BENG 221: Mathematical Methods in Bioengineering

Lecture 11

Heat and Diffusion Equation in Space and Time

References

Haberman APDE, Sec. 11.3.

http://en.wikipedia.org/wiki/Heat_equation
http://en.wikipedia.org/wiki/Diffusion_equation
 Cartesian Box Value Boundary Conditions

h11 = 19.74/L2 h12 = 49.35/L2 h13 = 98.7/L2 h14 = 167.8/L2 h15 = 256.6/L2

h21 = 49.35/L2 h22 = 78.96/L2 h23 = 128.3/L2 h24 = 197.4/L2 h25 = 286.2/L2

h31 = 98.7/L2 h32 = 128.3/L2 h33 = 177.7/L2 h34 = 246.7/L2 h35 = 335.6/L2

h41 = 167.8/L2 h42 = 197.4/L2 h43 = 246.7/L2 h44 = 315.8/L2 h45 = 404.7/L2

h51 = 256.6/L2 h52 = 286.2/L2 h53 = 335.6/L2 h54 = 404.7/L2 h55 = 493.5/L2
 Cartesian Box Flux Boundary Conditions

h00 = 0/L2 h01 = 9.87/L2 h02 = 39.48/L2 h03 = 88.83/L2 h04 = 157.9/L2

h10 = 9.87/L2 h11 = 19.74/L2 h12 = 49.35/L2 h13 = 98.7/L2 h14 = 167.8/L2

h20 = 39.48/L2 h21 = 49.35/L2 h22 = 78.96/L2 h23 = 128.3/L2 h24 = 197.4/L2

h30 = 88.83/L2 h31 = 98.7/L2 h32 = 128.3/L2 h33 = 177.7/L2 h34 = 246.7/L2

h40 = 157.9/L2 h41 = 167.8/L2 h42 = 197.4/L2 h43 = 246.7/L2 h44 = 315.8/L2
BENG 221
M. Intaglietta
Lecture 9
Time dependent solution of the heat/diffusion equation

Derivation of the diffusion equation

The diffusion process is describe empirically from observations and

measurements showing that the flux of the diffusing material Fx in the x direction is
proportional to the negative gradient of the concentration C in the same direction, or:

dC
Fx = − D 1
dx

where D, the diffusion constant, is a coefficient that may be constant, or a function of

time, location and concentration.

With reference to Figure 1, the flux of material through the face of the element of
volume at x, minus the flux through the face at x + dx equals the rate at which the
concentration changes in the volume, assuming that fluxes occur only in the x-direction,
or:

∂Fx ∂C ∂F
Fx − ( Fx + dx) = =− x
∂x ∂t ∂x

Figure 1. Flux balance along the x-direction in a region of space described in Cartesian
coordinates.

This can be readily extended to effects in all directions yielding:

∂C ∂Fx ∂Fy ∂Fz

+ + + =0
∂t ∂x ∂y ∂z

And applying (1) we obtain:

∂C ∂ ⎛ ∂C ⎞ ∂ ⎛ ∂C ⎞ ∂ ⎛ ∂C ⎞
= ⎜D ⎟+ ⎜D ⎟+ ⎜D ⎟ 2
∂t ∂x ⎝ ∂x ⎠ ∂y ⎝ ∂y ⎠ ∂z ⎝ ∂z ⎠

which in the nomenclature of vector analysis is expressed by:

∂C
= div( D grad C ) = D∇ 2C the Laplacian operator for D = constant
∂t

Solution for constant diffusion coefficient from a plane source

Straight forward differentiation shows that:

1 x2
− −
C = At e 2 4 Dt
3

is a solution of:

∂C ∂ 2C
=D 2 4
∂t ∂x

⎛ A −3/2 − x Ax −5/2 ⎞
2 2
A −3/2 − 4xDt Ax −5/2
t e + t = D⎜ t e 4 Dt + t ⎟
2 4 Dt ⎜ 2D 4 D 2
t ⎟
⎝ ⎠

This solution for C is symmetrical relative to x = 0, tends to 0 as x tends to

infinity, and is everywhere zero for t = 0, except for x = 0 where it is infinite. This
solution shows the concentration of the diffusing material originating from a plane source
with an amount of material M at zero time. The diffusing material is not consumed and
the amount of material is constant at all times. To evaluate A we assume that material is
diffusing in an infinite cylinder from a plane located at x = 0. Mass balance requires that
for all times:
∞
M= ∫ Cdx
−∞
5

Changing variables and substituting in 3 and 5:

1
x2 2x x x
ξ =
2
; 2ξ d ξ = dx; 1
dξ = dx; dx = 2( Dt ) d ξ
2
4 Dt 4 Dt 2 ( Dt ) 2 4 Dt

1 ∞ 1 1 ∞ 1
−
∫e ∫ e d ξ = 2 A (π D ) 2
−ξ 2 −ξ
M = At 2( Dt ) d ξ = 2 AD
2
2 2 2

−∞ −∞
and substituting A in 3 we obtain:

M x2
C= exp(− ) 5
4π Dt 4 Dt

In this solution half of the material diffuses in the positive x direction and the
other half in the negative x. This solution is also valid for a semi infinite cylinder where
diffusion takes place in the positive x-direction only from a plane located at x = 0.
Clearly the concentration will be double of that of the infinite cylinder. In this case we
indicate that the solution is reflected at the boundary and superposed. Note that the
gradient of concentration at x = 0 is zero in both cases, indicating that in either case no
material crosses the plane source (or boundary).

Diffusion from a finite region consisting of a volume source

The solution for the diffusion of material occupying a volume in space can be
obtained by assuming that the region is composed of an infinite number of plane sources
and superposing the infinite number of related solutions. This problem describes effects
taking place in an infinite cylinder filled with water, where the concentration of a solute
is C = Co for x < 0, C = 0 for x > 0, t = 0. Consider in the geometry of Figure 2 a plane
of unit surface area containing diffusible material in a quantity Codξi located at ξi
according to 5 will produce a distribution of concentration at any time t given by:

Co d ξi ⎛ ( x − ξi ) 2 ⎞
Ci ( x, t ) = exp ⎜ − ⎟ 6
4π Dt ⎝ 4 Dt ⎠

Figure 2. Material at a concentration Co occupies the region along the negative x-axis.

Therefore the effect due to the infinite number of planes at any given time t is obtained
by adding the effect of each plane solution from 0 to - ∞ or:

0 0
Co ⎛ ( x − ξ )2 ⎞
C ( x, t ) = ∑ Ci = ∫ 4π Dt
exp ⎜ − ⎟ dξ 7
i =−∞ −∞ ⎝ 4 Dt ⎠
and making the substitution of variables:

x −ξ
=η and differentiating dξ = − 4 Dt dη
4 Dt

Changing variables and limits of integration in 7 we obtain:

for ξ = 0 η = x / 4 Dt and for ξ = −∞ η = ∞

Substituting in 7 we obtain:

x / 4 Dt 0 x / 4 Dt
Co Co Co
C ( x, t ) = −
π ∫
−∞
exp(−η 2 )dη = −
π ∫
−∞
exp(−η 2 )dη −
π ∫
0
exp(−η 2 )dη

∞ x / 4 Dt
Co Co
= ∫ exp(−η )dη − ∫ exp(−η 2 )dη
2

π 0 π 0

Co ⎛ π π x ⎞ Co ⎛ x ⎞
= ⎜⎜ − erf ⎟⎟ = ⎜1 − erf ⎟
π⎝ 2 2 4 Dt ⎠ 2 ⎝ 4 Dt ⎠

Note that:
x
2
∫ e dη
−η
erfx =
2

π 0

Diffusion from and in confined regions

The methods described allow to describe diffusion from a substance confined

between –h < x < h along the x-axis. Solution of this problem gives the concentration in
terms of:

1 ⎛ h−x h+ x ⎞
C ( x, t ) = C0 ⎜ erf + erf ⎟
2 ⎝ 4 Dt 4 Dt ⎠

This solution is symmetrical about x = 0 therefore the system can be cut in half,
providing the solution for the semi-infinite system.

One dimensional diffusion from a finite system into a finite system that extends
up to x = l can be analyzed by the method of reflection and superposition, where in this
case the reflection (and superposition) occurs at x = l and x = 0. In this system the
solution reflected at x = l is reflected again at x = 0, at infinitum, resulting in an infinite
series of error functions, namely:

1 ∞
⎛ h + 2nl − x h − 2nl + x ⎞
C ( x, t ) = C0 ∑ ⎜ erf + erf ⎟
2 n =−∞ ⎝ 4 Dt 4 Dt ⎠

This solution is useful for calculating the distribution of concentration at early

times, when the series converges rapidly. A solution of this type can also be obtained
using the Laplace transform.

Non-dimensionalization of the diffusion equation

Given the diffusion equation in one dimension (4) over a one dimensional region
of total length L we introduce a non-dimensional spatial coordinate as the ratio x = x’L
so that the diffusion equation becomes:

∂C ∂ 2C D ∂ 2C
=D =
∂t ∂ ( Lx′)2 L2 ∂x′2

We can also define a non-dimensional time as t = t’t0 where t0 is an arbitrary time scale;
the new equation is:

∂C D ∂ 2C ∂C Dt ∂ 2C
= 2 ∴ = 20
∂ (t0t ′) L ∂x′2 ∂ (t ′) L ∂x′2

Therefore setting t0 = L2/D the diffusion equation becomes:

∂C ∂ 2C
=
∂t ′ ∂x′2

This result indicates that all diffusion problems are the same. This requires
scaling the geometry so that the basic dimension ranges from zero to one. The
combination of the size and the diffusivity yield the appropriate time unit. On the scaled
domain and in the proper time units, problems of different size and diffusion constants
will have the same solution. Thus solving the diffusion equation for one set of boundary
conditions solves it for all cases. As an example the time that it takes for diffusion to
change concentration by a given amount is directly proportional to the size of its principal
dimension. Thus doubling its size quadruples the time.