Definition of Mass Flux:

NA is the number of moles of A that go through a unit area in unit time, the unit area being fixed in
space. The molar flux is related to the concentration by:
NAz = – c DAB + xA(NAz + NBz) --- [1]
for two component systems. (DAB is mass diffusivity) NAz is the z-component of the vector NA. In each of the
problems the relation of NAz / NBz is to be specified on physical grounds to enable elimination of NBz.
Diffusion both in non-reactive and reactive systems can be studied. In the reacting systems one can distinguish
between two types: homogeneous, wherein the chemical species takes place in the entire volume of the fluid
and heterogeneous, in which the chemical change takes place only in a restricted region in the system, such as
at the surface of a catalyst. Apart from the difference in the physical picture between these two situations, even
their description methodology is different. The rate of production by a homogeneous reaction appears in a
source term in he differential equation obtained from the shell mass balances. In the case of heterogeneous
reaction the rate of production appears not in the D.E. but in the boundary condition.
One need to know some information about the rate at which the various species appear or disappear by
reaction in order to set up the problem. For our analysis we shall have to stipulate that the mechanisms of the
reactions are known and that the reaction rates are describable by means of simple functions of the
concentrations of the reacting species.
For homogeneous reactions, the volume rate of production of species A may be given by an expression of
the form
RA = kn’’’ CAn --- [2]
In which [=] moles/cm3. Sec. and CA [=] moles/cm3. The index ‘n’ indicates the order of the reaction. For a I
order reaction k1’’’ [=] sec-1.
For heterogeneous reactions, the rate of reaction at the catalytic surface may be specified by a relation of the
form
NAz |surface = kn’’ CAn|surface --- [3]
In which NAz [=] moles/cm2.Sec. and CA [=]moles/cm3. Here k1’’[=] cm/sec. again for a I order reaction. It is to
be noted that ’’ indicates a rate constant related to a surface source.

SHELL MASS BALANCES AND BOUNDARY CONDITIONS

We solve the diffusion problems presently by making mass balances for a specific chemical species over a
thin shell of solid or fluid. We select an appropriate system and then write down the law of conservation of
mass.

– + = 0 --- [4]

This conservation statement can also be expressed in terms of moles also.

The chemical species A may enter or leave the system by means of diffusion and by virtue of the overall
motion of the fluid. Also species A may be produced or destroyed by homogeneous chemical reactions.
After the balance is made on a shell of a finite thickness, the dimensions of the system are allowed to become
infinitesimally small. The resulting differential equation on solving yields the distribution of species A in the
system. As in the case of momentum transfer and heat transfer the differential equation generates constants of
integration which have to be evaluated by use of boundary conditions.
a] The concentration at a surface can be specified: for example, xA = xA0.
b] The mass flux at a surface can be specified (if the ratio NA /NB is known, this is tantamount to giving the
concentration gradient); for example NA = NA0.
c] If diffusion is occurring in a solid, it may happen that at the solid surface substance A is lost to a surrounding
fluid stream according to the relation: NA0 = kc (CA0 – CAf) --- [5]
in which NA0 is the mass flux at the surface, CA0 is the surface concentration, CAf is the concentration in the
fluid stream and the proportionality constant ‘kc’ is a mass transfer coefficient. Equation [5] is analogous to
Newton’s law of cooling.
d] The rate of chemical reaction at the surface can be specified e.g. if substance ‘A’ disappears at a surface by a
I order chemical reaction, NA0 = k1’’ CA; i.e. the rate of disappearance at a surface is proportional to he surface
concentration, the proportionality constant k1’’ being a I order rate constant.
1
DIFFUSION THROUGH A STAGNANT FILM
Considering the diffusion system as shown in the figure below where liquid A is evaporating into gas B and
imagining that by some device the liquid level at z = z1 is maintained despite the evaporation, we shall tackle
the problem. Right at the liquid-gas interface the gas phase concentration of A corresponding to equilibrium
with the liquid at the interface; i.e. xA1 is the vapour pressure of A divided by the total pressure, (pA / P)
provided that A and B form an ideal gas mixture. Let us also assume that solubility of B in in liquid A is
negligible.
At the top of the tube at (z=z2) a stream of gas mixture of A & B of concentration xA2 flows past slowly.
Thus the mole fraction of A at the top of the column is maintained at x A2. It is also presumed that the entire
system is held at constant temperature and pressure. Also to be assumed is that gases A and b are ideal.
When such an evaporating system attains steady state, there is a net motion of A away from the evaporating
surface and the vapour B is stationary. Thus Equation [1] can be written for this system with NBz = 0 as follows:

NAz = – c DAB + xA NAz --- [6]

Or NAz = DAB --- [7]

A mass balance over the incremental z states that at steady state:

S NAz|z – S NAz|z+z = 0 --- [8]
where S is the cross sectional area of the column. Division by S z and taking limit as z approaches zero, we
get:
– = 0 --- [9]
Substituting equation [7] into equation [9] results in:
--- [10]

2
For ideal gas mixtures at constant pressure and temperature c is constant and DAB is very nearly independent of
concentration. Thus we have: --- [11]

Integration of this II order Differential Equation with respect to z gives:

= C1 ---[12]

The second integration yields:

– Ln (1 – xA ) = C1 z + C2 --- [13]
The two boundary conditions for evaluating the two integration constants are:
BC 1: at z = z1 ; xA = xA1 --- [14]
BC 2: at z = z2 ; xA = xA2 --- [15]
On evaluating the integration constants by applying the boundary conditions and substituting the the values for
the constants in equation [13] we get

= --- [16]

or = --- [17]

The constants if integration are determined as:

C1 = = ; C2 =
To get equation [16] first we have to rewrite C2 thus:
C2 = = –

Equations [16] & [17] are the expressions for the concentration profiles drawn it can be seen that slope (dx A/
dz) is not constant with respect to z even though the molar flux NAz is constant:

Average concentration = --- [18a] Expressing  =

= = |01 -- [18]

Whence:
xBavg = --- [19]

similarly the rate of mass transfer at the liquid gas interface (.e. the net rate of evaporation) is obtained by using
equation [7].
NAZ|z=z1 = – DAB |z=z1 = DAB |z=z1 = DAB Ln --[20]
Equations [19] & [20] can be combined to give:
NAZ|z=z1 = DAB --- [21]
(xA1 – xA2) is the characteristic driving force. It can also be expressed in terms of total and partial pressure as
follows:
NAZ|z=z1 = (p DAB / RT) Ln = (p DAB / RT) --- [22]

3
DIFFUSION WITH HETEROGENEOUS CHEMICAL REACTION
Let us consider a catalytic reactor in which a dimerization reaction 2A  A2 is carried out. This is such
a complex system that it cannot be described exactly by any theoretical development.
Analysis made of a highly simplified model can provide some information about the behavior of the system.
E.g. e may consider that each catalyst particle is surrounded by a stagnant gas film through which A has to
diffuse in order to arrive at the catalytic surface. We consider that the reaction 2A  A2 occurs instantaneously
at the catalyst surface, and that the product A2 then diffuses back out through the gas film to the main turbulent
gas stream composed of A and A2. Our interest is to get an expression for the local rate of conversion from A to
A2 when the effective gas-film thickness  and gas stream compositions xA0 and xA20 are known. Let us assume
that the gas film is isothermal despite the fact that in many catalytic reactions the heat generated by the reaction
cannot be neglected.
For the situation as described in the figure, there will be one mole of A2 moving in the negative z-direction for
every two moles of A moving in the +ve z-direction. Thus we know that at steady state:
– (1/2) NAZ = NA2Z --- [23] at any value of z.
We can substitute this relation in equation [1] and solve it for NAZ to give:
NAZ = – DAA2 --- [24]

Thus the law of diffusion and the stoichiometry of the reaction have led to to an expression for NAZ in terms
of the concentration gradient.

Let us now make a mass balance on species ‘A’ over a thin slab of the gas film of thickness z. The
procedure is exactly the same used in connection with equation [8] which results into:

4
 = 0 --- [25] ; Insertion of expression for NAz by equation [24] into [25] gives

= 0 --- [26]

Twice integration yields: – 2 ln = C1z + C2 --- [27]

The constants of integration are then evaluated by using the boundary conditions:
BC 1: at z =0; xA = xAo --- [28] and BC 2: at z = ; xA = 0 --- [29] to yield:

= --- [30]
For the concentration profile in the gas film we can get the molar flux through the film from equation [24]

NAZ = 2c DAA2 ln --- [31]

This quantity is also the local rate of Dimerization per unit area of catalyst surface. Even though the reaction of
A A2 is instantaneous at the catalyst surface, it can proceed only at a finite rate because of the diffusion
process which is ‘in series’ with the reaction process and thus we consider that the conversion of A to A 2 is
“Diffusion Controlled.”

DIFFUSION WITH HOMOGENEOUS CHEMICAL REACTION

Presently another system in which gas ‘A’ dissolves in liquid ‘B’ and diffuses into the liquid phase is
considered for setting up a mass balance. As the gas diffuses, it also undergoes an irreversible first order
chemical reaction: A + B  AB. Hence the mass balance (input – output + production = 0) will have the form:
NAZ|z S – NAZ|z+z S – k1’’’CA Sz = 0 --- [32]
in which k1’’’ is a first order rate constant for the chemical decomposition of A and S is the cross sectional area
of the liquid. The quantity k1’’’CA represents the moles of A disappearing per unit volume per unit time.

Dividing by Sz and taking the limit as z tends to zero gives:

+ k1’’’ CA = 0 --- [33]
If A and AB are present in small concentrations then one can write Equation [1] in an approximate form as
follows:
NAZ = – DAB --- [34]

5
Equations [33] and [34] combine to give: – DAB + K1’’’CA --- [35]
The boundary conditions for solving this II order Differential Equation are:
BC 1: at z = 0; CA = CA0 --- [36] & BC 2: at z = L; NAz = 0 or = 0 --- [37]
The BC1 states that the surface concentration is maintained at a fixed value CA0. BC2 states that no A diffuses
through the bottom of the container. The solution by using these boundary conditions can be obtained as:

= --- [38] In which b1 =

The average concentration of A in the liquid phase is obtained as:

= = =

= tanh (b1) --- [39] where  = [ 1 – ]

The molar flux of A at plane z=0 can also be computed from equation [38]
NAz |z=0 = – DAB |z=0

= – DAB CA0 |z=0 = + DAB CA0 tanhb1 --- [40]

In which b1 again is b1 = .
Our assumptions have been that A is present in small concentrations and that AB (the product) does not
interfere with the diffusion of A through B.

THE EQUATIONS OF CHANGE FOR ISOTHERMAL SYSTEMS

[1] Equation of Continuity; [2] Equation of Motion

If we have to report the concentration of fish in a river, the fish are moving and hence the fish concentration
‘C’ will be a function of position (x, y, z) and time (t).
The Partial time Derivative:
Suppose a person stands upon the bridge on the river and notes the concentration of fish just below him. The
concentration of fish just below him changes with time. He will be observing as to how the concentration
changes with time at a fixed position in space.
Thus by we mean the partial of ‘C’ with respect to t, holding x, y and z constant.

Total Time Derivative :

Let us suppose that the observer chooses to get into a motor boat and speeds around on the river, at times
going upstream, sometimes across the current and perhaps sometimes downstream. If the observer has to report
the change of fish concentration with respect to time, the numbers reported must also reflect the motion of the
boat. The total time derivative is given by:
= + + +

in which , and are the x, y and z components of the velocity of the boat respectively.

Substantial time Derivative :

Let us suppose that the observer, feeling a little lazy, gets into a canoe (Parrisal) and decides to float along
counting the fish. Presently the velocity of the observer is just the same as the speed or velocity of the stream
‘v’. If he reports the change of fish concentration with respect to time, the numbers depend on the local stream

6
velocity. This derivative is a special kind of total time derivative and is called the ‘Substantial Time Derivative’
or (more logically) sometimes the derivative following the motion. It is related to the total time derivative as
follows:
= + vx + vy + vz
in which, vx, vy and vz are the x, y, and z components of the local fluid velocity v.
It is essential to thoroughly master the physical meaning of these three derivatives. It is to be specifically
remembered that is the derivative at a fixed point in space and is a derivative computed by an
observer floating downstream with the fluid.

The Equation of Continuity:

Let us begin by considering the pair of faces  (perpendicular) to x-axis. The rate of mass in through the face at
x is ( vx )x y z and the rate of mass out through the face at x+x is ( vx )x+x y z . Similar expressions
can be written for the other two pairs of faces. The rate of mass accumulation within the volume element is (x
y z) (d/dt).
The mass balance thus becomes:
(x y z) = yz[( vx)x - (vx)x+x] +xz[(vy)y - (vy)y+y]
+ y x [(vz)z - (vz)z+z] --- [2]
Let us divide both the sides of this equation by x y z and take limits as these dimensions approach zero, we
get
= -{ (vx)+ (vy) + (vz)} --- [3]
This is the equation of continuity, which describes the rate of change of density at a fixed point resulting
from the changes in the mass velocity vector (v). In the vector notation the equation reads as follows:
= - {.(v)} --- [4] {  has the dimensions of reciprocal length}
{.(v)} is called the divergence of (v) sometimes written as div(v).
It is to be noted that the vector v is the mass flux and its divergence has a simple significance that it is the
net rate of mass efflux per unit volume. Then equation [4] simply states that the rate of (change of) increase of
the density within a small volume element fixed in space is equal to the net rate of mass influx to the element
divided by its volume. Equation [3] is generally modified by expanding the RHS derivatives t and rearranging
the terms to get:
+ vx + vy + vz = - -- [5]

7
The LHS is the Substantial time derivative of density, i. e. the time derivative for a path following the fluid
motion. Hence Eqn.[5] may be abbreviated as:

= -  (.v) --- [6]

The operator is as described by substantial time derivative. In this form the equation of continuity describes
the rate of change of density as seen by an observer “floating along” with the fluid. It is to be remembered that
this equation in any form is simply a statement of conservation of mass. For incompressible fluids the equation
of continuity takes the special form:
(.v) = 0 --- [7]

THE EQUATION OF MOTION

Let us consider a three dimensional volume element same as the one considered in respect of the derivation of
equation of continuity.
As in the case of Equation of continuity we have to write the momentum balance for the volume element x y
z. The form of the momentum balance equation is:
= – + --[1]
This equation can be noted to be just an extension to the unsteady state systems of the momentum balance
equation studied for the steady state systems.
– + =0
In laminar flow situations only one directional flow was considered. Herein, in addition to considering
unsteady state behavior, the fluid is considered to move through all six faces of our volume element in any
arbitrary direction. Eqn.[1] above is vector equation with components in each of the three coordinate directions
x, y and z. For simplicity the x-component of each term in Eqn.[1] will be considered. The y- and z-
components can be handled analogously.
Let us consider the rates of flow of the x-component of momentum into and out of the volume element as
depicted in the figure below. Momentum flows into and out of the volume element x y z by two
mechanisms: by convection (i. e. by virtue of the bulk fluid flow and by molecular transfer (i.e. by virtue of the
velocity gradients).

8
 Let us consider the rates of flow of the x-component of momentum into and out of the volume element as
depicted in the figure above. Momentum flows into and out of the volume element x y z by two
mechanisms: by convection (i. e. by virtue of the bulk fluid flow and by molecular transfer (i.e. by virtue of the
velocity gradients).
The rate at which the x-component of the momentum enters the face at x by convection is: [(vx vx)x]yz ,
and
the rate at which it leaves at x+x is [(vx vx)x+x]y z
the rate at which it enters at y, is [(vy vx)y] x z
the rate at which it leaves at y+y is [(vy vx)y+y]x z
the rate at which it enters at z, is [(vz vx)z]
the rate at which it leaves at z+z is [(vz vx)z+z]x y
We see that convective flow of x-momentum must be considered across all six faces and hence the net
convective x-momentum flow into the volume element is:
yz{vxvx)x – vxvx)x+x} + xz{vyvx)y – vyvx)y+y} +
yx{vzvx)z– vzvx)z+z} ---[2]
Similarly the rate at which the x-component of momentum enters the face at x- by molecular transport is:
yz[xx)x] and
the rate at which it leaves at x+x, is: yz[xx) x+x]
the rate at which it enters the face at y, is: xz[yx)y]
the rate at which it leaves at y+y, is: xz[yx) y+y]
9
Similar expressions for the other two faces can also be written.(To keep in mind is
the fact that yx is the flux of x-momentum through a face  to the y-axis.)
The other expressions are: xy[zx)z] and xy[zx) z+z]
The six contributions can be summed up to write:
yz[xxx - xxx+x ] + xz[yxy - yxy+y] + yx[zxz - zx z+z] ---[3]
These momentum fluxes can be considered stresses. Thus xx is the normal stress on the x-face and yx is the x-
directed tangential (shear) stress on the y-face resulting from viscous forces.
In most of the cases the only important forces will be those arising from the fluid pressure ‘p’ and
gravitational force per unit mass ‘g’. Hence, clearly the resultant of these forces in the x-direction will be:
yz[px - px+x ] +  gx x yz --- [4]
We always define the pressure in a moving fluid by the equation of state:
p= p( , T) and it is a scalar quantity.
Finally, the rate of accumulation of x-momentum within the element is:
x yz --- [4a]
All the above expressions can be inserted into the general momentum balance Eqn.[1]. We will then have:
yz{vxvx)x–vxvx)x+x} + xz{vyvx)y–vyvx)y+y} +
yx{vzvx)z–vzvx)z+z} + yz [xxx - xxx+x]+xz [yxy - yxy+y] + yx[zxz - zx z+z]+ yz[px -
px+x ] +gx x yz =x yz
On dividing by x yz and taking the limit as x, y and z approach zero, the x-component of the
equation of motion is arrived at as below:
=– – + .gx –

--- [5]
Likewise the y and z components which can be obtained in similar manner will be:

=– – + .gy –

--- [6]

=– – + .gz –

--- [7]
The quantities vx, vy and vz are the components of the mass velocity vector v; similarly gx, gy and gz
are the components of the gravitational acceleration g. Further , and are the components of a
vector p, known as the ‘ gradient of p ’ (sometimes written as grad p). The terms vxvx , vxvy , vxvz , etc. are
the (9) nine components of the convective momentum flux ‘vv ’ which is the “dyadic product” of v and v.
Similarly the terms xx , xy , xz , yx , yy , yz , zx, zy, zz are the nine components of ( ), known as the
“Stress Tensor” Since Eqns.[5,6 and 7] occupy so much space. It is convenient to combine them to give a single
vector equation:
[(v)/t] = - [.vv] - p - [.] + g --- [8]

It is to be noted thoroughly (cautiously) that [.vv] and [.] are not simple divergences because of the
tensorial nature of ‘vv’ and ‘’. Though the physical interpretation is analogous to that of [.v] in respect of
Equation of Continuity; it is to be understood that while [.v] represents the rate of loss of mass (a scalar) per
unit volume by fluid flow, the quantity [.vv] represents the rate of loss of momentum (a vector) per unit
volume by fluid flow.
The x-component of the equation of motion as expressed by Eqn.[5] can be rearranged with the help of the
equation of continuity as follows;
10
 =– – + .gx –

--- [5]

– + .gx – = +

---[5a]

– + .gx – = + vx + vx + vx +

vy + vx + vz + vx ---[5b]

– + .gx – =

+ vx [5c]

The equation of continuity says: =0 ---[5d]

By the definition of substantial time derivative:
=
Thus Eqn.[5c] reduces to
 =– + . gx – --- [9]
Similar rearrangement of equations [6] & [7] can be made to obtain
 =– + . gy – --- [9a]

 =– + . gz – --- [9b]
When all the three components of the momentum balance equations 9,9a & 9b are added together we get:

 = – +  [gx + gy + gz]

– --- [9c]

It is to be noted that these additions have vectorial significance and we write them in the vectorial notation as
below:
 = - p + g - [.] ----[10]

In this form the equation of motion states that a small volume element moving with the fluid is accelerated
because of the forces acting upon it. In other words, this is a statement of Newton’s Law of motion in the form:
Mass x acceleration = Sum of forces. Thus it can be seen that momentum balance is completely equivalent to
Newton’s second law of motion.
Equation[8]
= - [.vv] - p - [.] + g --- [8]

11
corresponds to the equation of continuity as given by: = -{.(v)}
represents a balance over a volume element fixed in space and Equation[10]
 = - p +  g - [.] -----[10]
corresponds to the equation of continuity as given by: = -  (.v)
represents a description of the changes taking place in an element following the fluid motion. It is to be kept in
mind that all the equations from [5] to [10] are valid for any continuous medium.
To use these equations for the determination of velocity distributions for different flow situations, one
should insert expressions for the various stresses in terms of velocity gradients and fluid properties. For
Newtonian fluids these expressions are: xx , xy , xz , yx , yy , yz , zx, zy, zz

xx = - 2 + (2/3)  (.v) --- [11]

yy = - 2 + (2/3)  (.v) ---[12]

zz = - 2 + (2/3)  (.v) ---[13]

xy = yx = -  ---[14]

yz = zy = -  --- [15]

zx = xz = -  ---[16]

We shall stipulate these expressions since the reasoning involved in arriving at them is quite lengthy. They
constitute a more general statement of Newton’s law of viscosity than that is used by us earlier in the flow
(viscous) problems. They apply to complex flow situations with the fluid flowing in all directions. As in the
earlier example, if the fluid flows in the x-direction between two plates  to y-direction so that vx is a function
of y alone, then the set of 6 equations 12 to 13 gives:
xx = xz = yy = yz = zz = 0 and yx = -  (vx/x)
This is the same as the viscosity defined then. Substituting Eqns. 11 to 16 into Eqns. 9, 9a and 9b
(corresponding y- and z- equations) gives the general equations of motion for a Newtonian fluid with varying
density and viscosity.
 =– + .gx + +

 +  --- [17]

 = – + .gy +  +

+  --- [18]

 = – + .gz + +

12
  + --- [19]

These equations along with eqn. of continuity, the eqn. of state p= p(), the density dependence of viscosity
 =  () and the boundary and initial conditions determine completely the pressure, density and viscosity
components in a flowing isothermal fluid.
These equations are rarely used in their complete form to set up flow problems. Usually restricted forms of
the equations of motion are used for convenience.
1] For constant  and  the equation of continuity deduces to (.v) = 0 and the equations 17, 18 and 19 (the x,
y and z components) of equation of motion simplify to give:

(Dv/Dt) = - p + g +  2v] -----[20]

The operator 2 = + + is called the Laplace Operator. The

Cartesian components of this equation [20] will be:

 =– +.gx +

 ---[A]

 =– +.gy +

 ---[B]

 =– + .gz +

 ---[C]

Equation[20] above is the well known Navier-Stokes equation (Nah-Vyay].

2] For [.] = 0; Equation [10] reduces to (Dv/Dt) = - p + g ---[21]

This Eqn.[21] is the famous “Euler’s” equation (first derived in 1755). It has been widely used for describing
flow systems in which viscous effects are relatively unimportant.

13