Mass transfer

Introduction
Mass transfer as the term implies deals with transport of material. A simple example
of mass transfer is the movement of a scent from one end of a room to the other. Mass
transfer basically deals with transport of species:
within a medium.
across an interface, i.e. from one medium to another.
The medium could be stationary or mobile. There are two types of mass transfer:
1. Purely diffusive mass transfer (molecular diffusion)
2. Convective mass transfer (occurs in moving fluids)
Molecular diffusion is governed by a random walk process and involves the transport
of molecules from a region of high concentration to one where its concentration is
lower. Steady state molecular diffusion of species A (e.g. sucrose) in a diffusion
medium B (e.g. water) can be expressed by Fick's first law (see Fig. 3.1):

The flux is the amount or concentration of solute carried by a fluid past a plane
perpendicular to direction of flow or velocity, and having a 1 cm2 of surface area. JA
unit is mol/cm2 s.

Our study of the molecular diffusion type of mass transfer will cover the following
three cases:

1. Molecular diffusion in liquid medium

Molecular diffusion in the context of bioseparations mainly involves the transport of
dissolved species in a liquid medium. An example of this is the transport of an
antibiotic from an aqueous solution to the surface of an ion-exchange resin. Gaseous
diffusion is far less important in bioseparation except in very specific types of
separation e.g. freeze drying, pervaporation and molecular distillation. Molecular
diffusion in liquid medium is significantly slower than that in a gaseous medium and
hence the diffusion coefficients are significantly lower.

If we consider a simple case of molecular diffusion e.g. that of an amino acid (A) in
water (B), the flux of the A from point 1 to 2 in the liquid medium is given by (see
Fig. 3.3):
Hence:

This is referred to as equimolar counter-diffusion. This means that the molar flux of
the amino in a certain direction is matched by the molar flux of water in the opposite
direction.

Estimation of diffusivity
The diffusivity of a solute in a liquid medium at a particular temperature can be
estimated using different mathematical correlations. These correlations link diffusivity
to solute and liquid properties such as molar volume, molecular weight and liquid
viscosity. The three most widely used correlations are:
1. Stokes-Einstein correlation:

VA = solute specific molar volume at its normal boiling point (m3/kg mol)
2. Wilke-Chang Correlation:

Where
= association parameter of solvent (equals 2.6 for water)
MB = molecular weight of liquid medium
3.

Polson Correlation:

Where MA is the molecular weight of solute.

Example
Estimate the diffusivity of the protein lysozyme in water at 25 degrees celsius.
Solution
The diffusivity of a solute can be calculated from its molecular weight using Polson
correlation above. From table, the molecular weight of lysozyme is 14,100 kg/kgmole. The viscosity of water at 25 degrees centigrade is 0.001 kg/m s. Therefore:

2. Diffusion of solutes in a dense solid

Solute molecules can diffuse through a dense solid medium after dissolving in it. An
example of this is the diffusion of solutes through dense membranes. The molecules
of the solid medium do not counter-diffuse because of their limited mobility.
However, Fick's law can still be used to describe the diffusion of solute molecules in a
solid medium. An example of this is the pervaporation separation of dissolved volatile
solutes from solvents.
3. Diffusion of solutes in a porous solid
Solute molecules can diffuse through the pores present in porous solids. In order for
this to happen, the pores have to be filled with some liquid medium. Therefore no
diffusion takes place through the solid material itself. All it does is hold the liquid
medium in place. However, the solid material can have an influence on the diffusion
within the liquid medium. It can for instance increase the effective diffusion path
length of the solute if the pores are tortuous in nature. When the pores have dimension
of the same order of magnitude as the solute, the pore wall can cause hindrance to
diffusion. An example of un-hindered diffusion in a porous medium is the transport of
sodium chloride through a microfiltration membrane (which has micron sized pores)
while an example of hindered diffusion is the transport of albumin through an
ultrafiltration membrane (which has nanometer sized pores). The steady state equation
for unhindered diffusion of a solute from point 1 to 2 within a slab of porous solid is
given by:

Where

The hindered diffusion of a solute through a porous solid from point 1 to 2 is given
by:

Where Deff = effective hindered diffusivity (m2/s).

The effective hindered diffusivity of a solute in a pore can be obtained by:

Where
ds = solute diameter (m)
dp = pore diameter (m)
It should be appreciated that when the pore size is very large as compared to particle
or molecular size, the term in the parenthesis will essentially equal 1, and thus:
Deff = D
Which is of course the case for unhindered diffusion.
Example
Glucose is diffusing at 25 degrees centigrade in water within a porous medium having
a porosity of 0.5, tortuosity of 1.8 and average pore diameter of 8.6x 10-3 microns.
Determine the steady state flux of glucose between two points within the medium
separated by a distance of 1 mm and having concentrations 1.5 g/1 and 1.51 g/1
respectively.
Solution
Table 2.2, gives the molecular weight of glucose as being 180 kg/kgmole. The
diffusivity of glucose at 25 degrees centigrade can be determined using Poison
correlation:

The size of glucose can be obtained from Table 2.1. The effective diffusivity of
glucose in the porous structure can be obtained using the equation:

The steady state flux of glucose can be obtained using equation:

The molar concentration of glucose = (0.01/180) mol/L

Convective mass transfer

Convective transport occurs when a constituent of the fluid is carried along with the
fluid. The amount carried past a plane of unit area perpendicular to the velocity (the
flux) is the product of the velocity and the concentration

NA = kcA(C0 CA), or
NA = kcACA
Where
NA is the flux in convective mass transfer
kcA is the convective mass transfer coefficient
(C0 CA) is the solute concentration difference between two points
Convective mass transfer is observed in flowing fluids can be emphasized for two
situations:
1. Transport of a solute in a liquid flowing past a solid surface (see Fig. 3.5)
2. transport of a solute in a liquid flowing past another immiscible liquid (see
Fig. 3.6).
An example of the first type is the transfer of urea from blood towards the surface of a
dialyser membrane in haemodialysis.

An example of the second type is the transfer of penicillin G within filtered aqueous
media flowing past an organic solvent in a liquid-liquid extractor.

When a liquid flows past a solid surface a stagnant boundary liquid layer is formed
close to the surface. Similarly when two liquids flow past one another, two boundary
liquid layers are generated on both sides of the interface.

Within these boundary layers, the transport of solute mainly takes place by molecular
diffusion. If the flow of liquid is laminar, the transfer of solute in the directions
indicated in Figs. 3.5 and 3.6 would be by molecular diffusion. However, if the flow
were turbulent in nature, mass transfer would take place by a combination of
molecular diffusion and eddy diffusion. This is referred to as convective mass
transfer.
Another argument of convective mass transfer can be mentioned for a solute band
progresses along a column, where the solute molecules are continually transferring
from the mobile phase into the stationary phase and back from the stationary phase
into the mobile phase. This transfer process is not instantaneous, because a finite time
is required for the molecules to traverse (by diffusion) through the mobile phase in
order to reach, and enter the stationary phase. Thus, those molecules close to the
stationary phase will enter it almost immediately, whereas those molecules some
distance away from the stationary phase will find their way to it a significant interval
of time later. However, as the mobile phase is moving, during this time interval while
they are diffusing towards the stationary phase boundary, they will be swept along the
column and thus dispersed away from those molecules that were close and entered it
rapidly. The dispersion resulting from the resistance to mass transfer in the mobile
phase is depicted in the figure below. The diagram shows 6 solute molecules in the
mobile phase and those closest to the surface (1 and 2) enter the stationary phase
immediately. During the period, while molecules 3 and 4 diffuse through the mobile
phase to the interface, the mobile phase moves on. Thus, when molecules 3 and 4
reach the interface they will enter the stationary phase some distance ahead of the first

two. Finally, while molecules 5 and 6 diffuse to the interface the mobile phase moves
even further down the column until molecules 5 and 6 enter the stationary phase
further ahead of molecules 3 and 4. Thus, the 6 molecules, originally relatively close
together, are now spread out in the stationary phase. This explanation is a little oversimplified, but gives a correct description of the mechanism of mass transfer
dispersion.

The same phenomenon occurs when solutes transfer between two immiscible moving
liquids in contact with one another.

Experimental determination of mass transfer coefficient

1. In flowing liquids
Mass transfer coefficient of a solute in a flowing liquid can be determined by carrying
out steady-state experiments based on the equation:

Where
NA = molar flux of solute A
m = amount of solute transferred (kg-moles)
a = mass transfer area (m2)
t = transfer time (s)
Also we have:
NA = kcA(C0 CA), or
NA = kcACA
Therefore, the mass transfer coefficient can be obtained from the relation:

The general approach in these experiments is to measure the amount of solute

transferred in a given period of time across a given surface area for a particular
concentration difference across the transfer zone. Solute mass transfer coefficient in a
liquid flowing within a tube can be determined by coating the solute over the inner
wall of the tube followed by measurement of the amount of solute removed.
Example
Water is flowing past one of the rectangular sides of a slab of benzoic acid. The
surface area exposed to water is 0.01 m2 and it is estimated that in 600 seconds, 5x10-4
kg of benzoic acid is lost from the slab by dissolution. If the molecular weight of
benzoic acid is 121.1 kg/kg-mol, calculate its average molar flux. If the solubility of
benzoic acid in water is 0.2 kg/m3, what is its mass transfer coefficient? Assume that
the concentration of benzoic acid in the bulk flowing water is negligible.
Solution
The amount of solute transferred in 600 s is:

Assuming that the benzoic acid concentration on the surface of the slab is the same as
its solubility, the concentration difference across the transfer zone is calculated and
inserted in the equation:

2. Mass transfer between phases (equilibrium state)

Instead of a fluid in contact with a solid, suppose we now consider two immiscible
fluids, designated 1 and 2, in contact with each other. If fluid 1 has dissolved in it a
substance A that is also soluble in fluid 2, then as soon as the two fluids are brought
together, substance A will begin to diffuse into fluid 2. As long as the two phases
remain in contact, the transport of A will continue until a condition of equilibrium is
reached.

10

In the system of fluids 1 and 2 with solute A, CA1 and CA2 are the bulk phase
concentrations in fluids 1 and 2, respectively, CAi is the concentration of A at the
interface, and NA is the molar flux of A. For steady state conditions, we can define the
flux of A as: NA = Kc (CA1 CA2)
Where
kc = individual mass transfer coefficient defined in terms of the concentration
difference in a single phase.
Kc = overall mass transfer coefficient defined in terms of the overall difference in
composition.
Since:
NA = kc1(CA1 CAi)
NA = kc2(CAi CA2)
Kc is defined by the equation:
Kc = (1/kc1) + (1/kc2)
This equation can be rewritten as:
Kc = kc1*kc2/(kc1+kc2)
Substituting for kc1 and kc2 from NA = kc1(CA1 CAi) and NA = kc2(CAi CA2) in the
equation of Kc above, we get:
NA = Kc (CA1 CA2)
Unsteady state mass transfer
In many separations, particularly in rate processes, complete equilibration does not
really take place, and the concentrations do change with time. Therefore, it is
important to understand what happens in an unsteady state mass transfer process. The
nature of the process is such that no identifiable equilibrium is reached. An example
of this is haemodialysis where maintenance of a concentration gradient throughout the
process is essential. Such processes are referred to as non-equilibrium or rate
processes
Using the following equation the solute concentration at any point x, at any time t, can
be calculated. The flux of solute across the interface at any time t is given by:

11