Scribd
Upload
65 views

Learn How To Set Up A Problem With Differential Balance: P.C. Chau (UCSD, 1999)

This document discusses setting up differential balances to solve transport problems. It begins by summarizing the key is to associate the physics of a problem with the mathematics. It then works through examples of: 1) Steady state diffusion in a planar geometry, deriving equation 5. 2) Transient diffusion in a planar geometry with chemical reaction, deriving equation 7. 3) Steady state diffusion in a spherical geometry, deriving equation 12. 4) Transient diffusion in a spherical geometry with chemical reaction, deriving equation 14. It concludes by transforming equation 14 using a new variable to simplify it, resulting in equation 16.

Uploaded by

tarhuni
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
65 views

Learn How To Set Up A Problem With Differential Balance: P.C. Chau (UCSD, 1999)

This document discusses setting up differential balances to solve transport problems. It begins by summarizing the key is to associate the physics of a problem with the mathematics. It then works through examples of: 1) Steady state diffusion in a planar geometry, deriving equation 5. 2) Transient diffusion in a planar geometry with chemical reaction, deriving equation 7. 3) Steady state diffusion in a spherical geometry, deriving equation 12. 4) Transient diffusion in a spherical geometry with chemical reaction, deriving equation 14. It concludes by transforming equation 14 using a new variable to simplify it, resulting in equation 16.

Uploaded by

tarhuni
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

P.C.

Chau (UCSD, 1999)

Learn how to set up a problem with differential balance


A puzzling experience that we have from teaching is that many students can derive (or rather
memorize?) the general balance equations with very rigorous steps, but they cannot solve a simple
problem. The route to continuity via the transport theorem is mathematically elegant, but most
beginners to the subject tend to be overwhelmed by the math and lose sight of the physics.
Sometimes, this difficulty is compounded by instructors who encourage the approach of taking a
general equation and throwing away the unnecessary terms.
The key is that we have to learn how to associate the physics of a problem with the
mathematics. This is a point that we easily miss if we are preoccupied with the mathematics. We
may avoid this pitfall if we are competent in setting up a problem with the most basic tool. This
is what many texts call "doing the differential balance or the shell balance." We should appreciate
that many problems, including those that you may encounter in real life, can be tackled this way.
In these notes, we will stick with a general statement, and go through the derivation steps,
which are repetitive, but hopefully youd find instructive. Afterward, you should be much better
prepared to tackle the actual examples in the text.
___________________________
A. Steady state diffusion in a planer geometry
Let's consider the scenario that
some real problem can be modeled as
dV = Adz, rA
steady state one-dimensional diffusion.
We identify a differential volume (the
d
DAB d CA
DAB CA
so-called shell) dV over the distance dz
dz z
d z z + dz
(Fig. 1). The cross-sectional area is
denoted by A. (Warning: we are also
using A to denote the chemical
species.) For the moment, let's consider
z
z + dz
that there is no chemical reaction and
Figure 1. Differential volume in planer geometry.
thus rA = 0.
We assume that species A is dilute and there is no convection. With Fick's law, the diffusive
flux of species A at constant temperature and pressure is

J A = D AB

dCA
dz

(1)

where DAB is the binary diffusion coefficient and CA is the concentration of species A.
At steady state, the transport of A into the differential volume at the plane z must be matched
by the transport of A exiting at z+dz:

JA z = JA

(2)

z + dz

How can we rewrite this equation in a more usable form? Without any fancy mathematics, we
can simply use a first term Taylor approximation: 1

JA

z + dz

dJ
J A z + dzA dz
z

(3)

which we substitute back in Eq. (2) to obtain

dJ A
dz = 0
1

(4)

We can just as well take the expansion of the flux definition in Eq. (1), and write
dC
dC
d 2 CA
D AB A
D AB A D AB
dz
dz z + dz
dz z
dz 2 z
1

P.C. Chau (UCSD, 1999)

This is a simple statement saying that the flux of A is a constant in the z direction. No surprise
here. The result is general in the z direction since our choice of this location "z" at the beginning
is arbitrary. If we now substitute Eq. (1) in (4), we obtain

d 2 CA
=0
dz 2

(5)

where we have taken the diffusion coefficient to be a constant. We gather that in this simple case,
the solution CA(z), the so-called concentration profile, is linear with respect to z.

B. Transient diffusion in a planer geometry with chemical reaction


We now repeat the exercise without the restrictions that we have imposed on the previous
step. With a transient problem, we may accumulate (or deplete) species A in the differential
volume. The differential balance of species A is now

(Adz)

CA
= A JA z A JA
t

Accumulation

z + dz

Net transport
(in out)

+ r A(Adz)

(6)

Source/Sink
due to reaction

A few words on unit are necessary here. If we use cgs units, the molar concentration CA is
gmol/cm3. The differential volume Adz is cm3, and every term in the balance equation has the
units of gmol/s. To have consistent units (and quantities), we must use the rate of transport on the
right side, hence the multiplication of the flux by the cross-sectional area.
Finally, the rate of chemical reaction is usually given in volumetric units of gmol.cm3s1.
Thus we have to multiply it by the differential volume. In many illustrative problems, the species
A is a reactant and the chemical reaction is first order, in which case rA = kCA, where k is the first
order rate constant.
Next, we can substitute for the diffusive flux at z+dz as in Eq. (4) to obtain

J
CA
= A + rA
t
z
or more commonly with further substitution of Eq. (1) to arrive at

C A
t

= D AB

CA
z 2

+ rA

(7)

In Chapter 2, we look at simple steady state problems without the transient and chemical
reaction terms. In Chapter 3, we include chemical reactions. In Chapter 4, we look at problems
where we have the transient and diffusion terms, but without chemical reaction. Of course, there
are more complex problems in which we need all the terms.
___________________________

P.C. Chau (UCSD, 1999)

C. Steady state diffusion in a spherical geometry


We now repeat Step A with diffusion in a
problem which requires the use of spherical
geometry. The radial diffusive flux of species A at
constant temperature and pressure is

dC
J A = D AB drA

dV = 4r2 dz, rA

NA r + dr
NA r

(8)
r

where the flux notation JA can be more formally


written as the radial component JA,r.
In this case, the differential volume is a very
"thin" spherical shell. At steady state, the transport of
A into the differential volume at r must be matched
by the transport of A exiting r+dr:

(4r 2 J A) r = (4r 2 J A)

r + dr

Figure 2. Differential volume in


spherical geometry. (Notation is drawn
for the general case, so NA is JA in
these examples.)

(9)

r + dr

It is important to note that the rate of transport is flux times the area, which in this case varies
with the radial position. In other words, the radial diffusive flux JA,r itself cannot be a constant in
this case. We can easily miss this point in Step A where both the cross-section and the diffusive
flux are constants.
Next, we once again do the first term Taylor approximation:

(4r 2 J A)

r + dr

d (4r 2 J ) dr
(4r 2 J A) r + dr
A
r

(10)

which we substitute back in Eq. (9) to obtain

d 2
2
dr (r J A) = 0 , or r J A = constant

(11)

This is a statement saying that the rate of transport of A is a constant in the r-direction. (Note
again: the radial flux is not constant.) If we now substitute Eq. (8) in (11), we obtain

d r 2 dC A = 0
dr
dr

(12)

D. Transient diffusion in a spherical geometry with chemical reaction


To repeat Step B with spherical geometry, the equivalent form of Eq. (6) is now

(4r 2 dr)

CA
= (4r 2 J A) r (4r 2 J A)
t

r + dr

+ r A(4r 2 dr)

(13)

The substitution of Eq. (10) and cancellation of terms lead to

CA

= 12 (r 2 J A) + r A
t
r r
A final substitution of Eq. (8) results in

C A
D 2 C A
= AB
(r
) + rA
t
r
r 2 r

(14)

You should by now acquire the confidence that you can derive your own balance equations and
eliminate the reliance on the simple-minded table-lookup style of learning.
3

P.C. Chau (UCSD, 1999)

E. A simple transformation to simplify Eq. (14)


We can define a new variable such that
u = rCA

(15)

With freshmen calculus, we have

dC A d u 1 du u
= r =r
,
dr
dr
dr r 2

or r 2

du
dC A
= r
u
dr
dr

and
2
d (r 2 dC A ) = d (r du u) = r d u
dr dr
dr
dr
dr 2

Now, if the reaction rate term in Eq. (14) is first order such that rA = kCA, and we substitute the
calculus steps into (14), we have

2
= D AB d u2 ku
dr
t

(16)

The point is that problems in spherical geometry could be transformed to planar geometry and
this move simplifies the equation solving. You will find the use of Eq. (15) in several instances in
later chapters.
___________________________
3. You take over!
Well, you can't possibly expect us to spoonfeed you everything. Here are a few exercises that
you should do before you read the text.
Derive the general solutions to Eqs. (5) and (12). They should help to derive specific
solutions to a given problem with appropriate boundary conditions.
Repeat Steps C and D with cylindrical coordinates. Sovle for the general concentration profile
too.

You might also like