large time. Thus
for r0 s D O >> 1 (short times), γ (s ) = 1 and

nFAc∗O D O
I(s ) = (50)
(1 + ξθ ) s
This exactly matches with semi-infinite plane electrode eq (25) and hence the solution
is

nFAc∗O D O
I (t ) = (51)
(1 + ξθ ) πt
The reason for this is that at short time the diffusion layer thickness is much smaller
than the radius of the electrode and electrode appears flat. On the other hand if we
have r0 s D O << 1 (long time solution), eq 49 reduces to

nFAc∗O D O
I(s ) =
( )
1 + ξ 2 θ r0s
(52)

The inverse of this is

nFAc∗O D O
I=
(
1 + ξ 2θ r0) (53)

This implies after a sufficiently long time, steady state is reached at the spherical
electrode. This is in contrast to the plane electrode, where the current progressively
decreases to zero. It is seen that electrode behaves as if there is a diffusion film of

11
thickness r0 through which the diffusion occurs at steady state. The reason is at this
distance, the net flux of current going to the particle exactly matches with that coming
from the outside.
As we reduce the radius of the electrode, the condition r0 s D O << 1 is

satisfied at progressively shorter times, i.e., steady state is attained faster. For ultra-
microelectrodes (UME), the steady state occurs so quickly that electrode remains at
quasi-steady state even when the imposed potential varies with time. The current
potential relationship can be obtained as
nFAc∗O D 0
I (t ) = (54)
⎧ ⎡ nF ⎤⎫
r0 ⎨1 + ξ 2 exp ⎢ φ(t ) − φ0′ ( )
⎥⎦ ⎬
⎩ ⎣ RT ⎭
Note that unlike multiple steps used in the case of a plane electrode, we are using a
continuous variation of the potential of UME. At high negative potentials, I(t )
reaches a constant value of the limiting current
nFAc∗O D 0
Id = (55)
r0
At highly positive potentials, the current has zero value.
Case-3: Plane Electrode, Non-equilibrium Reaction, step change in the electrode
potential - Let us consider a one electron reaction O + e ⇔ R conducted at an infinite
dimensional plane electrode. The reaction is slow and the net cathodic current is given
by
I (t ) ⎛ ∂c (x , t ) ⎞
= DO ⎜ O ⎟ = k f c O (0, t ) − k b c R (0, t ) (56)
FA ⎝ ∂x ⎠ x =0

−
(
(1− α ) F φ − φ0′ ) (
αF φ − φ0′ )
where kf = k e 0 RT
and k b = k e 0 RT
(57)
In the Laplace domain eq 56 transforms to
⎛ ∂ c (x , s ) ⎞
DO ⎜ O ⎟ = k f cO (0, s ) − k b cR (0, s ) (58)
⎝ ∂x ⎠ x =0
Eq 12 and 17 are still valid and are rewritten as
c∗O
cO (x , s ) = + A(s )e
−x s DO
(59)
s

cR (x, s ) = −A (s )ξe
− x s DR
(60)
We substitute these into eq 58 and obtain

12
 k f c∗O
A (s ) = −
(
s DOs + k f + k bξ ) (61)

The concentration profiles in the Laplace domain are

−x s D
c∗O k f c∗O e O

cO (x , s ) = −
(
s s D Os + k f + k b ξ ) (62)

−x s D
k f c∗O ξe R

c R (x , s ) =
(
s D Os + k f + k b ξ ) (63)

and the current in the Laplace domain is

I(s ) s
=
∗
FAk f c O s s + H ( ) (64)

kf k
where H= + b (65)
DO DR

The inverse transform of eq 64 is (see Appendix-A for the derivation)

I (t )
FAk f c O∗
2
= e H t erfc H t ( ) (66)

The current is highest at the beginning, with a magnitude of FAk f c∗O . It monotonically
decays to zero with time.
If the reductant is present in the bulk at zero time, then the above equation gets
modified to
I(s )
FA
( )
= k f c∗O − k b c∗R e H t erfc H t
2
( ) (67)

Case-4: Plane Electrode, equilibrium Reaction, ramp change in the electrode

potential- In this case the potential of the electrode changes linearly with time
according to equation
φ(t ) = φi − vt (68)
In the above equation φi is the initial potential of the electrode and v is the sweep rate.
We assume that the reaction at the electrode is so fast that it reaches equilibrium as
dictated by the Nernst equation
RT c R (0.t )
φ( t ) = φ0′ − ln (69)
nF c O (0, t )
Combining eq 68 and 69, we obtain

13
 c O (0.t ) ⎡ nF ⎤
= exp ⎢ (
φi − vt − φ 0′ ⎥ ) (70)
c R (0, t ) ⎣ RT ⎦
The solutions in the Laplace domain are (see eq 12 and 17)
c∗O
c O (x , s ) = + A(s )e
−x s DO
(71)
s

cR (x , s ) = − A(s )ξe
− x s DR
(72)
The current in the Laplace domain is
⎛ ∂ c ( x , s) ⎞
I(s ) = nFADO ⎜ O ⎟ (73)
⎝ ∂x ⎠ x =0
We note from equation 71 that

⎛ ∂cO (x , s ) ⎞ s
⎜ ⎟ = − A(s ) (74)
⎝ ∂x ⎠ x =0 DO

Combining eq 73 and 74, we obtain

I(s )
A (s ) = − (75)
nFA D O s

We use this equation to eliminate A(s ) from eq 71and obtain

c∗O I(s )
c O (x , s ) =
−x
−
s DO
e (76)
s nFA D Os

At x = 0 , the above equation reduces to

c∗O I(s )
cO (0, s ) = − (77)
s nFA D Os

The above equation, on inversion, yields

t
1
c O (0, t ) = c∗O − ∫ I(τ)(t − τ)
−1 2
dτ (78)
nFA πD O 0

In a similar manner, we can derive

t
1
c R (0, t ) = ∫ I(τ)(t − τ)
−1 2
dτ (79)
nFA πD R 0

Combining eq 78 and 79 with eq 70 gives

t
1
∫ I(τ)(t − τ)
−1 2
c∗O − dτ
nFA πD O ⎡ nF ⎤
0
= exp ⎢ (
φi − vt − φ0′ ⎥ ) (80)
⎣ RT ⎦
t
1
∫ I(τ)(t − τ)
−1 2
dτ
nFA πD R 0

14
The above equation, on rearrangement yields
t
(nFA )
πD O c∗O
∫ I(τ)(t − τ)
−1 2
dτ = (81)
⎡ nF ⎤
0 1 + ξ exp ⎢ (
φi − vt − φ0′ ⎥ )
⎣ RT ⎦

where ξ = (D O D R ) . The solution of the above integral equation yields I(t ) as a

12

function of time. The equation does not have analytic solution and needs to be
numerically solved. Before solving the equation, it is convenient to convert it to
dimensionless form. This is achieved by the following substitutions
⎛ nF ⎞
σ=⎜ ⎟v and z = στ (82)
⎝ RT ⎠
Note that z is the dimensionless variable. From this we have I(τ) = I(z σ ) , τ = z σ
dτ = dz σ
σt
dz (
nFA πD O c∗O )
∫ I(z σ)(σt − z )
−1 2
= (83)
0 σ 1 + ξθ exp(− σt )
where
⎡ nF ⎤
θ = exp ⎢ (φi − φ0′ )⎥ (84)
⎣ RT ⎦
Writing
I(z σ )
χ(z ) =
(nFA πD O σ c∗O ) (85)

We finally obtain
σt
1
∫ χ(z )(σt − z )
−1 2
dz = (86)
0
1 + ξθ exp(− σt )

Note that
⎛ nF ⎞ ⎛ nF ⎞
σt = ⎜ ⎟ vt = ⎜ ⎟(φi − φ) (87)
⎝ RT ⎠ ⎝ RT ⎠
σt is the measure of dimensionless potential. The solution of eq 28 gives χ(σt ) as a
function of σt . From this, the current is obtained as
(
I(t ) = nFAc∗O πD O σ χ(σt ) ) (88)

We note that

⎡ nF ⎤ ⎡⎛ nF ⎞ ⎤
ξθ exp(− σt ) = (D O D R ) exp ⎢
12
(φi − φ0′ )⎥ exp ⎢⎜ ⎟(φi − φ)⎥ (89)
⎣ RT ⎦ ⎣⎝ RT ⎠ ⎦

15
We define the half wave potential as

RT ⎛ D R ⎞
φ1 2 = φ0' + ln⎜ ⎟ (90)
2nF ⎜⎝ D O ⎟⎠

Combining eq 89 and 90, we obtain

⎡⎛ nF ⎞ ⎤
ξθ exp(− σt ) = exp ⎢⎜ ⎟(φ − φ1 2 )⎥ (91)
⎣⎝ RT ⎠ ⎦

It is thus customary to plot πχ(σt ) versus n (φ − φ1 2 ) . The values of these quantities

are listed below

n (φ − φ1 2 ) πχ(σt ) n (φ − φ1 2 ) πχ(σt ) n (φ − φ1 2 ) πχ(σt )
mV (250C) mV (250C) mV (250C)
120 0.009 20 0.269 -28.50 0.4463
100 0.020 15 0.298 -30 0.446
80 0.042 10 0.328 -35 0.443
60 0.084 5 0.355 -40 0.438
50 0.117 0 0.380 -50 0.421
45 0.138 -5 0.400 -60 0.339
40 0.160 -10 0.418 -80 0.353
35 0.185 -15 0.432 -100 0.312
30 0.211 -20 0.441 -120 0.280
25 0.240 -25 0.445 -150 0.245

From the table above, we see that the current increases with increase in the cathodic
potential, reaches a peak and decreases again. The peak occurs at
n (φ − φ1 2 ) = 28.50 mV , the corresponding value of πχ(σt ) = 0.4463 , hence the peak

current is ( see eq 88)

⎛ n 3 F3 ⎞
I p = 0.4463Ac∗O D O ⎜⎜ ⎟⎟ v (92)
⎝ RT ⎠
Note that for a reversible reaction, the peak potential is independent of the scan rate.
Case-5: Plane Electrode, equilibrium Reaction, cyclic voltammetry
In this case, the electrode potential is varied as per that following equation.
φ(t ) = φi − vt 0<t≤λ (93)

16
 φ(t ) = φi − 2λv + vt λ < t ≤ 2λ (93)
Eq 81 is valid in the time span 0 < t ≤ λ

t
(nFA )
πD O c∗O
∫ I(τ)(t − τ)
−1 2
dτ = (94)
⎡ nF ⎤
0 1 + ξ exp ⎢ ( )
φi − vt − φ0′ ⎥
⎣ RT ⎦
whereas during the time span λ < t ≤ 2λ , the following equation is valid
t
(nFA )
πD O c∗O
∫ I(τ)(t − τ)
−1 2
dτ = (95)
⎡ nF ⎤
0 1 + ξ exp ⎢ (
φi − 2λv + vt − φ0′ ⎥ )
⎣ RT ⎦
The reason for the appearance of the anodic peak at a higher positive potential than
the cathodic peak is that at the beginning of the reversal, the region in the vicinity of
the electrode has very high concentration of the reductant. A higher positive potential
is needed to reduce its concentration to a sufficiently low value so as to induce the
peak.

17
 Appendix-A
Alternative form of the Butler-Volmer equation
In the cases where concentrations of both the oxidant and reductant vary with time, a
different form of Butler-Volmer equation is useful. To derive this alterative form, we
begin with Nernst equation

RT ⎛ a R ⎞ RT ⎛ c R γ R ⎞
Ee = E0 − ln⎜⎜ ⎟⎟ = E 0 − ln⎜ ⎟ (A1)
nF ⎝ a O ⎠ nF ⎜⎝ cO γ O ⎟⎠

In the above equation, E e is the Nernst potential, E 0 is the standard potential (at unit
activity), cO and c R are concentrations and γs are the activity coefficients. It is

customary to define a quantity called the formal potential E 0′ as follows

RT ⎛ γ R ⎞
E 0′ = E 0 − ln⎜ ⎟ (A2)
nF ⎜⎝ γ O ⎟⎠

Although, this formal potential is dependent on composition, the effect is small and
we can treat it as constant. Hence we can write eq 1 as

RT ⎛ c R ⎞
E e = E 0′ − ln⎜ ⎟ (A3)
nF ⎜⎝ c O ⎟⎠

Introducing the above in the Butler-Volmer equation, we find

⎛ − (1− α ) nF ⎡⎢ E − E 0′ + RT ln ⎛⎜⎜ c R ⎞⎟⎟ ⎤⎥ α
nF ⎡ 0′ RT
⎢E −E + ln ⎜ R ⎟ ⎥ ⎞
⎛ c ⎞⎤
⎜ RT ⎣⎢ nF ⎝ c O ⎠ ⎦⎥ RT ⎣⎢ nF ⎜⎝ c O ⎟⎠ ⎦⎥ ⎟
i cath = i0 ⎜ e −e ⎟⎟ (A4)
⎜
⎝ ⎠
On simplification, we obtain
⎛ ⎛ c ⎞ − (1− α ) − (1− α ) nF (E − E 0′ ) ⎛ c ⎞α α nF (E − E 0′ ) ⎞
i cath = i 0 ⎜ ⎜⎜ R ⎟⎟ e RT
− ⎜⎜ R ⎟⎟ e RT ⎟ (A5)
⎜ ⎝ cO ⎠ ⎝ cO ⎠ ⎟
⎝ ⎠
We write the exchange current density in terms of concentration (eq 45 of E-Lecture
6)
K L
i 0 = nF kcO ( ) (kc )α
R
1− α
(A6)
Combining this equation with eq A5, we get
K L 1− α ⎛ (E − E 0′ ) (E − E 0′ ) ⎞
( ) (k )
nF nF
α − (1− α ) α
i cath = nF k ⎜ cO e RT
− c e RT ⎟ (A7)
⎜ R ⎟
⎝ ⎠

18
We write
K 1− α
⎛k⎞
A = ⎜⎜ L ⎟⎟ (A8)
⎝k⎠
and using this rewrite eq A5 as
K⎛ − (1− α )
nF
(E − E 0′ ) α
nF
(E − E 0′ ) ⎞
i cath ⎜
= nFAk⎜ cO e RT
− cR e RT ⎟ (A9)
⎟
⎝ ⎠
This is the final form we need. Writing the Butler-Volmer equation in this manner has
an advantage that we obtain the explicit dependence on concentrations of the
reactants.

19
 Appendix-B
Derivation of Equation 66
We begin with eq 64

I(s ) s
=
∗
FAk f c O s s + H( ) (B1)

This can be rearranged to give

I(s ) 1 H
= −
∗
FAk f c O s s s + H ( ) (B2)

Using entry 65 of the table of Special Laplace Transforms, we have

L−1
1
s+H
=
1
πt
2
− He H t erfc H t ( ) (B3)

Using rule 11 in the table of General Properties of Laplace Transforms, we obtain

⎡ 1 ⎤
( )
t
1
L−1 = ∫⎢
2
− He H t erfc H t ⎥ dt
( )
s s + H 0 ⎣ πt ⎦

( )
t
t 1 H2t 1 2 1 H2 t −H2 t H
π ∫0 H
=2 − e erfc H t + − e e dt . (B4)
π H H 2 t

=2
t 1 H2t
( 1
− e erfc H t + − 2
π H H
t
π
) 1 2
= − e H t erfc H t +
H
1
H
( )
Substituting eq B4 into B2 gives
I (t ) H
= 1−
FAk f c O∗
(
s s +H ) (B5)

I (t )
erfc(H t ) + ⎥ = e ( )
⎡ 1 2 1⎤ H2t
∗
= 1 − H ⎢− e H t erfc H t (B6)
FAk f c O ⎣ H H ⎦

20
 Lecture-9
Instrumental Techniques in Electrochemistry-II
9.1 Voltammetry at Rotating Disk Electrode
This is an instrument, where the hydrodynamics is well characterized and
hence the interaction between hydrodynamics and the electrode kinetics can be
studied with precision. It is also compact instrument and very simple to operate and
hence is popular for studying electrode kinetics. Its construction is as shown below.

Brush Contact

Brass Shaft

Teflon Insulator Platinum Disk

It consists of a shaft which is insulated from the solution by a Teflon sleeve. A

platinum disk is welded to the end of the shaft. The disk is exposed to the solution.
The sleeve is sufficiently tight so as not to allow the solution to seep through the gap
between the shaft and the sleeve. The shaft is rotated about its axis at a high speed.
The electric connection is made to the electrode through a carbon brush contact. The
noise level in the current depends on this contact.
9.2 Analysis of Rotating Disk Electrode
Consider a large disk rotating at a constant speed of ω (rad.s-1) about an axis through
its centre, in an incompressible Newtonian fluid. We use cylindrical coordinates,
r, θ, z . The equation to be solved is the equations of continuity and equation of motion
at steady state, i.e.
K K
∇⋅v = 0 (1)
K KK K K K
ρv.∇v = ∇p + μ∇ 2 v + ρg (2)
K
Here v is the fluid velocity, ρ is the fluid density, p is the hydrodynamic pressure, μ is
K
the dynamic viscosity and g , the gravitational acceleration.
In cylindrical coordinates, the equation of continuity takes the following form

1
 0
z

1 ∂ (rv r ) 1 ∂v θ ∂v z
+ + =0 (3)
r ∂r r ∂θ ∂z
Symmetry with respect to θ allows us to drop the corresponding derivative to get
1 ∂ (rv r ) ∂v z
+ =0 (4)
r ∂r ∂z
The three components of the equation of motion can be written as

∂v r v θ2 ∂v 1 ∂P ⎡ ∂ ⎛ 1 ∂ (rv r ) ⎞ ∂ 2 v r ⎤
vr − + vz r = − + ν⎢ ⎜ ⎟+ 2 ⎥ (5)
∂r r ∂z ρ ∂r ⎣ ∂r ⎝ r ∂r ⎠ ∂z ⎦

∂v θ v r v θ ∂v ⎡ ∂ ⎛ 1 ∂ (rv θ ) ⎞ ∂ 2 v θ ⎤
vr + + vz θ = ν⎢ ⎜ ⎟+ 2 ⎥ (6)
∂r r ∂z ⎣ ∂r ⎝ r ∂r ⎠ ∂z ⎦

∂v z ∂v 1 ∂P ⎡ ∂ ⎛ 1 ∂(rv z ) ⎞ ∂ 2 v z ⎤
vr + vz z = − + ν⎢ ⎜ ⎟+ 2 ⎥ (7)
∂r ∂z ρ ∂z ⎣ ∂r ⎝ r ∂r ⎠ ∂z ⎦
μ
In the above equation ν = is the dynamic viscosity and P = p − ρgz is the dynamic
ρ
pressure.
The boundary conditions required to solve these equations are
at z = 0 : v r = 0, v z = 0, v θ = rω (8)

as z → ∞ : v r = 0, v θ = 0 (9)
The set of equations is nonlinear and there is no obvious method to solve it
analytically. However it is possible to convert these into ODE by the following
transformation
v r = rωF(ς ) , v θ = rωG (ς ) , v z = (νω) H(ς ) , P(ς ) = νρωΠ (ς )
12
(10)

2
 ω
where ς=z (11)
ν
The resulting set of the equations is (see Appendix-A)
2 F + H′ = 0 (12)
F2 − G 2 + HF′ = F′′ (13)
2FG + HG′ = G ′′ (14)
HH′ = −Π′ + H′′ (15)
Here the primes denote differentiation with respect to ς . The boundary conditions
transform to
H = F = 0, G = 1 at ς = 0 (16)
and G = F = 0 at ς → ∞ (17)
These equations can be solved using collocation technique to obtain H, F, G
(dimensionless velocities)as functions of ς the dimensionless distance. These plots are
shown in Fig -1
The solutions for small values of ς are as shown below

⎛ ς2 1 ⎞
v r = rωF(ς ) = rω⎜⎜ aς − − bς 3 + "⎟⎟ (18)
⎝ 2 3 ⎠
⎛ 1 ⎞
v θ = rωG (ς ) = rω⎜1 + bς + aς 3 + "⎟ (19)
⎝ 3 ⎠

1 2⎛ ς3 ς4 ⎞
v z = (νω) H (z ) = (νω) ⎜⎜ − aς 2 + + "⎟⎟
12
` (20)
⎝ 3 6 ⎠

with a = 0.51023 and b = −0.6159 .

Very near the surface of the electrode, we can simplify eq 18 to 20 further to get

ω
v r = rωF(ς ) = arως = 0.51rωz = 0.51ω3 2 ν −1 2 rz (21)
ν
v θ = rωG (ς ) = rω (22)

v z = −a (νω) ς 2 = −0.51ω3 2 ν −1 2 z 2
12
` (23)
On the other hand, for large values of ς , The following expansion is valid

1 2⎛ A −ας ⎞
v z = (νω) H (z ) = (νω) ⎜ − α + e + "⎟
12
(24)
⎝ 2α ⎠
with α = 0.88447 and A = 0.934 . The asymptotic value of v z is

3
 v ∞ = −0.88447(νω)
12
(25)

v z = 0.8v ∞ at ς = 3.6 or z = 3.6(ω ν )

−1 2
. This distance is called the hydrodynamic
boundary layer thickness and represents the thickness of the liquid layer dragged by
the rotating disc.
The flow around the disk remains laminar as long as the Reynolds number
defined below does not exceed 2x105
R 2ω
Re = (26)
ν
R in the above equation represents radius of the disk. Thus for a disk of 5 mm radius,
rotating in water ν = 10 −6 m 2 ⋅ s −1 , the maximum value of ω allowed for laminar flow

is ωmax = (2 x105 )x10 − 6 (5x10 −3 ) = 8x103 rad.s −1 , i.e. about 1000 rev.s-1. This provides
2

a very large range of speeds within which the flow remains laminar.
In the dimensional form the z-directional velocity could be expressed as
⎛ ω⎞
v z = νω H⎜⎜ z ⎟
⎟ (27)
⎝ ν ⎠
We now consider transfer of mass of component i from the bulk solution to the
surface. Let the concentration of the component in the bulk be c ∞ and that on the disk
surface be c 0 . We begin with the equation of continuity.

∂ci G K K G
= D i∇ 2ci + Fz i u i∇ ⋅ (ci∇φ) − v ⋅ ∇ci + R i (28)
∂t
Following assumptions are made (i) steady state (ii) absence of homogeneous reaction
(iii) High ionic strength so that electric potential gradient is small enough to allow
omission of migration term. The resulting equation is
K G
v ⋅ ∇c i = D i ∇ 2 c i (29)
Since concentration gradient exists only in z direction, we can write the equation as
dc i d 2c
vz = D i 2i (30)
dz dz
The boundary conditions are: ci = c 0 at z = 0 and ci = c ∞ at z = ∞

dc i
Eq 30 can be solved as follows. Let = p , then we can recast eq 30 as
dz
dp
v z p = Di (31)
dz

4
Integration gives,
z
p 1
v z (η)dη
p 0 D i ∫0
ln = (32)

dci
where p0 = (33)
dz p=0

Rearrangement yields

dci dci ⎡1 z ⎤
= exp ⎢ ∫ v z (η)dη⎥ (34)
dz dz z=0 ⎣ Di 0 ⎦
The second integration gives

dc i
z
⎡1 ξ ⎤
ci = c 0 + ∫ exp ⎢ ∫ v z (η)dη⎥dξ (35)
dz z =0 0 ⎣⎢ D i 0 ⎦⎥
Using the boundary condition, ci = c ∞ at z = ∞ , we obtain

dci c∞ − c0
= (36)
dz ∞
⎡1 ξ ⎤
z=0
∫0 ⎢ Di ∫0
exp ⎢ v z (η )dη⎥dξ
⎣ ⎥⎦

dci
Substitution of from eq 36 into eq 35 and subsequent rearrangement gives the
dz z =0

concentration profile of i in the vicinity of the disk as

⎡1 ξ z
⎤
∫ exp ⎢ ∫ v z (η)dη⎥dξ
ci − c 0 0 ⎣ Di 0 ⎦
=∞ (37)
c∞ − c0 ⎡1 ξ ⎤
∫0 exp⎢ Di ∫0 v z (η)dη⎥dξ
⎣ ⎦

⎛ ω⎞
v z (η) = νω H⎜⎜ η ⎟
⎟ (38)
⎝ ν ⎠
The flux of component i at the disk surface is given by
dc i D i (c ∞ − c 0 )
N i = Di = (39)
dz ∞
⎡1 ξ ⎤
∫0 ⎢⎢ Di ∫0 v z (η)dη⎥⎥dξ
z =0
exp
⎣ ⎦
If a single reaction is occurring at the electrode and if n electrons are transferred
during the reaction, per one molecule of i, then the current density is given by

5
 D i nF(c ∞ − c 0 )
i = nFN i = (40)
∞
⎡1 ξ ⎤
∫0 ⎢⎢ Di ∫0 v z (η)dη⎥⎥dξ
exp
⎣ ⎦
And the current is given by
D i AnF(c ∞ − c 0 )
I = iA = (41)
∞
⎡1 ξ ⎤
∫0 exp⎢⎢ Di ∫0 v z (η)dη⎥⎥dξ
⎣ ⎦
We use eq 23 for v z

` v z = −0.51ω3 2 ν −1 2 z 2 (42)
Then
ξ ξ
1 0.51ω3 2 ν −1 2 0.51ω3 2 ν −1 2ξ3 ξ3
( )
D i ∫0 ∫0
v z η d η = − η 2
dη = − = − (43)
Di 3D i 3B

Di
where B= (44)
0.51ω3 2 ν −1 2
Substituting the expression from 43 into eq 41, we obtain
D i AnF(c ∞ − c 0 )
I = iA = − ∞
(45)
⎡ ξ3 ⎤
∫0 ⎢⎣− 3B ⎥⎦dξ
exp

Integral in the denominator can be solved by substitution ξ3 3B = y

∞ ∞
⎡ ξ3 ⎤ 1
∫0 ⎢⎣ 3B ⎥⎦ 3
exp − dξ = (3B )13
∫0 y exp(− y )dy
−2 3
(46)

We now make use of the definition of Gamma function

∞
Γ(p ) = ∫ y p −1 exp(− y )dy (47)
0

Comparison of eq 45 and 46 yields p = 1 − 2 3 = 1 3 and hence

∞
⎡ ξ3 ⎤ 1
∫0 ⎢⎣− 3B ⎥⎦dξ = 3 (3B) Γ(1 3) = 0.8934(3B)
13 13
exp (48)

Substituting the above expression in eq 42, and using eq 41, we obtain the current as
D i AnF(c ∞ − c 0 )
I = iA = 13
= 0.62nFADi2 3ω1 2 ν −1 6 (c ∞ − c 0 ) (49)
⎛ 3D i ⎞
0.8934⎜ 3 2 −1 2 ⎟
⎝ 0.51ω ν ⎠
The equation is called the Levich equation. We can abbreviate eq 49 as

6
 I = k SL A (c ∞ − c 0 ) (50)

Here k SL is the mass transfer coefficient between the electrode and the liquid and is
given by
k SL = 0.62D i2 3ω1 2 ν −1 6 (51)

It is seen that the mass transfer coefficient is proportional to 2 / 3 power of the

diffusion coefficient and 1 / 2 power of the disk speed.
The concentration profile of component i in the vicinity of the electrode can be
obtained by combining eq 37 with eq 43. Thus
z
⎡ ξ3 ⎤
c i − c 0 ∫0
exp ⎢− 3B ⎥dξ
= ⎣ ⎦ (52)
c∞ − c0 ∞ ⎡ ξ3 ⎤
∫0 exp⎢⎣− 3B ⎥⎦dξ
Using eq 48 in eq 37, we find
z (3 B ) −1 3
⎡ ξ3 ⎤
ci − c 0
[ ]
z
1 1
c ∞ − c 0 0.8934(3B)1 3 ∫0 ∫
= exp ⎢ − ⎥dξ = exp − u 3 du (53)
⎣ 3B ⎦ 0. 8934 0

The surface concentration c 0 can be obtained by matching the diffusion current with
reaction current. Foe example, let us consider a reaction O + ne ⇔ R and that the
reaction is fast enough to be considered as equilibrium (Nernstian) reaction. The
following equation is valid on the electrode surface.
c O (0.t ) ⎡ nF ⎤
θ= = exp ⎢ (
φ( t ) − φ0′ ⎥ ) (54)
c R (0, t ) ⎣ RT ⎦
From eq 52
I
Ni = = 0.62D i2 3ω1 2ν −1 6 (c ∞ − c 0 ) (55)
nFA
From this we get

NO =
I
nFA
(
= 0.62D O2 3ω1 2 ν −1 6 c∗O − c O (0, t ) ) (56)

and
N R = −0.62D 2R 3ω1 2 ν −1 6c R (0.t ) (57)

But the flux balance on the electrode surface yields N O + N R = 0 , or

( )
0.62D O2 3ω1 2ν −1 6 c∗O − c O (0, t ) = 0.62D 2R 3ω1 2ν −1 6 c R (0.t ) (58)
Simplification and use of eq 54 gives

7
 ( )
D O2 3 c∗O − c O (0, t ) =
1 23
θ
D R c O (0.t ) (59)

Elimination of c O (0.t ) between eq 49 and 59 yields

⎛ θ(D O D R )− 2 3 ⎞
I = 0.62nFAD ω ν 23 12
c ⎜⎜
−1 6 *
⎟
−2 3 ⎟
(60)
i O
⎝ 1 + θ(D O D R ) ⎠
9.3 Polarography
Polarography uses a dropping mercury electrode as a measuring instrument. It
is shown in Figure-5. It consists of a capillary of 50 μm internal diameter. It is fed
from a reservoir of mercury so that the available head of mercury is 200-1000 mm.
Mercury issues from the capillary in the form of nearly spherical drops. The Drop
grows until its weight is no longer supported by the capillary. The size of the drop is
about 1 mm before it detaches from the capillary. The drop disengaged from the
capillary falls in the pool of mercury at a short distance below. The capillary is
immersed in the electrolyte solution under study. The mercury drop is made the
working electrode and the mercury pool the counterelectrode. The most important
aspect which distinguishes the dropping mercury electrode from the rest of the
electrodes is that when drop falls through the electrolyte at its terminal velocity, it
agitates the liquid and helps to attain uniformity. If the potential is kept fixed, then
each drop behaves exactly as any other drop. The experiments are, therefore, highly
reproducible.
If m is the mass flow rate of mercury, then at time t after the drop begins to
form, mass of the drop will be equal to mt and is related to the drop radius by the
following equation
4 3
mt = πr ρ Hg (61)
3
The drop radius and the surface area of the drop are then given by
13
⎛ 3mt ⎞
r =⎜ ⎟ (62)
⎜ 4πρ ⎟
⎝ hg ⎠
23
⎛ 3mt ⎞
and A = 4π⎜ ⎟ (63)
⎜ 4πρ ⎟
⎝ hg ⎠

Substitution into Cottrell relation which is valid for limiting current condition gives

8
 ⎡ ⎛ ⎞ ⎤ 1 2 ∗ 2/3 1 6
23
1 2 ⎜ 3mt ⎟
⎢
I d = 4π F ⎥ nD c m t (64)
⎢ ⎜ 4πρ ⎟ ⎥ 0 O
⎣ ⎝ hg ⎠
⎦
This equation does not include the effect due to expansion of the drop. In an
approximate analysis it is assumed that when drop expands the diffusion film

Potentiaostat

electrolyte

Figure-3 Dropping Mercury Electrode

stretches over a greater surface area and is therefore thins down. This effect is
included by increasing the diffusion coefficient by factor of 7/3. Hence
⎡ 7 1 2 ⎛ 3mt ⎞ 2 3 ⎤
⎛ ⎞
Id = ⎢4⎜ π ⎟ F⎜ ⎟ ⎥ nD10 2c∗O m 2 / 3 t1 6 (65)
⎢ ⎝ 3 ⎠ ⎜⎝ 4πρ hg ⎟⎠ ⎥
⎣ ⎦
The above equation is simplified to
Id = 708nD10 2c∗O m 2 / 3 t1 6 (66)

where Id is in amperes, D0 is in cm2.s-1, c∗O is in mol.cm-3, m is in mg.s-1 and t is in s.

The above equation is called Ilkovic equation.

9
 A typical polarogram is shown in Figure-4. Polarography is mainly
used for estimation of a small concentration of ions (cations) in the solution. The
electrode potential is so adjusted that it works in the region of limiting current, so that
the Cottrell equation applies. Then the current drawn by the system is proportional to
the concentration of the discharging ion in the solution, which allows estimation of
the concentration of the ion in the solution. If more than one type of ions is present,
and if their reduction potentials are different, one can estimate them sequentially.

Drop Fall

Figure-4 A typical polarogram

Dropping mercury electrode has several characteristics which make it a unique

electrode for polarography. The dropping action is very reproducible and the surface
is continually renewed. Hence measurements are both accurate and reproducible.
Also, the electrode is not modified by reactions such as deposition of metal on the
electrode. Yet another advantage of mercury is that it has high overpotential for
hydrogen ion discharge. Hence, it is possible to go highly cathodic potentials without
fear of water being electrolyzed. Hence one can analyse cations which high negative
reduction potential e.g. alkali metals. Additional advantage here is that mercury forms
amalgam with alkali metal and since this reaction has negative free energy, it helps in
reducing the overall reduction potential of the alkali metal.
The main disadvantage of mercury electrode is that it cannot work at potential
greater than 0 Volt with respect to SCE, since at this potential mercury begins to
oxidize. For example consider the reaction between mercury and chloride ions

10
 Hg 2Cl 2 + 2e − ⇔ 2Hg + 2Cl− ΔE 0 = +0.27V (67)
Similar reactions could occur with other anions at positive potentials with respect to
SCE.
The instrument is calibrated with known concentration of the solution of the
ion to be analyzed and then used with the solution with unknown concentration. The
accuracy of measurement is general is better than 1%, but with certain precautions,
could be even better than 0.1%. Analysis works best in the range of 0.1 to10 mM
concentration. The cause of the lower limit is due to charging current, i.e. non-
faradaic current required for charging the double layer which expands during the drop
formation. The lowest limit set by this current is 0.01 mM. This limit is comparable
with other competing method such as atomic absorption spectrometry. Three methods
are used to improve the sensitivity
1. Tast polarography: This method is based on the fact that charging current decreases
with time during the drop growth and is least before the drop falls. The current is
sampled just before the drop falls when the noise is at the minimum.

nFA(7 3) D10 2c∗O

12
(Id )tast = (68)
π1 2 (τ)
12

The disadvantage of the Tast method is that it measures the faradaic current when it
is at the least value (current decreases as square root of the age of the drop). Thus
sensitivity of the method is reduced.
2. Pulse polarography: In this method the drop is held at a potential well above that
which the reaction occurs. This allows the double layer to get established. Then a
potential pulse is imposed on the drop. The current is measured at the end of the
pulse. During the short time of application the surface of the drop could be
considered as planar. The Cottrell equation gives
12 ∗ 2/3
(Id )Pulse = nFAD 0 cO m
(69)
π (Δτ )
12 12

where Δτ is the time interval of the pulse

Comparison of eq 68 and 69, shows that

(Id )Pulse 12
⎛3⎞ ⎛ τ ⎞
12

=⎜ ⎟ ⎜ ⎟ (70)
(Id )tast ⎝ 7 ⎠ ⎝ Δτ ⎠

11
The typical value of τ is 5sec and that of Δτ is 50 ms. This gives the ratio of the two
currents as about 6. Thus measurement sensitivity of the pulse polarography is higher
than the Tast polarography.
9.4 Theory
1. Cottrell equation
We first derive the Cottrell equation for the limiting current. We consider the
following reaction
O + ne − ⇔ R (71)
We assume electrode to be infinite plane situated at x = 0 . If the electrode is spherical
in shape, we can add the correction for this geometry later. However, this effect is
small unless the electrode is very small in dimension (high curvature). The electrolyte
solution is assumes be an aqueous solution of O and is assumed to extend in positive x
direction to infinity. We assume that the overpotential is kept sufficiently high that
electrode reaction is fast enough to drop the concentration of O on the electrode
surface to zero.
The continuity eq for O can be written as
∂c O ( x , t ) ∂ 2c O ( x , t )
= DO (72)
∂t ∂x 2
The initial condition is
cO (x ,0 ) = c∗O (73)
The boundary condition corresponding to the bulk of the fluid is
c O (∞, t ) = c∗O (74)
and that at the electrode surface is
cO (0, t ) = 0 (75)
We apply Laplace transformation to eq 72 and use the initial condition given by eq
70, to obtain
c∗O
cO (x , s ) = + A(s )e
−x s DO
(76)
s
Here, the overbar on cO (x , s ) indicates the Laplace transform of the original

quantity cO (x , t ) , s being the Laplace transform parameter. The Laplace transform is

defined as follows

12
 ∞
cO (x , s ) = ∫ cO (x , t )e − st dt (77)
0

Eq 76 automatically satisfies the boundary condition at infinity as given by eq74. The

boundary condition at the electrode surface (eq 75) gives
c∗O
A (s ) = − (78)
s
Substituting the above expression into eq 76, we get

cO (x , s ) =
c∗O
s
(
1− e
−x s DO
) (79)

Current is given by the product of the interfacial flux and electrode area
⎛ ∂c ( x , t ) ⎞
I(t ) = nFADO ⎜ O ⎟ (80)
⎝ ∂x ⎠ x = 0
where A is the area of the electrode. Eq 80 can be written in Laplace domain as
⎛ ∂c ( x , s) ⎞
I(s ) = nFADO ⎜ O ⎟ (81)
⎝ ∂x ⎠ x = 0
Substitution of the expression for cO (x , s ) from eq 79 into eq 81 gives

nFAD1O2C∗O
I(s ) = (82)
s1 2
Laplace inversion of the above equation yields
nFAD1O2C∗O
I( t ) = (83)
π1 2 t1 2
This is the Cottrell equation.
2. Reversible (Nernstian Reaction)
The Butler-Volmer equation is
K⎛ − (1− α )
nF
(φ−φ0′ ) α
nF
(φ−φ0′ ) ⎞
i cath = nFAk ⎜⎜ c O e RT
− c R e RT ⎟
⎟ (84)
⎝ ⎠
In the above equation φ0′ is called the formal potential and is defined as

RT ⎛ γ R ⎞
φ0′ = φ0 − ln⎜ ⎟ (85)
nF ⎜⎝ γ O ⎟⎠

We assume that the kinetics of the reaction is fast compared to diffusion so

that at any time, the equilibrium exists at the electrode. Such a behavior is called the
Nernstian behavior. The electrode potential under this condition is related to the
surface concentration by Nernst equation

13
 RT c R (0.t )
φ( t ) = φ0′ − ln (86)
nF c O (0, t )
It is assumed that initially the entire solution contains only species O at
concentration c∗O . The diffusion equations for O and R at any time t are

∂c O ( x , t ) ∂ 2cO ( x, t ) ∂c R ( x , t ) ∂ 2cR (x, t )

= DO and = D (87)
∂t ∂x 2 ∂t ∂x 2
R

The initial conditions are

cO (x ,0 ) = c∗O c R (x ,0 ) = 0 (88)
The flux balance at the interface is
⎛ ∂c ( x , t ) ⎞ ⎛ ∂c ( x , t ) ⎞
DO ⎜ O ⎟ + DR ⎜ R ⎟ =0 (89)
⎝ ∂x ⎠ x = 0 ⎝ ∂x ⎠ x = 0
The other boundary condition is provided by eq 86. These initial and boundary
conditions are sufficient to solve the equation.
We apply Laplace transformation to the first part of eq84 and use the initial
condition given by eq 88, to obtain
c∗O
cO (x , s ) = + A(s )e
−x s DO
(90)
s
The second part of eq 87 gives

cR (x, s ) = B(s )e
x s DR
(91)
Transformation of eq 89 gives
⎛ ∂ c ( x , s) ⎞ ⎛ ∂ c ( x , s) ⎞
DO ⎜ O ⎟ + DR ⎜ R ⎟ =0 (92)
⎝ ∂x ⎠ x = 0 ⎝ ∂x ⎠ x = 0
Substituting eq 90 and 91, into eq 91, we obtain
− A(s)D1O2s1 2 − B(s)D1R2s1 2 = 0 or B(s ) = A(s )ξ (93)
12
⎛D ⎞
where ξ = ⎜⎜ O ⎟⎟ (94)
⎝ DR ⎠
Substitution of eq 93 into eq 91, we obtain

cR (x, s ) = − A(s )ξe

−x s DR
(95)

c O (0.t ) ⎟ (E )
⎛ RT ⎞
⎜
i
− E 0′
= e⎝ nF ⎠ =θ (96)
c R (0, t )
where θ is a constant. Taking Laplace transform, we obtain

14
 cO (0.s )
=θ (97)
cR (0, s)
Substitution of eq 90 and 95 into 97 yields
c∗O
A(s) = − (98)
s(1 + ξθ)
Eq90 and 95 become
c∗O c∗O
cO (x, s ) =
−x
−
s DO
e (99)
s s(1 + ξθ)

c∗O
cR (x, s ) =
−x
ξe
s DR
and (100)
s(1 + ξθ)
The current in the Laplace domain is given by
⎛ ∂ c ( x , s) ⎞ nFAD1 2 C∗
I(s ) = nFADO ⎜ O ⎟ = 12 O O (101)
⎝ ∂x ⎠ x = 0 s (1 + ξθ)
Laplace inversion gives
nFAD1O2C∗O
I( t ) = (102)
π1 2 t1 2 (1 + ξθ)

15
 Appendix-A
Derivation of Eq 12 to 15
We begin with eq 4 to 7 which are rewritten below
1 ∂ (rv r ) ∂v z
+ =0 (A-1)
r ∂r ∂z
The three components of the equation of motion can be written as

∂v r v θ2 ∂v 1 ∂P ⎡ ∂ ⎛ 1 ∂ (rv r ) ⎞ ∂ 2 v r ⎤
vr − + vz r = − + ν⎢ ⎜ ⎟+ 2 ⎥ (A-2)
∂r r ∂z ρ ∂r ⎣ ∂r ⎝ r ∂r ⎠ ∂z ⎦

∂v θ v r v θ ∂v ⎡ ∂ ⎛ 1 ∂ (rv θ ) ⎞ ∂ 2 v θ ⎤
vr + + vz θ = ν⎢ ⎜ ⎟+ 2 ⎥ (A-3)
∂r r ∂z ⎣ ∂r ⎝ r ∂r ⎠ ∂z ⎦

∂v z ∂v 1 ∂P ⎡ ∂ ⎛ 1 ∂v z ⎞ ∂ 2 v z ⎤
vr + vz z = − + ν⎢ ⎜ ⎟+ 2 ⎥ (A-4)
∂r ∂z ρ ∂z ⎣ ∂r ⎝ r ∂r ⎠ ∂z ⎦
We perform the following transformations
v r = rωF(ς ) , v θ = rωG (ς ) , , v z = (νω) H(ς ) , P = νρωΠ (ς )
12
(A-5)

ω
where ς=z (A-6)
ν
Eq A-1 transforms to

[ +
] [
1 ∂ r 2ωF(ς ) ∂ (νω) H(z )
12
] ω
=0
r ∂r ∂ς ν
The above equation on simplification yields eq 12
2 F + H′ = 0 (A-7)
Where the prime represents derivative with respect to ς .
Eq A-2 transforms to
∂[rωF(ς )] [rωG(ς )] ∂[rωF(ς )] ω
[ ]
2
rωF(ς ) − + (νω) H(ς )
12

∂r r ∂ς ν
(A-8)
=− + ν ⎢ ⎜⎜
[
⎟⎟ +
]
1 ∂[νρωΠ (ς )] ⎡ ∂ ⎛ 1 ∂ r 2 ωF(ς ) ⎞ ∂ 2 [rωF(ς )] ⎛ ω ⎞⎤
⎜ ⎟⎥
ρ ∂r ⎣ ∂r ⎝ r ∂r ⎠ ∂ς 2 ⎝ ν ⎠⎦
This, on simplification yields eq 13
F2 − G 2 + HF′ = F′′ (A-9)
Eq A-3 transforms to

16
 ∂[rωG (ς )] rωF(ς )[rωG (ς )] ∂[rωG (ς )] ω
rωF(ς ) + + (νω) H (ς )
12

∂r r ∂ς ν
⎡ ∂ ⎛ 1 ∂ (r[rωG (ς )]) ⎞ ∂ 2 [rωG (ς )] ⎛ ω ⎞⎤
= ν⎢ ⎜ ⎟+ ⎜ ⎟⎥
⎣ ∂r ⎝ r ∂r ⎠ ∂ς 2 ⎝ ν ⎠⎦
This on simplification yields eq 14
2FG + HG′ = G ′′ (A-10)
Eq A-4 transforms to

∂ (νω) H(ς ) ∂ (νω) H(ς ) ω

12 12
rωF(ς ) + (νω) H (ς )
12

∂r ∂ς ν
1 ∂[νρωΠ (ς )] ω ⎡ ∂ ⎛ 1 ∂(νω)1 2 H (ς ) ⎞ ∂ 2 (νω)1 2 H (ς ) ω ⎤
=− + ν ⎢ ⎜⎜ ⎟+
⎟ ⎥
ρ ∂ς ν ⎢⎣ ∂r ⎝ r ∂r ⎠ ∂ς 2 ν ⎥⎦

This on simplification yields eq 15

HH′= −Π ′ + H′′ (A-11)

v r = rωF(ς ) , v θ = rωG (ς ) , , v z = (νω) H (ς ) , P = νρωΠ (ς )

12
(A-5)

2 F + H′ = 0 (12)
F2 − G 2 + HF′ = F′′ (13)
2FG + HG′ = G ′′ (14)
HH′ = −Π′ + H′′ (15)

17
 Lecture-10
Instrumental Techniques in Electrochemistry-III
Techniques based on Impedance Measurement
10.1 Introduction
In these techniques, we perturb an electrochemical system, operating at steady
state/equilibrium state, by superimposing an alternating (sinusoidal) potential of small
amplitude on the driving potential. The frequency of the alternating potential can be
varied over a wide range. The impedance (resistance in complex domain) of the
circuit is measured and is related to the system characteristics. Since the perturbations
from the steady state are small, the response to them is almost linear. It is therefore
simple to analyse. By combining this technique with polarography and cyclic
voltammetry, one can obtain more detailed information about dynamics of both
faradaic( electrode reactions) and nonfaradaic ( double layer charging, diffusion of
ions) processes. By varying the frequency of the alternating potential over wide range
(10-6 to 107 Hz) it is possible to study the process in the entire set of dynamic regimes
during a single experiment. These advantages make impedance measurement as one
of the important techniques of electrochemistry. The technique is further classified as
follows
a. Electrochemical Impedance spectroscopy (EIS): Here either the cell or the electrode
impedance is measured as a function of the frequency of the AC potential. A
frequency response analyzer is used for this purpose. The cell is usually operating
either under steady/equilibrium state.
b. AC Voltammetry and AC polarography: In these experiments, three electrode cell is
used. The alternating potential is superimposed on the working electrode, which has
an imposed DC potential that is dictated by the protocol of the cyclic voltammetry or
polarography.
c. Higher Harmonic AC Voltammetry: If the electrochemical system is highly
nonlinear, then it retains, albeit, a small amount of nonlinearly in the AC impedance
output. In such a case, response to higher harmonics ( 2ω,3ω," ) helps in assessing
this nonlinearly. Technique which incorporates higher harmonics in the output is
called the higher harmonic AC voltammetry. Since the double layer capacitance is
much more linear than the faradaic impedance, it is eliminated in the nonlinear

1
analysis and hence nonlinear analysis helps us to more effectively separate the double
layer charging process from the rest.
d. Intermodulation Voltammetry: Here two AC potential signals having different
frequencies ω1 and ω 2 are simultaneously imposed on a steady DC potential.

Outputs produce beats at combination frequencies ω1 + ω 2 and ω1 − ω 2 .

Impedance at these beat frequencies is measured.
e. Faradaic rectification: Here purely sinusoidal potential (no base DC potential) is
imposed and the DC component of the current is measured. This allows one to
compute the symmetry factor (transfer coefficient) of the faradaic process.
10.2 AC Impedance
A sinusoidal potential can be expressed as
1
φ = φ m sin (ωt ) = Im φ m e jωt
j
( ) (1)

In the above equation, j = − 1 , φm is amplitude of the AC potential (V) and ω is its

frequency ( rad.s-1). Im ( ) is the imaginary part of the complex variable enclosed in
the bracket. We now define complex potential as
φ = φ m e jωt (2)
We also define the complex current as
I = I m e j ( ωt + ψ ) (3)
Note that since I and φ are linearly related, the frequency of oscillations of both are
the same. Also, if ψ is positive, current leads the potential and if it is negative, the
current lags the potential.
We define the impedance (complex resistance) as
φ
Z= = Z e jψ Z (4)
I
Where Z is the amplitude of Z and ψ Z is its phase angle. We can also write Z as

Z = Z Re + jZIm (5)
then comparing 4 and 5, we deduce that
⎛Z ⎞
Z = Z 2Re + Z 2Im and ψ Z = tan⎜⎜ Im ⎟⎟ (6)
⎝ Z Re ⎠
For a pure resistor R, the current is

2
 φ φ m jω
I= = e (6)
R R
Thus the current is in phase with the potential. The impedance is real
φ
Z= = R = R + j(0 ) (7)
I
⎛0⎞
Z = Z 2Re = R and ψ Z = tan ⎜ ⎟ = 0 (6)
⎝R⎠
For a pure capacitor with capacity C, we have
q
C= (8)
φ
Where q is charged stored in the system. The current input to the system equals the
time rate of increase in the charge, or
dq
I= (9)
dt
Elimination of q between 7 and 8 and then using eq 2, we obtain

I=C
dφ
= Cφ m
d e jωt ( )
= Cjωφ m e jωt = Cjωφ (10)
dt dt
φ 1 j
Hence Z= = =− = − jX C (11)
I jCω ωC
1
where XC = (12)
ωC
X C is called the capacitive reactance. Thus for a pure capacitor

⎛ − XC ⎞ π
Z = Z 2Im = X C and ψ Z = tan⎜ ⎟=− (13)
⎝ 0 ⎠ 2
Note that the capacitive reactance depends on frequency. It decreases with increase in
the frequency of the AC potential.
We can view an electrochemical system as a network of pure resisters and
pure capacitors. The impedance of a circuit depends on how the resistance and
capacitance are interconnected. We consider two simple cases. The first is where two
are in series

R C

3
Let the potential difference across the system is φ . The current in the two elements is
the same and is denoted by I. Potential drop across the resistance is IR and that across
the capacitance is − IjX C . Hence

φ = IR − IjX C

φ
or Z= = R − jX C (14)
I
The impedance has the real component R and imaginary component − X C . The
amplitude and the phase angle of Z are respectively

1 ⎛ 1 ⎞
Z = R 2 + X C2 = R 2 + and ψ Z = − tan −1 ⎜ ⎟ (15)
ωC22
⎝ ωRC ⎠
It is seen that as frequency increases the contribution due to capacitor decreases. Thus
at very high frequencies, Z → R and ψ Z → 0 .

In general if two impedances Z1 and Z2 are in series, we can write

φ = Z1I + Z 2 I
φ
or Z= = Z1 + Z 2 (16)
I
Thus, impedances in series are added in order to find the total impedance. More
generally, for impedances Z1 , Z 2 , " , in series, the total impedance is

Z = ∑ Zi (17)
i

We now consider the second case where the resistor and the capacitor are in
parallel.
R

The potential across each element is the same ( φ ). The current in the resistor is
φ
I1 = (18)
R
and that in the capacitor is
φ
I2 = (19)
− jX C
The total current is

4
 φ φ
I = I1 + I 2 = + (20)
R (− jX C )
−1
φ ⎡1 1 ⎤
or Z= =⎢ + ⎥ (21)
I ⎣ R (− jX C ) ⎦

This expression can be rewritten as

1 1 1
= + (22)
Z R (− jX C )
We see from this that
1 1
Z= = (23)
1 1 1
2
+ 2 + ω2 C 2
R XC R2

⎛ (1 X C ) ⎞
and ψ Z = − tan −1 ⎜⎜ ⎟⎟ = − tan −1 (RωC ) (24)
⎝ (1 R ) ⎠
It is seen that as frequency increases, amplitude of the impedance decreases and the
phase lag increases. At very high frequencies, Z → 0 and ψ → − π 2 .

In general if two impedances Z1 and Z2 are in series, we can write

φ φ
I= +
Z1 Z 2
−1
φ ⎛ 1 1 ⎞ 1 1 1
or Z = = ⎜⎜ + ⎟ or = + (25)
I ⎝ Z1 Z 2 ⎟⎠ Z Z1 Z 2

Thus, reciprocals of impedances (reciprocal of impedance is called admittance and is

denoted by Y) in parallel are added in order to find the total impedance. More
generally, for impedances Z1 , Z 2 , " , in parallel, the total impedance is
1 1
=∑ (26)
Z i Zi

Eq 17 and 26 allow us to obtain the impedance of any network containing resistors

and capacitors.
10.2 Graphical representation of AC impedance
There are two ways of graphically representing AC impedance of a system as
function of the frequency. The first is the set of two plots, log Z vs. log (frequency)

and ψ Z vs. log (frequency). They are known as Bode plots. The other is the plot of the

5
imaginary part of Z (i.e. Z Im ) versus the real part of Z (i.e. Z Re ). It is called the
Nyquist plot.
1
For a series RC circuit, Z Re = R and Z Im = −
ωC
1 ⎛ 1 ⎞
Z = R2 + and ψ = − tan −1 ⎜ ⎟ (27)
ω C2
2
⎝ ωRC ⎠
1
At low frequencies ( ω << 1 RC ), Z = and hence the plot of log Z versus log ω
ωC
is a straight line with slope -1. and the intercept on log Z axis of − log C . On the other

hand − ψ Z equals π 2 and the Bode plot of − ψ versus log ω is a straight line parallel

to the frequency axis at − ψ Z = π 2 . At high frequencies ( ω >> 1 RC ), Z = R and the

Bode plot is a straight line parallel to the frequency axis at Z = R . Similarly ψ = 0 .

Thus − ψ varies from π 2 to zero in the region 1 RC >> ω >> 1 RC

Since R, the real part of Z, is independent of ω, the Nyquist plot is vertical
line at Z Re = R .
For a parallel RC circuit,
1
Z= (28)
1
2
+ ω2 C 2
R

⎛ (1 X C ) ⎞
and ψ Z = − tan −1 ⎜⎜ ⎟⎟ = − tan −1 (RωC ) (29)
⎝ (1 R ) ⎠
At low frequencies ( ω << 1 RC ), Z = R and hence the plot of log Z versus

log ω is a horizontal line at Z = R . On the other hand ψ Z = 0 and the Bode plot of

ψ versus log ω is a straight line along the frequency axis. At high frequencies
1
( ω >> 1 RC ), Z = and the Bode plot is a straight line with slope of -1. It
ωC
intersects the frequency axis at ωC = 1 . Similarly ψ Z = − π 2 . Thus − ψ Z varies from
0 to π 2 to zero in the region 1 RC >> ω >> 1 RC .
1 1 1 + jRωC
= + jωC = (30)
Z R R
R R (1 − jRωC )
Or Z= = (31)
1 + jRωC 1 + R 2ω2 C 2

6
This gives us
R R 2 ωC
Z Re = and Z = − (32)
1 + R 2 ω2 C 2 1 + R 2 ω2 C 2
Im

From the first equation we obtain

1 R
ωC = −1 (33)
R Z Re

Elimination of ωC from the second equation gives

R
Z Im = − Z Re − 1 = − RZ Re − Z 2Re (34)
Z Re

Squaring the equation we obtain

Z 2Im + Z 2Re − RZ Re = 0 (35)
This equation can be rearranged to give
2 2
⎛ R⎞ ⎛R⎞
Z + ⎜ Z Re − ⎟ = ⎜ ⎟
2
Im (36)
⎝ 2⎠ ⎝2⎠
Thus the plot of Z Im versus Z Re is a circle with centre (R 2 ,0 ) and radius R 2 . Since

Z Im is negative, we plot − Z Im versus Z Re . The Nyquist plot is a semicircle. As ω → 0 ,

ZIm → 0 and Z Re → R . As ω → ∞ , ZIm → 0 and ZRe → 0 . Thus the semicircle is

traced from the right to the left as we go from ω → 0 to ω → ∞ .

10.3 Equivalent electric circuit of an electrochemical cell
An equivalent electric circuit of an electrochemical cell is as shown in the
figure below
Ic

RΩ
Cd

Zf
Ic+If
If

The double layer impedance is a pure capacitor and is denoted by Cd. The faradaic
resistance is frequency dependent and hence is considered to be general impedance.
The two impedances are in parallel since the two processes occur independently and
require separate current. The current through the double layer is denoted by Ic, the
charging current and that through the electrode as faradaic current If. R Ω is the ohmic

7
resistance of the bulk solution. It is a pure resistance. Since the entire current i.e.,
I c + I f , passes through the bulk solution, the resistance R Ω is in series with the
impedance of the electrode.
It is the faradaic impedance of the electrode, which is most difficult to
represent. It can be represented in the form of RC circuit only as a first
approximation. The resulting circuit elements show that R and C values change with
the frequency.
We derive the expression for the impedance for a simplest single step
electrode reaction occurring at an electrode under stagnant condition.
O + ne = R (37)
We assume that the potential perturbation is introduced on the equilibrium potential.
We represent Zf as a resistance and a capacitance in series
j
Zf = R S − (38)
ωC s

We try to obtain R S and Cs from the actual kinetics. Let

I = I e jωt (39)

Then the potential across the impedance is

⎛ j ⎞ jωt
φ = IZf = ⎜⎜ R S − ⎟⎟ I e (40)
⎝ ω C s ⎠

Hence its time derivative is

dφ ⎛ j ⎞ ⎛ 1 ⎞
= ⎜⎜ R S − ⎟⎟ jω I e jωt = ⎜⎜ jωR S + ⎟⎟ I e jωt (41)
dt ⎝ ωC s ⎠ ⎝ Cs ⎠

Since we have
φ = φ(I, c O (0, t ), c R (0, t )) (42)
Hence

dφ ⎛ ∂φ ⎞ dI ⎛ ∂φ ⎞ dc O (0, t ) ⎛ ∂φ ⎞ dc O (0, t )
=⎜ ⎟ +⎜ ⎟ + ⎜⎜ ⎟⎟ (43)
dt ⎝ ∂I ⎠ dt ⎜⎝ ∂c O (0, t ) ⎟⎠ dt ⎝ ∂c O (0, t ) ⎠ dt
We define the following three parameters. The first is the charge transfer resistance
⎛ ∂φ ⎞
R ct = ⎜ ⎟ (44)
⎝ ∂I ⎠ cO (0, t ),c R (0, t )
The second and the third parameters are

8
 ⎛ ∂φ ⎞
βO = ⎜⎜ ⎟⎟ (45)
∂
⎝ Oc (0, t ) ⎠ I ,C R (0 , t )

⎛ ∂φ ⎞
and β R = ⎜⎜ ⎟⎟ (46)
⎝ ∂c R (0, t ) ⎠ I,CO (0, t )
Eq 43 can now be rewritten as
dφ dc (0, t ) dc (0, t )
= R ct jω I e jωt + βO O + βR R (47)
dt dt dt
dc O (0, t ) dc R (0, t )
We obtain and as follows. The continuity equations for O and R
dt dt
can be written as
∂c O ( x , t ) ∂ 2c O ( x , t )
= DO (48)
∂t ∂x 2
∂c R ( x , t ) ∂ 2c R ( x, t )
= DR (49)
∂t ∂x 2
The initial conditions are
c O (x ,0 ) = c∗O , c R (x ,0 ) = c∗R (50)
The boundary conditions corresponding to the bulk of the fluid are
c O (∞, t ) = c∗O , c R (∞, t ) = c∗R (51)
The flux balance equation is
⎛ ∂c ( x , t ) ⎞ ⎛ ∂c ( x , t ) ⎞
DO ⎜ O ⎟ + DR ⎜ R ⎟ =0 (52)
⎝ ∂x ⎠ x =0 ⎝ ∂x ⎠ x =0
and the expression for the current is
⎛ ∂c ( x , t ) ⎞
I = nFADO ⎜ O ⎟ (53)
⎝ ∂x ⎠ x =0
The Laplace transforms of the diffusion equations, after incorporating the initial
conditions yield
c∗O
c O (x , s ) = + A(s )e
−x s DO
(54)
s
c∗R
c R (x , s ) = + B(s )e
−x s DR
(55)
s
Incorporating the interfacial flux continuity and the current conditions, i.e., eq 52 and
53 in the above two equations, we obtain

9
 c∗O I(s )
cO (0, s ) = + (56)
s nFA D Os

c∗R I(s )
and cR (0, s ) = − (57)
s nFA D R s

The above equations on inversion yield

I (t − u )
t
1
c O (0, t ) = c +
π∫
∗
O du (58)
nFA D O 0 u

I (t − u )
t
1
c R (0, t ) = c∗R −
π∫
and du (59)
nFA D R 0 u

We are interested in the (cyclical) steady state solution of the equations an hence we
allow the limit in the integral to reach ∞ . The equations transform to

I (t − u )
∞
1
c O (0, t ) = c∗O +
π∫
du (60)
nFA D O 0 u

I (t − u )
∞
1
c R (0, t ) = c∗R −
π∫
and du (61)
nFA D R 0 u

Substitution of eq 39 into the concentration profiles yield

e j ω( t − u )
∞
I
c O (0, t ) = c∗O +
nFA D O π ∫0
du
u
(62)
I e jωt ∞

∫u
∗ −1 2 − jωu
=c +O e du
nFA D O π 0

Using jωu = y 2 , we obtain

2 I e j ωt ∞
c O (0, t ) = c∗O +
π∫
− y2
e dy (63)
nFA jωD O 0

∞
π 1 1
But ∫ e − y dy = = e − jπ 4 . Hence
2
and =
2 jπ 2
0 j e

I e j(ωt − π 4 )
c O (0, t ) = c +∗
O (64)
nFA ωD O

I e j(ωt − π 4 )
Similarly c R (0, t ) = c∗R − (65)
nFA ωD R

Differentiation with respect to time yields

10
 j ( ωt − π 4 )
dc O (0, t ) jω I e
= (66)
dt nFA ωD O

dc R (0, t ) jω I e j(ωt − π 4 )
Similarly =− (67)
dt nFA ωD R

dφ jω I e j(ωt − π 4 ) jω I e j(ωt − π 4 )
= R ct jω I e jωt + βO − βR (68)
dt nFA ωD O nFA ωD R

We equate the right hand side of eq 68 with eq 47and obtain the following equation

j ω I e j(ωt − π 4 ) j ω I e j(ωt − π 4 ) ⎛ 1 ⎞
R ct jω I e jωt + βO − βR = ⎜⎜ jωR S + ⎟⎟ I e jωt (69)
nFA D O nFA D R ⎝ Cs ⎠

We simplify the equation to

ω ( j + 1) ω ( j + 1) ⎛ 1 ⎞
R ct jω + β O − βR = ⎜⎜ jωR S + ⎟⎟ (70)
nFA 2D O nFA 2D R ⎝ Cs ⎠

Defining σ as

1 ⎛⎜ βO β ⎞
σ= − R ⎟ (71)
nFA 2 ⎜⎝ D O D R ⎟⎠

we can rewrite eq 70 as
⎛ 1 ⎞
R ct jω + σ ω ( j + 1) = ⎜⎜ jωR S + ⎟⎟ (72)
⎝ Cs ⎠

Comparing real and imaginary parts, we obtain

1
Cs = (73)
σ ω
σ
R s = R ct + (74)
ω
σ
The term arises because of the diffusional mass transfer and hence cannot be
ω
treated either are true resistance or capacitance. It is called the Warburg impedance.
σ
Rw = (75)
ω
1
Cs = is called the pseudo capacitance.
σ ω
10.4 Estimation of kinetic parameters from impedance measurements

11
 Impedance measurement s provide us the values of R ct and σ , from which we
can determine the kinetic parameters as follows.
We begin with the following form of the Butler-Volmer equation
i K ⎡ (1 − α )nF ⎤ H ⎡ αnF ⎤
= kc O (0, t ) exp ⎢− η⎥ − kc R (0, t ) exp ⎢ η⎥ (76)
nF ⎣ RT ⎦ ⎣ RT ⎦
The exchange current density is
i0 K ∗ H ∗
= kc O = kc R (77)
nF
Combining eq 76 and 77, we obtain
i c O (0, t ) ⎡ (1 − α )nF ⎤ c R (0, t ) ⎡ αnF ⎤
= exp ⎢− η⎥ − exp ⎢ η (78)
⎣ RT ⎥⎦
∗ ∗
i0 cO ⎣ RT ⎦ cR
Since the overpotential is small, we can linearized the above equation to the following
form
i c O (0, t ) ⎡ (1 − α )nF ⎤ c R (0, t ) ⎡ αnF ⎤
= 1− η⎥ − 1+ η (79)
i0 c∗O ⎢⎣ RT ⎦ c∗R ⎢⎣ RT ⎥⎦
The above equation can be rearranged to

i ⎛ c O (0, t ) c R (0, t ) ⎞ c∗O + δc O ⎡ (1 − α )nF ⎤ c∗R + δc R ⎡ αnF ⎤

=⎜ − ⎟− ⎢⎣ RT η⎥⎦ − c∗ ⎢⎣ RT η⎥⎦ (80)
i 0 ⎜⎝ c∗O c∗R ⎟⎠ c∗O R

We neglect the product of δc O η and δc R η . This allows us to simplify the above

equation to the following form

i ⎛ c O (0, t ) c R (0, t ) ⎞ nF
=⎜ − ⎟− η (81)
i 0 ⎜⎝ c∗O c∗R ⎟⎠ RT

Rearrangement gives

RT ⎛ c O (0, t ) c R (0, t ) i ⎞
η= ⎜ − − ⎟⎟ (82)
nF ⎜⎝ c∗O c∗R i0 ⎠

Noting that η = φ and i i 0 = I I 0 , we obtain

RT ⎛ c O (0, t ) c R (0, t ) I ⎞
φ= ⎜ − − ⎟⎟ (83)
nF ⎜⎝ c∗O c∗R I0 ⎠

Using eq 44 to 46
⎛ ∂φ ⎞ RT
R ct = ⎜ ⎟ =− (84)
⎝ ∂I ⎠ cO (0, t ),c R (0, t ) nFI 0

12
 ⎛ ∂φ ⎞ RT
βO = ⎜⎜ ⎟⎟ = ∗ (85)
⎝ ∂c O (0, t ) ⎠ I,C R (0, t ) Fc O

⎛ ∂φ ⎞ RT
and β R = ⎜⎜ ⎟⎟ =− ∗ (86)
⎝ ∂c R (0, t ) ⎠ I,CO (0,t ) Fc R

From the impedance measurement allow us to determine the exchange current density
and allow us to obtain diffusion coefficients of the species from σ . Note that R ct is
inversely proportional to the exchange current density and hence one can measure
small exchange current densities very accurately.
Although, one needs a single frequency to determine the exchange current
density, it is important to conduct the experiments at different frequencies and check
whether Cs and R s vary linearly as ω−1 2 . If not, the electrode process is more
complicated than is assumed.
10.4 Effect of the other impedances in the system
The actual impedance measurements include all the impedances in the system,
i.e. ohmic resistance R Ω and the double layer capacitance C d . We now measure

overall resistance R B and the capacitance C B of the system, i.e.

j
Z = RB − (87)
ωC B

By obtaining Z at different frequencies, it is possible to obtain R Ω , Cd , Cs and

R s separately.
From the equivalent circuit of the cell, we get
−1
⎛⎛ j ⎞
−1
⎞ −1
⎜
Z = R Ω + ⎜⎜ − ⎟⎟ +
1 ⎟ = R + ⎛⎜ jωC + 1 ⎞
⎟⎟ (88)
⎜ ⎝ ωC d ⎠ ⎟ Ω ⎜ d
⎝
Zf
⎠ ⎝ Zf ⎠

From eq 38, we have

j
Zf = R S − (89)
ωC s
Combining the two equations, we get
−1
⎛ ωC s ⎞ ωC s R s − j
Z = R Ω + ⎜⎜ jωC d + ⎟⎟ = R Ω + 2 (90)
⎝ ωCs R s − j ⎠ jω C d Cs R s + ω(C d + Cs )

Further simplification yields

13
 Z = RΩ +
(ωCs R s − j){− jω2Cd Cs R s + ω(Cd + Cs )}
ω4 (C d C s R s ) + ω2 (C d + Cs )
2 2

ωCs2 R s − j[(C d + Cs ) + ω2 C d (C s R s ) ]
2
(91)
= RΩ +
ω3 (C d C s R s ) + ω(C d + Cs )
2 2

This gives
Cs2 R s
Z Re = R Ω + (92)
ω2 (C d Cs R s ) + (C d + Cs )
2 2

and Z Im = −
(Cd + Cs ) + ω2Cd (Cs R s )2 (93)
ω3 (C d Cs R s ) + ω(C d + Cs )
2 2

1 σ
Substituting Cs = and R s = R ct + in the above equations, we get
σ ω ω
⎛ σ ⎞
⎜ R ct + ⎟
⎝ ω⎠
Z Re = RΩ + (94)
( )
2
2 2⎛ σ ⎞ 2
ω C d ⎜ R ct + ⎟ + σ ωC d + 1
⎝ ω⎠

( )
2
σ ⎛ σ ⎞
C d σ ω + 1 + ωC d ⎜ R ct + ⎟
ω ⎝ ω⎠
and Z Im =− (95)
( )
2
2 2⎛ σ ⎞ 2
ω C d ⎜ R ct + ⎟ + Cd σ ω + 1
⎝ ω⎠
Consider two limiting cases. At low frequencies, eq 94 and 95 reduce to
σ
Z Re = R Ω + R ct + (96)
ω
σ
and − Z Im = 2σ 2C d + (97)
ω
Combining the last two equations, we obtain
− Z Im = Z Re − R Ω − R ct + 2σ 2 C d (98)

The Nyquist plot ( − Z Im versus Z Re ) should be a straight line with unit slope and

intercept on real axis of Z Re = R Ω + R ct − 2σ 2 C d . It is easy to see that the frequency

dependence arises due to Warburg impedance. Thus linear Nyquist plot corresponds
to the diffusion controlled regime. This is understandable since at low frequency, the
boundary layer is allowed to grow to a sizable thickness.
At very high frequencies, eq 94 and 95 reduce to

14
 R ct
Z Re = R Ω +
(
C ω R ct2 + ωσ 2
2
d
2
) (99)

σ 2 + ω(R ct )
2
Z Im = −
C d (ω2 R ct2 + ωσ 2 )
and (100)

The above two equations could be combined to yield

2 2
⎛ R ⎞ ⎛R ⎞
⎜ Z Re − R Ω − ct ⎟ + Z Im = ⎜ ct ⎟
2
(101)
⎝ 2 ⎠ ⎝ 2 ⎠
R ct R
The Nyquist plot is a semicircle centered at Z Re = R Ω + , Z Im = 0 and radius ct .
2 2

15
 Lecture-11
Modeling and simulation of Electrochemical Reactors
11.1 Introduction
An electrochemical reactor consists of cathodes, anodes, anolyte and catholyte
compartments and the diaphragms/membranes separating them. It is either connected
to a power source or a power sink (load). Catholyte and anolyte may be fed into the
cell in a batch, semibatch or continuous mode. A reactor could consist of a battery of
cells, which operate either in series or parallel. The geometry of the electrodes as well
as the electrolyte compartments can vary widely. These complications can be
analyzed accurately only using the modern simulation tools.Two different types of
problems need simulation. In the first situation, the design of the reactor is provided
and we are interested in evaluating its performance. Thus we are provided with the
geometry of the reactor, the operating temperature, modes of flow of the anolyte and
catholyte streams, the initial (or inlet) composition of these streams, the potential
difference across the cell. We are interested in finding the current and the current
distribution in the reactor ( and on the electrodes), catholyte and anolyte compositions
either as functions of time or location in the reactor, the current efficiency for the
desired reaction, capacity of the reactor and the power consumption per unit mass of
the desired product. The second and more difficult situation is when one has to evolve
an optimal design and operation of the reactor. Here, one has to screen a variety of
possible design strategies, first on the heuristic basis and then through detailed
analysis. After arriving at the best design strategy, one has to optimize the design
using an appropriate cost objective function. In the present lecture, we only consider
the first type of the situation.
11.2 Types of Models
The most important problem in the analysis of electrochemical reactor is to
determine the current distribution. This allows us to estimate the production capacity
of the reactor. Three models with increasing level of difficulty are used. They
respectively estimate (1) Primary current distribution (2) Secondary current
distribution and (3) Tertiary current distribution. In the primary current distribution
model, the spatial variation of the composition of the solution is ignored and it (the
solution) is assumed to be uniform in composition. The overpotential is assumed to be
zero and hence the solution potential is assumed to be uniform at the electrode

1
surface. The model reduces to solving the Laplace equation subject to appropriate
potential boundary conditions at the electrode surface. In the secondary current
distribution model, the spatial variation of the composition is still ignored, but the
overpotential is estimated using the electrode kinetics. In the tertiary current
distribution model, both the spatial variation of the solution composition and its effect
on the current is analyzed in conjunction with the electrode kinetics. The tertiary
current distribution model is the most correct model and the other two are only the
approximations. However, the latter models are still popular because of the following
two reasons. Firstly, in many situations, the primary current distribution model
provides sufficient information needed for the design. The secondary distribution is
needed only fewer times. The tertiary distribution needed only in some cases. It is
however important to recognize a priori which level of the model is appropriate,
otherwise one may settle for a lower level model when the situation really demands a
higher level model. Secondly, each model can form the first guess for the next level of
the model. Since the equations involved are complex, the proper first guesses allow
the solution of the model to converge quickly. Thus one can begin with primary
model. Use its solution as the first guess for the secondary distribution and used the
converged solution for the secondary distribution as the first guess for the tertiary
distribution. This stepwise procedure forms a systematic way for detailed simulation.
11.3 Primary Current distribution
Since the solution has uniform composition, the potential distribution is
obtained by solving the Laplace equation
∇ 2φ = 0 (1)
The boundary conditions are obtained as follows. If a surface is an insulator, we have
K K
∇φ ⋅ n̂ = 0 (2)
where n̂ is the unit outward normal to the surface.
The electrode surface is equipotential. If we assume zero overpotential, the
potential of the solution in contact with the electrode should also be uniform. The
difference in the solution potential at the two electrodes should correspond to ohmic
drop through the solution. Omission of the overpotential implies that the electrodes
are at the Nernst potentials with respect to the solution in contact. Thus we can write
ΔφΩ = Δφ − Δφ0 (3)

2
Δφ represents the actual cell potential (known), Δφ0 the equilibrium cell potential and

ΔφΩ , the ohmic potential drop across the cell. Since, inside the cell, the current flows
from the anode to the cathode, the potential of the solution in contact with anode is
always higher than that of the cathode. One can assume potential in the solution at the
anode as zero and assign ΔφΩ as the potential at the anode.
Once all the boundary conditions are known, it is possible to solve the Laplace
equation in order to obtain the potential distribution in the entire reactor. The current
density at any location in the reactor can be obtained by using the following equation
K K
i = − k∇φ (4)
where κ is the electrical conductivity of the solution and is given by.
κ = F2 ∑ z i2 u i c i (5)
i

Since the solution is assumed uniform in concentration, it will have a constant value
of κ .
Many simple problems for primary current distribution can be solved
analytically. As the first problem consider two infinite parallel electrodes shown
below

The Laplace equation in this case reduces to one dimensional equation

d 2φ
=0 (6)
dz 2
Let the potentials in the solution at z = 0 and z = h be respectively φ0 and φ h
respectively. The solution of eq 6, with these boundary conditions is
⎛z⎞
φ = φ0 + (φ h − φ0 )⎜ ⎟ (7)
⎝h⎠
The current density is given by

3
 κ(φ h − φ0 )
i=− (8)
h
If φ h > φ0 , current flows in negative z direction. Note that the current density is
uniform everywhere including that on the surface of the electrode.
As the next simple case, we consider two concentric cylinders of infinite
length. The outer radius of the inner cylinder is ri and the inner radius of the outer
cylinder is r0 . The corresponding solution potentials are φi and φ0 respectively.

1 d ⎛ dφ ⎞
⎜r ⎟ = 0 (9)
r dr ⎝ dr ⎠
The solution of this equation is
φ − φi ln (r ri )
= (10)
φ0 − φi ln (r0 ri )
The potential varies logarithmically in radial direction. The current density is given by
κ(φ 0 − φi )
i=− (11)
r ln (r0 ri )
Current density is lower on the outer cylinder than the inner one.
In both the above problems the current density is uniform on the
electrode. We now consider a more complex case, where it is not uniform. Let us
consider a case where two cylindrical electrodes, each of radius a, are placed parallel
to each other so that the centre to centre distance between them is 2d (see figure
y
below).

a
θ x
- +

d
If the solution in contact with the right hand side electrode is V / 2 and that on the left
side electrode is − V / 2 . The potential distribution can be obtained by the method of
images as

V ⎡ 2(d + p )(d + r cos θ ) + r 2 + p 2 − d 2 ⎤

φ= ln ⎢ ⎥ (12)
⎛ d + p ⎞ ⎣ 2(d − p )(d + r cos θ ) + r 2 + p 2 − d 2 ⎦
2 ln⎜⎜ ⎟⎟
⎝d−p⎠

4
 where p = d2 − a2 (13)
The current density on the electrodes is given by the following equation
apκV
i= (14)
(d + p )(d − p )(d + a cos θ) ln⎛⎜⎜ d + p ⎞⎟⎟
⎝d−p⎠
Note that the current density on the electrodes is a function of θ , which indicates that
it is nonuniform. The total current on any one of the electrodes of length l is obtained
by integration of eq 14 with respect to θ
π
I = 2πla 2 ∫ i sin θdθ (15)
0

The resulting current is

2πlκV
I= (16)
⎛d+p⎞
ln⎜⎜ ⎟⎟
⎝d−p⎠
The potential difference between the electrodes is V. Hence the ohmic resistance of
the cell is given by
V 1 ⎛d+p⎞
R= = ln⎜⎜ ⎟ (17)
I 2πlκ ⎝ d − p ⎟⎠
Now consider the cell below.

V=0 V=0

V = V0
x

In the above case, there are three electrodes. Two are parallel and one is perpendicular
to them. The solution of the Laplace equation for this case is obtained using the
method of conjugate functions and is given by

2V0 ⎡ sin (πx H ) ⎤

φ= tan −1 ⎢ ⎥ (18)
π ⎣ sinh (πy H ) ⎦
The current density at the horizontal electrode is given by

5
 2 κV0
iy = (19)
H sin (πx H )
Note again that the current is a function of x.
Following are the characteristics of the primary current distribution.
1. The primary current distribution is based on the Laplace equation, which is
linear. Moreover, boundary conditions involve surfaces having uniform
potentials (or zero normal potential gradients). As a result, the current
distribution remains unaltered if system is scaled, keeping geometry of the
scaled system identical to the original system. Scale-up is simple and
straightforward for the primary current distribution.
2. For a system having complex geometry, the current distribution is highly
nonuniform. The reason is that the electrolyte resistance is different in
different current paths. More current flows along the path with lesser
resistance. A point on cathode receives current from different points on the
anode. A greater contribution to this current is provided by the point on the
anode which is nearer to the given point on the cathode.
3. In a primary current distribution, current is always perpendicular to the
equipotential surfaces. Hence equipotential lines intersect an insulator surface
at right angle (i.e. current always flows along the insulator surface). On the
other hand, near the electrode, the equipotential lines closely follow the
electrode surface profile.
4. Current tends to concentrate near the convex electrode surfaces since flux
lines must be perpendicular to the surface. Thus current converges from all
direction onto the convex surface. In the extreme case, the electrode is a sharp
point. The current density becomes infinite at that point. However, the total
current remains finite since the electrode area becomes infinitesimally small at
the point.
5. Near a concave surface, current density is reduced as the curvature is
increased. Reason is that region from which it receives the current
progressively decreases as the curvature increases.
6. Intersection of an insulator and an electrode in important. If a planar electrode
intersects an insulator at obtuse angle, the current density at the point of
intersection is infinite. The reason is that the equipotential lines near the
electrode are parallel to it, but those near the insulator are perpendicular it.
6
 Hence the equipotential line must change its direction rapidly at the junction.
As a result it becomes sharply convex. Since flux line must cross the
equipotential surface at right angle, they concentrate at the junction. This
results in infinite current density.
7. If a planar electrode intersects a planar insulator at acute angle, the current
density at the point of intersection is zero. The reason is that the equipotential
lines near becomes sharply concave. This results in zero current density.
8. Only when the two planes intersect at right angle, the current is finite. Under
this situation, the equipotential lines satisfy both the required conditions at the
junction.
11.4 Secondary current distribution
In the secondary current distribution, electrode kinetics is incorporated. The
boundary condition for the potential at the electrode surface is given by
K
− κ∇φ ⋅ n̂ = i(η) (20)
Here i is the current density of reaction as a function of the electrode overpotential. It
could be Butler-Volmer equation or one of its asymptotic forms (Tafel equation or the
linear equation). This is the mixed boundary condition. The potential of the solution at
the electrode will now be nonuniform (even though the electrode is at a constant
potential).
It is necessary to provide a judgment as to whether the secondary current
distribution is needed to solve the given problem or not. If the electrode kinetics is
comparable in speed to the transport of current through the bulk against the ohmic
resistance, then secondary current distribution needs to be determined. On the other
hand if the electrode kinetics is very rapid compared to the transport of the through
the bulk, we need not consider the secondary current distribution. This can be
expressed mathematically using Wagner number, Wa , defined as
κ ⎛ ∂η ⎞
Wa = ⎜ ⎟ (21)
L ⎝ ∂i ⎠
The term in the bracket is the charge transfer resistance and is obtained from the
slope of the polarization curve. The ratio L κ represents the ohmic resistance of the
solution, L being the characteristic length of the system. In the case of parallel plate
electrodes, distance between the plates can be used as L. Thus Wagner number is the
ratio of the charge transfer resistance to the ohmic resistance. A small Wagner number

7
implies that the charge transfer resistance is small compared to the ohmic resistance
and hence primary current distribution is adequate to represent the system. On the
other hand if the Wagner number is of the order of one or greater, then charge transfer
resistance is important and primary current distribution is not enough to describe the
system correctly.
Inclusion of electrode kinetics evens out the current distribution compared to
that obtained using the primary distribution model. The reaction can be viewed as the
charge transfer resistance in series with ohmic resistance of the solution. Hence if
ohmic resistance is low in comparison to the charge transfer resistance, the current
density is controlled by the charge transfer resistance. In such cases, the current
density does not depend on the shape of the electrode. This renders the current
distribution to be uniform.
Behavior of the Wagner number depends on various factors. Increase in the
characteristic dimension reduces anger number, thereby making the current
distribution less uniform. Increasing the electrolyte conductivity makes the current
distribution more uniform by reducing the ohmic resistance. The nature of the
polarization curve is also important. If linear kinetics is being followed, i.e. if
⎛ nF ⎞
i = i0 ⎜ ⎟η (22)
⎝ RT ⎠
κ ⎛ ∂η ⎞ κRT
then Wa = ⎜ ⎟= (23)
L ⎝ ∂i ⎠ nFLi 0
Thus for linear kinetics, Wagner number is independent of the current density.
For the Tafel model,
⎛ nFα ⎞
i = i 0 exp⎜ η⎟ (24)
⎝ RT ⎠
∂η nF
Hence = (25)
∂i RTαi
κ ⎛ ∂η ⎞ κRT
and Wa = ⎜ ⎟= (26)
L ⎝ ∂i ⎠ nFLαi
The Wagner number is evaluated at the average current density. When all other
factors are constant, Wagner number varies inversely with the current density. Thus
the current distribution becomes less uniform as the current density increases.
As temperature increases, Wagner number increases due to increase in the
electrolyte conductivity and the direct effect of T.

8
11.5 Tertiary current distribution
The tertiary distribution includes the effect of concentration of the reacting
species on the current distribution along with effects included in the secondary
distribution i.e. it must include the effect of the concentration overpotential.
The concentration overpotential is given by

RT ⎛ c b ⎞
ηc = ln⎜ ⎟ (27)
F ⎜⎝ c s ⎟⎠

where c b is the bulk concentration of the reactant and cs is the concentration in the
solution in contact with the surface of the electrode. The solution potential (which is
required as boundary condition for the Laplace equation) is obtained by subtracting
the Nernst potential, the surface overpotential and the concentration overpotential
from the electrode potential. We can relate the concentrations to current density as
follows
i = nFk SL (c b − c s ) (28)

In the above equation k SL represents the solid-liquid mass transfer coefficient.

Limiting current, i L occurs when cs = 0 . Hence

i L = nFk SL c b (29)
From this we obtain
i c
= 1− s (30)
iL cb

cb
Eliminating between eq 27 and 30, we obtain
cs

RT ⎛ i ⎞
ηc = − ln⎜⎜1 − ⎟⎟ (31)
nF ⎝ i L ⎠

From this we obtain the concentration diffusion resistance as

∂ηc RT 1
= (32)
∂i nF (i L − i )
Wagner number for concentration diffusion can now be written as
⎛ κ ⎞ ∂η κRT 1
Wa = ⎜ ⎟ c = (33)
⎝ L ⎠ ∂i nFL (i L − i )
From this we see that the tertiary current distribution is important when either the
limiting current is low or the system is operating close to the limiting current.

9
 Since both the current density at the electrode as well as hydrodynamics affect
the concentration depletion of the reactant at the interface, high current density does
not necessarily implies high concentration overpotential. One must see the intensity of
agitation at a location. For example, if a surface is open and hence accessible to
agitation, and if the agitation is uniform, then diffusion film thickness is uniform and
high current density also corresponds to high concentration overpotential. In this case
tertiary current distribution is more uniform than the secondary distribution. However,
if there are regions which are not accessible to agitation, than those regions have high
concentration overpotential. But since they are also concave regions, current density
is lower in these regions. Thus concentration overpotential will tend to reduce the
current density further. In such cases, the tertiary current distribution is less uniform
than the secondary current distribution.

10
 Exercise-1
Q1: Devise electrochemical cells in which the following reactions could be made to
occur. If liquid junctions are necessary, note them in the cell schematic appropriately, but
neglect their effect. What half reactions take place at the electrodes in each cell? What is
the standard cell potential in each case? Which electrode is negative? Would the cell
operate electrolytically or galvanically in carrying out the net reaction from the left to the
right? (Bard and Faulkner)
(a) H 2O ↔ H + + OH −
(b) 2H 2 + O 2 ↔ H 2O

(c) Cu 2 + + Pb ↔ Pb 2 + + Cu
(d) 2Ce3+ + 2H + + BQ ↔ 2Ce 4 + + H 2Q (BQ = p-benzoquinone, H2Q =hydroquinone)

Q2: (a) What is the emf at 250C of a cell Pt H 2 (1atm ) HCl(m1 ) HCl(m 2 ) H 2 (1atm ) Pt

when m1 = 0.1mol ⋅ kg −1 and m 2 = 0.2 mol ⋅ kg −1 ? What would be the emf of the cell if the
pressure of the gas at the right hand electrode were increased to 10 atm? The mean
activity coefficients of HCl(aq) are γ1 = 0.798 and γ 2 = 0.790
(b) Hydrogen behaves imperfectly at high pressure, and its equation of state may be
summarized by the virial expansion
pVm RT = 1 + 5.37 x10 −4 (p atm ) + 3.5x10 −8 (p atm )
2

The e.m.f. of the cell Pt H 2 HCl(0.1mol ⋅ kg −1 ) Hg 2Cl 2 Hg has been measured up to

pressures of 1000 atm and good agreement between experiment and theoretical
expression was obtained up to 600 atm on the assumption that the only significant
volume change was that associated with the hydrogen. Above that pressure, the volume
change of the other components become significant. What is the e.m.f. of the cell when
the pressure is 500 atm? Take γ ± (HCl ) = 0.798 . (Atkins)

Q3: The variation with temperature, of EMF of the cell

α β ε φ
Pt (s), H 2 (1bar ) HBr (a = 1) AgBr (s) Ag(s)
is given by the following expression( t in 0C)
E(V) = 0.0731 − 4.99x10 −4 ( t − 25) − 3.45x10 −6 ( t − 25) 2
Find the free energy, enthalpy and the entropy of the following reaction at 250C.
H 2 (1bar ) + 2AgBr (s) → 2Ag(s) + 2HBR (a = 1)
(Glasstone)

Q4: Obtain an expression for the potential of the cell below. The nickel electrode can be
regarded as inert, and the solubility of oxygen and the mercuric oxide in the solution can
be ignored. Thus the solution can be regarded to be of uniform concentration.
α β δ ε φ α′
Pt (s) Ni(s), O 2 (g ) KOH in H 2O HgO(s) Hg (l) Pt (s)

The electrode reactions are

O 2 + 4e − + 2H 2O → 4OH −

and Hg + 2OH − → HgO + H 2O + 2e −

What is the expression for the standard cell potential and what is its value? (Newman)

Q5: Devise a cell in which the following reaction is the overall cell process
2 Na + + 2Cl − → 2 Na (Hg) + Cl 2 (aq)
where Na(Hg) symbolizes amalgam. Is the reaction spontaneous or not? What is the
standard free energy change? Take the standard free energy of formation of Na(Hg) as -
85 kJ.mol-1. From the thermodynamic standpoint another reaction should occur more
readily at the cathode of your cell. What is it? (Atkins)

Q6: A fuel cell converts chemical energy directly into electrical energy without involving
the wasteful process of combustion, heating of steam for use in a turbine, and the
conversion into electrical energy. What is the maximum e.m.f. obtainable from a cell in
which hydrogen and oxygen, both at 1 atm pressure at 298 K, combine on catalyst
surface? What is the maximum electrical work per mole of H2(g)? (Atkins)

Q7: Consider the following cell

 α δ ε φ α′
Pt (s), H 2 (g ) KOH in H 2O HgO(s) Hg (l) Pt (s)

where the reaction for hydrogen in alkaline media is regarded to be

2H 2O + 2e − → H 2 + 2OH −
What is the expression for the standard cell potential? What is its value? (Newman)

Q8: In order to deal with corrosion processes, we have to treat them as dynamical
processes. Nevertheless, the tendency of the system to corrode may be assesses
thermodynamically, and we look into that here. Metals corrode by being oxidized, the
driving reaction being reductions
Acid Conditions
2H 3O + + 2e − = 2H 2O + H 2 Eθ = 0 V

O 2 + 4H + + 4e − = 2H 2O E θ = 1.23 V
Alkaline Conditions
O 2 + 2H 2O + 4e − = 4OH − E θ = 0.40 V
Which of the following metals will corrode in water made slightly acid (pH = 6)? Which
will corrode in water made slightly alkaline (pH = 8.0)? Which corrode only in strong
acid (pH=1) and which corrode only in strong alkali (pH=14)? Take the criterion for
corrosion to be the thermodynamic tendency to form at least a 10-6 M solution of ions in
the vicinity of the metal surface. The metals to be considered are (a) Al (b) Cu (c) Fe (d)
Ag (e) Pb (f) Au (Atkins)

Q9: The reversible cell

α δ ε φ ε φ
Zn (s) ZnCl 2 (a 1 ) Hg 2 Cl 2 (s) Hg(l) Hg 2 Cl 2 (s) ZnCl 2 (a 2 ) Zn (s)

Was found to have EMF of 0.09535V at 250C. Determine the ratio of the mean ion
activities of the zinc chloride in the two solutions. (Glasstone)

Q10: The EMF of the cell

 α β ε φ
Pt (s), Cl 2 (1bar ) HCl(a = 1) AgCl(s) Ag(s)

is -1.1364 V at 250 C. The Ag, AgCl(s) electrode may be regarded as the chlorine
electrode with a gas at a pressure equal to the dissociation pressure of silver chloride;
calculate the value of the pressure at 250C. (Glasstone)
 Exercise-2
Q1: A 0.10 m CdSO4 solution is electrolyzed between a cadmium cathode and platinum
anode with a current density of 10 A.m-2. The hydrogen overpotential is 0.6 V. What will
be the concentration of Cd2+ ions when evolution of hydrogen just begins at the cathode?
Assume all activity coefficients as unity. (Atkins)

Q2: The kinetics of an electrochemical reaction is described by the Butler-Volmer

equation with the transfer coefficient of 0.5 and exchange current density of 100 A.m-2.
(a) Determine the error in using the Tafel approximation for overpotential of 1 mV and
100 mV. (b) Repeat part (a) for linear approximation (Prentice)

Q3: An electrochemical reaction follows the Butler-Volmer kinetics with the transfer
coefficient of 0.5 and exchange current density of 10 A.m-2. (a) Determine the increase in
the reaction rate when the overpotential is increased from 0.1 V to 1.1 V. (b) For a
chemical reaction following Arrhenius behavior, determine the temperature required to
achieve the same increase. Assume that the rate is initially measured at 250C and the
activation energy is 100 kJ.mol-1. (Prentice)

Q4: The standard potential of the Zn+/Zn electrode is -0.76 at 250C. The exchange
current density for H+ discharge at platinum is 7.9 A,m-2. Can Zn be plated on the
platinum at that temperature? (take unit activities) (Atkins)

Q5: Two large plane parallel copper electrodes are immersed in a 1 M solution of copper
sulfate at 250C. The gap between the electrodes is 1 cm. Estimate the potential difference
between the anode and the cathode when the average current density is 10 A.m-2.
(Prentice)

Q6: The transfer coefficient of a certain electrode in contact with M3+ and M4+ in
aqueous solution at 250 C is 0.42. The current density is found to be 17mA.cm-2 when
overvoltage is 125 mV. What is the overvoltage required for the current density of 75
mA.cm-2? (Atkins)
Q7: The following data were obtained for reduction of species R to R- in a stirred
solution at a 0.1 cm-2 electrode; the solution contains 0.01 M R and 0.01 M R-1.
η mV -100 -120 -150 -500 -600

i 0 mV 45.9 62.6 100 965 965

Find the exchange current density, the limiting current density and the transfer coefficient
for this reaction. (Bard and Faulkner)

Q8: Zinc is being deposited from acid electrolyte. The concentration of ZnCl2 is 1 M and
pH is 3. At 250C both zinc deposition and hydrogen evolution is possible. Assume that
the reaction follows the Tafel behavior in the region of interest. For zinc reaction
α c = 0.5 and i 0 = 1kA.m −2 . For hydrogen α c = 0.5 and i 0 = 10μA.m −2 . A voltmeter
connected to SCE reference electrode reads -1.2 V. (a) Estimate the current efficiency for
zinc deposition. (b) in qualitative terms explain the effect of raising the pH on the current
efficiency. (Prentice)

Q9: If α = 1 2 , an electrode interface is unable to rectify alternating current because the

current density curve is symmetrical about η = 0 .When α ≠ 1 2 , the magnitude of the
current density depends on the sign of the overpotential, and so some degree of faradic
rectification sis possible. Suppose that the overpotential varies as η = η0 cos ωt , derive an

expression for the mean flow of current (averaged over a cycle) for general α , and
confirm that the mean current is zero when α = 1 2 . In each case work in the limit of
small η0 but to second order in η0 F RT . Calculate the mean direct current for a 1 cm2

hydrogen-platinum electrode with α = 0.38 , when η0 = 10mV and frequency ω 2π is 50

Hz. (Atkins)

Q10: The potential of a metal oxide reaction is given by

E = −0.2 + 0.06 log(ix106 )
Here E is in Volts and i in Amperes. The potential for a reduction reaction in the same
solution is
E = 0.1 − 0.1log(ix107 )
If these are the only two reactions possible on the metal surface, estimate the corrosion
potential and current density.
 Exercise-3

Q1: 26 % w/w solution of MgSO4 has a density of 1292.2 kg.m-3. Find (a) molarity (b)
molality (c) mole fraction of MgSO4 in the solution. (Atkins)

Q2: 1 kg of 0.5 m solution of NaCl is mixed with 3 kg of 2 m solution of NaCl. What

will be the molality of the resulting solution? (Atkins)

Q3: Calculate the minimum distance at which the attractive electrostatic energy of the
ions having charges z + and z − in n electrolyte is greater than the thermal energy of an
interacting cation and anion. (Bokris and Reddy)

Q4: What will be the change in the chemical potential of MgCl2 at 300C, if 10 kg of
water is added to 10 kg of 2.5 m aqueous solution of MgCl2. The activity coefficient of
MgCl2 solutions are listed below (Atkins)
m 0.8 0.9 1.0 1.2 1.4 1.6 1.8 2.0 2.5
γ 0.521 0.543 0.569 0.630 0.708 0.802 0.914 1.051 1.538

Q5: Calculate the mean activity coefficient of thallous chloride, the solubility of which
has been measured in water and in the presence of various concentrations of potassium
chloride solutions at 250C, as given in the following table. The solubility of this salt in
pure water is 1.607x10-2 mol.kg-2. (Bokris and Reddy)
KCl 0.025 0.05 0.10 0.20
mol.kg-1
TiCl 8.69x10-3 5.90x10-3 3.96x10-3 2.68x10-3
mol.kg-1

Q6: Mean activity coefficients of HBr in three dilute aqueous solutions at 250C are:
0.927 (at 5mmol.kg-1), 0.902 (at 10.0 mmol.kg-1) and 0.816 (at 50 m.mol.kg-1). Estimate
the value of ‘a’ which fit these data. (Atkins)
Q7: Show that σ(x ) defined by eq 36 approaches 1 as x → 0 [Hint: expand both

ln(1 + x ) and (1 + x ) −1 by Taylor series]

Q8: Evaluate Debye-Huckel constants α and B for ethyl alcohol at 250C, taking
dielectric constant to be 24.3. Utilize this result together with known values of α and B
for water, to compare approximate activity coefficients for 1:1, 1:2 and 2:2 electrolytes in
ethyl alcohol and water at 250C at ionic strength of 0.01, 0.1m. Ionic diameter may be
taken as 0.3 nm in each case. (Glasstone)

Q9: The following freezing point depressions were observed when KCl was dissolved in
water. What is the mean activity coefficient for KCl in 0.05 mol.kg-1 solution? Assume
that limiting law is valid from m = 0 to m = 0.010 mol.kg −1 (Atkins).
m/ mol.kg-1 0.010 0.020 0.030 0.040 0.050
δT/K 0.0355 0.0697 0.1031 0.137 0.172

Q10: (a) The Whitecoats built the cell below and managed to measure its potential at
250C as 0.35 V.
Ag AgCl(s) AgCl(aq, saturated) AgCl(aq, saturated in 0.05M CaCl 2 ) AgCl(s) Ag

The solubility product of AgCl is known to be 1.77x10-10 at room temperature. Calculate

the individual activity coefficient for Ag+ in CaCl2 solution. (b) Is there a way to obtain
the individual activity coefficient for Cl-1 in the same solution? (Bockris and Reddy)
 Exercise-4
Q1: Estimate the electrical conductivity of pure water at 250 C. (Newman)

Q2: The diffusion coefficients of cupric ions and sulfate ions at infinite dilution are
0.713x10-9 and 1.065x10-9 m2.s-1, respectively at 250C. Estimate the transport number of
the cupric ion at infinite dilution and compare with the value of 0.363 at the concentration
of 0.1 M. (Newman)

Q3: Write down the expression for effective diffusion coefficient D of the electrolyte, the
cation transport number and the conductivity for solutions of sulfuric acid when the
electrolyte is assumed to dissociate either as
H 2SO 4 ⇔ H + + HSO 4
or as
H 2SO 4 ⇔ 2H + + SO 24 −
Compare the two set of values. (Newman)

Q4: At the negative electrode of the lead acid battery, the reaction is
Pb(s ) + SO 24 − ⇔ PbSO 4 (s) + 2e −
Regard the solution as a binary electrolyte of H2SO4 dissociated into H+ and SO42- ions
and show that the current density at the electrode surface is related to the concentration
gradient by
z − ν − FD dc
i=− at x = 0
1 − t − dx
(Newman)

Q5: At the positive electrode in the lead acid-battery, the reaction is

PbO 2 (s) + SO 24 − + 4H + + 2e − ⇔ PbSO 4 (s) + 2H 2O
Regard the solution as a binary electrolyte of H2SO4 dissociated into H+ and SO42- ions
and show that the current density at the electrode surface is related to the concentration
gradient by
 2FD dc
i=− at x = 0
2 − t + dx
(Newman)

Q6: Dissolved oxygen at a concentration of 9.5x10-4 M is reacted at limiting current from

a 5 M KOH solution where the diffusion coefficient of oxygen is estimated to be 2.3x10-9
m2.s-1 and that of KOH to be about 5x10-9 m2.s-1. The diffusion layer near the electrode
can be treated as a stagnant layer of thickness 50 μm . (a) Estimate the magnitude of the
limiting current density. (b) Obtain the expression for the concentration profile of
potassium ion. (c) Estimate a numerical value for the concentration of potassium ions
adjacent to the electrode surface (outside the diffuse part of the double layer). (d) Obtain
the expression for the concentration profile of hydroxide ions. (Newman)

Q7: For cathodic deposition of copper from 0.5 M CuSO4 and 0.5 M H2SO4 electrolyte,
the kinetic parameters are α c = 0.5 and i 0 = 10A.m −2 . Two copper parallel electrodes
with spacing of 2 cm between them are used (a) Calculate the current density at which
hydrogen evolution is expected if only kinetic limitations are taken into account (b)
Calculate the current density at which hydrogen evolution is expected if mass transport
effect are also taken into account. Assume that the liquid between the two electrodes is
stagnant. (Prentice)

Q8: The conductivity of aqueous sodium chloride solutions becomes proportional to the
concentration such that κ c approaches 126.45 S.cm2.mol-1 at infinite dilution and the
cation transference number approaches 0.396 at 250 C. Estimate the diffusion coefficient
of the salt at infinite dilution. (Newman)
 Exercise-5
Q1: In acetonitrile( η = 0.345mPa.s ), the equivalent conductivity of vary dilute solution
of KI is 198.2 S.cm2.eq-1 at 250C. Calculate the equivalent conductance of KI in a similar
concentration range in acetophenone( η = 0.280mPa.s ).

Q2: Calculate the mobility of sodium ion in 0.01 M NaCl, viscosity of the solution is
0.895 mPa.s and the Stokes radius of sodium ion is 260 pm.

Q-3: A student has to determine the equivalent conductivity at infinite dilution for KCl,
NaCl, KNO3 and NaNO3 solutions and the transference numbers of the ions in these
solutions. He managed to determine only λ0 (KNO3 ) , λ0 ( NaNO3 ) , t 0+ ( Na + / NaCl) and

t 0+ (K + / KCl) an wrote them in a table:

NaNO3 KNO3 NaCl KCl
λ0 , S.cm 2 .eq −1 121.4 144.9 _ _

t 0+ _ _ 0.396 0.490

t 0− _ _ _ _

Q-4: A current of 5 mA flows through a 2-mm inner-diameter glass tube filled with 1 N
CuSO4 solution in the anodic compartment and with 1 N solution of Cu(CH3COO)2
solution in the cathode compartment. The interface created between the two solution
moves 6.05 mm towards the anode in 10 min. Calculate the transport number of sulfate
ion in this solution.

Q-5 : Given the transport numbers of Ca2+ in CaCl2 as 0.438 and that of K+ in KCl as
0.490, calculate the transport number of Ca2+ in a solution containing both 1 mM CaCl2
and 10 mM KCl., neglecting variation of transport number with concentration.

K
Q-6: Let the transference number t i of species i with respect to the velocity v be defined
by the equation
 K K K
t i i = z i Fci (vi − v )
for a solution of uniform composition. This equation says that the flux of species i
K
relative to velocity v accounts for fraction t i of the current density.
K
(a) Let t′i be the transference number of species i with respect to the velocity v′ .
Show that the transference numbers of the two species i and j relative to the
K K
velocities v and v′ are related by
t′i − t i t′j − t j
=
z ici z jc j

(b) For a binary electrolyte, show that

t 0+ c t0
=− 0 +
z0 z +c+

Thus demonstrating that the ratio t i z i is not always zero for a neutral species. Here

t 0+ is the transference number of the solvent relative to the cation velocity.

(c) From (a) above show that

ti K t′i K
∑ z ∇μ = ∑ z ∇μ
i
i
i
i
i i

(Newman)

Q-7 Find the liquid junction potential for the following junction
0.1mFeCl2 // 0.1mFeCl2 + 0.1mFeCl3
(Prentice)

Q-8 (a) Show that for a single symmetric electrolyte (− z − = z + = z ) Henderson equation
simplifies to
RT(u + − u − ) ⎛ c(0) ⎞
φ(L ) − φ(0) = ln⎜ ⎟
zF(u + − u − ) ⎜⎝ c(L ) ⎟⎠
(b) Show that for 1:1 electrolyte, the Henderson equation reduces to

φ(L ) − φ(0) =
RT
(t + − t − ) ln⎛⎜⎜ c(0) ⎞⎟⎟
zF ⎝ c(L ) ⎠
Q-9 The equivalent conductivities of the aqueous solutions of KCl and MgCl2 at 250 C
were estimated as 146.95 and 124.11 (S-1.cm2.eq) respectively. Calculate the molar and
specific conductivities when concentrations of both solutions were 1 eq.m-3. What would
be the measured resistance of these two solutions when two planar platinum electrodes of
2 cm2 area and 0.5 cm apart are employed? Measurements of specific conductivity and
hence of the solution resistance are usually carried out under a small ac field. Explain
why a small ac field is used? (Bockris and Reddy)
 Exercise-6
Q1: Show that the equation for the potential profile in the diffuse double layer, that is

e( y 2 ) =
(
1 + tanh y s 4 e − κx )
(
1 − tanh y s 4 e − κx )
can also be written in the following alternative form

κx = ln
(e y 2
)(
+1 ey
s
2
−1 )
(e − 1)(e + 1)
s
y 2 y 2

(Adamson and Gast)

Q2: Calculate the surface charge density σ for 0.02 M 1:1 electrolyte and 2:2 electrolytes
at 250C and φs = 40mV (Adamson and Gast)

Q-3: From the Lippmann equation: dγ + ΓNa + dμ NaCl (α ) + σdΔφ = 0 , derive the Maxwell

relations
⎛ ∂Γ+ ⎞ ⎛ ∂σ ⎞ ⎛ ∂μ ⎞ ⎛ ∂σ ⎞
⎜⎜ ⎟⎟ = ⎜⎜ ⎟⎟ , ⎜⎜ ⎟⎟ = −⎜⎜ ⎟⎟
⎝ ∂Δφ ⎠μ ⎝ ∂μ ⎠ Δφ ⎝ ∂Δφ ⎠ Γ+ ⎝ ∂Γ+ ⎠ Δφ

⎛ ∂Γ+ ⎞ ⎛ ∂Δφ ⎞ ⎛ ∂μ ⎞ ⎛ ∂Δφ ⎞

⎜ ⎟ = −⎜⎜ ⎟⎟ , ⎜ ⎟ = −⎜⎜ ⎟⎟
⎝ ∂σ ⎠μ ⎝ ∂μ ⎠ Δφ ⎝ ∂σ ⎠ Γ+ ⎝ ∂Γ+ ⎠ Δφ

⎛ ∂σ ⎞ ⎛ ∂σ ⎞ ⎛ ∂Δφ ⎞ ⎛ ∂Δφ ⎞
Show that ⎜⎜ ⎟⎟ = −⎜⎜ ⎟⎟ ⎜⎜ ⎟⎟ = −C d ⎜⎜ ⎟⎟
⎝ ∂μ ⎠ Δφ ⎝ ∂Δφ ⎠ μ ⎝ ∂μ ⎠ σ ⎝ ∂μ ⎠ σ
(Newman)

Q-4: Show how to obtain the surface excess of cations ΓNa + for mercury NaCl solution

interface from the measurement of the differential capacity of the double layer as a
function of potential and NaCl concentration. In addition, the potential the interfacial
tension at the point of zero charge can be assumed to be known as functions of
concentration. (Newman)
Q-5: Let the surface excess Γi,d of an ionic species i, in the diffuse double layer be

defined as ( x 2 represents the distance of the outer Helmholtz plane from the electrode
surface. The corresponding potential is denoted by φ 2 and the charge density by σ 2 ).
∞
Γi,d = ∫ (ci − cib )dx
x2

Show, for the special case when all ions have the same magnitude of the charge ( z i = z ),

that the diffuse layer theory yields the result

Γi,d =
κ
e (
2cib −z i φ2 2
−1 )
From this result show that
⎛ ∂Γi ,d ⎞ 2z i cib
⎜⎜ ⎟⎟ =
( )
⎝ ∂σ 2 ⎠ μ F 1 + e ∑ z jc j
zi φ2 b

And that, consequently, in the absence of specific adsorption, the potential of zero
charge, Δφ z varies with composition as
dΔφz 1
=
dμ 2zF
where Δφ z is measured relative to a reference electrode reversible to the anion. Note that
for the repelling potentials, Γi,d shows a limiting amount of exclusion from the double

2cib
layer, Γi,d = − (Newman)
κ

Q-6: For a layer of water dipoles oriented perfectly at the interface estimate the
magnitude of the difference in the electrostatic potential across the layer. Take the dipole
moment of water to be 7.85x10-30 C.m , area per molecule to be 0.16 x10-18 m2, and the
permittivity of that of the free space. (Newman)

Q-7 A porous carbon material acts as an ideally polarizable electrode and has a double
layer capacity of 30 μF.cm −2 and surface area of 250 m2.g-1. Estimate the electric capacity
of a cubic centimeter of a cubic centimeter of electrode packed with this material. The
density of the pure solid carbon is 2250 kg.m-3, and the porosity of the material is
estimated to be 0.7 as packed. What difference in the applied potential is required to
effect a change in the surface charge density of 0.1 C.m-2? If this change in the surface
charge density is accompanied by the adsorption of chloride ions, what surface area,
expressed in nm2, is available to each chloride ion? (Newman)

Q-8 (a) The following data at 180 C are obtained for the interfacial tension of mercury in
NaCl solution at the point of zero charge. Estimate the surface excess of NaCl for a 0.3 M
NaCl solution at the point of zero charge.
CNaCl σ , mN.m-1
0 425
1 422

(b) What convention is used for the position of the Gibbs interface for the result
expressed in part (a)?

Q-9 A bulk solution of 0.06 M in NaCl and 0.001 in ZnCl2 is subjected to electrosorption
on a high surface-area carbon material. Assume that there is no specific adsorption and
estimate the selectivity of the diffuse layer for zinc ions relative to sodium ions. Assume
that the potential at the outer Helmholtz plane ( relative to the bulk solution) is so small
that the expression for the Boltzmann distribution can be linearized wherever it is
encountered.
(Newman)
 Exercise-4
Q1: Given n = 1 , c∗ = 1mM , A = 0.02cm 2 , and D = 10 −5 cm 2 .s −1 , calculate the current for
the diffusion controlled electrolysis at (a) a planar electrode (b) a spherical electrode at
t = 0.1, 0.5,1, 2, 3, 5,10s and as t → ∞ . How long can the electrolysis proceed before the
current at the spherical electrode exceeds that at the planar electrode by 10%?
(Bard and Faulkner)

Q2: Integrate Cottrell equation to obtain the total charge consumed in electrolysis at any
time. Calculate the value for t = 10s , for the data in Q-1. Use Faraday’s law to obtain the
number of moles reacted by that time. If the total volume of that solution is 10 mL, what
fraction of the sample has been altered by electrolysis? (Bard and Faulkner)

Q3: Derive equation below, which is valid for a finite rate of reaction at an infinite planar
electrode surface.
I(s )
FA
( ) (
= k f c∗O − k b c∗R e H t erfc H t
2
)

Q4: For a diffusion controlled reaction on disk electrode with radius r0 , the current is
given by
4nFADO c∗O
I= f (τ)
πr0

where τ = 4D 0 t r02 and

1 π π
f (τ ) = + + 0.094 τ τ <1
2 τ 4

1 1 1
f (τ ) = 1 + 0.71835 + 0.05626 3 − 0.00646 5 τ >1
τ τ τ
A disc UME gives a plateau current of 2.32 nA in the steady state voltammogram for a
species known to react with n = 1 and to have concentration of 1 mM and a diffusion
coefficient of 1.2 x10 −9 m 2 .s −1 . What is the radius of the electrode? (Bard and Faulkner)
Q5: Derive the equation for the transient current at a spherical electrode to step change
in the electrode potential, when both O and R are present in the bulk and the reaction
equilibrium prevails on the electrode surface. (Bard and Faulkner)

Q6: For an electrochemical system to be described by equation involving semi-infinite

boundary condition, the cell wall must be at least five diffusion layer thickness away
from the electrode. For a substance with D = 10−9 m 2 .s −1 , what distance between the
working electrode and the cell wall is required for a 100-s experiment?
(Bard and Faulkner)

Q7: Do problem number 6.6 from Bard and Faulkner

Q8: Do problem number 6.7 from Bard and Faulkner

Q9: Do problem number 6.8 from Bard and Faulkner

Q10: Do problem number 6.9 from Bard and Faulkner

 Exercise-8
Q1: Consider an RDE with a disk radius 2 mm, immersed in an aqueous solution of
substance A ( c∗A = 10 −2 M, D A = 5x10−10 m 2 .s −1 ) and rotated at 1000 rpm. A is reduced in a

one electron reaction. ν = 10 −6 m 2 .s −1 . Calculate v r and v z at the electrode surface and at a

distance a distance 10 microns normal to the disk surface at the edge of the disk. Also
compute v ∞ , thickness of the boundary layer and the mass transfer coefficient.
(Bard and Faulkner)
Q2: An electrochemical reaction I 2 + 2e − ↔ 2I − was investigated at rotating disk
electrode in solution that contains 6.6x10-4 M KI3 and a supporting electrolyte. Calculate
the interfacial concentration of an I 3− ion when the current density is 10 A.m-2 and the

rotation rate of the disk is 3000 rpm. (Take diffusion coefficient of I 3− ions as 1.14x10-9

m2.s-1 and ν = 1x10 −6 m 2 .s −1 . (Bockris et al.)

Q3: Show that for a ring electrode, having inner radius ri and the outer radius r0 , the
current is given by

(
I = 0.62nFπ r02 − ri2 )
32
ω1 2ν −1 6 (c ∞ − c 0 )

Q4: Copper is deposited at a rotating disk electrode from a solution containing 0.1 M
CuSO4. The rotation speed is 1000 rpm. Calculate the thickness of Cu deposit formed
during 30 minutes if the deposition current density is one third of the limiting current
density. Take D Cu 2+ = 5.2x10−10 m 2 .s −1 , ν = 1x10 −6 m 2 .s −1 and density of the deposit is 8900

kg.m-3. (Bockris et al.)

Q5: A copper-cadmium alloy is deposited onto a rotating disk electrode at 2000 rpm in a
solution containing 0.01 M CuSO4, 0.1 M CdSO4, 1x10-3 M H2SO4 and 1 M Na2SO4. It is
known from the literature that the exchange current density for Cu deposition in 0.01 M
CuSO4 is 5.5 A.m-2 and the exchange current density for Cd deposition in 0.1M CdSO4 is
84 A.m-2. The cathodic transfer coefficients for the deposition of both the metals are 0.50.
Calculate the current density at which an alloy with 25 mol% of Cu and 75 mol% Cd
should be deposited, assuming that the deposition current density of the alloy is the sum
of the current density of individual metals. What is the value of the electrode potential at
that current density? Take D Cu 2+ = D Cd 2+ = 5x10 −10 m 2 .s −1 , ν = 1x10 −6 m 2 .s −1 , T = 298K .

(Bockris et al.)

Q6: Electrochemical oxidation of hydroquinone was investigated on a rotating disk

electrode in a solution containing 0.01 M quinine and hydroquinone in 0.5M H2SO4 at
298 K. The following values of current density at different electrode potentials and disk
speeds were obtained
ω, rpm mA mA mA mA mA mA mA
i( ) i( ) i( ) i( ) i( ) i( ) i( )
m2 m2 m2 m2 m2 m2 m2
0.75V 0.77V 0.79V 0.81V 0.83V 0.87V 1.05V
500 0.9363 1.846 3.211 4.717 5.887 6.969 7.491
1000 0.9717 1.990 3.672 5.784 7.648 9.579 10.59
1500 0.9885 2.061 3.921 6.428 8.815 11.48 12.97
2000 0.9987 2.1o6 4.087 6.885 9.698 13.03 14.98
2500 1.006 2.137 4.208 7.236 10.41 14.34 16.75
3000 1.011 2.161 4.302 7.519 11.00 6.969 18.35
4000 1.019 2.196 4.442 7.956 11.97 6.969 21.19
The value of the equilibrium potential was 0.683 V vs. SHE. Assuming that the reaction
is first order in hydroquinone, determine the anodic transfer coefficient and the exchange
current density. (Bockris et al.)

Q7: The following measurements were made on a reversible polarographic wave at 250C.
The process could be written as O + ne ⇔ R .

φ V(versus SCE) I μA
-0.395 0.48
-0.406 0.97
-0.415 1.46
 -0.422 1.94
-0.431 2.43
-0.445 2.92

I d = 3.24μA . Calculate (a) the number of electrons involved in the electrode reaction,
and (b) the formal potential (vs. NHE) of the couple involved in the electrode reaction,
assuming D O = D R .

Bard and Faulkner

Q9: Do problem number 6.8 from Bard and Faulkner

Q10: Do problem number 6.9 from Bard and Faulkner

 Exercise-9
Q1: Find the impedance of the following RC circuit. All resistance are equal to R and
capacitances are equal to C and frequency equals ω . Find specific value of the impedance
for R = 100Ω , C = 1μF . Obtain real and imaginary parts of the impedance in terms of ω .

Q-2 Derive formula for converting a parallel RC network (Rp, Cp in parallel) to a series
equivalent ((Rs, Cs in series). (Bard and Faulkner)

Q3: The faradaic impedance is sometimes represented by a resistance and a capacitance

in parallel rather than in series. Find the expression for the parallel representation of the
impedance in terms of R ct , βO , β R and ω. {Hint: Use the result of Q-1}. (Bard and

Faulkner)

Q4: The faradaic impedance method is employed to study the reaction O + e ⇔ R by

imposing a small sinusoidal signal (5mV) to the cell, and measuring the equivalent series
resistance R B and capacitance C B of the cell. The following data are obtained for 250C,

c∗O = c∗R = 1mM and A = 1cm 2 .

Frequency ( ω 2π ) Hz RB , Ω CB, μF

49 146.1 290.8
100 121.6 158.6
400 63.3 41.4
900 30.2 25.6
In a separate experiment under the exactly same conditions, but in the absence of the
electroactive species, the cell resistance R Ω is found to be 10Ω , and the double layer
capacitance of the electrode Cd is found to be 20 μF . (a) From these data calculate, at
each frequency, Rs and Cs and the phase angle φ between the components of the faradaic
impedance. (b) Calculate i0 and k0 for the reaction and estimate D assuming D O = D R .
(Bard and Faulkner)

Q5: Find the charge transfer resistance R ct , the double layer capacitance C D , and the

solution resistance R Ω from the data listed in the following table. If the measurement was
carried out equilibrium potential, what is the exchange current?

Frequency Hz Amplitude, Z Phase angle, ψ

1 3116 -77.8
10 319 -87
100 33.4 -72.4
Bockris et al.
 Exercise-10
Q1: A plate of thickness L a two dimensional slot of width h through its thickness ( see
figure below. The origin of the rectangular coordinates is at the centerline of the plate and
the centerline of the slot. Two counterelectrodes are placed on either side of the plate. We
want to consider possibility that deep within the slot, the potential distribution is given by
⎛ λz ⎞ ⎛ λ y ⎞
φ = A cosh⎜ ⎟ cos⎜ ⎟
⎝ h ⎠ ⎝ h ⎠
where A is a constant.
(a) Does the potential distribution satisfy Laplace equation?
(b) What boundary condition would be satisfied along the centerline of the plate (z = 0)?
Is this reasonable from the point of view of the symmetry conditions prevailing in the
system? Is it reasonable from the point of view of the suggested analytic form for the
potential?
(c) What boundary condition would be satisfied along the centerline of the slot (y = 0)? Is
this reasonable from the point of view of the symmetry conditions prevailing in the
system? Is it reasonable from the point of view of the suggested form for the potential?
(d) Is there a saddle point in the potential at the origin? (Saddle point is defined as one at
which the function has a maximum in one direction and the minimum in the other
direction.)
(Newman)

Plate L

slot h z

Q-2 (continuation of Q-1): (a) Along the electrode surface (within the slot), what is the
distribution of the current density that contributes to the overall electrode current?
(b) For a deep slot, what is the penetration depth according to the suggested analytic
solution? (When quantities vary exponentially with distance, the penetration depth is the
distance over which such quantities vary by a factor of e.)
(c) For the primary current and potential distribution, determine parameter λ and
numerical value of the penetration depth for h = 40μm . The potential of the plate is zero.
(h) For the secondary distribution, where linear electrode kinetics is obeyed with
exchange current density of i 0 = 5 A.m −2 , discuss how to determine parameter λ and a
numerical value of the penetration depth. A clear graphical method for determination of
λ and an indication of how its value depends on the magnitude of i 0 is sufficient.

(Newman)

Q3: (a) For the two electrode system shown in the figure below, show that the equation
for the potential

V ⎡ 2(d + p )(d + r cos θ) + r 2 + p 2 − d 2 ⎤

φ= ln ⎢ ⎥
⎛ d + p ⎞ ⎣ 2(d − p )(d + r cos θ) + r 2 + p 2 − d 2 ⎦
2 ln⎜⎜ ⎟⎟
⎝d−p⎠
satisfies the Laplace’s equation in two dimensions
1 ∂ ⎛ ∂φ ⎞ 1 ∂ 2 φ
∇ φ=
2
⎜r ⎟ + =0
r ∂r ⎝ ∂r ⎠ r 2 ∂θ 2
and also the boundary conditions that the potential on the right electrode is V / 2 and on
the left electrode it is − V / 2 .
y

a
θ x
- +

d
(b) Show that the current density on the right electrode is given by
apκV
i= , p = d2 − a 2
(d + p )(d − p )(d + a cos θ) ln⎛⎜⎜ d + p ⎞⎟⎟
⎝d−p⎠
(c) Show that the total current on the right electrode is
2πlκV
I=
⎛d+p⎞
ln⎜⎜ ⎟⎟
⎝d−p⎠

Q4: (a) For the electrolytic cell shown below, the following equation for the potential

2V0 ⎡ sin (πx H ) ⎤

φ= tan −1 ⎢ ⎥
π ⎣ sinh (πy H ) ⎦
satisfies the Laplace equation in two dimensions in Cartesian coordinates
∂ 2φ ∂ 2φ
+ =0
∂x 2 ∂y 2
(b) Show that the current density on the electrode along x-axis is
2 κV0
iy =
H sin (πx H )

V=0 V=0

V = V0
x

(c) Find the expression for the total current crossing the electrode along x-axis.
(d) Determine the current density distribution and the total current on the electrodes at
x = 0 and x = H .
Q5: An electrode at angle θ = 0 meets an insulator at angle θ = α . Show how the primary
current and potential distributions vary in this corner region by solving the Laplace
equation in r − θ coordinates, with appropriate boundary conditions. Assume the

following form for the solution (n is a constant and f (θ) is a function of θ alone)

φ = r n f (θ) (Newman)

Q6: Two plane insulators meet at an angle α . It is desired to ascertain the distributions of
current density and potential near such a corner. In particular, how much current
penetrates into an acute corner, and might current densities reach infinity for some large
angles? Write down Laplace equation in cylindrical coordinates, where z-axis is the line
of intersection of the insulating planes. Seek the simplest, nontrivial solution for the
potential φ with no z dependence. (Newman)

Q7: The Hull cell is used to test the throwing power of plating baths by assessing how
nonuniform a deposit will form on an electrode, all parts of which are not accessible.
There are two planar electrodes at opposite ends of the cell; one electrode is
perpendicular to the four insulting boundaries of the cell, but the other is at an angle
α with one of the insulating walls, as shown in the figure below.

Insulator

cathode
Anode

Insulator

Top View of Hull Cell

The cell also reveals the nature of the deposit that will form at different current densities
that occur along its surface.
As an extreme case, it is desired to calculate the maximum potential variation that
can occur in the solution adjacent to the electrode, and it is assumed that this happens
when there is a uniform current density at each electrode. Determine this distribution of
potential in the solution under this condition and evaluate the maximum potential
variation in the solution adjacent to each electrode. Further assume for this calculation
that the solution is well stirred.
Obtain numerical values for the case where the anode is 10cm by 10cm and is 10
cm from the cathode at the nearest point and α = 450 , the current density on the anode is
0.2 A.m-2 and the solution conductivity is 6 S.m-1. (Newman)

Q8: A copper disk rotated in seawater is observed to corrode preferentially near the
periphery. An iron disk, on the other hand, is observed to corrode preferentially near the
centre. Discuss qualitatively these observations. (Newman)