Class Announcements

• Assignments
– Required reading: Callister chapter 6
– Homework 5 due Thursday, Nov. 7

• Tuesday, Nov 5
– Stanford holiday to encourage student voting: Democracy Day
– Nov 5 during our normal lecture time will be an optional review
for test 2
– Review will be held on Zoom (not in person) and recorded

• Last lecture:
– Phase diagrams help us predict which phases will form and what
their compositions will be

• This lecture:
– How do materials change their phases?
– What processes have to take place?
– How quickly do these processes occur?
What processes are occurring?
• Kinetics: the study of dynamic
processes that occur over time

a phase: 15 wt% Sn, 85 wt% Pb

a phase: 5 wt% Sn, 95 wt% Pb

b phase: 98 wt% Sn
Tin (Sn) atoms 2 wt% Pb
Lead (Pb) atoms
What processes are occurring?
1. Diffusion of atoms (mass transport)
2. Cooling of atoms (heat transport)
3. Formation of an interface

a phase: 15 wt% Sn, 85 wt% Pb

a phase: 5 wt% Sn, 95 wt% Pb

b phase: 98 wt% Sn
Tin (Sn) atoms 2 wt% Pb
Lead (Pb) atoms
 Outline for Mass Transport Discussion
1. Diffusion of atoms (mass transport)
2. Cooling of atoms (heat transport)
3. Formation of an interface

• How does diffusion occur?

• Why is diffusion important during

processing?

• How can we predict the diffusion

rate for some simple cases?

• How does diffusion depend on

temperature and structure?
 Diffusion
• What is it?
– The spontaneous movement of molecules

• What causes it?

– Thermal energy resulting in molecule movement

• Demonstration:
– Iodine (red) reacts with starch to change color (blue)

Iodine solution
Ziploc bag
Starch solution

http://youtube.com/watch?v=wO1yq6YTJDI
 Class Discussion
• What variables affect diffusion?

• Demonstration:
– Iodine (red) reacts with starch to change color (blue)

Iodine solution
Ziploc bag
Starch solution

http://youtube.com/watch?v=wO1yq6YTJDI
 Diffusion
• What variables affect diffusion?
– Temperature: higher temperature = more diffusion
– Viscosity of the solvent: higher viscosity = less diffusion
– Molecular weight: higher molecular weight = less diffusion
– Microstructure of the polymer:
diffusion through amorphous polymers is faster than crystalline
– Particle-polymer interactions:
if a particle can interact with the polymer (e.g. through
secondary interactions), then the diffusion will be slower

• Each particle-solvent interaction or particle-material

interaction will have a different diffusivity
 Diffusion Through a Solid
• Interdiffusion: In an alloy, atoms tend to migrate
from regions of large concentration.
Initially After some time
 Diffusion Through a Solid
• Self-diffusion: In an elemental solid, atoms
also migrate.

Label some atoms After some time

 Diffusion Mechanisms
Substitutional Diffusion:
• Applies to substitutional impurities
• Atoms exchange with vacancies
• rate depends on:
--number of vacancies
--temperature
--activation energy to exchange:
structure of lattice, size of impurity
 Diffusion Mechanisms
• Interstitial diffusion
• More rapid than vacancy diffusion
• Rate depends on:
--number of interstitial impurities
--temperature
--activation energy to exchange:
structure of lattice, size of impurity
 Processing Example using Diffusion

• Case Hardening:
--Diffuse carbon atoms
into the host iron atoms
at the surface.
--Example of interstitial
diffusion is a case
hardened gear.

• Result: The "Case" is

--hard to deform: C atoms
"lock" planes from shearing.
--hard to crack: C atoms put
the surface in compression.
 Processing Example using Diffusion
• Doping Silicon with Aluminum for semiconductors:
• Silicon: 4 valence e-, Aluminum: 3 valence e-
• Process:
1. Deposit Al rich
layers on surface.

2. Heat it.
3. Result: Doped
semiconductor
regions.
 Diffusion Measurements
• Diffusion is measured by molar flux, J
J = M/At where J: molar flux [mol/m2s]
M: moles diffusing [mol]
A: cross-sectional area [m2]
t: time [s]

Note: J can also be

written in terms of
mass flux, where M
is the mass diffusing
 Measuring Diffusion
• Flux in differential form:

• Directional Quantity

• Flux can be measured for:

--vacancies
--host (A) atoms
--impurity (B) atoms
 Flux and Concentration Profiles
• Concentration Profile, C(x): [kg/m3]

• Fick's First Law:

• The steeper the concentration profile (dC/dx),

the greater the flux (J)!
 Steady State Diffusion
• Steady State: the concentration profile doesn't
change with time.

• Apply Fick's First Law:

• If Jx(left) = Jx(right), then

• Result: the slope, dC/dx, must be constant

(i.e., slope doesn't vary with position)
 Steady State Diffusion
• Steel plate at
Steady state, so
700C with concentration
geometry c1
c2 profile is linear
shown:
Carbon rich At 5 mm:
c1 = 1.2 kg/m3

At 10 mm:
c2 = 0.8 kg/m3

x1 x2 D = 3x10-11 m2/s
• Q: How much
carbon transfers
from the rich to
the deficient side?
 Steady State Diffusion
How many carbon
atoms pass through a Steady state, so
concentration
1 cm2 plate in 1 ms? c1 profile is linear
c2
Carbon rich At 5 mm:
c1 = 1.2 kg/m3

At 10 mm:
c2 = 0.8 kg/m3

x1 x2 D = 3x10-11 m2/s

flux of atoms = J NA/MW = (2.4x10-9 kg/m2s)(6.02x1023 atoms/mol)/(12x10-3kg/mol)

flux of atoms = 1.2x1017 atoms/m2s

Number of atoms = flux of atoms*time*area = (1.2x1017 atoms/m2s)(1x10-3 s)(1x10-4 m2)

Number of atoms = 1.2x1010 atoms
 Steady State Diffusion Example
Micro-capillaries restrict fluid
Source channel motion, but allow stable
diffusion (not drawn to scale)
1 mm
Cell culture chamber
4 mm

500 µm Sink channel

Top View
Microfluidic Gradient Generator
• Maintains a stable concentration gradient for multiple days
• Cell culture chamber is shear-free (no bulk fluid motion)
• Allows exact quantification of the concentration gradient
Shamloo A, Ma N, Poo MM, Sohn L, Heilshorn SC. Endothelial cell polarization
and chemotaxis in a microfluidic device. Lab on a Chip, 2008, 8:1292-1299.
 Quantification of Concentration Gradient
• Computer simulation of device geometry using diffusion equations
• Comparison to experimental observations of gradients using
fluorescently-tagged molecules
• Validation of gradient with both small and large molecules
– Large proteins: bovine serum albumin, MW ~ 66 kD, Diffusivity ~ 3x10-7 cm2/s
– Small molecules: cAMP analogues, MW ~ 1.16 kD, Diffusivity ~ 3x10-6 cm2/s

Increasing concentration
 Steady State Diffusion Application
• Chemotaxis: Migration towards (or away from) a higher
concentration of a soluble stimulus (e.g. growth factor)
• Critical during tissue development and remodeling
• Example: New blood vessel growth
– Required for cancer tumor growth
– Required for tissue regeneration like skin or bone healing

Basement membrane Endothelial cell

Davis GE, Senger DR. Circ. Res. 2005. 97:1093.

 Endothelial Cell Chemotaxis
• The first step in new blood vessel
growth is endothelial cell chemotaxis
• Endothelial cells chemotax towards a
source of vascular endothelial growth
factor (VEGF)

*
*

*
*
N = 5 independent trials.

Shamloo A, Ma N, Poo MM, Sohn L, Heilshorn SC. Endothelial cell polarization

and chemotaxis in a microfluidic device. Lab on a Chip, 2008, 8:1292-1299.
 Diffusivity
• Diffusion through solids is a thermally activated process:
D = D0 exp (-Qd/RT) where D0: pre-exponential
Qd: activation energy

How can we find D0 and Qd from experimental data of D and T?

ln D = ln D0 - Qd/RT
y = b +mx where y: ln D
b: ln D0
m: -Qd/R
x: 1/T
 Diffusivity
• Activation energy will depend on:
– Lattice structure
– Size mismatch

ln D = ln D0 - Qd/RT
y = b +mx

where y: ln D
b: ln D0
m: -Qd/R
x: 1/T

• What about ceramics?

 Diffusivity
• Activation energy will depend on:
– Lattice structure
– Size mismatch

• What about ceramics?

– Charge neutrality must
be preserved
– Movement of an ion must
be paired with another
vacancy/impurity/electron
– Diffusion is often much
slower than in metals
 Predicting Diffusion by Comparing Materials

• Which of these materials would have faster diffusion:

– Open crystal structures or close-packed crystal structures

– Higher melting temperature or lower melting temperature materials

– Larger diffusing atoms or smaller diffusing atoms

– Cations or anions

– Lower density materials or higher density materials

 Effects of Structure on Diffusion

Diffusion FASTER for... Diffusion SLOWER for...

• open crystal structures • close-packed structures

• lower melting T materials • higher melting T materials

• smaller diffusing atoms • larger diffusing atoms

• cations • anions

• lower density materials • higher density materials

 Diffusivity in Polymers
• Example: Packaging material is often made from thin polymer films
with thickness L; e.g., a bag of chips, sandwich bag, or drug packaging
• Assume concentration outside and inside the package is constant
– Typically true soon after the product was manufactured
J = -D (c2 - c1)/(x2 - x1) = -D (c2 - c1)/L

Often rewritten in terms of partial pressures when talking about gases:

V/A = PM(p2 - p1)/L where V: volume flow rate of gas (m3/s)
PM: permeability (m4/sN)
A: cross-sectional area (m2)
pi: partial pressure of gas (N/m2)
 Steady State Diffusion
• Low-density polyethylene (LDPE) and
polystyrene (PS) are being considered to
package a pharmaceutical product that
degrades in oxygen.
• Both polymer films are 0.15 mm thick
• Partial pressure of O2 in the air is 0.21 atm
• Inside the package, partial pressure of O2 is
0.01 atm
• Which material should we use?

Material O2 PM (10-16 m4/sN) H2O PM (10-16 m4/sN)

LDPE 0.45 - 1.5 0.97 - 3.8

PS 0.023 - 0.088 18 - 25
 Steady State Diffusion
• Low-density polyethylene (LDPE) and
polystyrene (PS) are being considered to
package a pharmaceutical product that
degrades in oxygen.
• Both polymer films are 0.15 mm thick
• Partial pressure of O2 in the air is 0.21 atm
• Inside the package, partial pressure of O2 is
0.01 atm
• What if the drug was water sensitive?

Material O2 PM (10-16 m4/sN) H2O PM (10-16 m4/sN)

LDPE 0.45 - 1.5 0.97 - 3.8

PS 0.023 - 0.088 18 - 25
 Steady State Diffusion
• Steady State: the concentration profile doesn't
change with time.

• Apply Fick's First Law: = constant

• When is steady state not a good assumption?

 Steady State Diffusion
• Steady State: the concentration profile doesn't
change with time.

• Apply Fick's First Law: = constant

• When is steady state not a good assumption?

Non-steady state diffusion:

Any case where the concentration changes over time.
Jx ≠ constant; cannot use Fick’s First Law
 Non-Steady State Diffusion Example
• Diffusion calculations are commonly used in drug delivery products
• For a traditional pill, the drug concentration within the blood stream over time
is a “burst” profile (i.e., non-steady state diffusion)
• Goal: Deliver a sustained concentration of the drug
– MEC: minimum effective concentration
– MTC: minimum toxic concentration

MTC

MEC
 Non-Steady State Diffusion Example
• What are some ways to deliver drugs to a person besides a pill?
 Non-Steady State Diffusion Example
• What are some ways to deliver drugs to a person besides a pill?
 Non-Steady State Diffusion
• Drug eluting stents

New England Journal of Medicine

March 8, 2007
Special issue on drug-eluting stents

For more information:

http://www.ptca.org/des.html
 Non-Steady State Diffusion
• Concentration profile,
C(x), changes
with time

• To conserve matter: • Fick's First Law:

• Governing Equation, Fick’s 2nd Law:

 Non-Steady State Diffusion
• Concentration profile,
C(x), changes
with time

• To conserve matter: • Fick's First Law:

• Governing Equation, Fick’s 2nd Law:

 Non-Steady State Diffusion
• Governing Equation, Fick’s 2nd Law:

This is a second-order differential equation.

There are a variety of mathematical methods to solve this equation,

which is beyond the scope of this class.

Our textbook gives a mathematical solution to this equation for one

special case of one-dimensional, non-steady state diffusion:

At time=0, the concentration at the surface (Cs) is constant.

Cs
C0

At time=0, the concentration in the material (C0) is the

same everywhere (all values of x have the same C0).
 Non-Steady State Diffusion
• Governing Equation, Fick’s 2nd Law:

This is a second-order differential equation.

There are a variety of mathematical methods to solve this equation,

which is beyond the scope of this class.

Our textbook gives a mathematical solution to this equation for one

special case of one-dimensional, non-steady state diffusion:

At all times, the concentration at the surface (Cs) is constant.

Cs
C0

At all times, the concentration in the material very far

away from the surface (xè∞)is still C0.
 Non-Steady State Diffusion
• Copper diffuses into a bar of aluminum

• Solution for this

specific geometry:
2√Dt
"error function"
Values tabulated in Callister table
 Non-Steady State Diffusion
• Copper diffuses into a bar of aluminum
• 10 hours at 600°C gives desired C(x)
• How many hours would it take to get the same C(x)
if we processed at 500°C?

2√Dt
 Non-Steady State Diffusion
• Copper diffuses into a bar of aluminum
• 10 hours at 600°C gives desired C(x)
• How many hours would it take to get the same C(x)
if we processed at 500°C?
Key point 1: C(x,t500°C) = C(x,t600°C)
Key point 2: Both cases have the same Co and Cs

(Dt)500 = (Dt)600
2√Dt

Dt should be constant!
 Non-Steady State Diffusion
• Copper diffuses into a bar of aluminum
• 10 hours at 600°C gives desired C(x)
• How many hours would it take to get the same C(x)
if we processed at 500°C?
Key point 1: C(x,t500°C) = C(x,t600°C)
Key point 2: Both cases have the same Co and Cs

(Dt)500 = (Dt)600
2√Dt

• Answer: t500 = (Dt)600 (5.3x10 -13 m2/s)(10 hr)

= = 110 hr
D500 (4.8x10 m /s)
-14 2
 Non-Steady State Diffusion
• Transdermal patch -
sustained drug delivery
through the skin
• When the patch is first
applied, can assume 100%
of the drug is outside skin
(Cs = concentration of drug
in the patch) and 0% of the
drug is inside skin (C0 = 0)

Photo from MACtac.com

 Lecture 9 Summary
• Things you should now be able to do…
Calculate the flux of a material
Quantitatively describe 1-dimensional steady state diffusion
Qualitatively predict diffusivity for different pairs of materials
Determine the diffusive activation energy from experimental data
Quantitatively describe a special case of non-steady state diffusion

• Assignment
– Required reading: Callister chapter 6
– Homework set 5 due Thursday, Nov 7