Microemulsion Polymerization
M
M
M
M
P•
M
M
PM•
M
M
IM•
M
Eric W. Kaler
M
I•
M
M•
M
M
Department of Chemical Engineering
M
University of Delaware
M
Newark, DE 19716
M
M
M
M
M
Jen O’Donnell
Kevin Hermanson
CarlosMCo (U. Cincinnati)
Renko de Vries (Wageningen U.)
M
Characteristics and Structures
Emulsion:
•Two Phase
•Energy Needed to Form
•Opaque
•Monomer Drops > 1 µm
Microemulsion:
•One Phase
•Spontaneous
•Transparent/
•Swollen Micelles ( D < 20 nm)
Why Study Microemulsion Polymerization?
Produces nanosized (~15 nm) latex
particles smaller than those obtainable
by emulsion polymerization
Polymerization in a confined
environment may lead to unique
polymer morphologies, e.g. tacticity
and knotting.
“Paint” the walls of a
microporous material
“Knotted” polymer
chain in solution
Dry
“Seeds” for emulsion
polymerization
Extremely high MW
( ~20 million daltons)
are readily made
Outline
•
•
•
•
•
•
•
Problems and Model Mixtures
How to Make Microemulsions
Microstructures
Polymerization – Kinetics and Model
Structure Evolution
Multiple Additions
The End!
Ternary Phase Diagrams at 60ºC
Added degree of freedom
from mixing surfactants is
used to tune one phase
oil-in-water microemulsions
Gibbs Phase
Prism at 60ºC
Mixing ratio sets
surfactant
chemical potential
Free Radical Thermal Polymerization
Monomers
n-C6MA
n-C4MA
t-C4MA
Styrene
Water Solubility
60°C (mM)
~ 0.4
3.4
4.3
4.6
Tg (°C)
-5
20
128
106
kp 60°C
(L mol-1 s-1)
995
1015
1140
340
Reaction at 60°C
H2C
CH3
O
O
CH2
H2C
CH2
H2C
CH2
H3C
H2C
CH3
O
O
CH2
H2C
CH2
H3C
H2C
CH3
O
O
H3C C CH3
CH3
HC CH2
A Basic Recipe
• Surfactant mixture to tune phase behavior
Cationic surfactants
DTAB
DDAB
• Monomer with low water solubility - hexylmethacrylate (0.4 mM)
• Polymer with low Tg - polyhexylmethacrylate (-5°C)
• Radical initiator with simple dissociation kinetics - V50 (not persulfates)
A Simple View Of Microemulsion Polymerization
M
Initiation by IM•
M
M
M
M
Chain
Transfer
M
P•
M
M•
M
IM•
M
PM•
M Propagation
M
M
M
M
I•
M
M
M
M
Initiation
by M•
M
M
M
M
# Micelle
~ 103
# Particles
5 nm
V – 50 Polymerizations
1. Rapid polymerization
2. 100% conversions
3. Rate profile parabolic
4. Average maximum
rate at about 39%
conversion
Microemulsion Polymerization – John Morgan
Modeling the V-50 Rate Curves
•
Fundamental rate equation for addition polymerization:
c = monomer concentration at polymerization locus
[ R • ] = concentration of propagating radicals
∂c
−
= k p [ R • ]c
∂t
• Microemulsion conversion form:
∂f
=
∂t
k p N * (t )c( f )
Mo
Mo = monomer concentration at polymerization locus (M)
N* = propagating radical concentration in whole microemulsion (M)
Modeling the V-50 Rate Curves (cont.)
• Assumption: Monomer concentration within polymer
particles given by:
c = co (1 – f )
co = initial concentration of monomer, M
• Entry rate is constant; all radicals remain active
N*(t) = ρot
ρo = rate of radical entry, M s-1
Modeling the V-50 Rate Curves (cont.)
k p N * (t ) c ( f ) k p c o ρ o
∂f
=
=
t (1 − f )
• Rate Equation:
∂t
Mo
Mo
Or
Parameter:
Conversion:
∂f
= At (1 − f )
∂t
k p co ρ o
A=
Mo
1 2
f = 1 − exp(− At )
2
Modeling Implications
Rate Maximum :
f′=
A
e
Conversion at Maximum Rate :
Time of Maximum Rate :
f = 1 − e −0.5 = 0.39
t = A−1/ 2
• Dependence on Initiator Concentration:
Assume
ρo = 2kd [I]
then
A goes as [I]
Measured Rate Constants
– Propagation Rate Constant
• kp = 995 M-1s-1 (Pulsed Laser Polymerization)
– Initiator Decomposition Rate Constant
• kd = 3 x 10-5 M-1s-1 (Literature)
– Initial Monomer Concentration in Droplet
• C0 = 1.0 M (SANS)
– Initial Monomer Concentration in microemulsion
• M0 = 0.257 (Formulation)
N.B. No fitted parameters!
Comparison with Experiment
Quantitative Agreement Through to Full Conversion
Kinetic Modeling: Monomer Concentration at
Locus of Polymerization
Carlos Co, Renko de Vries
How Does the Microstructure Evolve?
Case II
Case III
Flory-Huggins for both
latex particles and micelles.
All monomer is taken up
by latex particles at ~ 5%
conversion.
Flory-Huggins for latex
particles and curvature
energy for micelles.
Monomer partitions
between latex and micelles.
No swelling of polymer
latex. Polymerization
occurs in shell of latex
particles with monomer
concentration equal to
that in the micelles.
Monomer
Concentration
Case I
Conversion
Conversion
Conversion
Monomer concentration is approximately linear with conversion.
Can differentiate only using Small Angle Neutron Scattering (SANS).
Small Angle Neutron Scattering (SANS)
2D Detector
Sample
Neutron Source
q=
λ
4π ⎡ θ ⎤
sin ⎢ ⎥
λ
⎣2⎦
θ
q~
I(q) = n P(q) S(q)
P(q) ⇒
Form
Single particle properties
⇒
Factor
(size, shape, composition)
Structure
⇒
S(q) ⇒
Factor
Relative positions
of particles due to
interactions
1
Length
q
What Can SANS Tell Us?
Particle
Scattering
Observed
Scattering
I(q)
Micelle
Scattering
q
q
q
How does the SANS spectra change as the microstructure
evolves from micelles to a mixture of polymer particles and micelles?
Reactor for Online
Scattering During
Polymerization
Online SANS / Kinetics Experiments
C6MA (DTAB/DDAB)
-1
Intensity (cm )
1000
100
Increasing
Particle
Size and
Number
Density
Decreasing
Micelle Size
0%
10%
38%
65%
85%
96%
100%
10
0.10
0.01
1
-1
q (Å )
Gradual shifts in SANS spectra indicate that monomer partitions
between the micelles and the polymer particles. Case I is incorrect.
Connection of particle size to MWD
• Basic Idea
– Growing chain of L segments was initiated at an
earlier time t1
– At t1, calculate Δt for one propagation event
– Number of chains initiated during Δt is N(L)
– Assume single chain particles
• Final result is analytical
• See Morgan and Kaler, Macromolecules, 1998
Predicted MW and Size Distributions
SANS Model for Online SANS Spectra
Effective HS Interactions
Form Factors
Polymer
Discretize Model-Predicted
Particle Size Distribution
Micelle
Calculate model intensities using Vrij’s analytical equation
Three Adjustable
Parameters
REHS
Rmin
Rmaj
Online SANS Modeling Results
Micelle Dimensions
C6MA (DTAB/DDAB)
(Fitted Parameters)
0%
100
0%
Minor Radius (Å)
29
23
20
Aspect Ratio
2.1
2.3
2.2
Minor/HS
Radius Ratio
1.6
1.7
1.6
65%
100%
10
Particle Size Distribution
(Model Predictions)
0.20
0.10
0.08
0.06
0.04
1
0.02
0%
0.01
-1
Intensity (cm )
1000
65% 100%
65% 100%
Avg Radius (Å)
160
130
Stdev (Å)
50
70
-1
q (Å )
To within the accuracy of the predicted particle size distributions,
the polymer particles are not swollen by monomer.
SANS Model Fitting Results
Particle size distribution
model is consistent with SANS
Micelle size decreases steadily
with increasing conversion
Validation of SANS Swelling Experiments
(n-C4MA)
-1
Intensity (cm )
1000
100
4%
Online
Swelling
33%
Online
Swelling
75%
Online
Swelling
10
1
0.00
0.05
0.10
0.15
-1
q (Å )
cryo-TEM by Stefan Burauer (Universitaet zu Koeln)
Molecular Weight Distribution
Conv. (%) Mw (106) Mw/Mn
4
18.6
1.9
37
26.1
1.2
100
22.4
1.4
By Pat Cotts, Dupont
Molecular Weight Distributions
n-C6MA
n-C4MA
dw/d(log M)
95%
94%
76%
79%
60%
43%
33%
Molecular Weight (106)
GPC/MALLS/RI by Patricia Cotts (DuPont)
Polystyrene Exceeds Chain Transfer Limit
Styrene
kp
dw/d(log M)
91%
PM •
kp/ktr ratio sets characteristic
P•+M
ktr
P+M•
limiting molecular weight
70%
53%
27%
Molecular Weight (106)
Styrene free-radical polymerizations (60°C)
limited to ~2·106 daltons.
MW~15 ·106 polystyrene is consistently
prepared by microemulsion polymerization
Summary: A Simple Model for Microemulsion Polymerization
Rate Equation
∂f
= At (1 − f )
∂t
⎛
t2 ⎞
Conversion : f = 1 − exp⎜⎜1 − A ⎟⎟
2⎠
⎝
k p c oρ o
⎛ 1⎞
Conversion at max rate : f = 1 − exp⎜ − ⎟ = 0.39
⎝ 2⎠
Mo
is known
V50 Concentration
Predictions for Particle Size and Molecular Weight Distributions
dw/d(log M)
A=
Kinetic Predictions
Molecular Weight (106)
Initial
Step 1
1
2
How to Increase Polymer
Loading?
M
M
M
Step 2
Refill
3
…Sequential Addition
Polymerization
4
M
M
Kinetics
•Radicals present from prior monomer additions
N* = ρο t + N*o
Additional Steps
5
∂f k p co (1 − f )
=
2 k d [I ] t + N * o
∂t
Mo
⎛ k pco
f = 1 − exp ⎜⎜ −
k d [I ] t 2 + N *o t
⎝ Mo
(
(
)
⎞
⎟⎟
⎠
)
Determine Cmon from Single Addition Kinetics
C6MA/DTAB/DDAB (5% Total Surfactant)
Single Addition Kinetics
Monomer Partitioning Map
Single Addition Polymerization
(Scaled*)
wt%mon=1.90
1.4
wt%mon=1.43
0.008
0.002
0.8
90
wt%mon=0.475
0.004
1.0
.
=1
wt%mon=0.95
0.006
1.2
on
m
Monomer Concentration
Cmon(M)
0.010
%
wt
d(Conversion)/d(time) (s-1)
0.012
0.6
0.4 Increase Total
Monomer wt%
mon =0
0.2 Concentration5
.47
0.0
0.000
0.0
0.2
0.4
0.6
0.8
1.0
Conversion (f)
*0.95% scaled 1.5X, 1.43% scaled 2X, 1.90% scaled
2.5X
0.0
0.2
0.4
0.6
Conversion
0.8
1.0
Determine Cmon for Multiple Addition Polymerization
Cmon=Co(1-f)
Monomer Partitioning Map
Multiple Addition Polymerization
Monomer Partitioning Map
Single Addition Polymerization
0.475% Monomer
0.475% Polymer
1.4
0.475% Monomer
0.95% Polymer
1.2
1.0
0.475% Monomer
1.43% Polymer
0.8
0.6
0.4
0.2
New Co
0.0
0.0
0.2
0.4
0.6
Conversion (f)
0.8
1.0
Monomer Concentration
Cmon(M)
Monomer Concentration
Cmon(M)
1.4
Addition 1 Co=0.66M
Addition 2 Co=0.56M
Addition 3 Co=0.47M
Addition 4 Co=0.37M
1.2
1.0
0.8
Ad
diti
1
on
0.6
0.4
0.2
Additio
n
4
0.0
0.0
0.2
0.4
0.6
Conversion (f)
0.8
1.0
Multiple Addition Kinetics
Multiple Addition Model
∂f k p c o (1 − f )
=
2 k d [I ] t + N *o
∂t
Mo
(
Measured vs. Predicted Kinetics
Measured Kinetics
0.004
0.003
0.002
Addition 1
Addition 2
Addition 3
Addition 4
0.001
0.000
0.0
0.2
0.4
0.6
Conversion (f)
0.8
(offset*)
Addition 1
0.008
Co decreases
lowering
reaction rate
d(Conversion)/d(time) (s-1)
d(Conversion)/d(time) (s-1)
0.005
)
1.0
Addition 2
0.006
Addition 3
0.004
Addition 4
0.002
0.000
0.0
0.2
0.4
0.6
0.8
1.0
Conversion (f)
The predicted data matched the measured data when N*o=0
* Addition 2 offset 0.001, Addition 3 offset 0.002, Addition 4 offset 0.003
Multiple Addition Particle Size Measurements
Particle Size Distribution
Contin Analysis
Particle Size
Predicted particle size
if no new particles are
formed
80
100
80
Experimentally
Measured Particle Size
60
60
40
40
20
20
QLS Measured Particle Size
Maximum Predicted Particle Size
0
0
0
10
4.6
20
30
Generation Number
8.8
12.6
% Polymer
40
16.2
50
19.4
*100)
100
25 0
3 5
3
Dia 4045
me 50 5
ter 5 60
65
1
12
22
32
42
0
max
Diameter (nm)
120
Intensity (I/I
140
n
itio
d
Ad
After 43 additions latex contains 17% polymer and 4% surfactant
Summary
• Microemulsion polymerization produces small
monodisperse particles
• Initiator charge plays no role (with pH control…)
• Reaction rate and MWD can be modeled with minimal
assumptions and no free parameters
• Microstructures “meter” monomer and control
polymerization
• Commercially interesting concentrations can be
produced by sequential (or continuous) polymerization