Lectures on Thermodynamics 1

Chapter: 6

TRANSPORT PHENOMENA

Introduction
In the kinetic theory of gases, it is assumed that the molecules of a gas are all identical and
perfectly elastic spheres of negligible but finite size. Due to haphazard motion, the molecules constantly
collide on each other. At room temperature, an air molecule suffers about 4  109 collisions per second
and travels a mean free path about 107m. When the gas is not in equilibrium, the random i.e., haphazard
motion of gas molecules give rise to viscosity, conductivity and diffusion, when it tries to attain an
equilibrium state by transporting momentum, heat and mass respectively.

Molecular Collisions
According to kinetic theory of gases, the molecules are of finite size. During the random motion of
gas molecules in all possible directions and with all possible velocities, they collide with each other, as
shown in Figure 1. Between two successive collisions, a molecule moves with a constant speed, along a
straight line. These collisions are known as molecular collisions.

Mean Free Path ()

At a given temperature, the molecules suffer continuous collisions with one another. Between any
two collisions, the molecule travels freely a certain distance in a straight line. This distance is known as
free path. Thus, the path covered by a gas molecule between any two consecutive collisions a straight
line is called the free path. The direction of the molecule is changed after every collision. After a number
of collisions, the total path appears to be zig-zag and free path is not constant as own in Figure 1.
Therefore, a term mean free path is used and is defined as the average distance travelled by a gas
molecule between two successive collisions. It is denoted by . If the total distance ravelled after N
collisions is S, then mean free path () is given by
S
=
N

Figure 1
2 Chapter-6. Vijay Kumar Jha, Pigauna (Jaleshwar -11), Madhesh Province, Mithila
Mean Free time ()
The average time taken by a gas molecule between two successive collisions is called mean free
time. It is given by

=
c

Sphere of Influence
In the kinetic theory for an ideal gas, it is assumed that all the gas molecules are identical and perfectly
elastic spheres moving randomly in all possible directions. In doing so, they are continuously colliding
against each other. For the sake of simplicity of calculations, it is assumed that the molecule under
consideration say A, is in motion, all other molecules are at rest (Figure 2). Let  be the diameter of each
gas molecule. Taking the centre of molecule A as a centre, if we draw a sphere of radius  , then
molecule A will collide with all those molecules whose centres lie within this sphere. Such a sphere is
called sphere of influence.

Figure 2

Collision Cross-section

If we take a cross-section along the diameter of a sphere of influence, we get a collision cros section. The
area =  ² is called the collision cross-section as shown by dotted lines in Figure 2. It is the effective area
which determines the probability that a molecule of diameter  will collide another molecule of the same
diameter.

6.1 Expression for Mean Free Path (Clausius' Expression)

Suppose the molecules of a gas are all identical perfectly elastic spheres, each of diameter .  assumed
that the molecules do not exert any force on each other. To simplify the calculations, we further assume
that only the molecule under consideration is in motion while all other molecules are at rest. The sphere
of influence of the molecule has a radius  i.e. equal to the diameter of the molecule (shown in Figure 3).

Let c be the average velocity of molecule A and n be the number of molecules per unit volume, then in
one second, A will collide with all the molecules whose centres lie within a cylinder of radius  and
length c.
 Lectures on Thermodynamics 3

Figure 3
No. of molecules in the cylinder =  2c n
No. of collisions made by molecule A in 1 second =  2c n
1
1 Collision takes place in seconds.
 2cn
1
The time interval between two successive collisions = seconds.
 2cn
1 1
Distance travelled between two successive collisions = speed  time = c  =
 2cn  2n
1
 Mean free path,  = (1)
 2n
This is Clausius' expression for the mean free path. This is not exact and hence the corrected. Maxwell’s
formula is to be applied for the steady state.

Maxwell's Formula
The above expression (1) is not exact, because it is based on the assumption that only one
molecule under consideration is in motion while all other are at rest. If the motion of all the molecules is
taken into account and the Maxwell's law of distribution of molecular velocities is applied, which must
hold in steady state, the corrected formula for mean free path is
1
= (2)
2  2n
This is Maxwell's formula for the mean free path. If  and n are known, the molecular diameter  can be
calculated.
c
Note: Collision frequency is f = where c is average velocity of the gas molecule.

-::o::-
Variation of  with Temperature and Pressure

Consider 1 mole of an ideal gas. Then

PV = RT
Dividing both sides by N, the Avogadro's number, we have,
PV RT
=
N N
4 Chapter-6. Vijay Kumar Jha, Pigauna (Jaleshwar -11), Madhesh Province, Mithila
N R
P= T
V N
 The pressure P of a gas is given by P = nkT (3)

N R
Where, n = , the no. of molecules per unit volume and k = , the Boltzmann’s constant. Equation (3)
V N
gives,
P
n=
kT
Substituting for n in equation (2), the mean free path
1
=
2  2n
kT
= (4)
2  2P
Thus, the mean free path varies directly as the absolute temperature (T) and inversely as the pressure.

6.2 Transport Phenomena

According to the kinetic theory of gases, the molecules of a gas are in a state of thermal agitation. The
gas, therefore, attains a steady state or equilibrium state by transporting momentum, heat (thermal energy)
and mass from one layer of a gas to another layer, giving rise to the viscosity, conductivity and diffusion
respectively and the phenomena is called transport phenomena. The transport phenomena occur only in
the non-equilibrium state of a gas.

6.2.1 Viscosity (Transport of Momentum)

The different layers of a non-equilibrium gas may have different velocities. This will result in the relative
motion of the different layers with respect to one another. In such a case, the layer moving faster will
impart momentum to the layer moving slower to bring about an equilibrium state. Thus, the transport of
momentum gives rise to the phenomenon of viscosity.

Consider a gas flowing over a horizontal surface OX from left to right. The velocity the layer in contact
dv
with the surface OX is zero and goes on increasing along OZ at a uniform rate known as velocity
dz
gradient (Figure 4).
 Lectures on Thermodynamics 5

Figure 4
Consider a layer AB moving with a drift velocity V, situated at a distance Z from O. All the molecules in
this layer have the same drift velocity. Consider two more layers EF and CD just above and below AB
respectively at a distance A equal to the mean free path of the molecules, so that the molecules moving
vertically up or down do not suffer any collision while moving between the two layers.
dv
The velocity of the gas in the layer EF = V +
dz
dv
and the velocity of the gas in the layer CD = V 
dz
Due to thermal agitation, the gas molecules are moving in all directions, thus, the average num ber of
1
molecules along any one (+ ve or ve) axis will be th of the total number of molecules in the gas.
6
Let n be number of molecules per unit volume, m be the mass of each gas molecule, and c be the average
velocity of a molecule at a given temperature of the gas, then

nc
Number of molecules passing downwards from EF to CD per unit area of the AB in one second =
6
.
nc dv
 Forward momentum lost per unit area per second by the layer EF = m 6 (V + d z )
Siimilarly,
nc
Number of molecules passing upwards from CD to EF per unit area of the layer AB in second = .
6
nc dv
Forward momentum gained per unit area per second by the layer EF = m 6 (V  d z )

Net momentum lost by the layer EF per unit area per second = m 6 (V + d z )  (V  d z )
nc dv dv

1 dv
= mnc .
3 dz
6 Chapter-6. Vijay Kumar Jha, Pigauna (Jaleshwar -11), Madhesh Province, Mithila
The layer CD below AB gains the same amount of momentum. Hence, the layer EF above AB tends to
accelerate its motion and the layer CD below AB tends to retard its motion.
The backward dragging force per unit area = gain or loss of momentum per unit area per second.
1 dv
F= mnc (5)
3 dz

This must be equal to the viscous or tangential force 

dv
 d z  acting per unit area of the layer AB;  being
the coefficient of viscosity of the gas.
dv 1 dv
  = mnc
dz 3 dz
1
or = mnc
3
1
or = c (6)
3
where  = m c is the density of the gas.
-::o::-
Effect of Temperature on η

The density () of the gas decreases with increase in temperature but , the mean free path increases in
the same proportion so that  remains constant. Since the average molecular speed c is directly
proportional to the square root of its absolute temperature (c  T ) the coefficient of viscosity () will
also be proportional to T
Therefore,  T (7)
In other words, the coefficient of viscosity of a gas increases with increase in temperature.

Effect of Pressure on η

The density () of a gas increases with increase in pressure but , the mean free path decreases in the
same proportion, so that  remains constant. Moreover, the average molecular speed c is independent of
pressure. Thus, using relation

1
=
3
c

, the coefficient of viscosity is independent of pressure.

6.2,2 Thermal Conductivity (Transport of Thermal Energy)

The different layers of a gas may have different temperatures. The molecules will carry kinetic energy
(heat) from regions of higher temperature to the region of lower temperature to bring about an
equilibrium state. Thus, the transport of thermal energy (heat) gives rise to the phenomenon of thermal
conductivity.
Consider a fixed mass of a gas at rest as shown in Figure 5. Suppose there is a uniform temperature
d
gradient along OZ. In the steady state, the temperature of each horizontal layer parallel to OX is
dz
 Lectures on Thermodynamics 7
constant. Let  be the temperature of layer AB. The temperature of the layer EF which is at adistance (the
d d
mean free path) above AB is  +  and of the layer CD at the distance  below AB    .
 dz  dz

Figure 5

This means the gas above the layer AB is at higher temperature than the gas below AB. Hence, molecules
of gas coming from EF and passing downward across AB possesses more K.E than the molecules coming
from CD and passing upward across AB. Thus, the average energy of molecules below AB is continuously
increasing and that of molecules above AB is continuously de creasing, till the equilibrium is reached.

nc
Number of molecules crossing unit area of the layer AB upward or downward per second = .
6
where n is the number of gas molecules per unit volume and c is the average molecular velocity. If m is
the mass of each gas molecule, then the mass of gas crossing unit area of layer AB upward or downward
mnc
per second = .
6
If Cv be the specific heat of the gas at constant volume, then

Heat energy carried by the molecules in crossing unit area of the layer AB in the downward direction per
d
 Cv   + 
mnc
second = mass  specific heat  temperature = .
6  dz
Similarly, heat energy carried by the molecules crossing in unit area of the layer AB in the upward
d
 Cv    
mnc
direction per second = = .
6  dz
 Net transfer of heat energy per unit area of layer AB in the downward direction per second,
d mnc d mnc d
 Cv   +   Cv    
mnc
Q=  =  Cv  2
6  dz 6  dz 6 dz
Since mn = , the density of the gas, we should have
8 Chapter-6. Vijay Kumar Jha, Pigauna (Jaleshwar -11), Madhesh Province, Mithila
1 d
Q=  c  Cv (8)
3 dz

The coefficient of thermal conductivity (K) of the gas is defined as the quantity of heat that flows per unit
area per second per unit temperature gradient.
d
or Q=K (9)
dz

Equating equation (8) and (9),

d 1 d
K =  c  Cv
dz 3 dz
1
or K =  c  Cv (10)
3

Note: The coefficient of thermal conductivity of a gas is directly proportional to the square root of its
absolute temperature but K it is independent of pressure.

6.2.3 Self Diffusion (Transport of Mass)

The phenomenon of self diffusion is due to the transport of mass from a region of higher concentration to
a region of lower concentration to bring about an equilibrium.
dn
Consider a a gas having a concentration gradient along OZ. Let n be the concentration of layer AB.
dz
Let two more layers EF and CD be above and below that of AB at a distance , equal to the mean free
path of the molecules of the gas, so that the molecules moving vertically upward or down ward do not
collide between these two layers (Figure 6).

Figure 6
 Lectures on Thermodynamics 9
dn
Concentration at the layer EF = n +
dz
dn
and concentration at the layer CD = n 
dz
Therefore,
the number of molecules coming from layer EF and crossing AB downward per unit
area per second = c n + 
1 dn
6  d z
and number of molecules coming from layer CD and crossing AB upward per unit area per second
= c n  
1 dn
6  dz
 Net number of molecules crossing per unit area per second of layer AB in downward direction
= c n + 
1 dn 1  dn 1 dn
 c n   = c
6  dz 6  dz 3 dz
6.3 Brownian Motion
It is experimentally observed that pollen suspended in water move to and fro continuously.
These particles are dancing about in a state of random motion. This irregular motion is called "Brownian
motion" after the name of the discoverer Brown. The suspended particles are called the Brownian
particles. The gas molecules are of same nature as the Brownian particles only the difference is that the
Brownian particles are larger than the gas molecules. Laws of Kinetic theory of gases are obeyed by
Brownian particles.

Einstein's theory of the translation Brownian motion:

Einstein's theory of translation Brownian motion explains the physical nature of the
phenomenon. Brownian particle due to their random motion tend to diffuse into the medium in the course
of time. Einstein, therefore related the diffusion coefficient to Brownian motion. He calculated the
diffusion coefficient in two ways:
(i) From the irregular motion of the suspended particles.
(ii) From the difference in osmotic pressure between different parts due to difference in the of
the suspended particles.

6.3.1 Expression for diffusion coefficient when the molecules are in random motion:
Consider a cylinder of cross-sectional area, A, having two faces S1 and S2 separated by distance
, as shown in the figure. Let n1 and n2 be the molecular concentration at ends S1 and S2 respectively.
Since half of the molecules contained in imaginary cylinder move toward right side (S2) and other half
move towards left (S1), we can write:

Figure 7
1
The number of molecules crossing end face S2, in time t, to the right = n A
2 1
10 Chapter-6. Vijay Kumar Jha, Pigauna (Jaleshwar -11), Madhesh Province, Mithila
1
The number of molecules crossing end face S1, in time t, to the left = n A
2 2
1
Hence, the excess of particles crossing a middle layer to the right = (n n )  A
2 1 2
dn
If  be the gradient of molecular concentration, then according to definition, diffusion coefficient, D
dx
is given by

number of molecules crossing per second per unit area

D=
 dn 
 dx 
dn
But number of molecules crossing in time t through area A = D tA
dx
So, we can write
1 dn
(n n )  A =  D tA
2 1 2 dx
dn
Since, n1  n2 =  , we should have
dx
1 dn  dn
 =D t
2 dx  dx
 2
 D= .
2t

6.3.2 Calculation of diffusion coefficient from the difference in osmotic pressure

Consider a cylinder of length A. Let P1 and P2 be the osmotic pressure at end faces S1 and S2 respectively,
as shown in the figure below. In this process, same temperature exists. So from the gas laws
P1 = n1 KT and P2 = n2KT
; where P and n refer to the pressure and the number of particles per unit volume at temperature T. K is
Boltzmann constant.

Figure 8

Due to the difference of pressure, the cylinder will experience a force equal to
(P1  P2) A = (n1 KT n2 KT) = (n1  n1) KTA
If n is the mean concentration then number of particles contained in the cylinder = n  A
(n1  n1) KTA dn n1  n2 dn
Force acting on each particle = [∵ n1  n2 =    = ]
nA dx  dx
 Lectures on Thermodynamics 11
KT  dn  KT  dn 
 = 
n  dx n  dx

But, this force is equal to the viscous force because the particles are in viscous medium. So from stokes
law, we have
Viscous force = 6   r v

where r is the radius of particle and v is the steady velocity

KT  dn 
 6rv=
n  dx
KT  dn 
or nv= (1)
6r  dx
This expression represents the number of particles moving to the right per unit area. Now according to the
definition of diffusion coefficient,

No. of particles per unit area/s

D=
 dn 
 dx
No. of particles per unit area/s = D  
dn
or (2)
 dx
D = 
dn KT  dn 
or
 dx 6r  dx
RT R
or D= [∵ K = ]
6Nr N
Also, we know that
2
D=
2t
So we can write
2 RT
=
2t 6Nr
RT 1
 2 = t.
N 3r

-::o::-