Unit 3 Mean Free Path and Transport Phenomena

UNIT 3
Diffusion is a transport

MEAN FREE PATH AND

phenomenon. You will learn about it
in this unit.

TRANSPORT PHENOMENA

Structure
3.1 Introduction 3.4 Summary
3.5 Thermal
TerminalConductivity:
Questions Transport of Energy
Expected Learning Outcomes
3.6 Diffusion:
SolutionsTransport of Matter
and Answers
3.2 Mean Free Path
3.4 Summary
Elementary Derivations
3.5 Terminal Questions Solutions and
Distribution of Free Paths
Answers
3.3 Transport Phenomena
Viscosity: Transport of Momentum
Thermal Conductivity: Transport of Energy
Diffusion: Transport of Matter

STUDY GUIDE
In the previous unit, you have learnt about Maxwell distribution function for molecular speeds and used
it to obtain expressions for average speed, most probable speed and root mean square speed. In this
unit, you will learn about mean free path and transport phenomena. The study of free paths involves
simple concepts of probability, which you have learnt in your school mathematics classes. You are,
therefore, advised to refresh your earlier knowledge of probability. Also, you are familiar with the
concept of viscosity, diffusion and thermal conductivity. It will be a good idea to revise your school
physics about these concepts. This unit is slightly longer but the concepts are simple. Yet we advise
you to master sub-topics one at a time. Moreover, if you solve numerical problems given here as also in
other texts, you should get extra rewards.

“Atoms and molecules from their very nature can never be Ernst Mach
made the objects of sensuous contemplations.”

55
Block 1 Kinetic Theory of Gases
3.1 INTRODUCTION
In the previous units you have learnt that molecules of an ideal gas can be
considered as point masses, which move randomly with varying speeds. We
have seen that even at room temperature, the molecules of a gas, say
oxygen, move with very large speeds: v rms  480 ms 1. It means that a small
amount of scent released in a large room from a vessel should spread
throughout the room in no time. But this is not supported by our observation;
when we open the lid of a perfume bottle in one corner of the room, the
perfume is smelt at the other corner after a considerable time. This apparent
paradox led to a serious objection in the early stages of development of
kinetic theory of gases. However, this paradox was resolved by Clausius when
he argued that gas molecules have finite size and as they move, they
collide with one another leading to increase in diffusion time. This
amounts to modifying the basic assumption about the size of gas molecules.

You may now like to know: What is the average distance travelled by a
molecule between successive collisions? What is the mechanism of these
collisions? To answer these and such other questions, we use the concept of
mean free path. We have derived an expression for mean free path in
Sec. 3.2 under simplified assumptions. These help us to get a feel of the basic
physics. The distribution of free paths is also discussed in this section.

Suppose that a gas moves with a flow velocity towards the right. If its
molecules experience velocity gradient in a direction normal to the direction of
flow, the gas experiences net transport of momentum across an imaginary
plane in the gas along the direction of flow. This is characterised by the
coefficient of viscosity. You will note that in gases, unlike the case of liquids,
random thermal motion associated with gas molecules rather than frictional
force between successive layers, gives rise to the phenomenon of viscosity.
We will just quote the expression for the coefficient of viscosity in terms of
mean free path and discuss its temperature as well as pressure dependence.

When a temperature gradient exists in the body of such a gas in a direction

perpendicular to the direction of flow, random motion of gas molecules leads
to net transport of thermal energy in the direction of flow. This gives rise to the
phenomenon of thermal conduction. This is characterised by thermal
conductivity. You will learn how to correlate it with the coefficient of viscosity
in Sec 3.3.2.

When concentration gradient exists inside a gas normal to the direction of

flow, there is net transport of mass in the direction of preferential motion. This
is characterised by the coefficient of diffusion. You will learn how to express
the coefficient of diffusion in terms of mean free path, discuss its temperature
dependence and correlate it with coefficient of viscosity in Sec. 3.3.3. These
processes (viscosity, thermal conduction and diffusion) are collectively
referred to as transport phenomena, since we observe net transfer of
momentum, energy or matter.
56
Unit 3 Mean Free Path and Transport Phenomena
Expected Learning Outcomes
After studying this unit, you should be able to:

 define mean free path and obtain its expressions using elementary
derivation;

 describe the law of free paths;

 explain the term transport phenomena;

 discuss physical implications of viscosity of a gas;

 discuss how thermal conductivity is related to the coefficient of viscosity;

and

 establish the relation between mean free path and coefficient of

self-diffusion.

3.2 MEAN FREE PATH

You may recall that in elementary kinetic theory, gas molecules are
considered point masses (Assumption 2, Sec. 1.2) and intermolecular
collisions are ignored. But as mentioned earlier, to explain the finite time taken
by a gaseous mass in an open container to disperse, Clausius assigned a
finite size to gas molecules. He argued that when a molecule moves within the
gas under equilibrium conditions, it necessarily collides with other molecules it
encounters in its path and undergoes frequent changes in its direction of
motion. This results in a series of zig-zag paths. These are called free paths.
The trajectory of a molecule moving with average speed v is shown in
Fig. 3.1. The concept of mean free path may be visualised if you consider a
person shooting aimlessly in a thick forest. The bullets fired by him may
eventually hit trees but some will travel farther than others.

Fig. 3.1: The trajectory of a molecule moving in a gas.

Now carefully examine Fig. 3.1. You will note that free paths are of unequal
length; some are short while others are long. The average length of the free
paths is referred to as mean free path. Thus, mean free path is the average
distance travelled by a molecule between two successive collisions. We
denote it by the symbol  (pronounce as lambda).
57
Block 1 Kinetic Theory of Gases
If 1, 2, 3, …, N are the successive free paths traversed by a molecule in
time t and N denotes the total number of collisions suffered in this period, we
can write
  2  3  ...  N
 1
N
total distance travelled

total number of collisions
If v is the average speed with which a molecule moves in the body of the gas,
then you can write
vt
 (3.1)
N
Under normal conditions, the mean free path of oxygen molecules is about
2  10 7 m. This is smaller than the wavelength of visible light, which is in the
range 4  10 7 m  6  10 7 m. However, mean free path is greater than the
average intermolecular separation (3  10 9 m).

If   t / N denotes the mean time between two successive collisions, then you
can also write
v
 v  (3.2)
Pc

Here Pc  ( 1) denotes the collision frequency, which is a measure of the

average number of collisions per second. You may now ask: How can we
relate  to the microscopic properties of a gas? To answer this question, we
first make an elementary calculation.

3.2.1 Elementary Derivations

Suppose that a gas consisting of a large number of molecules, each of mass
m and diameter d, is in thermal equilibrium. We assume that gas molecules
undergo random collisions. Mathematically speaking, we can say that the
probability of a molecule suffering a collision in a small interval of time dt is
independent of the history of past collisions made by it. As a simple analogy,
you can consider that if you throw a dice, the probability of getting a six does
not depend on the preceding throw, where a six may or may not have
appeared. Similarly, winning or losing toss before a match is independent of
earlier results.

From Eq. (3.2), we note that to derive an expression for , we must first know
the expression for Pc . We shall make this calculation by assuming that only
one molecule is in motion. We refer to it as Zeroth order approximation. You
will realise that essential physics involved in this phenomenon can be
understood well through simple arguments.

Case 1: Zeroth order approximation: Only one molecule in motion

We assume that only one molecule, labelled as A, criss-crosses a sea of

other molecules that are at rest. (As such, this is an over-simplification of the
actual situation.) You can easily visualise that for a collision to take place, the
58
Unit 3 Mean Free Path and Transport Phenomena
centre-to-centre distance between the moving and a stationary molecule
should be equal to 2r (  d ) , as shown in Fig. 3.2a.
Note that the centre-to-centre distance at the time of collision between the
moving and a stationary molecule will also be equal to the diameter d of a gas
molecule, if the stationary molecules were shrunk to geometrical points and
the moving molecule is taken to be of radius d (Fig.3.2b). You can visualise
this as if the moving molecule carries with it a circular disc of radius d.
Therefore, the moving molecule can be thought of as sweeping out a cylinder
of cross-sectional area d 2 and length v t in time t. This is illustrated in
Fig. 3.2c. During this time, it will collide with all other molecules whose centres
lie within a cylinder of volume v t d 2 . You must convince yourself about this
before proceeding further.

r r

(a)
d=2r
2r

(c)
(b)

Fig. 3.2: a) Collision of two molecules of radii r : Instantaneous representation;

b) equivalent representation of collision in (a) where radius of the
moving molecule is increased to 2r and the stationary molecule is
shrunk to a geometrical point; c) cylindrical volume mapped by the
moving molecule of radius 2r.

If the number of molecules per unit volume is n, the number of molecules

contained in the cylinder of volume v t d 2 will be equal to v t d 2n . You will
agree that this number also equals the number of collisions suffered by the
moving molecule in time t.
The collision frequency, which defines the number of collisions per second, is
given by
Pc  v d 2 n  n v (3.3)
where   d 2 is known as microscopic collision cross-section. It is expressed
in m 2.
Substituting this result in Eq. (3.2), we get
v 1 1
   (3.4)
Pc d n  n
2

The quantity n is called macroscopic collision cross-section and its unit is

m1. Note here that the unit of macroscopic collision cross-section is
reciprocal length, not an area.
Let us pause for a moment and ask: What is good about the relation
contained in Eq. (3.4)? What are its implications? To seek answers to these
questions, we note that Eq. (3.4) predicts that mean free path is inversely
proportional to the macroscopic collision cross-section or number density and
second power of diameter of the molecule. It means that mean free path will
59
Block 1 Kinetic Theory of Gases
be less for a denser and/or a heavier gas. This is in perfect agreement with
common observation and lends support to the basic tenets of kinetic theory.
The aesthetic beauty of kinetic theory is in its ability to relate experimentally
measurable macroscopic quantity such as mean free path to a microscopic
quantity – size of a molecule.

You may now logically ask: What is the typical magnitude of mean free path in
a gas at STP? The answer to this question is contained in the following
example. You should go through it carefully.

XAMPLE 3.1: COLLISION CROSS-SECTION

The average speed of hydrogen molecules is 1840 ms 1. The radius of a

hydrogen molecule is 1.37  1010 m. Calculate (i) collision cross section,
(ii) collision frequency and (iii) mean free path. Take n  3  1025 m3.

SOLUTION  Since the radius of a hydrogen molecule is 1.37  10 10 m

and number density n is 3  1025 m3, we can easily calculate the collision
cross-section, collision frequency and mean free path using Eqs. (3.3) and
(3.4):

(i)   d 2  4 (1.37)2  1020 m2  23.6  1020 m2

(ii) Pc  d 2n v  n v 

 (3  1025 m3 )  (1.84  103 ms 1)  (23.6  1020 m2 )

 1.3  1010 s1

1 1
(iii)   
3
nd 2
(3  10 25
m )  (23.6  10 20 m2 )

 141 109 m  141 nm

10
Note that the number of collisions per second is of the order of 10 , which
is a very large number. For this reason, the path of a molecule is made up
of so many kinks and zigzags that it is almost impossible to follow its
trajectory.
Also note that  is large compared to intermolecular distance, which is only
a few nanometer ( 3 nm).

To ensure that you have grasped the ideas and your progress is satisfactory,
we would like you to solve an SAQ before proceeding further.

SAQ 1 – Mean free path

The mean speed of oxygen molecules at room temperature is

1
450 ms . If the radius of an oxygen molecule is 1.8 Å, calculate , Pc ,  , and
. Take n  3  1025 m3.
60
Unit 3 Mean Free Path and Transport Phenomena
From ideal gas equation, we know that the pressure exerted by a gas can be
expressed in terms of its temperature as

p  nkBT

On using this result to substitute for n in Eq. (3.4), the expression for mean
free path takes the form

  kBT /  p (3.5)

This result is very interesting; it suggests that mean free path is directly
proportional to absolute temperature of a gas and inversely proportional to the
pressure exerted by it on the walls of the container. It means that the value of
 will increase as pressure decreases and/or temperature increases. Now
suppose we reduce pressure to a very small value using a vacuum pump.
According to Eq. (3.5), the mean free path can approach infinity. But in actual
practice, this is forbidden by physical considerations. At the most, the value of
 can equal the dimensions of the container.

You will agree that in deriving Eq. (3.4), we made a rather unrealistic
assumption that only one molecule moved while other molecules were at rest.
But in Unit 2 you have learnt that molecules in a gas move about randomly.
Therefore, a better way to represent the actual situation will be to assume that
all molecules move with the same speed. When we consider this, Eq. (3.5)
modifies to

3 0.75 k T
C    0.75 B (3.6)
4n n p

It may be mentioned here that this result was obtained by Clausius. That is
why we have put the subscript C with . It shows that mean free path of the
molecules in a gas decreases when all molecules are moving.

You will recall that in actual practice, we should have considered Maxwellian
distribution of velocities. However, the mathematical calculations are quite
involved and it will be sufficient to quote the result. (Those of you who are
interested in these details may refer to references given in the Further
Readings):

 0.707 
1 1
M 
2n n

k T
 (0.707 ) B (3.7)
p
M
Note that  0.94.
C

This result shows that there is a correction of only 6 per cent in the value of
mean free path when we go from uniform speed model to Maxwell distribution.
You will also note that for point molecules (d  0), collision cross-
section   0 and mean free path    .

Before proceeding further, we would like you to revise what you have learnt so
far in this unit. 61
Block 1 Kinetic Theory of Gases

MEAN FREE PATH

 The average distance covered by a gas molecule between two

successive collisions defines the mean free path.
 The mean free path of a gas molecule when only one molecule moves
with speed v is given by
1

 d 2n

Now go through the following example carefully to get a feel of the values
involved in real situations.

XAMPLE 3.2: MOLECULAR DIAMETER

A gas is assumed to obey Maxwell’s law of distribution of speeds.

Calculate the (i) molecular diameter of a gas and (ii) the number of
collisions per unit distance if mean free path at STP is 2.85  10 7 m. Take
n  3  1025 m3.

1 1
SOLUTION  From Eq. (3.7), we can write  
2 d 2n
(a) To determine d , we rewrite the expression for mean free path as
1 1
d2 
2  n
On substituting the given values in this expression, we get
1 1 1 1
d2  .  
2 3.14 2.85  10 7 m 3  10 25 m3
1
 m 2  2.63  1020 m2
18
37.96  10
so that

d  1.62  10 10 m

Note that the order of magnitude of mean free path is 103 times that
of molecular diameter.
(b) By definition, the number of collisions per unit distance is reciprocal
of the mean free path. Therefore, we can write
1 1
Nc  
 2.85  10 7 m

 3.51 10 6 m1

To test your understanding, you may like to answer the following SAQ.
62
Unit 3 Mean Free Path and Transport Phenomena

SAQ 2 – Pressure and mean free path

The mean free path of the molecules of an ideal gas at 15C is

6.28  10 8 m. If the radius of a gas molecule is 1.88 Å, calculate the pressure
exerted by the gas. Also calculate the number of collisions suffered by a
molecule in travelling a distance of one metre. Take kB  1.38  10 23 JK 1.

You now know that molecules undergo collisions very frequently. It means that
the distribution of free paths will be spread over a wide range. It is, therefore,
interesting to know their distribution. This forms the subject of discussion of
the following sub-section.

3.2.2 Distribution of Free Paths

Imagine a person shooting aimlessly in a thick forest. Every bullet eventually
hits a tree, but some travel farther than others. This situation is analogous to
the flights of gas molecules. We now wish to know the distribution of free
paths. To do so, let us consider a molecule at the start of its journey (point O).

dx

O
A B

Fig. 3.3: A molecule travels from O to A without colliding with another

molecule. It traverses a further distance AB in time dt.

We assume that past history of a molecule does not influence its subsequent
motion in any way. Suppose that a molecule moving with average speed v ,
travels a distance OA  x without suffering any collision. The probability of
this event will be a function of x . Let us denote it by f (x ) . The probability that 1. If the probability of
this molecule makes a collision in moving from A to B will be determined by occurrence of a
collision is p(x), the
the product of collision frequency and time interval in which it covers the given
probability of its non-
distance. So, we can take it to be equal to Pc dt , where Pc is the collision occurrence will be
frequency. Since dx  v dt , we can write 1  p(x).
2. The probability of
dx dx
Pc dt  Pc  occurrence of two
v  mutually exclusive
events is
where  is the mean free path defined by Eq.(3.2). Hence, the probability that multiplicative.
 dx 
a molecule traverses a distance dx without making a collision is 1  .
  
63
Block 1 Kinetic Theory of Gases
Since motion from O to A and from A to B can be considered as two
independent events, we can say that the probability f ( x  dx ) of this molecule
completing its journey from O to B without making any collision is equal to the
product of probabilities for individual events. Mathematically, we express it as
 dx 
f x  1  . Hence, we can write
The function of a real   
variable at a point is
 dx 
f x  dx   f x  1 
expressed in terms of
 (3.8)
the power series   
involving the value of
the function at a To simplify this expression, we use Taylor series expansion of f ( x  dx ) and
neighbouring point and
ignore second and higher order terms in dx . Then we can write
its derivatives using
Taylor series
df x 
expansion: f x  dx   f x   dx
dx
f (x + dx)
On using this result in Eq. (3.8), we get
= f (x) + x (df /dx) +
df x  f x 

( x )  d 2f 
2 dx 
 ... …
2!  dx 2 
You can easily integrate it to obtain

f ( x )  A exp(x / ) . (3.9)

The constant of integration A can be easily evaluated using the fact that
f (0)  1 . (Physically it means that the particle is yet to move or collide.) This
gives A  1. Hence, Eq. (3.9) takes a compact form:

f ( x )  exp( x / ) (3.10)

This is the law of distribution of free paths. It suggests that the probability of a
molecule travelling very large distance without making a collision will be
vanishingly small.

If we have a sample of N0 molecules to start with, only N0 exp(x / ) of these

will survive a collision in traversing a distance x . Let this number be denoted
by N(x ) . Then we can write

 x
N( x )  N0 exp    (3.11)
 

This equation is known as survival equation. You will come across similar
equations in other areas of physics. For example, radioactive decay obeys the
equation N( x )  N0 exp t  , where N(x ) is the number of atoms which
survive disintegration for time t and  is disintegration constant.

Now refer to Fig. 3.4. It shows the plot of distribution of free paths as a
Fig. 3.4: The plot of the function of x /  . Note that the fraction of molecules with free paths larger than
survival equation.  is only e 1  0.37.

We now give a solved example to reinforce our discussion of this section.

64
Unit 3 Mean Free Path and Transport Phenomena

XAMPLE 3.3: SURVIVAL EQUATION

4
In a sample of 10 gas molecules, each molecule is moving with the same
speed. Calculate the number of molecules that will continue to travel
undeflected after traversing distances of 0.4  , 0.5  ,  , 2  , and 10  .

SOLUTION  The values of N(x ) are tabulated below for different

distances traversed by the gas molecules and you can readily verify these
using Eq. (3.11). To illustrate, for x  0.4 , we can write

N( x )  104 e0.4  6703

x 0.4 0.5 1 2 10


N 6703 6065 3679 1353 0.4540

You will note that after traversing a distance of , 2 and 10 , nearly 37 %,
13.5% and 0.5% molecule, respectively, will survive a collision.

You may now like to solve an SAQ to assess your understanding.

SAQ 3 – Law of free paths

The mean free path of the molecules of a gas at pressure p and temperature
T is 5.2  10 8 m. Calculate the probabilities that a molecule will travel
10.4  108 m without making a collision, if (i) temperature is doubled, (ii) both
temperature and pressure are doubled, and (iii) pressure is tripled.

Before proceeding further, let us revise what you have learnt in this section.

LAW OF FREE PATHS

 The law of distribution of free paths gives the probability that a gas
molecule covers a distance x without making a collision:
f ( x )  exp( x / )
 The law of distribution of free paths is supported by experimental
results.

So far we confined ourselves to a gaseous system at uniform temperature and

pressure in equilibrium so that there is no preferential motion of any kind. This
is because the rate of migration of molecules across a given plane in the gas
is exactly balanced by an equal number moving in the opposite direction.
However, when the entire gas or a part of it moves as a whole in a particular
direction, the preferential motion and random molecular motion combine to
give rise to a new behaviour. Let us learn about it now. 65
Block 1 Kinetic Theory of Gases
3.3 TRANSPORT PHENOMENA
We know that each gas molecule has a finite mass and is characterised by
random molecular motion. Therefore, it possesses momentum as well as
energy. So, while moving from one part of the container to another, it is a
potential carrier of these physical quantities under certain peculiar conditions.
When a gas is in equilibrium, there is no net transport of matter, energy, or
momentum. However, when a gas is endowed with macroscopic motion, i.e.,
the entire gas or a part of it moves as a whole in a particular direction, the
following three cases may occur singly or jointly:
 The different parts of the gas move with different velocities. This will give
rise to relative motion between different layers of the gas. As a result, the
faster moving layers in the body of the gas will lose momentum to the
slower moving layers. Therefore, across an imaginary plane, there will be
net transport of momentum in the preferential direction. This results in a
frictional force, which is characterised by the coefficient of viscosity.
Note that viscosity in gases arises due to the random thermal motion of
molecules when it is endowed with macroscopic motion rather than a
frictional force between any two adjacent layers, as in the case of liquids.
 Different parts of a gas are at different temperatures. In this case, the
molecules of the gas will carry thermal energy from regions of higher
temperature to regions of lower temperature and tend to attain thermal
equilibrium. This gives rise to the phenomenon of thermal conduction.
 Different parts of a gas have different concentrations. The molecules from
regions of higher concentration will migrate to the regions of lower
concentration resulting in the transport of mass (matter). This leads to the
phenomenon of diffusion.
We thus find that viscosity, conduction and diffusion are bulk properties of
gases signifying transport of momentum, energy and mass, respectively.
These are collectively categorised under the title of transport phenomena.
These processes are of vital importance in physical sciences and find several
important applications. We will now discuss these in some detail on the basis
of the simple molecular model discussed in the preceding chapters.

3.3.1 Viscosity: Transport of Momentum

In your school physics classes, you have learnt that the property by virtue of
which a liquid opposes the motion between adjacent layers is referred to as
viscosity. It is quantitatively expressed in terms of the coefficient of viscosity,
. Can you recall its definition? The coefficient of viscosity is defined as the
tangential force per unit area when a unit velocity gradient exists in a direction
perpendicular to the direction of motion. Mathematically, for small velocity
gradient we write
du
F  A (3.12)
dy

Here F denotes the viscous force acting on an area A and du / dy is velocity

gradient along the +ve y-direction. The negative sign signifies that the viscous
66 force is directed against the velocity gradient. Note that  has dimensions
Unit 3 Mean Free Path and Transport Phenomena
1 1 1 1
ML T . Its SI units are kg m s . It is also denoted as pascal-second. The
CGS unit of  is poise (  10 1 kg m1s1).

Let us consider a gas enclosed between two planes depicted as aa and bb
in Fig. 3.5 and seprated through a distance h. Suppose that the gas is
endowed with mass motion from left to right and a positive velocity gradient
exists along the y-axis, which is normal to the direction of flow.

Fig. 3.5: Momentum transport across a plane in a gas having velocity gradient.

Mathematically, the coefficient of viscosity of a gas is given by

1 1
 mnv  v  (3.13)
3 3
On substituting the expression for  corresponding to Maxwellian distribution
of speeds in Eq. (3.13) from Eq. (3.7), we can express  in terms of :

1 mv
M  (3.14)
3 2 
This result provides us a useful way to estimate a microscopic property such
as molecular diameter in terms of a macroscopic property such as viscosity ,
which is directly measureable. This finding provided remarkable support to
elementary kinetic theory.

To be able to discuss pressure and temperature dependence of , we

8 kBT
substitute for v  from Eq. (2.34) in Eq. (3.14). This gives
m

2 m kBT 0.376
M   m kBT (3.15)
3  
Effect of Temperature and Pressure on Viscosity
 From Eq. (3.15) we note that the viscosity of a gas is directly proportional
to the square root of absolute temperature. So, if we plot  versus T 1/ 2 ,
we should obtain a straight line. However, experiments show that viscosity
of a gas increases with temperature at a slightly faster rate than T 1/ 2 .
67
Block 1 Kinetic Theory of Gases
The departure from the value 0.5 provides us an opportunity to review the
assumptions of elementary kinetic theory.

 According to elementary kinetic theory, the viscosity of a gas is indpendent

of pressure (or n). But in actual practice, pressure independence of  is
observed only for a limited range – from a few mm of mercury to a few
atmosphere. Experiments show that at very low pressures, viscosity
decreases and at very high pressures, viscosity increases.
We can explain these observations as follows:
At very low pressures, intermolecular collisions become rare and as
mentioned earlier, the mean free path becomes comparable to the
dimesions of the apparatus. However, the number density decreases
continuosly with pressure. This makes the coefficient of viscosity to
decrease as pressure decreases. On the other hand, at very high
pressures, the mean free path is comparable to molecular size and the
coefficient of viscosity increases with pressure (or n).
We now recapitulate important results of this sub-section.

COEFFICIENT OF VISCOSITY

 When different parts of a gas endowed with mass motion and its
molecules move with different speeds, there is net transport of
momentum in the direction of mass motion.
1 1
 The coefficient of viscosity of a gas is given by   mnv   v  .
3 3
 Theory predicts that  is independent of pressure but varies as T 1/ 2.

XAMPLE 3.4: COEFFICIENT OF VISCOSITY AND MEAN

FREE PATH

The coefficients of viscosity of argon and helium are 22  106 Pa s and

19  106 Pa s, respectively. Calculate the mean free paths for these
gases, if they are kept under identical conditions.
SOLUTION  Since the gases are under identical conditions, n is same
for both gases. Then from Eq. (3.13) we can write
Ar mAr v Ar  Ar
 (i)
He mHe v He He
8kBT
Using the relation v  , we can rewrite (i) as
m
Ar mAr  Ar

He mHe He

On re-arranging terms, we get

1/ 2  6 
 Ar mH e Ar  4   22  10 Pa s   0.367
  
He mAr He  39.9   19  10 6 Pa s 
 
68
Unit 3 Mean Free Path and Transport Phenomena

XAMPLE 3.5: COEFFICIENT OF VISCOSITY AND MEAN

FREE PATH

1
The molecules of helium gas move with an average speed of 1200 ms . If
  19  106 Pa s, calculate the mean free path for helium.

SOLUTION  We can rewrite Eq. (3.13) as

3 3V
 
v Mv

where M is molecular weight and V is volume occupied by one mole of the

gas. On inserting the given numerical values, we get

3  (19  10 6 kg m 1s 1 )  (22.4  10 3 m3 )


( 4  10  3 kg)  (1200 ms 1 )

 2.66  10 7 m

You may now like to solve an SAQ to assess your understanding.

SAQ 4 – Viscosity and molecular dimensions

Calculate the radius of an oxygen molecule, if coefficient of viscosity of

oxygen is 19.6  10 6 kg m1s1 at 15C and v  436 ms 1. Assume that it
obeys Maxwellian law of distribution of speeds.

3.3.2 Thermal Conductivity: Transport of Energy

When a gas is endowed with mass motion and a temperature gradient exists
between its different layers, a finite amount of energy is transported by gas You should not
molecules due to their random motion from regions of higher temperature to confuse the symbol K
regions of lower temperature. As mentioned earlier, this gives rise to the used for thermal
phenomenon of thermal conduction. The rate at which thermal energy is conductivity with the
transported across any surface per unit area is characterised in terms of symbol used for kelvin,
the unit of
thermal condctivity, K and we can write
temperature.
dT
Q  K (3.16)
dy
where (dT / dy ) is temperature gradient and temperature of the gas at an
imaginary surface xx within the gas is T.
The expression for thermal conductivity is given by
f
K n v  kB (3.17)
6

From this equation, we note that thermal conductivity is directly proportional to

average molecular speed. Therefore, theory predicts that thermal conductivity
69
Block 1 Kinetic Theory of Gases
is directly proportional to square root of temperature. However, in actual
practice, K increases somewhat more rapidly suggesting that when
intermolecular forces come into play, they begin to influence energy transport.
Relation Between  and K
When we closely examine the expressions for  and K, we expect some
connection between them. It is instructive to ascertain it as both these
quantities are physically measureable. Therefore, we divide Eq. (3.17) by
Eq. (3.13) and obtain
K f kB f kBNA f R
   (3.18)
 2 m 2 M 2 M
M
where m  and M is molecular weight of the substance.
NA

From Unit 2 you will recall that molar heat capacity at constant volume is
given by
f
cV  R
2
Using this result, we can rewrite Eq. (3.18) as
K cV

 M
KM
or 1 (3.19)
cV

Atomicity of a gas From this result we note that the ratio ( KM / cV ) is constant, same for all
defines the number of gases. But experiments show that this ratio is greater than one. A more
atoms in its molecule. rigourous calculation shows that this ratio varies between 1.5 and 2.5 and is
different for different gases; decreasing with atomicity. There can be several
reasons for this difference.

You should now go through the following example carefully.

XAMPLE 3.6: THERMAL CONDUCTIVITY AND

VISCOSITY

The coefficient of viscosity of helium is 18.6  106 Nsm2, M  4 kg kmol 1

and cV  12.5  103 Jkmol 1 K 1. Calculate the thermal conductivity of
helium.

cV
SOLUTION  From Eq. (3.19) we can write K  . On putting the
M
given values, we get

(18.6  10 6 Nsm 2 )  (12.5  10 3 J k mol 1K 1)

K
4 kg k mol 1

 5.8  10 2 Jm 1 s1 K 1 .
70
Unit 3 Mean Free Path and Transport Phenomena
Before proceeding further, we recapitulate important result in this sub-section.

THERMAL CONDUCTIVITY

 When a temperature gradient exists between different layers of a gas

endowed with mass motion, there is net transport of energy from
regions of higher temperature to regions of lower temperature. This
gives rise to phenomenon of thermal conduction.
 The thermal conductivity of a gas is given by
fnv
K  kB
6
KM
 In terms of the coefficient of viscosity, we can write  1.
cV

3.3.3 Diffusion: Transport of Matter

Inter-mixing of gases is known as diffusion. This phenomenon is responsible
for the smell of flowers/scent reaching us. This is also why while sitting in your
study room, you can know what is being cooked in the kitchen. It is a direct
consequence of random molecular motion when there are inequalities in
concentration. Molecules diffuse from regions of higher concentration towards
regions of lower concentration. Fig. 3.6a schematically depicts self-diffusion
(diffusion of like molecules). Diffusion of unlike molecules is shown in
Fig. 3.6b.

Fig. 3.6: a) Self-diffusion across a barrier; b) Diffusion of unlike molecules.

We describe intermixing of gases in terms of diffusion coefficient, D. If the

molecular concentration along a horizontal plane xx  is n and there is positive
concentration gradient dn / dy in the vertical plane, the number of particles
crossing the given surface per unit area per unit time is given by
dn
  D (3.20)
dy 71
Block 1 Kinetic Theory of Gases
The concentration of The calculation of diffusion coefficient is complicated due to the fact that the
235
U isotope in natural rates of diffusion of two gases may not be the same. However, we can
uranium is 0.71%. For understand the essential ideas by considering the diffusion of like molecules,
power generation in a i.e., self-diffusion. The diffusion of the isotopes of the same element (say
nuclear power plant, it 235
U and 238U) is an excellent example of self-diffusion. For self-diffusion, the
is desirable to increase
expression for diffusion coefficients is
its concentration to
2-3%, i.e., above its 1
natural value. Then we
D v (3.21)
3
say that uranium has
been enriched. If we now substitute the values of v and  for a Maxwellian gas, we get

0.376 kBT 3 / 2
D (3.22)
 p m

For air at STP,   100 nm, and v  450 ms 1. It means that the diffusion
coefficient for air is of the order of 105 m2 s1. Moreover, Eq. (3.22) implies
that diffusion coefficient will vary inversely with pressure and directly with
temperature as T 3 / 2 . The predicted variation with pressure is in agreement
with the observed results but the power of T lies between 1.75 and 2. The
reason for the more rapid increase with temperature is attributed to the
presence of intermolecular forces in real gases. To be precise, it is due to the
attractive part of the intermolecular potential.

Relation between D and 

If we combine Eqs. (3.13) and (3.21), we can write

D
  1 (3.23)


Though simple kinetic theory predicts that  is constant, equal to one for all
gases, its observed value lies between 1.3 and 1.5. It means that simple
calculations given here help us to know qualitative behaviour.

We now sum up what you have learnt in this unit.

3.4 SUMMARY

Concept Description

Mean free path  The mean free path is the average distance travelled by a molecule
between two successive collisions. If we assume that all but one
molecule moves with average speed v , the expressions for  is given
by

1 1
 
nd 2 n

where n is the molecular number density and  is the collision cross-

section. For a sphere of diameter d ,   d 2 .

72 
2
dNv  NA3e Bv d 3v
Unit 3 Mean Free Path and Transport Phenomena
Survival equation  For a sample of N0 molecules, the number of molecules which travel
a distance x without making any collision is given by survival
equation:

N  N0 exp(x / )

Transport phenomena 
 When a gas is endowed with mass motion, random molecular motion
can lead to transport of energy, momentum and mass, depending
on the physical conditions, which lead to the phenomena of thermal
conduction, viscosity and diffusion, respectively. These are
collectively referred to as transport phenomena.

Coefficient of  The coefficient of viscosity for a gas having velocity gradient is

viscosity given by
1 1
 m n v   v 
3 3
where m is molecular mass and v is average speed of a molecule.

Thermal conductivity  The thermal conductivity K is given by

f
K n v  kB
6
where f denotes the number of degrees of freedom.

 K and  are connected by the relation

KM
1
cV

Coefficient of  The coefficient of diffusion D is given by

diffusion v
D 
3

3.5 TERMINAL QUESTIONS

1. Calculate the diffusion coefficient of hydrogen molecules at 27C when
pressure is 2 atm. Assume that it behaves as a Maxwellian gas. Take

rH2  1.37  10 10 m and kB  1.38  10 23 JK 1

2. Calculate the radius of the nitrogen molecule using the following data:

K  23.7  103 J s1 m1 K 1, cV  29.1 103 Jkmol1 K 1 at 273 K

M  28.02 kg, NA  6.023  1026 and kB  1.38  10 23 JK 1

3. Calculate the mean free path for hydrogen molecules at STP. The

diameter of a hydrogen molecule is 2.9 A . Take
n  3  1025 molecules m3.
73
Block 1 Kinetic Theory of Gases
6
4. The coefficient of viscosity of a gas is 16.6  10 N s m2. Calculate the
diameter of the gas molecule at STP, when average molecular velocity
v  4.5 102 ms 1 ; number density  2.7  1025 molecules m3 , and
molecular weight of nitrogen  28. Take Avogadro’s number
 6.023  1023 mol 1.
5. Calculate the thermal conductivity of air using the following data:

v  470 ms 1,  1.29 kg m3,   5.6 10 8 m, M  29 ,   1.4, and

R  8.31 kJ mol 1 K 1.

3.6 SOLUTIONS AND ANSWERS

Self-Assessment Questions
1. We have

  4r 2  4(1.8  1010 m)2  40.7  10 20 m2

Pc  n v  (3  1025 m3 )  (40.7  1020 m2 )  (450 ms 1)

 5.49  109 s1

  Pc 1  1.8  10 10 s

1
and   82 nm
3
(3  10 25
m )  ( 40.7  10 20 m2 )

2. From Eq. (3.7), we recall that

k T
 B
p

Here   4r 2  4  3.14  (1.88  1010 m)2  44.42  1020 m2

On substituting the given values of Boltzmann constant and temperature

(in kelvin), we get

k BT (1.38  10 23 JK 1 )  (288 K)

p 
 M  (6.28  10  8 m)  ( 44.4  10  20 m 2 )

 1.43  10 5 Nm 2

1
For a Maxwellian gas, the value for p will be lower by a factor of .
2
Hence, its magnitude will be

p  1.01 10 5 Nm 2 .

The number of collisions suffered by a molecule per metre of the path

1
Ns   16  106 m1 .
74
6.28  10 8 m
Unit 3 Mean Free Path and Transport Phenomena
3. From Eq. (3.10) we recall that the probability of a molecule travelling a
 x
distance x without making a collision is given by f ( x)  exp   , where  is
 
a function of p and T. Here x  10.4  10 8 m and

 ( p,T )  5.2  10 8 m

i) Using Eq. (3.7) we can write ( p,2T )  2  ( p,T )  10.4  10 8 m.

 Required probability = e1 = 0.37

ii)  (2p, 2T) =  (p, T) = 5.2  108 m

 Required probability e2 = 0.14

1 5 .2
iii)  (3p, T) =  (p, T) =  108 m = 1.73  108 m
3 3

 Required probability e6 = 2.48  103

32
4. m   5.33  10  26 kg
26
6  10

From Eq. (3.13) we recall that

1 mv

3 d 2

On re-arrangement, we can write

mv
d2 
3 

On substituting the values of various physical quantities, we get

(5.33  10 26 kg)  ( 436 ms 1 )

d2 
3  (19.6  10  6 kg m 1 s 1 )

Hence,
1/ 2
 5.33  436 
d    10 10 m
 3  19.6 

= 3.55  10 10 m

and r  1.77  10 10 m

Terminal Questions
1. From Eq. (3.22), we recall that

0.376 (kBT )3 / 2
D
 p m
75
Block 1 Kinetic Theory of Gases
On substituting the given values, we get

0.376 [(1.38  10 23 JK 1)  (300 K )]3 / 2

D
(3.1417  4) (1.37  10 10 m) 2 (2.026  105 Nm 2 ) (3.3  10 27 kg )1/ 2

0.376 266.4  10 33 J3 / 2


(23.59  10 20 m2 ) (2.026  105 Nm 2 ) (5.74  10 14 kg1/ 2 )

 3.56  10 5 m2 s1

2. We have from Eq. (3.19)

cV
K  0.376
m
kBT 1/ 2 cV  0.376 kBT 1/ 2 cV
M  M  MN A

where we have used Eq. (3.15) for  and m  M / NA .

Hence,

  4r 2 
0.376 kBT 1/ 2 cV
K NM

or r 2 
0.376 kBT 1/ 2 cV
4K NM

On substituting the given values, we get

(0.376) (1.38  10 23 JK 1 )1 / 2  (273 K)1 / 2  (29.1 10 3 J k mol 1 K 1 )

r2 
4  (23.7  10 3 J m 1 s 1 K 1 )  [(6.02  10 26 k mol 1 )  (28.02 kg)]1 2

671.5  10 9
  1.74  10 20 m2
10
3869 .0  10

Hence r  1.32  1010 m

1
3. From Eq. (3.7) we recall that  
n

Here  = d 2 =  (2.9  1010)2 m2 = 26.42  1020 m2

and n = 3  1025 m3

Hence,

1

3
(3  10 25
m )  (26.42  10 20 m 2 )

 1.26  10 7 m

28 kg kmol 1
4. m 
6.023  10 26 kmol 1
76
Unit 3 Mean Free Path and Transport Phenomena

 4.649  10 26 kg

  16.6  106 Ns m2

v  4.5  102 ms 1

From Eq. (3.13), we can write

3

mnv

On inserting the values of various physical quantities, we get

3  (16.6  10 6 Ns m 2 )
  8.85  10 8 m
26 25 3 2 1
( 4.65  10 kg)  (2.7  10 m )  ( 4.5  10 ms )

Now, the diameter of a gas molecule can be determined by using the

1
relation  
2 d 2n

and rewriting it as

1/ 2
 1 
d 
 2 n 

On substituting the values of n and , we get

1/ 2
 1 
d  
 1.414  3.14  (2.7  10 25 m 3 )  (8.85  10 8 m) 
 

 3.07  10 10 m

and r  1.53  1010 m

Note that the values of radius calculated in TQ 2 and TQ 4 are slightly

different. These deviations point to limitations of theory, or values of
physical quantities used in two calculations. .

5. Using Eq. (3.17), we can write

f
K n v  kB
6

R
But we know that kB  , therefore we can rewrite the above
NA
expression as

f R
K nv  (i)
6 NA

77
Block 1 Kinetic Theory of Gases
You will recall that molar heat capacity

f
cv  R (ii)
2

Substituting the value of cv in Eq. (i), we get

c  n 
K  v  v 
3  NA 

1 cv
  v (iii)
3 M

Since air is primarily a mixture of two diatomic gases (N2 and O2),

5 5
cV  R   8.31  103 J kmol 1 K 1
2 2

41.55
  20.78 103 Jkmol 1 K 1
2

Now substituting the values of various physical quantities in Eq. (iii), we

get the required result:

1 (1.29 kg m3 )  (20.78  103 Jkmol 1K 1)  (470 ms 1)  (5.6  10 8 m)
K 
3 29 kg kmol 1

 8.1 10 3 Wm 1 K 1

78