Title: Diffusion in Solids: Point sources, infinite sinks
1Diffusion in SolidsPoint sources, infinite
sinks
27-216 Spring 2004 A. D. Rollett
2Objectives
- Use case studies to illustrate different types of
solutions. - Study the case of a point source coupled with an
infinite sink for the diffusant. - Study the case of a finite source, coupled with
an infinite sink. - Later lectures finite sources, finite sinks.
3Overview
- It is useful to keep in mind that there are
various characteristic classes of problem. Each
problem class has a generic set of boundary
conditions and a corresponding type of equation
as its solution (generally to Ficks 2nd Law). - 1D steady state linear for 1D Cartesian
logarithmic for 1D radial. - 1D finite point source, infinite sink Gaussian
(exp-x2) - 1D infinite source, infinite sink(s) error
function, erf or complementary error function,
erfc - 1D, finite sink series with both exponential and
cosine functions (in each term).
Examinable
4So far
Examinable
5 Transient Case 1 Finite thin film source
continued
6Case Study doping of semiconductor silicon
In order to make a electronic component, Boron
has to be implanted into Si. In order to do this,
a thin film of Boron is deposited on the surface
of the 2 mm thick silicon wafer as shown below.
The thickness of the film is 5 ?m and Boron
concentration is 105 atoms/m3. Diffusion
coefficient of B in Si is 10-10 m2/s. What
concentration of Boron would there be at a depth
of 1 mm in the Si wafer after 1 hr?
7How to solve the problem?
- In general, one simply looks up the appropriate
solution. However, in this case, we will examine
the solution procedure because it is useful to
know about it. - Then we examine the boundary conditions.
- Finally we can apply the solution to the
particular problem. - As always, beware of units in diffusion problems
here we deal with atom densities.
8Concentration profiles
t0
C(x,t)
x
Boundary Conditions
Examinable
9BCs
Solve through Laplace transformation
Examinable
10At a later time
- Does this solution fulfill our boundary and
initial conditions? - How does the flux change with time and position?
- How does the rate of Accumulation change with
time and position? - Is this solution valid forever?
Examinable
11Boron doping of Si
- First lets define the junction depth as the
diffusion penetration depth for a given amount of
time at which the concentration from diffusion
alone is equivalent to the initial concentration
of the diffusant. - Calculating the junction depth requires an
approximation to the thin film equation in which
we ignore the initial concentration.
- Choosing particular conditions of B in Si PG,
p483 T1150C (1423K) anneal time
2hr7200sfrom Table 13.3, DB0.16
µm2.hr-14.44.10-19m2.s-1Cinitial5.1021atoms.m-
3 C04.5.1026atoms.m-3.
12Boron doping of Si, contd.
Therefore the junction depth, xjunction, 3.19
µm To obtain the actual concentration at this
depth, we have to use the following more exact
equation
Naturally, the surface concentration at any
given time can also be calculated with the same
equation, setting x 0.
13 Deriving one solution from another
14(No Transcript)
15Instantaneous planar diffusion source in an
infinite medium
Initial state
These diffusion problems concern placing a finite
amount of diffusant that spreads into the
adjacent semi-infinite solid.
Final state
16Instantaneous planar diffusion source in an
infinite medium
These diffusion problems concern placing a finite
amount of diffusant that spreads into the
adjacent semi-infinite solid.
17Transform Methods
where
18Laplace Transform of the Diffusion Equation
Linear Diffusion Equation
19Laplace Transforms
The Laplace transform of Ficks second law when
C(x,0)0
20Instantaneous planar diffusion source
Application of the Laplace transform to Ficks
second law gives
The diffusion process is subject to the mass
constraint for a unit area
21Instantaneous planar source
Reduction of the Laplace transform
The general solution for which is
22Instantaneous planar source
Mass constraint for the field
23Instantaneous planar source solution
The transform solution
M units must be compatible with units of C,
taking note of the vDt in the denominator. For
example, M can be mass per unit area (e.g.
Cdepth).
24Normalized plot of the planar source solution
C (x) M length -1
x distance
25Transient Case 2 Large source large
destination
26Follow this outline
- Choose a coordinate system
- Find boundary and initial conditions
- Decide if the problem is transient or steady
state - If transient, solve PDE (Ficks 2nd law)
- Solve analytically
- Find solution in diffusion/heat transfer/math
textbooks - Solve numerically
- In this case, we treat the problem as a series of
thin film sources, all of equal strength, that
are spaced out along the bar. We then superpose
the solutions for each source.
27Superposition of solutions
28DIFFUSION COUPLE IN AN AIRCRAFT
I need to predict how much Ni I am losing from
the superalloy. T 500oC. Assume that properties
of the steel and superalloy are quite similar
with DNi 10-5 cm2/s at 500oC.
29How do we measure a diffusion coefficient?
Answer perform a thin-film diffusion penetration
experiment, measure the concentration profile,
and derive the coefficient by assuming the
solution.
30Thin-film configuration
31Procedure for Analysis of Thin-Film Data
Take logs of both sides of
A plot of lnC versus x2 yields a slope-1/4Dt
32Thin-Film Experiment
Geiger counter data after microtoning 25 slices
from a thin-film specimen in which a radioactive
species was diffused in to the specimen.
33Log Concentration versus x2
Slope-1/4Dt
34Infinite source, infinite sink
- What if we have a material that is exposed to a
source of diffusant in a situation where the
concentration is constant at one surface? - This corresponds to systems where a chemical
reaction at the surface maintains a certain
composition, e.g. carburization of steels, rapid
diffusion into a grain boundary (and then
diffusion into the bulk from the boundary). - This is known as the infinite source- infinite
sink problem, for cases where the penetration
distance into the material is small compared to
the dimensions of the material.
35Initial state
Linear diffusion into a semi-infinite medium
Final state
Examinable
36Boundary Conditions
The diffusion equation is a 2nd-order PDE and
requires two boundary or initial conditions to
obtain a unique solution.
1) Initial state C 0, for x gt 0, t 0.
2) Left-hand boundary At x 0, C0 is
maintained for all t gt 0.
Examinable
37Laplace Transforms
Transform of the boundary condition at the left
hand edge
38Laplace Transforms
The concentration field associated with the image
field is found by inverting the transform either
by formal means, a look-up table, or using a
computer-based mathematics package.
39Estimation of the Error Function
For small arguments
40Estimation of the Error Function
Piecewise approximations for restricted ranges
of the argument
Useful for spreadsheet calculations.
41Error Function
Examinable
42Complementary Error Function
Examinable
43Initial state
Linear diffusion into a semi-infinite medium
Final state
Examinable
44Concentration versus the similarity variable
DimensionlessVariable
Examinable
45 Concentration field versus distance
? 2(Dt)1/2 is the time tag
(Note ? has the units of distance!)
46Diffusion Penetration X
X K t1/2
Relative Conc. C/C0
Relative Conc. C(x)/C0
Distance, x units of ?1
47Penetration versus square-root of time
0.4
0.6
Penetration Distance, X(C/C0)
0.8
0.9
48Summary
Examinable