REVERSE-SELECTIVE GAS-SEPARATION MEMBRANES PREPARED BY ATOM
TRANSFER RADICAL POLYMERIZATION
By
Xiaojie Dong

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Chemistry
2012

ABSTRACT
REVERSE-SELECTIVE GAS-SEPARATION MEMBRANES PREPARED BY ATOM
TRANSFER RADICAL POLYMERIZATION
By
Xiaojie Dong

About 95% of the H2 synthesized in the United States is produced by steam reforming of
hydrocarbons followed by the water gas shift reaction. However, this process gives ~25
mol% CO2 as a byproduct, and the energy and capital-intensive pressure swing
adsorption (PSA) procedures for CO2 removal exhibit some unavoidable H2 loss.
Membrane separations, especially those with reverse-selective membranes that
selectively permeate CO2 from H2 streams, are a promising alternative to pressure swing
adsorption. The purified H2 on the high-pressure feed side of the membrane could
directly go to storage and transportation without re-pressurizing.
This research aims to develop reverse-selective membranes containing thin poly(ethylene
oxide) (PEO)-based polymer films grown from porous substrates via surface-initiated
atom transfer radical polymerization (ATRP). PEO has an excellent CO2 solubility, but
crystallization of PEO chains leads to low CO2 permeability and minimal CO2/H2
selectivity. To prevent crystallization, we copolymerized poly(ethylene glycol)methyl
ether methacrylate (PEGMEMA) monomers containing PEO side chains with 23-24
(PEGMEMA-1100) and 8-9 (PEGMEMA-475) PEO unites. The shorter PEO chains

prevent crystallization, and the copolymer membranes still exhibit a CO2/H2 selectivity
of 12 with a CO2 permeability of about 20 Barrers.

Cross-linking of poly(PEGMEMA) films may slightly decrease CO2 permeability and
CO2/H2 selectivity, but it should also enhance the membrane’s chemical and physical
durability. Cross-linked copolymer films prepared by polymerization of PEGMEMA1100, PEGMEMA-475 and poly(ethylene glycol)diacrylate (PEGDA-700) on RC
membranes showed CO2/H2 selectivities ranging from 6.5 to 19.9, but a CO2
permeability of 5-15 Barrer. Efforts to increase permeability included embedding SiO2
nanoparticles into the cross-linked poly(PEGMEMA-1100-co-PEGMEMA-475-coPEGDA-700) film to increase fractional free volume (FFV).

Unfortunately,

nanoparticles didn’t enhance the CO2 permeability, perhaps because the non-rigid chains.
During the course of membrane preparation, growth of poly(PEGMEMA-1100-coPEGMEMA-475) films in water resulted in an exceptionally rapid polymerization (200
μm-thick films in just 30 min of polymerization). The remarkable thickness does not stem
from precipitation, but seems to arise from a unique increase in catalyst activity in the
presence of water and PEO side chains.
Future work should include further examination of the kinetics of the very rapid
polymerization as well as studies of nanoparticle-containing membranes made with
glassy, reverse-selective polymers. The nanoparticles embedded in glassy polymer should
give more FFV, and finally increase the permeability and selectivity of the membrane.

To My Husband

iv 
 

ACKNOWEDGEMENT

Through the entire time that I spend in graduate school here in Michigan State University,
the Department of Chemistry, I am fortunate to encounter some great people who helped
me and worked with me professionally.
I would like to express my gratitude to my advisor Dr. Merlin L. Bruening, who was an
example of a dedicated scholar, a humors teacher and a down to earth researcher. I have
learned countless lessons from him, and for that I thank him. Dr. Gregory L. Baker
offered me great advices on polymerization, and he offered me encouragement when my
research is going not so well. More importantly, I admire his dedication toward a good
cup of coffee, which he brews from beans to start. I would also like to thank Dr.
Volodymyr V. Tarabara and Dr. Gary Blanchard for their suggestions, support and
guidance. I am very fortunate to have them on my guidance committee. Dr. Kathy
Severin also desderved my sincere thanks, she spend a lot of time helped me with the gas
chromatography separation system. I would also like to thank Dr. Stanley Flegler and the
entire staff of the Center for Advanced Microscopy, their assistance and friendship
means more than I could express.
I thank every member from Dr. Bruening’s lab that I am fortunate to encounter in the
past a few years. My friends Dr. Wenjing Wang, Yong Guan, Dr. Li Huang, Hui Zhao,
who offered me a great deal of confidence and motivation, not to mention they all helped
when my research reaches out to the area of organic chemistry.

v 
 

Last but not least, I would like to thank my parents, my in-law parents, and my brother
for their support. I love you all and your support carried me through the toughest times.
My husband is the person that stayed late with me in the lab for countless nights, and his
love is the greatest thing in my life. I thank him for every achievement I have ever made
and will make in the future.

vi 
 

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. x
LIST OF FIGURES ............................................................................................................ xi
LIST OF SCHEMES ........................................................................................................ xvi
LIST OF ABBREVATIONS .......................................................................................... xvii
Chapter 1. Introduction and Background ...................................................................... 1
1.1. Hydrogen as an Energy Carrier. ..................................................................................1
1.2. Hydrogen Purification Techniques..............................................................................3
1.3. Membrane Separations ................................................................................................7
1.4. H2 Separation Membranes and the Solution-diffusion Model . ..................................9
1.5. Solubility of Gases in Polymers and PEO-based Materials as Reverse-selective
Membranes. ......................................................................................................................14
1.6. Growth of Polymer Brushes by SI-ATRP. ................................................................18
1.6.1. Overview of Strategies for Brush Growth..............................................................18
1.6.2. ATRP Mechanism ..................................................................................................20
1.6.3. ATRP Catalyst........................................................................................................22
1.6.4. Ligands for ATRP Catalyst ....................................................................................22
1.6.5. ATRP Initiators ......................................................................................................23
1.6.6. ATRP Monomers ...................................................................................................25
1.6.7. Substrats for SI-ATRP ...........................................................................................28
1.6.7.1. SI-ATRP from Silicon Oxide ..............................................................................28
1.6.7.2. SI-ATRP from Gold ............................................................................................35
1.6.7.3. SI-ATRP from Polymeric Substrates ..................................................................35
1.7. Outline of This Dissertation ......................................................................................37
References ........................................................................................................................39

Chapter 2. Incorporation of Long and Short PEO side chains in polymer brushes to
prevent crystallization and provide CO2/H2 selectivity .............................................. 52
2.1. Introduction. ..............................................................................................................52
2.2. Experimental .............................................................................................................57
2.2.1. Chemicals and Materials. .......................................................................................57
2.2.2. Polymerization .......................................................................................................60
vii 
 

2.2.3. Characterization Methods ......................................................................................61
2.2.4. Gas Permeation ......................................................................................................62
2.3. Results and Disussion................................................................................................63
2.3.1. Polymerization Rates as a Function of Monomer Composition. ...........................63
2.3.2. Determining the Composition of Copolymer Films ...............................................66
2.3.3. Crystallization of PEG Side Chains .......................................................................75
2.3.4. Formation of Gas-Separation Membranes by ATRP .............................................77
2.3.5. Gas Permeation through Poly(PEGMEMA) Membranes ......................................78
2.4. Summary. ..................................................................................................................85
2.5. Acknowledgements & Copyright ..............................................................................85
References ........................................................................................................................86

Chapter 3. Coating of Regenerated Cellulose Membranes for CO2/H2 Separations 91
3.1. Introduction. ..............................................................................................................91
3.2. Experimental .............................................................................................................95
3.2.1. Chemicals and Materials ........................................................................................95
3.2.2. Polymerization .......................................................................................................96
3.2.2.1. Initiator Attachment ............................................................................................96
3.2.2.2. Catalyst Solution .................................................................................................98
3.2.2.3. ATRP ...................................................................................................................99
3.2.3. Characterization Methods ....................................................................................100
3.2.4. Gas Sepearation ....................................................................................................100
3.3. Results and Discussion. ...........................................................................................104
3.3.1. FTIR Studies of Cross-linked Copolymer Films..................................................104
3.3.1.1. Reflectance FTIR Spectra of Cross-linked Copolymer Brushes on a Au-coated
Silicon Wafer..................................................................................................................104
3.3.1.2. ATR-FTIR Study of Cross-linked Copolymer Grafted from RC .....................104
3.3.2. SEM Characterization of RC Membranes and Cross-linked Polymer Films ......107
3.3. Gas Separations .......................................................................................................109
3.4. Conclusion ...............................................................................................................112
References ......................................................................................................................113
Chapter 4. Synthesis of Nano-composite Membranes by ATRP and Investigation of
CO2/H2 Separation ........................................................................................................ 116
4.1. Introduction.. ...........................................................................................................116
4.1.1. Fraction of Free Volume Impact on Selectivity and Permeability. ......................116
4.1.2. Mixed Matrix Membranes for Gas Separation .....................................................118
4.2. Experimental ...........................................................................................................120
4.2.1. Chemicals and Materials ......................................................................................120
4.2.2. Polymerization . ...................................................................................................122
4.2.2.1. Initiator Attachment ..........................................................................................122
4.2.2.2. Catalyst Solution. ..............................................................................................123
4.2.2.3. ATRP. ................................................................................................................124
4.2.3. Characterization Methods and Gas Separation Tests ...........................................125
viii 
 

4.3. Results and Discussion ............................................................................................126
4.3.1. SEM Characterization of Nanoparticle-containing Membranes ..........................126
4.3.2. Gas Separations with Nanoparticle-containing Membranes ................................127
4.3.2.1. Effect of Nanoparticle Loading on CO2/H2 Selectivity and Permeability. ......127
4.3.2.2. Influence of Nanoparticle Initiator Density on Gas Permeation Through
Nanoparticle-containing Films .......................................................................................134
4.4. Conclusions. ............................................................................................................137
References. .....................................................................................................................139
 

Chapter 5. An Investigation of Rapid Surface-initiated Polymerization ................. 144
5.1. Introduction.. ...........................................................................................................144
5.1.1. Growth of Thick Polymer Films from Porous Alumina. .....................................144
5.1.2. Aqueous ATRP ....................................................................................................147
5.2. Experimental... ........................................................................................................149
5.2.1. Chemicals and Materials. .....................................................................................149
5.2.2. Polymerization .....................................................................................................150
5.2.2.1. Initiator Attachment ..........................................................................................150
5.2.2.2. Catalyst Solution ...............................................................................................150
5.2.2.3. SI-ATRP ............................................................................................................151
5.2.2.3. Solution ATRP ..................................................................................................151
5.2.3. Characterization Methods ....................................................................................152
5.3. Results and Discussion ............................................................................................153
5.3.1. Does Precipitation Increase Film Thickness ........................................................153
5.3.2. The Effect of Monomer Concentration on Film Thickness .................................157
5.3.3 In-solution Polymerization of PEGMEMA...........................................................157
5.4. Conclusion ...............................................................................................................164
References ......................................................................................................................165

Chapter 6. Summary and Future Work...................................................................... 168
References ......................................................................................................................170

 

 

ix 
 

LIST OF TABLES

Table 1.1. Physical properties of H2 and CO2.. ...............................................................10
Table 1.2. Initiators employed for growing polymer brushes from silicon oxide... .........29
Table 2.1. Absorption coefficients, k, of homopolymer films. The k values were
calculated using equations 2.5-2.10 and the peak absorbances corresponding to C-O-C
and C=O stretches. The uncertainty is one standard deviation of the values obtained
from films with a range of thicknesses.............................................................................72
Table 2.2. Composition of poly(PEGMEMA-475-co-PEGMEMA-1100) films as a
function of the PEGMEMA-475 mole percentages in the monomer solution... ..............74
Table 5.1. Characterization of the polymer resulting from copolymerization of
PEGMEMA-1100 and PEGMEMA-475 initiated by CH3(OCH2CH2)3OCOC(CH3)2Br
The reaction was catalyzed by [CuCl]- 2 mM, [CuBr2]- 0.67 mM and [HMTETA]- 6
mM, in water at room temperature. Each monomer initially was present at 0.34 M. The
conversions are calculated via integration of the NMR signals from the –CH3 group on
the polymethacrylate back bone and the vinyl protons of the unreacted monomer.. .....160
Table 5.2. Different conditions for PEGMEMA polymerization initiated by
CH3(OCH2CH2)3OCOC(CH3)2Br in solution. Reactions 1 and 2 employed
PEGMEMA-475 as the monomer, but reaction 3 used equimolar PEGMEMA-475 and
PEGMEMA-1100 with the total monomer concentration listed... .................................162

x 
 

LIST OF FIGURES

Figure 1.1. Schematic illustration of one separation cycle in a PSA vessel. .....................5
Figure 1.2. Conceptual drawing of membrane separation, which involves selective
transport through a membrane between two phases...........................................................6
Figure 1.3. Two representative categories of membrane-based separation of CO2 and H2.
The reverse-selective membrane relies on solubility differences between the gases.........9
Figure 1.4. Illustration of the solution-diffusion model for gas transport across a
membrane. The dashed line represents the concentration profile of a gas transporting
through the membrane. Cm,1,A and Cm,2,A are the concentrations of a gas in the
membrane at the feed side surface and permeate side surface, respectively....................12
Figure 1.5. Illustration of the lattice positions of a polymer segment and solvent. Black
dots shows polymer segments that have same size as solvent molecules (hollow dots). 15
Figure 1.6. "Grafting from" (left) and “grafting to” (right) methods for the generation of
polymer brushes from surface. “i” represents initiator molecules. . ..............................19
Figure 1.7. The mechanism of ATRP.. ............................................................................21
Figure 2.1. Proton NMR spectra of (A) PEGMEMA-300, (B) PEGMEMA-475, and (C)
PEGMEMA-1100. The integrals of the peaks with a chemical shift between 3.4 ppm
and 4.5 ppm (arising from the four hydrogen atoms on the ethylene oxide repeat units)
were summed and divided by 4 to give the average number of EO units per monomer.
(The intensity of the alkene protons was set to 1 for normalization.) Peak areas in these
spectra were averaged with areas from two additional replicate spectra to generate
values for the number of average ethylene oxide units per molecule... ...........................58
Figure 2.2. Evolution of film thickness with polymerization time for pure and mixed
monomer solutions. For copolymers, the legend shows the solution PEGMEMA-475
mole percentage with respect to the total monomer (PEGMEMA-475 plus PEGMEMA1100). The overall monomer concentration was 0.67 M, and the catalyst system was 2.0
xi 
 

mM CuCl, 0.60 mM CuBr2, and 6.0 mM HMTETA. The error bars, which in many
cases are obscured by the symbols, represent one standard deviation.. ...........................64
Figure 2.3. FTIR spectra of 152 nm-thick poly(PEGMEMA-475) (black line) and 118
nm-thick poly(PEGMEMA-1100) (dashed red line) films on gold-coated substrates.
Spectra were taken immediately after polymerization before the PEG side chains in
poly(PEGMEMA-1100) crystallized. The spectra are offset for clarity.. .......................67
Figure 2.4. Ratios of C-O-C to C=O peak heights (A) and peak areas (B) in reflectance
FTIR spectra of poly(PEGMEMA-1100), poly(PEGMEMA-475), poly(PEGMEMA-300)
and poly(PEGMEMA-475-co-PEGMEMA-1100) films on gold-coated substrates. The
ratios are plotted as a function of the ellipsometric thickness of the film, and for the
copolymers, the legend shows the mole fraction of PEGMEMA-475 (relative to total
monomer) in the polymerization solution. All polymerizations employed a total
monomer concentration of 0.67 M, and the experimental section describes the
polymerization conditions... .............................................................................................68
Figure 2.5. Schematic drawing of the reflection of light from (A) a bare gold surface and
(B) a film-covered gold surface.. .....................................................................................70
-1

-1

Figure 2.6 Ratios of the C-O-C (1149 cm ) and C=O (1731 cm ) absorption
coefficients in poly(PEGMEMA) homo and copolymer films as a function of film
thickness. In the case of copolymer films, the legend shows the PEGMEMA-475 mole
percentage of total monomer in the polymerization solution, and the rest of the monomer
was PEGMEMA-1100. The polymerization was performed as described in the
experimental section .. ......................................................................................................72
-1

-1

Figure 2.7. Ratios of C-O-C (1149 cm ) and C=O (1731 cm ) absorption coefficients,
k, as a function of the average number of ethylene oxide units per repeat unit, EOavg, in
poly(PEGMEMA-300), poly(PEGMEMA-475), and poly(PEGMEMA-1100) films.
The ethylene oxide units per monomer were calculated from 1H-NMR spectra of
monomers (Figure 2.1), and the standard deviations in the absorption coefficient ratios
were determined using films with thicknesses ranging from 5 to 454 nm. The line is a fit
to the data with a zero intercept... ....................................................................................73
Figure 2.8. Reflectance FTIR spectra of (A) 60 nm-thick poly(PEGMEMA-1100)
immediately after polymerization (solid line) and 24 hours later (dashed line) and (B) 97
nm-thick poly(PEGMEMA-475-co-PEGMEMA-1100) immediately after
polymerization (solid line) and 404 days later (dashed line). The copolymer was
prepared from a solution containing 25 mole% PEGMEMA-475 and 75 mole%
PEGMEMA-1100.............................................................................................................76
xii 
 

Figure 2.9. SEM image of the cross section of a gold-coated porous alumina membrane
modified with a poly(PEGMEMA-475-co-PEGMEMA-1100) film. The alumina
exhibits unfilled vertical pores covered by a 130 nm-thick film that was grown from an
aqueous solution containing 25 mole% PEGMEMA-475 and 75 mole% PEGMEMA1100.. ................................................................................................................................78
Figure 2.10. Single-gas fluxes through gold-coated porous alumina membranes capped
with (A) poly(PEGMEMA-1100) or (B) poly(PEGMEMA-475-co-PEGMEMA-1100)
grown from a solution containing 50 mole% PEGMEMA-475. (Fluxes are normalized
by multiplying by the SEM thickness of the membrane skin.) The circles represent CO2
data, and the triangles represent H2 data; filled symbols represent measurements taken
immediately after synthesis, and open symbols represent data obtained 24 h later. The
inset in Figure A shows the data on an expanded ordinate for clarity.. ...........................80
Figure 2.11. Single-gas CO2/H2 selectivity vs. mole fraction of PEGMEMA-475 in the
monomer solution used to create amorphous poly(PEGMEMA-475-co-PEGMEMA1100) films on porous alumina membranes. The triangles represent films grown from
silylated alumina, and the squares represent films grown from initiators on gold-coated
alumina. For the poly(PEGMEMA-1100) homopolymer on silylated alumina, the open
circle near the origin represents selectivity after crystallization. Error bars are one
standard deviation.. ..........................................................................................................83
Figure 3.1. Apparatus for testing mixed gas permeabilities. ..........................................101
Figure 3.2. Reflectance FTIR spectrum of a cross-linked co-polymer brush (top, 25 mole%
crosslinker, 37.5 mole% PEGMEMA-1100, and 37.5 mole % PEGMEMA-475 in the
polymerization solution that contained 0.67 M total monomer), and spectra of spincoated films of the three monomers. The co-polymer film was grown by ATRP from
intiators immobilized on Au-coated Si, and the monomers were spin coated on the
surface of Au-coated wafers. Compare the cross-linked copolymer absorbance with
poly(PEGMEMA-1100) and poly(PEGMEMA-475) (Chapter 2, Figure 2.3.) the
position of C=O peak and O-C-O peak was in match.. ..................................................103
Figure 3.3. ATR-FTIR of a bare RC membrane (before, bottom), and after (top)
modification by growth of a poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA)
film.. ...............................................................................................................................106
Figure 3.4. SEM images of cross-sections of an RC membrane. The black lines indicate
the RC layer, which resides on support of polyethylene fibers. .....................................107

xiii 
 

Figure 3.5. SEM images of cross-sections of cross-linked copolymer films grafted on
RC membranes. The black lines indicate the polymer film on top of the RC membrane.108
Figure 3.6. Gas chromatogram of the feed mixed gas. The molar ratio of CO2 to H2 is
79.7: 20.3. .......................................................................................................................110
Figure 3.7. Gas chromatogram of the permeate of a mixed-gas separation with a RC
membrane modified with a poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA)
film. (The splitting of the H2 peak is probably due to a mechanical disturbance on
sample injection, but this does not affect the integral of peak area).. ............................111
Figure 4.1. TEM images of SiO2 nanoparticles spotted on a carbon-coated Cu grid.. ..120
Figure 4.2. SEM images of the top (A) and cross-section (B) of porous alumina
membranes with nominal 20 nm surface pores. .............................................................121
Figure 4.3. (A) and (B) SEM images of cross-sections of polymer films prepared by
polymerization of a solution containing 0.25 M PEGMEMA-1100, 0.25 M PEGMEMA475, 0.17 M PEGDA, 2.0 mM CuCl, 0.60 mM CuBr2, 6.0 mM HMTETA and 0.18
mg/mL SiO2 nanoparticles. (C) and (D) are the cross-sectional images of similar films
of poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700) prepared without
nanoparticles...................................................................................................................127
Figure 4.4. SEM images of cross-sections of nanoparticle-containing membranes. The
nanoparticle concentrations in monomer solutions were 0 mg/mL (A), 0.18 mg/mL (B)
and (C), 0.45 mg/mL (D), 0.9 mg/mL (E), and 2.7 mg/mL (F). The film in (B) was
grown from a RC membrane, and all other nanocomposite membranes were grafted
from alumina. The black lines indicate the beginning and ending of the grafted polymer
film. For samples with seemingly unclear polymer-substrate boundaries, the beginning
of the film is determined by checking the difference of layer stability in the electron
beam.. .............................................................................................................................130
Figure 4.5. The influence of nanoparticle concentration during polymerization on CO2
permeability (left) and CO2/H2 selectivity (right) of membranes coated with
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700). The monomer ratio of
PEGMEMA-1100, PEGMEMA-475 and PEGDA is 3:3:2. ..........................................132
Figure 4.6. Illustration of the space between nanoparticles and the polymeric matrix
embedding the particles. A) The nanoparticles contain grafted polymer prepared by
xiv 
 

surface-initiated ATRP; B) the nanoparticles are blended in the polymer matrix, there is
no covalent bond between nanoparticle and polymer; C) the nanoparticles contain
grafted polymer grown by surface-initiated ATRP with lower initiator density. ..........134
Figure 4.7. SEM images of cross-sections of alumina membranes coated with a
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700) film containing
nanoparticles. The polymerization solution contained 0.25 M PEGMEMA-1100, 0.25 M
PEGMEMA-475, 0.17 M PEGDA, 2.0 mM CuCl, 0.60 mM CuBr2, 6.0 mM HMTETA
and SiO2 nanoparticles. The concentrations of nanoparticles in the monomer suspension
were 0.18 mg/mL (A) and 1.35 mg/mL (B). The polymerization time was 30 min. Figure.
........................................................................................................................................136
Figure 4.8. The effect of initiator attachment time on CO2 permeability (left) and
CO2/H2 selectivity (right) of porous alumina membranes coated with poly(PEGMEMA1100-co-PEGMEMA-475-co-PEGDA-700) nanoparticle-containing films. The open
circles indicate membranes prepared with SiO2 nanoparticles reacted with silane
initiator for over 6 hr, and the open triangles indicate membranes prepared with SiO2
nanoparticles reacted with silane initiator for only 10 min.. ..........................................137
Figure 4.9. The upper bound line of CO2/H2 separations and data for membranes
composed of alumina coated with poly(PEGMEMA-1100-co-PEGMEMA-475-coPEGDA-700) nanoparticle-containing films. Filled diamonds represent nanocomposite
membranes synthesized with nanoparticles reacted with initiator for 6 h, and open
triangles represent corresponding membranes with nanoparticles reacted with initiator
for only 10 min.. .............................................................................................................138
Figure 5.1. SEM images of cross-linked poly(PEGMEMA)-based films grown from
porous alumina. (A) poly(PEGMEMA-co-PEGDA), PEGDA molar fraction of 25%. (B)
poly(PEGMEMA-co-PEGDA) film with SiO2 nanoparticles, PEGDA mole fraction of
10%. (C) poly(PEGMEMA-1100-co-PEGMEMA-475) The total monomer
concentration in the polymerization solution was 0.67 M, and 0.34 M for (C); and
catalyst concentrations were CuCl- 2 mM, CuBr2- 0.67 mM and HMTETA- 6 mM.
Polymerization times were 45 min for (A) and 30 min for (B) & (C).. .........................146
Figure 5.2. NMR spectrum of the monomer and catalyst mixture (contains CuCl, CuBr2
HMTETA, PEGMEMA-1100 and PEGMEMA-475) after surface-initiated
polymerization from a porous alumina membrane. The absence of a PPEGMEMA signal
in region b and the large peak a for PEGMEMA show that there is little polymer in
solution. ..........................................................................................................................154

xv 
 

Figure 5.3. Illustration of different positions of an initiator-modified alumina substrate in
a monomer-catalyst mixture during polymerization. On the left only the top of the
membrane is exposed to monomer and catalyst, whereas on the right both sides are
exposed to the solution. ..................................................................................................155
Figure 5.4. SEM images of the cross sections of membranes modified with
poly(PEGMEMA-1100-co-PEGMEMA-475) brushes. The condition for
polymerizations are [CuCl]=2 mM, [CuBr2]=0.67 mM, [HMTETA]=6 mM, and
PEGMEMA-1100:PEGMEMA-475=1:1, monomer total concentration is 0.34 M. (A)
Only the top side of the membrane was exposed to the monomer solution (B) Both sides
of the membrane were exposed to the monomer solution... ...........................................155
Figure 5.5. Evolution of poly(PEGMEMA-1100-co-PEGMEMA-475) thickness with
polymerization time for films grown from both the top and bottom faces of initiatormodified porous alumina. The diamonds indicate the side of the membrane facing up,
and the circles represent the side facing down. The error bars indicate the standard
deviation of multiple SEM thicknesses on at least two membranes prepared under
identical conditions. The conditions for polymerization were [CuCl]=2 mM,
[CuBr2]=0.67 mM, [HMTETA]=6 mM, and PEGMEMA-1100:PEGMEMA-475=1:1,
total monomer concentration is 0.17 M. ........................................................................157
Figure 5.6. SEM images of poly(PEGMEMA-1100-co-PEGMEMA-475) films grafted
from porous alumina. The monomer concentrations and film thicknesses are A) 0.34 M,
150.0 ± 9.2 μm, (B) 0.17 M, 93.7 ± 2.7 μm, (C) 0.084 M, 19.7 ± 1.2 μm and (D) 0.042
M, 775 ± 120 nm. The monomer concentration is the sum of PEGMEMA-1100 and
PEGMEMA-475. The monomers were present in a 1:1 ratio, and polymerization times
were 30 min. ...................................................................................................................158
Figure 5.7. Thickness of poly(PEGMEMA-1100-co-PEGMEMA-475) brushes as a
function of monomer concentration. The brushes were grown for 30 min from initiators
immobilized on porous alumina. The monomer concentration is the sum of the two
monomers, which were present in a 1:1 ratio. The error bars indicate the standard
deviation of multiple thickness measurements on at least two membranes prepared under
identical conditions, [CuCl]=2 mM, [CuBr2]=0.67 mM, [HMTETA]=6 mM, and
PEGMEMA-1100:PEGMEMA-475=1:1. The polymerization time was 30 min in all
cases.. .............................................................................................................................159
Figure 5.8. Evolution of monomer conversion with time for polymerization of
PEGMEMA-475 (diamonds-bpy catalyst, reaction 1, and circles-HMTETA catalyst,
reaction 2) and a PEGMEMA-1100, PEGMEMA-475 mixture (triangles, reaction 3).
Table 5.2 lists the reaction conditions.. ..........................................................................162
xvi 
 

LIST OF SCHEMES

Scheme 1.1. Selected ligands for copper-mediated ATRP. .............................................23
Scheme 1.1. Common ATRP initiators. ...........................................................................25
Scheme 1.2. Representative monomers amenable to ATRP. ...........................................27
Scheme 2.1. Methods for growth of a poly(PEGMEMA) membrane from porous
alumina.. ...........................................................................................................................56
Scheme 2.2. Copolymerization of PEGMEMA-475 and PEGMEMA-1100 to provide a
membrane skin.. ...............................................................................................................57
Scheme 3.1. Production of regenerated cellulose from through chemical treatment of
cellulose. ...........................................................................................................................92
Scheme 3.2 Growth of a cross-linked copolymer brush from a RC membrane...............98
Scheme 4.1. Initiator attachment to SiO2 nanoparticles or alumina membranes.. ........124
Scheme 4.2. Preparation of nanocomposite membranes by ATRP. The Reaction is
performed an in oxygen-free glove bag. The pink dots represent nanoparticles, but they
are greatly expanded to allow visualization.. .................................................................125

xvii 
 

LIST OF ABBREVIATIONS

ATRP

Atomic Transfer Radical Polymerization

BiBB

2-bromoisobutyryl bromide

b.p.

Boiling Point

bpy

2,2'-bipyridine

DMAEMA

2-(dimethyl amino) ethyl methacrylate

DMF

dimethylformamide

FFV

Fractional Free Volume

GC

Gas Chromatography

GMA

glycidyl methacrylate

HEMA

2-hydroxyethyl methacrylate

HMTETA

1,1,4,7,10,10-hexamethyltriethylenetetramine

MES

2-(methacryloyloxy) ethyl succinate

MMMs

Mixed Matrix Membranes

MMA

methyl methacrylate

MWCOs

Molecular Weight Cut-offs

xviii 
 

NMR

Nuclear Magnetic Resonance

NMDG

N-methylglucamine

PAA

poly(acrylic acid)

PALS

Positron Anihilation Lifetime Spectroscopy

PDI

Polydispersity Index

PDMAEMA

poly(2-dimethylamino)ethyl-methacrylate

PEG

poly(ethylene glycol)

PEGDA

poly(ethylene glycol diacrylate)

PEGMA

poly(ethylene glycol methacrylate)

PEGMEMA

poly(ethylene glycol methyl ether methacrylate)

PEO

poly(ethylene oxide)

PES

poly(ether sulfone)

PET

poly(ethyleneterephthalate)

PGMA

poly(glydicyl methacrylate)

PMA

poly(methyl acrylate)

PMMA

poly(methyl methacrylate)

PMP

poly(4-methyl-2-pentyne)

xix 
 

PPEGMA

poly[poly(ethylene glycol methacrylate)]

PPEGMEMA

poly[poly(ethylene glycol methyl ether methacrylate)]

PS

poly styrene

PSA

Pressure Swing Adsorption

PVDF

poly(vinylidene fluoride)

RO

Reverse Osmosis

SEM

Scanning Electron Microscopy

SI-ATRP

Surface-Initiated Atomic Transfer Radical Polymerization

tBMA

tert-butyl methacrylate

TCD

Thermal Conductivity Detector

TEA

triethylamine

TEM

Transmission Electron Microscopy

xx 
 

xxi 
 

Chapter 1 Introduction and Background

This dissertation reports the growth of polymer brushes on porous supports to
create membranes that allow selective permeation of CO2 from CO2/H2 mixtures.
Specifically, atomic transfer radical polymerization (ATRP) of poly(ethylene glycol
methyl ether methacrylate) (PEGMEMA) monomers, sometimes in the presence of silica
nanoparticles, yields selective membrane skins.

My research investigates both film

formation and gas-permeation properties of these films. To put the work in perspective,
this chapter describes related previous studies and recent developments. First I introduce
the potential ‘Hydrogen Economy’ and the need to develop cheaper methods to remove
CO2 from H2. Second, I discuss membrane-based gas separations with a focus on
reverse-selective separations, where larger gases permeate through a membrane more
readily than smaller gases. Flory-Huggins theory provides a guide for selecting materials
for reverse-selective membranes. Third, I present our method to develop the thin skins of
membranes through surface-initiated ATRP (SI-ATRP). This includes discussion of the
mechanism and successful examples of ATRP. Finally, I give a brief summary of the
following chapters of this dissertation.
1.1 Hydrogen as an Energy Carrier.
Despite being the most abundant element in the universe, hydrogen does not exist
on earth as H2 in significant quantities. Instead, hydrogen reacts with oxygen, carbon and
1
 

other elements to form molecules such as water. The molar mass of diatomic H2 is only 2
1

g/mol, but the H-H bond energy is 436 kJ/mol, making H2 one of the most weightefficient energy carriers. Moreover, because combustion of H2 yields only water, delivery
of energy via H2 is environmental friendly. The concept of a ‘hydrogen energy economy’
emerged in the 1970’s, but numerous barriers currently limit such an economy.
The production of H2 requires large quantities of compounds that contain
releasable hydrogen. Currently, fossil fuels, water and biological matter are the main
resources for H2 production. As of 2010, annual H2 production in the United States was
2

nine million tons. However, over 95% of this H2 comes from fossil fuels, mainly
through steam reforming of natural gases, crude oil and coal. Electrolysis and other H2
3

production methods provide the other 4% of H2 production. Reactions (1.1) and (1.2)
show the steps in H2 production from hydrocarbons. In the first step (steam reforming),
hydrocarbons react with steam to yield CO and H2.

CnHm+nH2O→nCO+(m/2+n)H2

(1.1)

CO+H2O→CO2+H2

(1.2)

The second step, the water gas shift reaction provides additional H2 and generates CO2.
2
 

The above reactions demonstrate that n moles of CO2 result from a hydrocarbonbased synthesis of (2n+m/2) moles of H2. Even for the most hydrogen rich hydrocarbon
resource, CH4, 20% of the final product is CO2. Considering this high CO2 percentage
along with the current annual H2 production and the possible hydrogen economy of the
future, even small improvements in H2/CO2 separations could save a huge amount of
energy and capital.
If the H2 produced from reactions (1.1) and (1.2) delivers energy by its reaction
with oxygen, the amount of CO2 released to the environment during H2 production is
equivalent to the amount of CO2 produced from direct combustion of the hydrocarbon.
Moreover, some energy is lost during H2 synthesis. Thus, with current techniques, a
hydrogen economy would not reduce CO2 emissions. The potential benefit of a
hydrocarbon-based hydrogen economy is that CO2 and other pollutants could be
sequestered at H2 production plants.

1.2 Hydrogen Purification Techniques
The most common H2 purification processes are: (1) pressure swing adsorption
(PSA), (2) fractional/cryogenic distillation, and (3) membrane separations. Combinations
of these processes are also possible.

4

3
 

PSA operates in large vessels packed with materials that selectively adsorb CO2.
Initially CO2 adsorbs to the packing and purified H2 exits the vessel. When the amount
of adsorbed CO2 approaches the packing binding capacity, the high pressure feed is
switched off, and the adsorbing vessel releases CO2 and small amounts of H2 at a much
lower pressure to regenerate the packing. In PSA plants, a number of adsorbing vessels
work alternatively to allow continuous separation (Figure 1.1), because one vessel goes
through an adsorbing-purging cycle and cannot continuously adsorb CO2. PSA
separation yields high purity CO2, but it loses some CO2 during regeneration and is
relatively expensive.

4
 

 

Figure 1.1. Schematic illustration of one separation cycle in a PSA vessel.
 

Fractional/cryogenic distillation relies on the significant boiling point difference
between CO2 (194.75 K) and H2 (20.28 K).

1

The process includes cooling and

pressurizing the gas mixture to below the boiling point of both CO2 but above that of H2.
Then CO2 in the liquid phase can then be separated from H2 in the gas phase. This
method is in commercial operation by itself or in combination with PSA. Unfortunately,
cooling the gases to such low temperatures and high pressures is energy intensive.
5
 

In membrane separations (Figure 1.2), mass transport takes place through a
barrier between two phases. The separation can arise from differences in the size, charge
density, or solubility of the permeating species, as well as combinations of these factors.
Over the past 50 years, membrane separation has emerged as an attractive industrial
alternative to traditional separation approaches such as distillation, because membrane
5

separation usually requires much less energy.

Figure 1.2. Conceptual drawing of membrane separation, which involves selective
transport through a membrane between two phases.
Nevertheless, membrane separation is not in commercial operation for H2/ CO2
separations. Compared to PSA and fractional/cryogenic separations, membrane
separation is promising because of low energy consumption, the possibility of continuous
4

operation, low capital costs, and ease of operation. The major challenge to using

6
 

membrane for gas separation at industrial plants is that the performance of membranes
6

declines at high pressures and in high-humidity environments. If the hydrogen economy
develops to a larger scale in the future, research and development on membrane-based H2
purification will become even more important.
1.3 Membrane Separations
A membrane is a thin film that separates two phases and acts as a selective barrier
to the transport of the matter. Studies of membranes began as early as the eighteenth
century, but it was not until the twentieth century that research on synthetic membranes
7

started to spawn a multi-billion dollar per year industry. In 1907, Bechhold developed
the earliest device to control solvent evaporation rates and created nitrocellulose
membranes with graded pore sizes. With the advent of reproducible membrane
separations, more research appeared. By the 1930s, microporous collodion (nitrocellulose)
membranes were commercially available.

7

Perhaps the most notable breakthrough in membrane research was the
introduction of asymmetric membranes.

In the early 1960’s Loeb and Sourirajan
8

developed defect-free, anisotropic cellulose acetate reverse osmosis (RO) membranes.

These materials consisted of a dense, selective layer on the surface of a relatively
permeable material. The small thickness of the dense layer allowed for a reasonable flux
along with high salt rejection. Not long after this breakthrough, asymmetric
microfiltration and ultrafiltration membranes became commercially available. Research
on RO membranes for water desalination has been an ongoing effort since that time. By
7
 

2009, there were 14,451 desalination plants in operation worldwide, producing 59.9
million cubic meters of water per day (15.8 billion gallons a day), with an annual increase
in capacity of 12.3%.

9

Rapid advances in polymerization also occurred in the 1950s and 1960s, and this
promoted subsequent research in membrane technology. In addition to desalination, a
second branch of membrane technology focused on biological applications, such as
artificial kidneys. In biological and medical areas, membranes were also used in artificial
organs and controlled drug delivery systems. The market of membranes sales in the
7

biological area now exceeds that for traditional separation membranes.

Because gas molecules are much smaller than typical solvated molecules and ions,
gas-separation membranes appeared after membranes for solution-based applications. In
®

1980, DuPont introduced the Prism

10

hollow fiber membrane for H2 separations.

The

isolation of H2 was based mainly on its small size relative to nitrogen, ammonia, and
other gases. By 1982, Generon produced a membrane to purify N2 from air. Multiple
companies including Ube, Medal, and Generon launched advanced membrane materials
for O2/N2, H2/N2, and H2/CH4 separations by 1987.

11

Starting in 1988, multiple

companies installed membrane gas separation plants. Recently, the market for gas
separation membranes has grown to a $500 million per year business, totaling 24% of the
12

entire membrane market.

8
 

1.4 H2 Separation Membranes and the Solution-diffusion Model.

The mechanisms of membrane-based separation of H2 from CO2 fall into size-based and
solubility-based (Reverse-selective) categories (Figure 1.3). Table 1.1 presents some
physical properties of CO2 and H2. Importantly, H2 is lighter, smaller and less
condensable than CO2. Thus in general, H2 will interact less strongly with membranes
than CO2. The transport rates in size-sieving membranes depend on either the rate of gas
diffusion or entry into a pore. Because gas-diffusion rates decrease with molecular mass,
H2 permeates through the membrane faster than CO2 in both cases. Although this may
provide selectivity and useful membranes, one disadvantage of these mechanisms is that
the purified H2 on the permeate side of the membrane is at low pressure so compression
may be necessary.

9
 

 

Figure 1.3. Two representative categories of membrane-based separation of CO2 and H2.
The reverse-selective membrane relies on solubility differences between the gases.

Table 1.1. Physical properties of H2 and CO2.

13

Size

Condensability

Critical Volume
3
(cm /mol)

Kinetic
Diameter (nm)

b.p. (K)

Critical Temp (K)

H2

65.1

0.289

20.4

33.2

CO2

93.9

0.330

195

304.2

Reverse-selective membranes (Figure 1.3, right) allow selective passage of the
larger of two gases. Instead of taking advantage of size differences, the separation
mechanism in reverse-selective membranes relies on selective dissolution of a gas at the
10
 

membrane-feed interface prior to diffusion across the membrane, and desorption at the
low-pressure permeate side (Figure 1.4). According to Fick’s law of diffusion, with a
linear concentration gradient the flux of gas A through a membrane with thickness l is:

J A   DA

Cm ,2, A  Cm ,1, A
l

                                                                                                      (2)

where Cm,1,A and Cm,2,A are the concentrations of gas A in the membrane at the feed
side surface and permeate side surface, respectively. We define the solubility, SA, of gas
A in the membrane, as shown in equation (3)

Cm ,1, A  S A p A,1
Cm ,2, A  S A p A,2

(3)

where pA,1 and pA,2 are partial pressures of gas A at the feed side and permeate side of
the membrane, respectively. Under most circumstances, the feed side has a higher
pressure than the permeate side, and ΔpA as defined in equation (4) is positive.

p A,1  p A,2  p A                                                                    

(4)

Substitution of (3) and (4) into (2) gives (5).

J A  DA S A

p A
l

                                                             

11
 

(5)

 

Figure 1.4 Illustration of the solution-diffusion model for gas transport across a
membrane. The dashed line represents the concentration profile of a gas transporting
through the membrane. Cm,1,A and Cm,2,A are the concentrations of a gas in the
membrane at the feed side surface and permeate side surface, respectively.

Defining the permeability, PA, of a membrane for gas A using equation (6),

PA  DA S A                                                                                                         

12
 

(6)

we finally get equation (7), which is the usual expression for flux through a membrane in
the solution-diffusion model.

J A  DA S A

p A
l

                                                     

(7)

For a polymer membrane that separates two gases (A and B), the ideal permeability
selectivity, αAB, in equation (8) frequently characterizes the separation.

 AB 

PA DA S A


PB DB S B

                                                                

(8)

Component A is the more permeable species in the above equation, and DA/DB is the
diffusivity selectivity whereas SA/SB is the solubility selectivity between the two
components. Because permeability depends on the product of solubility and diffusivity,
gas A may be more permeable than gas B in a particular polymer film because: (i) A is
more soluble in this polymer than B, (ii) A has a larger diffusion coefficient than B, or (iii)
both. Thus, gas separation membranes typically separate gases based on differences in
gas molecule diffusivity in a polymer (so-called strongly size-sieving or high diffusivityselective materials) or on differences in gas solubility in polymers (so-called solubilityselective or weakly size-sieving materials).

5

In reverse-selective membranes, larger gases selectively permeate through the
membrane.

14

Although diffusivity decreases with molecule size, equation (8) shows that
13

 

reverse selectivity is possible by taking advantage of solubility selectivity. Generally,
penetrant solubility increases with increasing condensability (i.e. higher critical
temperature or higher normal boiling point) and more favorable interactions with the
polymer, whereas gas diffusivity increases with decreasing penetrant size, increasing
polymer fractional free volume, increasing polymer chain flexibility, and decreasing
15

polymer–penetrant interactions.

CO2 should have a higher solubility in most polymers

and liquids compared to H2 because of higher condensability, but H2 has a higher
diffusivity than CO2 because of its smaller physical size.

Separation of H2 from CO2 with reverse-selective membranes affords purified H2
at the high-pressure feed side of the membrane and would save significant energy by
avoiding H2 re-compression. Thus reverse-selective membranes may be more economical
than size-sieving membranes for purifying H2. Additionally, Reverse-selective
membranes could remove higher hydrocarbons from methane in the purification of
natural gas, separate organic monomers from nitrogen in the production of polyolefins,
and remove hydrocarbons from H2 in refinery applications.

14

1.5 Solubility of Gases in Polymers and PEO-based Materials as Reverse-selective
Membranes.
In the solution-diffusion model, the CO2 solubility in a polymer film directly
affects the CO2/H2 selectivity. The Flory-Huggins model describes penetrant solubility in

14
 

rubbery polymers

16-20

based on three assumptions for polymer-solution interactions: (i)

the process of dissolving polymer in solvent has no volume change, i.e. the total solution
volume equals the volume of polymer plus volume of solvent, (ii) the mixing process is
athermal (i.e. ΔHmix=0), and (iii) the solvent molecules and polymer segments, which
have the same size as solvent molecules, are considered as a lattice after the formation of
16

the solution.

Figure 1.5 illustrates this lattice. Based on the three assumptions above,

we can analyze statistically how many ways one can place the polymer into this lattice,
and the total possibility that this lattice could be arranged without breaking polymer chain
is defined as this lattice’s thermodynamic probability Ω.

 

Figure 1.5. Illustration of the lattice positions of a polymer segment and solvent. Black
18
dots shows polymer segments that have same size as solvent molecules (hollow dots).
15
 

Based on the Boltzman Equation, (9), the entropy of mixing polymer and solvent,
ΔSm, is given by equation (10)

S  k ln                                                                                                              

(9)

S m   R  x1 ln 1  x2 ln 2                                                                                      (10)
where R is the gas constant, xi is the molar fraction of solvent or polymer, and ϕi is the
corresponding volume fraction. Subscript 1 refers to the solvent and subscript 2 to the
polymer. Based on assumption (ii) that the enthalpy of mixing is zero, the Gibbs free
energy of this mixing process is:

Gm  T S m  RT  x1 ln 1  x2 ln 2 

(11)

where T is temperature. Because ϕi is always less than one, ΔGm is negative. Physically,
this means that the mixing process is thermodynamically favorable when ΔHm=0.
However, under most circumstances ΔHm is not close to zero, and equation (12)
describes the enthalpy of mixing.

H m  N1 x2  RT                                                             

(12)

In this equation, N1 is the number of solvent molecules in the system and χ is the FloryHuggins interaction parameter. This parameter describes the relative energy change
difference between mixing solvent molecules and the polymer-solvent molecules. Using

16
 

equation (12), the Gibbs free energy can be written as equation (13), which is also called
the Flory-Huggins equation.

21

Gm  H m  T S m  RT  N1 x2   x1 ln 1  x2 ln 2        

(13)

In the system of gas penetrating through a polymer film, the Flory-Huggins
interaction parameter χ depends on the penetrant and polymer solubility parameters, as
22

shown in equation (14),

V (1   2 )2
 
                                                               
RT

(14)

where β is a constant, V  (cm3/mol) is the partial molar volume of the gas penetrant in the
polymer, δ1 is the solubility parameter of the polymer, and δ2 is the solubility parameter
of penetrant gas. Based on this model, maximum values of gas solubility (minimum of
Gm) occur when the solubility parameters of the gas and polymer are the same. Equation
(15) describes the solubility parameters of liquids where



 H

vap

 RT  / V

                                                                

(15)

ΔHvap is the enthalpy of vaporization, and V is the molar volume of this liquid at 25 °C.
Thus, the solubility parameter is the square root of the cohesive energy density. However,
the determination of solubility parameters of polymers is much more complicated than
determination of the solubility parameters of liquids because the polymers are not volatile.
17
 

Thus accurate enthalpy of vaporization data is hard to obtain. Freeman and co-workers
summarized the experimental data for CO2 solubility in various solvents. The maximum
solubility occurred when the solvent solubility parameter is close to that of CO2, around
22 MPa

0.5 15

.

This suggests that the theoretical approach by Flory and Huggins is

applicable in studying the system of gas molecules dissolving in liquids.
Polymers containing ether oxygens exhibit high CO2 solubility presumably due to
15

the quadrupole-quadrupole interacttions.
oxide) is 19.8 MPa

0.5 22

.

The solubility parameter for poly(ethylene

Currently, polymer films containing poly(ethylene glycol)

(PEG), chains show great potential for removing CO2 from H2.

5,6,15,23-27
23

However,

and the high

most studies employed relatively thick (100-300 μm) membranes,

thickness limits flux. Thus, the main subject of my research is to facilitate formation of
thin, PEG-based polymer films on porous supports to increase flux in CO2/H2
separations.

1.6 Growth of Polymer Brushes by SI-ATRP
1.6.1

Overview of Strategies for Brush Growth
The most common ways to prepare composite membranes with a thin skin on a

permeable

support
30

polymerization,

include

solution

28

casting,

plasma polymerization

31,32

18
 

solution

29

coating,

and UV-induced grafting.

interfacial
33,34

The

principle challenge in these methods is the difficulty of controlling film thickness.

35

Surface-initiated ATRP allows controlled growth of polymer brushes, and it should
provide access to membranes with well-defined polymer skins.

 

Figure 1.6. "grafting from" (left) and “grafting to” (right) methods for the generation of
polymer brushes from surface. “i” represents initiator molecules.
A polymer brush is an assembly of polymer chains linked by one end to an
interface. Typically, there are two strategies of creating polymer brushes covalently
36

linked to surfaces, namely “grafting to” and “grafting from” methods.

Figure 1.6

illustrates these two approaches. The “grafting from” method employs polymerization
37

from immobilized initiators.

In controlled, surface-initiated polymerization, the brush

thickness varies approximately linearly with time, and the polymer film has low
polydispersity index (PDI=Mw/Mn). At the same time, when initiation occurs only at the
surface there is minimal solution polymerization.
In contrast, the “grafting to” method employs solution polymerization to form
polymer chains prior to covalent linking of these chains to a surface. One drawback of

19
 

this strategy is that steric hindrance prevents incoming polymer chains from diffusing
through the film and leads to a low polymer density on the surface.

38

Nevertheless the

“grafting to” method affords more control of polymer chain length and PDI. Thus allows
better characterization of these chains.
Surface-initiated polymerization can occur on a wide range of substrates including
39

Si, gold-coated surfaces, nanoparticles, and carbon nanotubes.

One of the most

common methods of initiator attachment involves adsorption of thiol-terminated initiators
on gold.

38

Over the past two decades, numerous reports examined surface-initiated
40,41

polymerization mechanisms including cationic,
46-48

opening methathesis,

49-51

radical,

and ATRP.

Anionic,
37,52-56

42-44

45

ring opening,

ring

This dissertation primarily

employs ATRP, which I discuss in more detail below.

1.6.2

ATRP Mechanism
ATRP is perhaps the most versatile process for creating well-defined polymers. 57

As Figure 1.8 shows, a transition metal complex extracts a terminal halogen atom from
the initiator or polymer chain, R-X, to create a radical. The equilibrium for this step
favors radical deactivation because the rate constant for recovering a halogen atom, kdeact,
is much greater than the constant for atom extraction, kact. Thus the concentration of
radicals is much smaller than the concentration of dormant species. Because termination
by radical recombination or disproportionation is proportional to the square of radical

20
 

concentration, low termination rates in ATRP lead to high molecular weight polymers, as
well as a low PDI.

 

Figure 1.7. The mechanism of ATRP.
Copper complexes are the most common and versatile ATRP catalysts. When
using such catalysts, equation (16) describes the ATRP rate for the ideal case of no
polymerization termination and 100% initiation efficiency.

RATRP

k

kact
p kdeact

[Cu I ]

[M][I]0 [XCu II ]                                                             (16)

The rate of ATRP is proportional to the monomer concentration in the solution [M], as
well as the ratio of activation and deactivation rate constants (kact/kdeact). Importantly,
(kact/kdeact) is a function of the Cu-binding ligand, the reaction solvent, and temperature.
I

Increasing the concentration of the Cu complex will increase the polymerization rate,
II

wherease incrasing the concentration of the Cu

complex should slow down ATRP.

Overall the catalyst system and monomer reactivity define the control and rate of ATRP.

21
 

1.6.3

ATRP Catalysts
Effective ATRP catalysts include a transition metal ion with accessible oxidation

states separated by one electron to facilitate halogen removal from the ATRP initiator.
Also, the coordination sphere around the metal should expand upon oxidation to
selectively accommodate a halogen.
Cu,

57,59-62

Pd,

63

Ni,

64-67

Rh,

68

Fe,

58

69,70

A wide range of transition metals including
Ru,

71-73

Re,

74

Co

75

and Mo

76

have served as

ATRP catalysts. Nevertheless, Cu complexes are perhaps the most economical and
versatile catalysts for polymerization of different types of monomers. This versatility
I

II

stems in large part from the large number of ligands that bind Cu and Cu .

1.6.4

Ligands for ATRP Catalysts
Ligands solubilize and stabilize the transition metal ions to form an ATRP
I

catalyst system. Selection of ligands depends mostly on the metal ion. For example, Cu

complexes tend to form square planar or tetrahedral geometries, so the most common
ligands for Cu catalysts provide tetradentate or bidentate coordination. Moreover, the
ligand should allow fast initiation, minimum side reactions, and an equilibrium that
I

I

II

favors the lower oxidation state, such as Cu . As mentioned above, Cu / Cu is the most
common ATRP catalyst system, and nitrogen-containing ligands bind strongly to Cu and
I

stabilize the Cu complex. Sulfur, oxygen, or phosphorus ligands are less effective due to
inappropriate electronic effects or unfavorable binding constants.

58

some examples of ligands used for ATRP with Cu catalyst systems.
22
 

Scheme 1.1 shows

 

Scheme 1.1. Selected ligands for copper-mediated ATRP.

1.6.5

58,35

ATRP Initiators
The initiator has two main functions in ATRP. Firstly, the amount of initiator

defines the maximum number of polymer chains if chain transfer is negligible. Secondly,
the structure of the initiator plays a role in ATRP mechanism. The leaving halogen group
should have affinity to the carbon of the polymerizing chain to promote deactivation.
Additionally, the R-X bond should always give homolytic rather than heterolytic
23
 

cleavage in the presence of catalyst. Last but not least, the initiation step should be
relatively faster than the propagation to insure low PDI. When selecting an initiator in
ATRP, several issues are worth consideration. (a) In the order of the stabilizing ability,
CN > COR > COOR > Ph > Cl > Me are the common groups that stabilize radicals when
halogen is removed.

77

(b) The C-halogen bond strength decreases in the order of R-Cl >

R-Br > R-I. Thus R-I initiators offer the most rapid initiation rates, but iodine-containing
compounds are sensitive to light and temperature.

78

Bromine and chlorine-terminated

molecules are the typical initiators. In most studies, the halogen in the initiator is also the
anion that couples with the metal cation when preparing the catalyst system. However,
some studies indicate that halogen exchange can offer better control over the
79,80

polymerization.

For polymer growth from surfaces, ATRP initiators need a

functional group for anchoring to a surface in addition to a group that initiates the
polymerization. As the anchoring group differs for various substrates, the section
introduces specific initiators for polymerization from a surface.

24
 

58,35

Scheme 1.2. Common ATRP initiators.

 

1.6.6

ATRP Monomers
A wide range of monomers can polymerize via ATRP. Scheme 1.3 provides some

examples from styrenes, methacrylates, and acrylates. The common features of ATRP
58

monomers is that they contain functional groups that help stabilize the radical.

25
 

This

dissertation focuses on ATRP of PEGMEMA and poly(ethylene glycol diacrylate)
(PEGDA) because the PEO side chains favor solubility selectivity for CO2 over H2. For a
given monomer, the rate of polymerization depends on the catalyst system, and the
reaction media.

81

The solubility of the ligand, the procedure of preparing the catalyst, the

solvent, the solubility of the initiator, and the temperature all play a role in ATRP, mainly
because each component impacts the equilibrium illustrated in Figure 1.7, and
temperature increases the polymerization rate constant.

26
 

58

Scheme 1.3. Representative monomers amenable to ATRP.

 

27
 

Scheme 1.3 (cont’d)

1.6.7

Substrates for SI-ATRP
Huang and Wirth initially reported SI-ATRP when they successfully grafted
82

poly(acrylamide) (PAM) brushes from benzyl chloride-derivatized silica particles.

Shortly after this discovery, a number of studies described SI-ATRP on various substrates.
Growth of polymer brushes from a surface requires initiator immobilization, and the
discussion below describes initiator anchoring and polymerization from different
substrates.
1.6.7.1 SI-ATRP from Silicon Oxide
Silicon oxide surface is the most extensively studied substrate for SI-ATRP.
Physically, such substrates include lightly oxidized silicon wafers, glass, quartz sides,
nanoparticles, capillaries, membranes or silicon oxide coatings on other substrates.
Formation of organosilane monolayers or multilayers is the most common method for
28
 

anchoring initiators on SiO2. After treatment with oxygen plasma or piranha solution (a
mixture of sulfuric acid and H2O2), the SiO2 surface contains a high density of silanol
groups (Si-OH) that can react with organosilane molecules such as R'SiRxX3-x (X=Cl,
83

OR, or NMe2) to form a Si-O-Si bond.

The silanes may also react with H2O and

neigboring organosilane molecules to form organosilane clusters instead of monolayers.
Practically, organosilane molecules with one hydrolyzable group tend to form better
defined monolayers than those with more than one hydrolyzable group. Table 1.2
illustrates common initiator examples for ATRP from silicon oxide.
Table 1.2. Initiators employed for growing polymer brushes from silicon oxide.
Initiator

Substrate

58

Reference

Oxidized silicon wafer;
glass slide; SiO2 deposited

84-86

Au wafer
Oxidized silicon wafer;
glass slide; silicon ATR
crystal; quartz; fused silica

56, 87-101

capillary column; porous
silica; silica nanoparticles
Oxidized silicon wafer,
37, 102, 103
silica nanoparticles

 
 

29
 

Table 1.2 (cont’d)

Oxidized silicon wafer

104, 105

Silica nanoparticles

106

Silica nanoparticles

107, 108

 

 

Cl
Cl Si
Cl

Cl
 

Porous silica
109
microparticles

 

Oxidized silicon wafer;
110-114
quartz; silica nanoparticles
 

Si/SiO2 core-shell
nanowire

 

Si/SiO2 core-shell

115

nanowires

 

Oxidized silicon wafer
 

30
 

124

52, 116-118

Table 1.2 (cont’d)
Oxidized silicon wafer;
glass side; silica
nanoparticles; Si/SiO2
 

75, 115,
120-126,
119

core-shell nanowires
Oxidized silicon wafer;
silica nanoparticles;
CdS/SiO2 core-shell

 

127-133

nanoparticles

Silica nanoparticles

134

Oxidized silicon wafer

135

Oxidized silicon wafer

136-138

 

 

 

Silica nanoparticles;
porous silica
 

microparticles

 
 
 
 

31
 

139-142

Table 1.2 (cont’d)
Oxidized silicon wafer;
silica nanoparticles;
silsesquioxane microgel
 

143-153

nanoparticles; silicone
nanofilaments
Porous silica
142
microparticles
 

Silica nanoparticles

154

Oxidized silicon wafer

155

Silica nanoparticle

156

 

Cl Si (CH2)3 O

Br
 

32
 

Table 1.2 (cont’d)

Silica nanoparticle

157

Oxidized silica wafer

158

 

 
 
 
 
 
 
 
 
 
 
 

33
 

Table 1.2 (cont’d)

 

Oxidized silicon wafer;

+ 

159
glass slide

 

Organosilane-based ATRP initiators also offer an opportunity to control the
136

polymer brush density and pattern polymer brushes. Bao et al

studied the effect of

initiator density on polymer brush thickness by simultaneously depositing a
monochlorosilane initiator and a similar monochlorosilane molecule with no halogen.
Jonas successfully printed polymer brush patterns via UV-light removal of initiator areas
144

followed by ATRP.

34
 

157-159

Armes and coworkers developed macroinitiators for ATRP from silicon.

Positively charged polyelectrolytes attach to the negatively charged silicon dioxide
surface through electrostatic forces, and macroinitiator adsorption can occur from water.
The initiators appear at the bottom of Table 1.2.
1.6.7.2 SI-ATRP from Gold
Similar to the monolayers on cleaned silicon oxide, the most common method of
attaching initiators on Au is adsorption of functionalized thiols or disulfides.
BrC(CH3)2COO-(CH2)11SH

81

55

and (BrC(CH3)2COO(CH2)11S)2

are two extensively

used initiators in ATRP from gold. Physically, typical gold surfaces are silicon wafers
coated with gold or gold nanoparticles.
For the same monomer and catalyst system, the surface and method of initiator
attachment may affect the polymer brush thickness. Saha and coworkers studied the
growth of poly(methyl methacrylate) (PMMA) from gold surfaces and gold surfaces
coated with a cross-linked silane. Their study implicates thiols in terminating growing
polymer brushes to give polymer chains with low molecular weights and low film
160

thicknesses.

1.6.7.3 SI-ATRP from Polymeric Substrates
ATRP from polymeric substrates is not as common as from SiO2 or Au because
some polymers do not tolerate organic solvents or high temperature. However, studies of
ATRP from polymers are increasing and there are successful examples. Polymeric

35
 

substrates fall roughly into two categories: functional and inert surfaces. Functional
polymeric surfaces have exposed groups that are either nucleophilic or electrophilic,
while inert polymeric surfaces are unreactive.
Cellulose offers pendant –OH groups, and Carlmark and Malmstrom immobilized
initiators by allowing these group to react with 2-bromoisobutyryl bromide in the
presence of triethylamine. They successfully polymerized poly(methyl acrylate) (PMA)block-poly(2-hydroxyethyl
paper.

161

methacrylate)

(poly(HEMA))

copolymers

from

filter

Later, Lindqvist and Malmstrom successfully polymerized polystyrene from

filter paper and regenerated cellulose. Based on FTIR spectra, the regenerated cellulose
162

surface offers lower polymer grafting density than natural cellulose.

Singh and

Husson performed ATRP of poly(poly(ethylene glycol) methacrylate) (PEGMA) and
poly(acrylic acid) (PAA) on regenerated cellulose membranes using similar initiator
attachment.
OH

163,164

groups,

Poly(ethylene terephthalate) (PET) membranes also offer pendant –

and

Demoustier-Champagne

and

co-workers

polymerized 

N-

isopropylacrylamide (NIPAM) from the walls of track-etched, initiator-modified PET
165

membranes.

Polyvinylidene fluoride (PVDF) films expose C-F bond that could

initiate ATRP, and Chen successfully polymerized poly[poly(ethylene glycol methyl
ether methacrylate)] (PPEGMEMA) and PPEGMEMA-b-PS block copolymer brushes
from PVDF.

166

Inert polymeric substrates that require activation through oxidization likely
produces some –OH or –COOH groups on the surface. Subsequent reaction with 236
 

bromoisobutyryl bromide in the presence of a base provides immobilized initiators for
ATRP. This strategy successfully anchored ATRP initiating groups onto ozone-treated
167

polypropylene hollow fiber membranes.

Some inert polymeric substrates do not survive exposure to organic solvents, and
extensive oxidation may alter their pore size. Jain reported layer by layer deposition of
macroinitiators on polyethersulfone (PES) films. This initiator attachment technique and
168

polymerization of PHEMA both occur in aqueous conditions.

The method offers an

opportunity to attach ATRP initiators on other inert polymeric substrates.

1.7 Outline of This Dissertation.
This work focuses on development of polymeric films for CO2/H2 separation.
Chapter 2 describes collaborative work on the synthesis and characterization of
poly(PEGMEMA-1100-co-PEGMEMA-475) copolymer films. The presence of both long
and short side PEO chains in the copolymer successfully prevented crystallization of the
long chains (23-24 PEO units) from PEGMEMA-1100. FTIR studies indicated that the
molar ratio of both monomers in the polymer film is comparable to the molar ratio of the
monomers in the polymerization solution. The CO2/H2 selectivity of the copolymer film
(~13) is comparable to an amorphous poly(PEGMEMA-1100) film, but the copolymer
film shows no sign of crystallization in over a year. This selectivity is among the highest
values for reverse-selective membranes.

37
 

Chapter 3 discusses ATRP of PEGMEMA-1100, PEGMEMA-475, and PEGDA
from regenerated cellulose. FTIR spectra is used to study the extent of cross-linking by
PEGDA, and the gas separation tests suggest that cross-linking due to a 25% mole
fraction of PEGDA causes a slight decline in CO2/H2 selectivities ranging from 6.5 to
19.9, but the CO2 permeability decreased to only 5-15 Barrer.
Chapter 4 focuses on nanocomposite membrane coatings prepared by SI-ATRP of
PEGMEMA and PEGDA in the presence of initiator-modified SiO2 nanoparticles.
Electron microscopy provides evidence for nanoparticle incorporation in the membrane
coating. Unfortunately, inclusion of nanoparticles did not significantly increase
selectivity. The rubbery nature of the polymer many limit the formation of free volume
around the nanoparticles and minimize their effect on permeation.
Chapter 5 investigates why aqueous SI-ATRP of PEGMEMA results in films with
extraordinarily high thicknesses. SI-ATRP yields ~100 μm thick films in only 30 min.
Preliminary studies suggest that the most likely hypothesis is that high concentrations of
PEGMEMA-1100 change the polarity of the solvent and the activity of the catalyst.
Finally, Chapter 6 discusses future work and what impact these studies may have
on forming coatings for membranes or other potential applications.  

 

 

38
 

REFERENCES

39
 

References.

1. Engel, T.; Reid, P. Physical Chemistry, 2nd ed.; Pearson Education, Inc.: Upper
Saddle River, 2010.

2. Hydrogen Production Technology.
http://www.fossil.energy.gov/programs/fuels/hydrogen/currenttechnology.html. July,
2010.

3. US Department of Energy, Hydrogen Production.

http://www.hydrogen.energy.gov/pdfs/doe_h2_production.pdf. Nov, 2010.

4. Ockwig, N. W.; Nenoff, T. M. Chem. Rev. 2007, 107, 4078-4110.
5. Bernardo, P.; Drioli, E.; Golemme, G. Ind. Eng. Chem. Res. 2009, 48, 4638–4663.
6. Lin, H.; Van Wagner, E.; Freeman, B. D.; Toy, L. G.; Gupta, R. P. Science 2006, 311,
639-642.
7. Baker, R. W. Membrane Technology and Applications; John Wiley & Sons: New
York, 2004.

8. Loeb, S. “The Loeb-Sourirajan Mermbrane” How It Came About. ACS Symposium
Series 1981, 153, 1-9.
9. Henthorne, L. The Current State of Desalination.
http://www.idadesal.org/PDF/the%20current%20state%20of%20desalination%20remarks
%20nov%2009%20by%20lisa%20henthorne.pdf. Nov, 2009.
10. Henis, J.; Tripodi, M. Sep. Sci. Technol. 1980, 15, 1059-1068.
11. Baker, R. W. Ind. Eng. Chem. Res. 2002, 41, 1393-1341.
12. Li, N. N.; Fane, A. G.; Winston Ho, W. S.; Matsuura, T. Advanced Membrane
Technology and Applications; Wiley: New York, 2011

40
 

13. Reid, R. C.; Prausnitz, J. M.; Polling, B. E. The properties of gases and liquids;
Mcgraw-Hill: New York, 1987.
14. Freeman, B. D., Pinnau, I., Eds. Advanced Materials for Membrane Separations (ACS
Symposium); American Chemical Society, 2004.
15. Freeman, B. D.; Lin, H. J of Mol. Struc.2005, 739, 57-74.
16. Flory, P. J. J. Chem. Phys. 1941, 9, 660-661.
17. Flory, P. J. J. Chem. Phys 1942, 10, 51-61.
18. Flory, P. J. Principles of Polymer Chemistry; Cornell University Press: Ithaca, 1953
19. Huggins, M. L.. J. of Chem. Phys. 1941, 9, 440.
20. Freeman, B. D.; Pinnau, I. Trends in Polym. Sci. 1997, 5, 167–173.
21. Chanda, M. Introduction to Polymer Science and Chemistry: a Problem Solving
Spproach; CRC Press, 2006.
22. Sperling, L. H. Introduction to Physical Polymer Science; Wiley: New York, 1992.
23. Lin, H.; Freeman, B. D. J. Membr. Sci. 2004, 239, 105-117.
24. Lin, H.; Kai, T.; Freeman, B. D.; Kalakkunnath, S.; Kalika, D. S. Macromolecules
2005, 38, 8381-8393.
25. Lin, H.; Freeman, B. D. Macromolecules 2006 39, 3568-3580.
26. Freeman, B. D.; Lin, H. Macromolecules 2005, 38, 8394-8407
27. Patel, N. P.; Hunt, M. A.; Lin-Gibson, S.; Bencherif, S.; Spontak, R. J. J. Membr. Sci.
2005, 251, 51-57.

41
 

28. Staudt-Bickel, C.; Koros, W. J. J. Membr. Sci. 1999, 155, 145-154.
29. Yanagishita, H.; Kitamoto, D.; Haraya, K.; Nakane, T.; Okada, T.; Matsuda, H.;
Idemoto, Y.; Koura, N. J. Membr. Sci. 2001, 188, 165-172.
30. Li, J.; Hitchcock, A. P.; Stöver, H. D. H. Langmuir 2010, 26, 17926–17935.
31. Hayakawa, Y.; Terasawa, N.; Hayashi, E.; Abe, T. J. Applied. Polym. Sci. 1996, 62,
951-954.
32. Ulbricht, M.; Belfort, G. J. Membr. Sci. 1996, 111, 193-215.
33. Kim, J. H.; Ha, S. Y.; Rhim, J. W.; Baek, K. H.; Lee, Y. M. J. Membr. Sci. 2001, 186,
97-107.
34. Pieracci, J.; Crivello, J. V.; Belfort, G. Chem. Mater. 2002, 14, 256-265.
35. Zheng, Y. Poly(ethylene oxide) Based Polymer Brushes: Synthesis, Properties and
Applications; Michigan State University, 2006.
36. Jordan, R. . U. A.; Kang, J.; Rafailovich, M.; Sokolov, J. J. Am, Chem, Soc. 1999,
121, 1016-1022.
37. Hussman, M.; Malmstrom, E.; McNamara, M.; Mate, M.; Mecerreyes, D.; Benoit, D.;
Hedrick, J.; Mansky, P.; Huang, E.; Russell, T.; Hawker, C. Macromolecules 1999, 32,
1424–1431.
38. Kim, J.; Bruening, M.; Baker, G. J. Am, Chem, Soc. 2000, 122, 7616-7617.
39. Pyun, J.; Kowalewski, T.; Matyjaszewski, K. Macromol. Rapid. Commun. 2003, 24,
1043-1059.
40. Jordan, R.; Ulman, A. J. Am. Chem. Soc. 1998, 120, 243-247.
41. Jordan, R.; West, N.; Ulman, A. . C. Y.; Nuyken, O. Macromolecules 2001, 34, 16061611.
42
 

42. Advincula, R. J. Dispersion Sci. Tech. 2003, 24, 343-361.
43. Zheng, L.; Xie, A.; Lean, J. Macromolecules 2004, 37, 9954-9962.
44. Sakellariou, G.; Park, M.; Advincula, R.; Mays, J.; Hadjichristidis, N. J. Polym. Sci.
A: Polym. Chem. 2006, 44, 769-782.
45. Husemann, M.; Mecerreyes, D.; Hawker, C.; Hedrick, J.; Shah, R.; Abbott, N. Angew.
Chem., Int. Ed. 1999, 38, 647-649.
46. Hermann, W.; Stumpf, A.; Bogdanovic, S.; Dufaud, V.; Basset, J. Angew. Chem. Int.
Ed. Engl. 1996, 35, 2803-2805.
47. Jeon, N.; Choi, I.; Whitesides, G.; Kim,.; Laibinis, P.; Harada, Y.; Finnie, K.;
Girolami, G.; Nuzzo, R. Appl. Phys. Lett. 1999, 75, 4201-4202.
48. Buchmeiser, M.; Sinner, F.; Mupa, M.; Wurst, K. Macromolecules 2000, 33, 32-39.
49. Prucker, O.; Ruhe, J. Macromolecules 1998, 31, 592-601.
50. Peng, B.; Johannsmann, D.; Ruhe, J. Macromolecules 1999, 32, 6759-6766.
51. Niwa, M.; Date, M.; Higashi, N. Macromolecules 1996, 29, 3681-3685.
52. Ejaz, M.; Yamamoto, S.; Ohno, K.; Tsujii, Y.; Fukuda, T. Macromolecules 1998, 31,
5934-5936.
53. Huang, X.; Wirth, M. Macromolecules 1999, 32, 1694-1696.
54. Huang, X.; Doneski, L.; Wirth, M. Anal. Chem. 1998, 70, 4023-4029.
55. Shah, R.; Merreceyes, D.; Husemann, M.; Rees, I.; Abbott, N.; Hawker, C.; Hedrick,
J. Macromolecules 2000, 33, 597-605.

43
 

56. Matyjaszewski, K.; Miller, P.; Shukla, N.; Immaraporn, B.; Gelman, A.; Luokala, B.;
Siclovan, T.; Kickelbick, G.; Vallant, T.; Hoffmann, H.; Pakula, T. Macromolecules 1999,
32, 8716-8724.
57. Matyjaszewski, K.; Nanda, A.; Tang, W. Macromolecules 2005, 38, 2015-2018.
58. Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921-2990.
59. Wang, J.; Matyjaszewski, K. J. Am. Chem. Soc. 1995, 117, 5614-5615.
60. Wang, J.; Matyjaszewski, K. Macromolecules 1995, 28, 7901-7910.
61. Wang, J.; Matyjaszewski, K. Macromolecules 1995, 28, 7572-7573.
62. Matyjaszewski, K.; Jo, S.; Paik, H.; Gaynor, S. Macromolecules 1997, 30, 6398-6400.
63. Lecomte, P.; Drapier, I.; Dubois, P.; Teyssie, P.; Jerome, R. Macromolecules 1997,
30, 7631-7633.
64. Granel, C.; Dubois, P.; Jerome, R.; Teyssie, P. Macromolecules 1996, 29, 8576-8582.
65. Moineau, G.; Minet, M.; Dubois, P.; Teyssie, P.; Senninger, T.; Jerome, R.
Macromolecules 1999, 32, 27-35.
66. Uegaki, H.; Kotani, Y.; Kamigaito, M.; Sawamoto, M. Macromolecules 1997, 30,
2249-2253.
67. Uegaki, H.; Kotani, Y.; Kamigaito, M.; Sawamoto, M. Macromolecules 1998, 31,
6756-6761.
68. Moineau, G.; Granel, C.; Dubois, P.; Jerome, R.; Teyssie, P. 1998, 31, 542-544.
69. Matyjaszewski, K.; Wei, M.; Xia, J.; McDermott, N. Macromolecules 1997, 30,
8161-8164.

44
 

70. Teodorescu, M.; Gaynor, S.; Matyjaszewski, K. Macromolecules 2000, 33, 23352339.
71. Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T. Macromolecules 1996, 28,
1721- 1723.
72. Nishikawa, T.; Ando, T.; Kamigaito, M.; Sawamoto, M. Macromolecules 1997, 30,
2244-2248.
73. Ando, T.; Kamigaito, M.; Sawamoto, M. Macromolecules 2000, 33, 5825-5829.
74. Kotani, Y.; Kamigaito, M.; Sawamoto, M. Macromolecules 1999, 32, 2420-2424.
75. Gill, C.; Venkatasubbaiah, K.; Phan, N.; Weck, M.; Jones, C. Chem. Eur. J. 2008, 14,
7306-7313.
76. Brandts, J.; Geijn, P.; Faassen, E.; Boersma, J.; Gerard, G. J. Organomet. Chem. 1999,
584, 246–253.
77. Tang, W.; Matyjaszewski, K. Macromolecules 2007, 40, 1858-1863
78. Jo, S.-M.; Lee, W.-S.; Ahn, B.-S.; Park, K.-Y.; Kim, K.-A.; Paeng, I.-S. R. Polym.
Bull. 2000, 44, 1-8.
79. Matyjaszewski, K.; Shipp, D.; Wang, J.; Grimaud, T.; Patten, T. Macromolecules
1998, 31, 6836-6840
80. Huang, W.; Kim, J.; Bruening, M.; Baker, G. Macromolecules 2002, 35, 1175-1179.
81. Jones, D.; Brown, A.; Huck, W. Langmuir 2002, 18, 1265-1269.
82. Huang, X.; Wirth, M. Anal. Chem. 1997, 69, 4577-4580.
83. Zhao, X.; Kopelman, R. J. Phys. Chem. 1996, 100, 11014-11018.
84. Yamamoto, S.; Ejaz, M.; Tsujii, Y.; Fukuda, T. Macromolecules 2000, 33, 5608-5612.
45
 

85. Tsujii, Y.; Ejaz, M.; Yamamoto, S.; Fukuda, T.; Shigeto, K.; Mibu, K.; Shinjo, T.
Polymer 2002, 43, 3837-3841.
86. Urayama, K.; Yamamoto, S.; Tsujii, Y.; Fukuda, T.; Neher, D. Macromolecules 2002,
35, 9459-9465.
87. Boyes, S. G.; Brittain, W. J.; Weng, X.; Cheng, S. Z. D. Macromolecules 2002, 35,
4960-4967.
88. Tomlinson, M.; Efimenko, K.; Genzer, J. Macromolecules 2006, 39, 9049-9056.
89. Xu, C.; Wu, T.; Mei, Y.; Drain, C. M.; Batteas, J. D.; Beers, K. L. Langmuir 2005, 21,
11136-11140.
90. Wu, T.; Gong, P.; Szleifer, I.; Vlcek, P.; Subr, V.; Genzer, J. Macromolecules 2007,
40, 8756-8764.
91. Tomlinson, M.; Genzer, J. Macromolecules 2003, 3, 3449-3451.
92. Yu, K.; Wang, H.; Han, Y. Langmuir 2007, 23, 8957-8964.
93. Bhat, R.; Genzer, J. Appl. Surf. Sci. 2006, 252, 2549-2554.
94. Ma, H.; Li, D.; Sheng, X.; Zhao, B.; Chilkoti, A. Langmuir 2006, , 22, 3751-3756.
95. Slim, C.; Tran, Y.; Chehimi, M.; Garraud, N.; Roger, J.; Combellas, C.; Kanoufi, F.
Chem. Mater. 2008, 20, 6677-6685.
96. Bell, N. S.; Piech, M. Langmuir 2006, 22, 1420-1427.
97. Miller, M.; Baker, G.; Bruening, M. J. Chromatogr., A 2004, 1044, 323-330.
98. Constable, A.; Brittain, W. Colloids Surf., A 2007, 308, 123-128.
99. Constable, A.; Brittain, W. Colloids Surf., A 2011, 380, 128-134.

46
 

100. Granville, A.; Brittain, W. Macromol. Rapid Commun. 2004, 25, 1298-1302.
101. Piech, M.; George, M.; Bell, N.; Braun, P. Langmuir 2006, 22, 1379-1382.
102. Mori, H.; Seng, D.; Zhang, M.; Muller, A. Langmuir 2002, 18, 3682-3693.
103. Mori, H.; Boker, A.; Krausch, G.; Muller, A. Macromolecules 2001, 34, 6871-6882.
104. Edmondson, S.; Huck, W. J. Mater. Chem. 2004, 14, 730-734.
105. Hamelinck, P.; Huck, W. J. Mater. Chem. 2005, 15, 381-385.
106. Parvole, J.; Laruelle, G.; Guimon, C.; Francois, J.; Billon, L. Macromol. Rapid
Commun. 2003, 24, 1074-1078.
107. Fu, G.; Zhao, J.; Sun, Y.; Kang, E.; Neoh, K. Macromolecules 2007, 40, 2271-2275.
108. Fu, G.; Shang, Z.; Hong, L.; Kang, E.; Neoh, K. Macromolecules 2005, 38, 78677871.
109. Mandal, T.; Fleming, M.; Walt, D. Chem. Mater. 2000, 12, 3481-3487.
110. Kobayashi, M.; Terayama, Y.; Hosaka, N.; Kaido, M.; Suzuki, A.; Yamada, N.;
Torikai, N.; Ishihara, K.; Takahara, A. Soft Matter 2007, 3, 740-746.
111. Uekusa, T.; Nagano, S.; Seki, T. Langmuir 2007, 23, 4642-4645.
112. Morinaga, T.; Ohkura, M.; Ohno, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2007,
40, 1159-1164.
113. Ohno, K.; Morinaga, T.; Koh, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2005, 38,
2137-2142.
114. Ohno, K.; Morinaga, T.; Takeno, S.; Tsujii, Y.; Fukuda, T. Macromolecules 2007,
40, 9143-9150.

47
 

115. Mulvihill, M.; Rupert, B.; He, R.; Hochbaum, A.; Arnold, J.; Yang, P. J. Am. Chem.
Soc. 2005, 127, 16040-16041.
116. Ejaz, M.; Yamamoto, S.; Tsujii, Y.; Fukuda, T. Macromolecules 2002, 35, 14121418.
117. Ejaz, M.; Ohno, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2000, 33, 2870-2874.
118. Kawakita, H.; Masunaga, H.; Nomura, K.; Uezu, K.; Akiba, I.; Tsuneda, S. Porous
Mater. 2007, 14, 387-391.
119. Kim, Y.; Lee, B.; Kim, E.; Choi, I.; Moon, W.; Lee, T.; Kim, H. Anal. Chem. 2008,
80, 5094-5102.
120. Tugulu, S.; Barbey, R.; Harms, M.; Fricke, M.; Volkmer, D.; Rossi, A.; Klok, H.
Macromolecules 2007, 40, 168-177.
121. Tugulu, S.; Harms, M.; Fricke, M.; Volkmer, D.; Klok, H.-A. Ind Eng Chem Res
2006, 45, 7458-7461.
122. Tugulu, S. a. K. H.-A. Biomacromolecules 2008, 9, 906-912
123. Tugulu, S.; Arnold, A.; Sielaff, I.; Johnsson, K.; Klok, H. Biomacromolecules 2005,
6, 1602-1607.
124. Tugulu, S.; Silacci, P.; Stergiopulos, N.; Klok, H.-A. Biomaterials 2007, 28, 25362546.
125. Murata, H.; Koepsel, R.; Matyjaszewski, K.; Russell, A. Biomaterials 2007, 28,
4870-4879.
126. Zhang, Z.; Chao, T.; Chen, S.; Jiang, S. Langmuir 2006, 22, 10072-10077.
127. Hoven, V.; Srinanthakul, M.; Iwasaki, Y.; Iwata, R.; Kiatkamjornwong, S. Colloid
Interface Sci. 2007, 314, 446-459.

48
 

128. Chen, X.; Randall, D.; Perruchot, C.; Watts, J.; Patten, T.; von Werne, T.; Armes, S.
J. Colloid Interface Sci. 2003, 257, 56-64.
129. Perruchot, C.; Khan, M. A.; Kamitsi, A.; Armes, S. P.; von Werne, T.; Patten, T. E.
Langmuir 2001, 17, 4479-4481.
130. Wang, T.; Ou, C.; Yang, C. J. Appl. Polym. Sci. 2008, 109, 3421-3430.
131. von Werne, T.; Patten, T. J. Am. Chem. Soc. 2001, 123, 7497-7505.
132. Perruchot, C.; Khan, M.; Kamitsi, A.; Armes, S.; Watts, J.; von Werne, T.; Patten, T.
Eur. Polym. J. 2004, 40, 2129-2141.
133. Farmer, S.; Patten, T. Chem. Mater. 2001, 13, 3920-3926.
134. von Werne, T.; Patten, T. J. Am. Chem. Soc. 1999, 121, 7409-7410.
135. Raghuraman, G. K.; Dhamodharan, R.; Prucker, O.; Ruhe, J. Macromolecules 2008,
41, 873-878.
136. Bao, Z.; Bruening, M.; Baker, G. Macromolecules 2006, 39, 5251-5258.
137. Sanjuan, S.; Tran, Y. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 4305-4319.
138. Sanjuan, S.; Perrin, P.; Pantoustier, N.; Tran, Y. Langmuir 2007, 23, 5769-5778.
139. Carrot, G.; Diamanti, S.; Manuszak, M.; Charleux, B.; Vairon, J. J. Polym. Sci., Part
A: Polym. Chem. 2001, 39, 4294-4301.
140. Audouin, F.; Blas, H.; Pasetto, P.; Beaunier, P.; Boissiere, C.; Sanchez, C.; Save, M.;
Charleux, B. Macromol. Rapid Commun. 2008, 29, 914-921.
141. Li, D.; Jones, G.; Dunlap, J.; Hua, F.; Zhao, B. Langmuir 2006, 22, 3344-3351.
142. Save, M.; Granvorka, G.; Bernard, J.; Charleux, B.; Boissiere, C.; Grosso, D.;
Sanchez, C. Macromol. Rapid Commun. 2006, 27, 393-398.
49
 

143. Ramakrishnan, A.; Dhamodharan, R.; R., J. Macromol. Rapid Commun. 2002, 23,
612–616.
144. Jonas, A.; Hu, Z.; Glinel, K.; Huck, W. Nano Lett. 2008, 8, 3819-3824.
145. Iwata, R.; Suk-In, P.; Hoven, V.; Takahara, A.; Akiyoshi, K.; Iwasaki, Y.
Biomacromolecules 2004, 5, 2308-2314.
146. Dong, R.; Krishnan, S.; Baird, B.; Lindau, M.; Ober, C. Biomacromolecules 2007, 8,
3082-3092.
147. Cullen, S.; Liu, X.; Mandel, I.; Himpsel, F.; Gopalan, P. Langmuir 2008, 24, 913920.
148. Pyun, J.; Jia, S.; Kowalewski, T.; Patterson, G.; Matyjaszewski, K. Macromolecules
2003, 36, 5094–5104.
149. Goel, V.; Chatterjee, T.; Bombalski, L.; Yurekli, K.; Matyjaszewski, K.;
Krishnamoorti, R. J. Polym. Sci., Part B: Polym. Phys. 2006, 44, 2014-2023.
150. Savin, D. A.; Pyun, J.; Patterson, G. D.; Kowalewski, T.; Matyjaszewski, K. J.
Polym. Sci., Part B: Polym. Phys. 2002, 40, 2667-2676.
151. Ramakrishnan, A.; Dhamodharan, R.; Ruhe, J. J. Polym. Sci., Part A: Polym. Chem.
2006, 44, 1758-1769.
152. Iwata, R.; Satoh, R.; Iwasaki, Y.; Akiyoshi, K. Colloids Surf., B 2008, 62, 288-298.
153. Jakuczek, L.; Gutmann, J.; Muller, B.; Rosenauer, C.; Zuchowska, D. Polymer 2008,
49, 843-856.
154. Iwasaki, Y.; Omichi, Y.; Iwata, R. Langmuir 2008, 24, 8427-8430.
155. Jeyaprakash, J. D.; Samuel, S.; Dhamodharan, R.; Ruhe, J. Macromol. Rapid
Commun. 2002, 23, 277-281.
156. Li, J.; Sheng, X.; Zhao, B. J. Am. Chem. Soc. 2005, 127, 6248-6456.
50
 

157. Chen, X.; Armes, S.; Greaves, S.; Watts, J. Langmuir 2004, 20, 587-595.
158. Edmondson, S.; Vo, C.-D.; Armes, S. P.; Unali, G.-F. Macromolecules 2007, 40,
5271-5278.
159. Fulghum, T.; Estillore, N.; Vo, C.; Armes, S.; Advincula, R. Macromolecules 2008,
41, 429-435.
160. Saha, S.; Bruening, M.; Baker, G. Facile Appl. Mater. Interfaces 2011, 3, 3042–
3048.
161. Carlmark, A.; Malmstrom, E. Biomacromolecules 2003, 4, 1740-1745.
162. Lindqvist, J.; Malmstrom, E. J. Appl. Polym. Sci. 2006, 100, 4155-4162.
163. Singh, N.; Wang, J.; Ulbricht, M.; Wickramasinghe, S.; Husson, S. J. Membr. Sci.
2008, 309, 64-72.
164. Singh, N.; Chen, Z.; Tomer, N.; Wickramasinghe, S. R.; Soice, N.; Husson, S. J.
Membr. Sci. 2008, 311, 225-234.
165. Alem, H.; Duwez, A.; Lussis, P.; Lipnik, P.; Jonas, A.; Demoustier-Champagne, S. J.
Membr. Sci. 2008, 308, 75-86.
166. Chen, Y.; Liu, D.; Deng, Q.; He, X.; Wang, X. J. Polym. Sci., Part A: Polym. Chem.
2006, 44, 3434-3443.
167. Yao, F.; Fu, G..; Zhao, J.; Kang, E.; Neoh, K. J. Membr. Sci. 2008, 319, 149-157.
168. Jain, P.; Dai, J.; Grajales, S.; Saha, S.; Baker, G.; Bruening, M. Langmuir 2007, 23,
11360–11365.

51
 

Chapter 2 Incorporation of Long and Short PEO side chains in
polymer brushes to prevent crystallization and provide CO2/H2
selectivity

This work was performed in collaboration with Dr. Sebastian T. Grajales and
1

recently published in Chemistry of Materials. Dr. Grajales and I are equal coauthors of
this work.

2.1 Introduction
Selective removal of CO2 from gas streams is becoming increasingly important
2

3

4

for applications such as CO2 sequestration, fuel cell operation, and H2 synthesis.

In

the case of H2 production, steam reforming of methane produces 9 million tons of H2 per
5

year in the U.S. alone. However, the product stream from steam reforming theoretically
contains 20 mole% CO2, so purification of the H2 is an important and expensive part of
the production process.6 A number of studies examined membrane-based gas separation
as a possible alternative, or as a pre-purification step, to the pressure swing adsorption
7,8

process currently used in H2 purification.

The viability of membrane processes for

purifying H2 streams depends critically on achieving both high flux and high selectivity,
9

and a recent review summarizes the types of membranes employed for H2 purification.
52
 

Processes such as H2 recycle in ammonia synthesis have employed H2-selective
10

membranes for several decades.

In the case of CO2 removal from H2 streams, however,

CO2-selective membranes are attractive because purified H2 remains on the high pressure
side of the membrane and does not need to be recompressed.
The solution diffusion model for transport through membranes (equation 2.1)
illustrates how the flux, J, of a gas through a membrane depends on the pressure drop
across the membrane, Δp, the membrane thickness, l, and the gas permeability, P, which
11

is the product of the solubility coefficient, S, and diffusivity, D, for the gas of interest.

Equation 2.2 shows that the ideal selectivity of gas A over gas B, αA/B, depends on the
solubility coefficients and diffusivities of the two gases. Because of its small size, H2 has
a greater diffusivity than CO2, so CO2/H2 selectivity requires high solubility selectivity
for CO2 over H2. Fortunately, CO2 is the more condensable of the two gases, and
11

solubility selectivity typically favors CO2 transport.

J

Pp
P  SD
l

 A/ B 

(2.1)

PA S A  DA

PB S B  DB

(2.2)

Membrane materials employed for selective passage of CO2 over H2 should thus
have a higher affinity for CO2 than H2 along with a relatively high free volume that
53
 

minimizes the difference between CO2 and H2 diffusivities. One method of increasing
12, 13

CO2 solubility is the incorporation of protonated amines within the membrane.

In

this case, facilitated transport can lead to CO2/H2 selectivities over 100, but carrier
saturation limits the application of these systems to low feed pressures (less than 0.05
14-16

bar).

At room temperature, amorphous poly(ethylene glycol) (PEG)-containing

membranes exhibit CO2/H2 selectivities ranging from 5 to 12, presumably because of
quadrupole-dipole interactions between CO2 and the ether functional groups of PEG.
20

17-

Unfortunately, however, crystallization of PEG chains decreases both the permeability

and selectivity of such membranes.

To overcome this challenge, several studies

employed cross-linked, solution-cast, PEG-containing membranes to maintain a high free
8, 21-23

volume, decrease crystallinity, and increase chemical stability.

In the best case,

the cross-linked membranes show CO2 permeabilities of up to 600 Barrers [1 Barrer =
10

−10

3

2

cm (STP) •cm / (cm •s•cmHg) = 7.5×10

−18

3

2

m

(STP) •m/(m •s•Pa)] and
24

CO2/H2 selectivities of approximately 12 at room temperature.

Nevertheless, the high thickness (>100 μm) of solution-cast membranes yields a
permeance (permeance is defined as flux/∆p) too low for practical separations. This
work examines the use of atom transfer radical polymerization (ATRP) from porous
substrates to create composite membranes with thin (50-500 nm) skins of PEGcontaining films that allow selective removal of CO2 from H2 streams.
54
 

ATRP is

attractive for synthesizing these membrane skins because it often affords control over
25,26

film thickness along with polymer chains with relatively low polydispersity.

Scheme 2.1 shows two permutations of the ATRP process for forming membranes on
porous alumina substrates, and similar processes on porous polymer supports are also
possible. Initiator attachment occurs either via adsorption of a disulfide initiator to goldcoated alumina to generate A (scheme 2.1), or attachment of a silane directly to the
alumina to generate B.

Subsequent atom transfer radical polymerization with a

+

Cu /1,1,4,7,10,10-hexamethyltriethylenetetramine

(HMTETA)

catalyst

yields

the

membrane skin. The permeance of the modified alumina membranes is several orders of
magnitude greater than that of solution-cast membranes.9
Additionally, this work examines the use of copolymers prepared from
poly(ethylene glycol methyl ether methacrylate) (PEGMEMA) monomers with different
PEG chain lengths (Scheme 2.2) to inhibit the crystallization of the PEG side chains. The
combination of monomers with 8-9 and 23-24 ethylene oxide repeat units yields films
that do not crystallize over many months and membrane skins with a room temperature
CO2/H2 selectivity of around 13.

55
 

Scheme 2.1. Methods for growth of a poly(PEGMEMA) membrane from porous alumina.
For interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation.

56
 

Scheme 2.2. Copolymerization of PEGMEMA-475 and PEGMEMA-1100 to provide a
membrane skin.

2.2 Experimental
2.2.1

Chemicals and Materials
PEGMEMA monomers with Mn = 300 Da (PEGMEMA-300), Mn = 475 Da

(PEGMEMA-475), and Mn = 1,100 Da (PEGMEMA-1100) were obtained from Aldrich.
These monomers contain both 100 ppm 4-methoxyphenol and 300 ppm 2,6-di-tert-butyl4-methylphenol as inhibitors. Proton NMR spectra (Figure 2.1) show that the average
numbers of ethylene oxide units in PEGMEMA-300, PEGMEMA-475, and PEGMEMA1100 are 4.4, 8.8, and 23.1, respectively. In some cases, the inhibitors were removed
from the monomers using flash column chromatography with basic alumina, but this did
not have a significant effect on polymerizations, so in most reactions the monomers were
used

as

received.

CuCl2,

CuCl,

CuBr,

CuBr2,

and

1,1,4,7,10,10-

hexamethyltriethylenetetramine (HMTETA) were purchased from Aldrich, and deionized
water was obtained from Millipore system (Milli-Q, 18.2 MΩ•cm). THF was distilled
57
 

over sodium metal with benzophenone, and DMF was distilled over 4 Å molecular sieves.
Silicon (100) wafers were obtained from NOVA Electronic Materials and sputter-coated
with 20 nm of chromium followed by 200 nm of gold by LGA Thin Films (Santa Clara,
CA). The porous alumina substrates were Anodisc membrane filters (25 mm disks with
0.02 µm surface pores) purchased from Whatman, and the disulfide, [Br-C(CH3)2COO(CH2)11S]2, and silane, SiCl(CH3)2(CH2)11OCOC(CH3)2Br, initiators were
27-30

synthesized as described previously.

e

(A) PEGMEMA -300
 

g

h 

a,b

CH3 

H2C
O

O

g
O 

CH 3
n 

c,d,e,f 
h 
c
a

b 

d

f

 6.5         6.0        5.5        5.0        4.5        4.0         3.5       3.0         2.5        2.0          ppm
1.00            1.00                                2.03           2.24  1.94                                            2.76 
                                                                                   11.39 2.63 

Figure 2.1. Proton NMR spectra of (A) PEGMEMA-300, (B) PEGMEMA-475, and (C)
PEGMEMA-1100. The integrals of the peaks with a chemical shift between 3.4 ppm and
4.5 ppm (arising from the four hydrogen atoms on the ethylene oxide repeat units) were
summed and divided by 4 to give the average number of EO units per monomer. (The
intensity of the alkene protons was set to 1 for normalization.) Peak areas in these spectra
were averaged with areas from two additional replicate spectra to generate values for the
number of average ethylene oxide units per molecule.
58
 

Figure 2.1 (cont’d)

   11       10       9          8        7         6         5         4         3         2         1         0       ppm
     1.00  1.00        2.02  28.32 8.48      2.64
                                      2.63  1.87 

     9           8           7           6           5           4           3           2          1            0          ppm
1.00  1.01         2.27    89.41 3.66          5.18
                                    1.92  2.55 

 

59
 

2.2.2

Polymerization
Porous alumina substrates were initially rinsed with ethanol, dried with N2,

cleaned with UV/O3 (Boekel UV-Clean Model 135500) for 15 min, and sputter coated
with 5 nm of gold in a Pelco SC-7 sputter coater. The thickness of the gold coating was
monitored with a Pelco FTM-2 quartz crystal microbalance. Attachment of a monolayer
of the disulfide initiator to gold-coated Si wafers or porous alumina occurred during an
overnight immersion of the substrate in an ethanolic solution containing 1 mM disulfide
initiator. The resulting sample was rinsed with 5 mL of ethanol and dried in a N2 stream.
Attachment of the silane initiator occurred during overnight immersion of an uncoated,
UV/ozone-cleaned porous alumina substrate in a solution containing 8 mM silane in THF.
The resulting sample was rinsed with 5 mL THF and dried in a N2 stream. The two
different types of initiator-modified membranes were handled identically in the ensuing
polymerization steps.
Following a previous procedure, the catalyst stock solution was prepared by first
dissolving 0.06 g (0.6 mmol) of CuCl and 0.04 g (0.2 mmol) of CuBr2 in 30.0 mL of
31, 32

distilled, degassed DMF.

This solution was further degassed via three freeze, pump,

thaw cycles. In a N2-filled glove bag, 490 µL (1.8 mmol) of degassed 1,1,4,7,10,10hexamethyltriethylenetetramine (HMTETA) ligand was added to this solution, which
subsequently turned a dark green color during several hours of stirring. During this
initial stirring, a small amount of precipitation occurred, but the resulting catalytic

60
 

activity of the solution remained constant for at least several weeks. The nominal mole
+

2+

ratio of Cu :Cu :HMTETA was 3:1:9.
The monomer solution, which contained 0.75 M PEGMEMA in water, was
initially degassed via three freeze, pump, thaw cycles. (When different molecular weight
monomers were used, the total monomer concentration was maintained at 0.75 M). The
polymerization solution was prepared in a N2-filled glove bag by combining degassed
monomer and catalyst solutions in a 9:1 volume ratio to produce a blue-green solution
containing 0.67 M monomer, 2.0 mM CuCl, 0.60 mM CuBr2, and 6.0 mM HMTETA.
The initiator-modified membranes or wafers were immersed in the polymerization
solution for designated periods of time, and the resulting polymer films were rinsed with
5 mL water and soaked in water for at least two hours before rinsing with 5 mL of
ethanol and drying under a N2 stream.

2.2.3

Characterization methods
Reflectance FT-IR spectroscopy was performed using a Nicolet Magna-IR 560

spectrometer with a Pike grazing angle (80°) attachment. A UV/ozone-cleaned, goldcoated Si wafer was used to obtain the background spectrum. Thicknesses of films
formed on gold-coated silicon were determined with a rotating analyzer ellipsometer (J.A.
Woollam model M44) at an incident angle of 75°. A two-term Cauchy equation was
employed to simultaneously fit film thickness and the Cauchy constants needed to model
the wavelength-dependence of the film refractive index. Thicknesses were determined at
a minimum of three locations on each substrate. Scanning electron microscopy was
61
 

performed with the JEOL semi-in-lens cold cathode field-emission scanning electron
microscope, model JSM-7500F, operating with a r-filter in signal maximum mode.
Samples were sputter coated with 5 nm of gold prior to imaging.
2.2.4

Gas Permeation
Single-gas permeation experiments were performed by exposing the membrane to

individual gases at varying pressures in an Advantec/MFS UHP-25 cell with a pressure
relief valve. The gas flow rate through the membrane was measured with either an
Optiflow 420 electronic soap bubble meter or a manual soap bubble meter. Mixed-gas
experiments were performed by loading the membrane into a custom membrane holder
(all connections utilized Swagelok fittings and were tested to ensure that they maintained
pressure over a time-scale longer than the experiment) that allowed cross-flow of the feed
gas as well as a sweep gas on the permeant side. A backpressure valve was employed to
sustain the feed gas pressure, and the feed flow rate was high enough to maintain constant
composition at the face of the membrane (the stage-cut, or ratio of permeant flow to feed
flow, was <1%). The N2 sweep gas/permeant stream was connected to an automated sixport injector valve on a Hewlett-Packard 6890 GC equipped with a thermal conductivity
detector (TCD) and an Agilent GS-CarbonPLOT capillary column (i.d. = 0.53 mm,
length = 30 m, 3 µm coating).

The mixed gas permeances as well as pure gas

permeances were calculated from GC results using equation 2.3,

Permeance 

B   N

2

 N  A  Δp
2

62
 

(2.3)

where χB is the mole fraction of a particular permeate gas, B, in the gas mixture injected
into the GC column, ΦN2 is the sweep gas flow rate, χN2 is the mole fraction of N2 in the
gas injected into the GC column, Δp is the transmembrane partial pressure difference for
the gas of interest, and A is the membrane area. (ΦN2/ χN2 is the sum of the permeate and
sweep gas flow rate, and χN2 was determined by subtracting the mole fractions of the
permeates from unity. This assumes that the amount of N2 passing from permeate to feed
was negligible.) The sweep gas flow rate was programmed into a mass flow controller,
pressure and area were measured, and the mole fractions were determined from the
integrated GC spectra and a calibration curve.

2.3 Results and Discussion
2.3.1

Polymerization Rates as a Function of Monomer Composition
The first step in developing surface-initiated ATRP for creating membrane skins,

especially for forming copolymer skins, is examination of polymerization from model flat
substrates such as gold-coated Si wafers.

33

Such substrates facilitate ellipsometric and

spectroscopic characterization of polymer growth and composition. We are particularly
interested in the relative growth rates of PEGMEMA-475 and PEGMEMA-1100 because
the short side chains of PEGMEMA-475 do not crystallize, whereas the long chains of
PEGMEMA-1100 should promote CO2/H2 selectivity.

34

As Figure 2.2 shows,

polymerization of PEGMEMA-475 by itself gives the most rapid initial growth in film
63
 

thickness, but the growth rate decreases dramatically after 15 min of polymerization. In
contrast, polymerization of PEGMEMA-1100 is slower, but the growth rate is essentially
constant for 120 min. Interestingly, polymerizations of mixtures in which PEGMEMA475 constitutes either 25 or 75 mole% of the total monomer show relatively high and
steady growth rates.

 

Figure 2.2. Evolution of film thickness with polymerization time for pure and mixed
monomer solutions. For copolymers, the legend shows the solution PEGMEMA-475
mole percentage with respect to the total monomer (PEGMEMA-475 plus PEGMEMA64
 

1100). The overall monomer concentration was 0.67 M, and the catalyst system was 2.0
mM CuCl, 0.60 mM CuBr2, and 6.0 mM HMTETA. The error bars, which in many
cases are obscured by the symbols, represent one standard deviation.
Decreases in growth rates with polymerization time, particularly for
poly(PEGMEMA-475), suggest that the HMTETA-Cu

+/2+

catalyst system gives a high

concentration of radicals that results in significant termination by radical-radical coupling.
The large side chains of PEGMEMA-1100 may provide steric hindrance to both
polymerization and termination. In solution polymerization studies, Haddleton et al.
attributed the high ATRP activity of PEGMEMA-475 relative to PEGMEMA-1100 to
35

ethylene oxide coordination with the Cu catalyst in close proximity to the double bond.
34

However, steric issues seem equally likely to affect polymerization.

In the case of

homopolymerization, the PEGMEMA-1100 solutions are also more viscous and contain
less water than PEGMEMA-475 solutions, and both of these factors should also lead to
lower polymerization rates for the PEGMEMA-1100. Mixtures of PEGMEMA-475 and
PEGMEMA-1100 have intermediate polymerization rates, and the side chains are
apparently still sufficiently large to reduce termination relative to pure PEGMEMA-475.
For PEGMEMA-475, PEGMEMA-1100, and their mixtures, film growth is rapid
compared

to

polymerizations.

polymerization
36-38

of

most

monomers

and

non-aqueous

A number of studies demonstrated that the presence of water
39, 40

frequently enhances the rate of ATRP.

65
 

other

2.3.2

Determining the Composition of Copolymer Films
The molar ratio of PEGMEMA-475 to PEGMEMA-1100 in a film may not

correspond to the monomer ratio in the polymerization solution. In particular, the smaller
monomer may be more reactive because of faster diffusion to the surface or less steric
hindrance to polymerization. In an attempt to determine the relative amounts of each
monomer incorporated in poly(PEGMEMA-475-co-PEGMEMA-1100) films, we
-1

-1

compared the intensities of carbonyl (1731 cm ) and C-O-C (1149 cm ) stretches in
both homopolymer and copolymer films. The reflectance FTIR spectra in Figure 2.3
show that for amorphous homopolymers, the intensity of the C-O-C stretch relative to the
C=O stretch is much higher for poly(PEGMEMA-1100) than poly(PEGEMEMA-475), as
we expected because of the longer PEG chain in the higher molecular weight monomer.

66
 

2000

0.05

C=O

Absorbance

poly(PEGMEMA-1100)

C-O-C

poly(PEGMEMA-475)

1750

1500

1250

1000

Wavenumbers (cm-1)

750

 

Figure 2.3. FTIR spectra of 152 nm-thick poly(PEGMEMA-475) (black line) and 118
nm-thick poly(PEGMEMA-1100) (dashed red line) films on gold-coated substrates.
Spectra were taken immediately after polymerization before the PEG side chains in
poly(PEGMEMA-1100) crystallized. The spectra are offset for clarity.
We initially planned to estimate the mole ratios of monomers incorporated into a
copolymer film using equation 2.4, where χ475 is the mole fraction of PEGMEMA-475 in
a copolymer film and Rco, R1100, and R475 are the ratios of the C-O-C peak height (1149
-1

-1

cm ) to the C=O peak height (1731 cm ) for copolymer, poly(PEGMEMA-1100), and
poly(PEGMEMA-475) films, respectively.

Rco  x475 R475  (1  x475 ) R1100

(2.4)

However, Figure 2.4 suggests that in some cases peak height and peak area ratios for the
C-O-C to C=O absorbance increase with thickness for homopolymer and copolymer

67
 

C-O-C / C=O Peak Height Ratio

8

18

(A)

(B)

16
C-O-C / C=O Peak Area Ratio

poly(PEGMEMA-1100)
25 % PEGMEMA-475
50 % PEGMEMA-475
75 % PEGMEMA-475
poly(PEGMEMA-475)
poly(PEGMEMA-300)

10

6

4

2

14
12
10
8
6
4
2

0

0

0

100
200
300
Dry Thickness (nm)

400

0

100
200
300
Dry Thickness (nm)

400

 

Figure 2.4. Ratios of C-O-C to C=O peak heights (A) and peak areas (B) in reflectance
FTIR spectra of poly(PEGMEMA-1100), poly(PEGMEMA-475), poly(PEGMEMA-300)
and poly(PEGMEMA-475-co-PEGMEMA-1100) films on gold-coated substrates. The
ratios are plotted as a function of the ellipsometric thickness of the film, and for the
copolymers, the legend shows the mole fraction of PEGMEMA-475 (relative to total
monomer) in the polymerization solution. All polymerizations employed a total
monomer concentration of 0.67 M, and the experimental section describes the
polymerization conditions.

films. This increasing ratio of C-O-C to C=O absorbance likely occurs because
reflectance measurements plot log(Ro/R), where Ro and R are the reflectivities of bare
and film-covered substrates, respectively. These reflectivities are complicated functions

of the film thickness and the complex refractive indices, n  n  ik , of the substrate and

the film. To account for these complications in reflectance FTIR spectroscopy, it is
necessary to use Fresnel calculations to determine the absorption coefficient, k, for
different films at the wavelengths of interest.

68
 

We plot reflectance IR spectra as log(Ro/R), where Ro is the fraction of the
incident light intensity reflected by a bare gold substrate (Figure 2.5A) and R is the
fraction of incident light intensity reflected by a film-coated substrate (Figure 2.5B). The
value of Ro is simply the Fresnel reflection coefficient, r13, multiplied by its complex
conjugate (equation 2.5), where the Fresnel coefficient is given by equation 2.6. In

~
equation 2.6, n 3 is the complex refractive index of gold, n-ik, (assumed to be 3-30i), θ1
is the angle of incidence (80 º), n1 is the refractive index of air (assumed to be 1.000), and

~

~

 au is the complex angle of refraction in the gold film. The value of  au was
calculated using Snell’s law, equation 2.7. (Tilda superscripts indicate a complex number,
and the IR beam was polarized with the electric field parallel to the substrate.)

 
Ro  r13 r13*

(2.5)



n3 cos 1  n1 cos  Au
r13 


n3 cos 1  n1 cos  Au

(2.6)



n1 sin 1  n3 sin  Au

(2.7)

69
 

A

B

Air, n1

Io

IR
θ1

Air, n1

Io

IR

θ1

~
Film, n2

~
Au, n3

...

~
Au, n3
Figure 2.5. Schematic drawing of the reflection of light from (A) a bare gold surface and
(B) a film-covered gold surface

Equation 2.8 allows calculation of the reflectance, R, of the film-covered substrate,
where the * represents the complex conjugate. In this equation,

~
r12

is the Fresnel

~
reflection coefficient for the air-film interface (equation 2.9), r23 is the Fresnel reflection
coefficient for the film-gold interface (equation 2.10), λ is the wavelength of the incident

~
light in air, n2 is the complex refractive index of the film, l is the film thickness, and cos
~

2

~

and cos  3 can be calculated using Snell’s law for the appropriate interfaces.

 r12  r23e 2i  r12  r23e 2i *
 
 
R

2 i 
 
  2i
 1  r12 r23e  1  r12 r23e 

2 n2

l cos  2

n1

(2.8)

r12 



n2 cos 1  n1 cos  2


n2 cos 1  n1 cos  2

(2.9)

r23 




n3 cos  2  n2 cos 3



n3 cos  2  n2 cos 3

(2.10)
70

 

To calculate the absorption coefficient, k, for a specific polymer film, we
-1

determined the absorbance peak heights at 1731 cm

-1

(C=O stretch) and at 1149 cm

(C-O-C stretch) from reflectance IR spectra. It is important to note that the presence of a
film on the gold substrate changes the reflectivity of the surface even in the absence of
absorption (when k=0), meaning that the film-covered gold gives rise to a baseline shift
that must be taken into consideration when calculating peak heights. Specifically, this
correction was made by calculating the absorbance when k=0 and subtracting this from
the calculated peak absorbance to generate a corrected, calculated peak height. This
correction was used along with the above equations to iteratively calculate the absorption
coefficient, k, that gives rise to the measured peak height value for a particular film. The
ratios of calculated absorption coefficients, kC-O-C/kC=O, were reasonably independent of
thickness for a given film composition, as Figure 2.6 shows.

71
 

20

poly(PEGMEMA- 1100)
25 % PEGMEMA- 475
50 % PEGMEMA- 475
75 % PEGMEMA- 475
poly(PEGMEMA- 475)
poly(PEGMEMA- 300)

kC-O-C / kC=O

15

10

5

0
0

100
200
300
Dry Thickness (nm)
-1

400
-1

Figure 2.6. Ratios of the C-O-C (1149 cm ) and C=O (1731 cm ) absorption coefficients
in poly(PEGMEMA) homo and copolymer films as a function of film thickness. In the case
Figure S2: .
of copolymer films, the legend shows the PEGMEMA-475 mole percentage of total
monomer in the polymerization solution, and the rest of the monomer was PEGMEMA1100. The polymerization was performed as described in the experimental section

Table 2.1. Absorption coefficients, k, of homopolymer films. The k values were
calculated using equations 2.5-2.10 and the peak absorbances corresponding to C-O-C
and C=O stretches. The uncertainty is one standard deviation of the values obtained from
films with a range of thicknesses

PEGMEMA-1,100

0.30 ± 0.08

1731

C=O

0.14 ± 0.03

C-O-C

0.42 ± 0.05

1731

C=O

0.11 ± 0.01

1149

PEGMEMA-475

C-O-C

1149

PEGMEMA-300

Absorption
Coefficient, k

1149

Homopolymer

Bond
Stretch

C-O-C

0.48 ± 0.08

1731

C=O

0.045 ± 0.006

-1

Wavenumber, cm

72
 

We used the absorption coefficients k values (see Table 2.1) for three
homopolymer

films,

poly(PEGMEMA-300),

poly(PEGMEMA-475),

and

poly(PEGMEMA-1100), to create a calibration curve (Figure 2.7) of the kC-O-C / kC=O
ratio versus the average number of ethylene oxide units per repeat unit, EOavg, and then
calculated EOavg in copolymer films. For poly(PEGMEMA-475-co-PEGMEMA-1100),
this EOavg value is a direct function of the fraction of each monomer in the copolymer
and the average number of ethylene oxide units in each monomer (equation 2.11).

kC-O-C / kC=O

EOavg  8.8 x475  23.1(1  x475 )                                

(2.11)

14
12
10
8
6
4
2
0
0

5
10
15
20
25
Ethylene Oxide Units per
Monomer
 
-1

-1

Figure 2.7. Ratios of C-O-C (1149 cm ) and C=O (1731 cm ) absorption coefficients, k,
as a function of the average number of ethylene oxide units per repeat unit, EOavg, in
poly(PEGMEMA-300), poly(PEGMEMA-475), and poly(PEGMEMA-1100) films. The
ethylene oxide units per monomer were calculated from 1H-NMR spectra of monomers
(Figure 2.1), and the standard deviations in the absorption coefficient ratios were
determined using films with thicknesses ranging from 5 to 454 nm. The line is a fit to the
data with a zero intercept.

73
 

The reflectance FTIR data and the Fresnel calculations show that the ratios of
PEGMEMA-475 to PEGMEMA-1100 in poly(PEGMEMA-475-co-PEGMEMA-1100)
films are approximately the same as those in the solutions from which they grow (Table
2.2). This is somewhat surprising considering the large differences in the initial growth
rates as well as termination rates for surface-initiated homopolymerization of the two
monomers (Figure 2.2). However, in the case of copolymers, solutions containing 25
mole% and 75 mole% PEGMEMA-475 give essentially the same initial growth rate
(Figure 1), which is consistent with the similarity between solution and film compositions.
This direct correlation between solution and film compositions affords straightforward
control over the ratio of long and short side chains to minimize PEG crystallization and
maximize CO2/H2 selectivity in membranes.
a

Table 2.2. Composition of poly(PEGMEMA-475-co-PEGMEMA-1100) films as a
b

function of the PEGMEMA-475 mole percentages in the monomer solution.

 

Monomer Solution, mole%

Polymer Composition, mole%

PEGMEMA-475

PEGMEMA-475

0

0

23 ± 1

25

33 ± 2

18 ± 1

50

54 ± 3

15 ± 1

75

71 ± 7

13 ± 1

100

100

8.7 ± 0.6

a

EOavg

Film compositions were determined from reflectance FTIR spectra and Fresnel
calculations and are expressed as both mole percent and the average number of ethylene
74

 

oxide (EOavg) groups per repeat unit. The uncertainties, which were calculated using
propagation of error, represent one standard deviation.
b

The mole percentage in the monomer solution is relative to the total amount of monomer
in solution.

2.3.3

Crystallization of PEG Side Chains
PEG chains crystallize in helices that contain two turns per 7 EO units over a
41-43

length of 19.3 Å,

-1

and helix formation leads to splitting of the 1150 cm
-1

band into peaks at 1120 and 1148 cm
-1 41, 42

1360 cm .

C-O-C IR

and a narrowing of the CH2 wagging peak at

In PEGMEMA films, long PEG side chains crystallize similarly, and

Figure 2.8A demonstrates the changes in reflectance FTIR spectra that occur upon
crystallization of a poly(PEGMEMA-1100) film. Previous reports of PEG crystallization
in related coatings suggest that the helical axis preferentially orients perpendicular to the
surface in films less than 100 nm thick, and both perpendicular and parallel in thicker
films.

28, 44-47

75
 

poly(PEGMEMA-1100) (freshly fresh
poly(PEGMEMA-1100) prepared)
poly(PEGMEMA-1100) (1 day 1 day
poly(PEGMEMA-1100)later)

25 % PEGMEMA-475 (freshly prepared)
25% PEGMEMA-475 fresh
25 % PEGMEMA-475 (404 404 days
25% PEGMEMA-475 days later)

(B)
(B)

1600
1600

Absorbance

Absorbance

(A)
(A)

0.02
0.02

1400
1400

1200
1200

1000
1000

800 1600
800 1600

-1

Wavenumbers (cm-1)

0.02
0.02

1400
1400

1200
1200

1000
1000

Wavenumbers (cm-1)

Wavenumbers (cm )

800
800

-1

Wavenumbers (cm )

 

Figure 2.8. Reflectance FTIR spectra of (A) 60 nm-thick poly(PEGMEMA-1100)
immediately after polymerization (solid line) and 24 hours later (dashed line) and (B) 97
nm-thick poly(PEGMEMA-475-co-PEGMEMA-1100) immediately after polymerization
(solid line) and 404 days later (dashed line). The copolymer was prepared from a
solution containing 25 mole% PEGMEMA-475 and 75 mole% PEGMEMA-1100.

However, the presence of crystallinity, rather than its orientation, is the primary
concern for creating selective membranes with high permeabililties, and prior studies
reveal that the PEG chain length, not film thickness, is the primary factor in determining
48,

whether crystallization occurs.

49

In contrast to poly(PEGMEMA-1100),

poly(PEGMEMA-475) films do not show any evidence of crystallization in reflectance
FTIR spectra, presumably because neighboring PEG side chains are too short to
50

crystallize in this environment.

76
 

Importantly, copolymers prepared from PEGMEMA-1100 and PEGMEMA-475
show no sign of crystallization, even when the fraction of PEGMEMA-475 in the film is
only ~25%.

Figure 2.8B shows that the reflectance FTIR spectrum of a

poly(PEGMEMA-475-co-PEGMEMA-1100) film remains essentially unchanged even
after more than a year at room temperature. The presence of the short chains is apparently
sufficient to inhibit crystallization of the longer side chains, and this effect occurs for
films prepared from solutions containing PEGMEMA-475 as 25, 50, and 75 mole% of
the total monomer.

2.3.4

Formation of Gas-Separation Membranes by ATRP
Composite membranes are attractive for gas separations because the minimal

thickness of the thin, selective skin allows for high flux, and the underlying support
provides mechanical strength. Hence, the goal in using ATRP to create a membrane skin
is to completely cover a porous substrate with a thin film without filling the underlying
pores. The first step in the formation of membranes by ATRP is attachment of an
initiator to the surface. To be consistent with our work on gold wafers, we sputtered 5
nm of gold on porous alumina and subsequently adsorbed a monolayer of the disulfide
initiator on the surface (Scheme 2.1A). In a separate procedure, to avoid the deposition
of gold on the membrane surface, we also attached a silane initiator directly to the
alumina (Scheme 2.1B).
Subsequent polymerization of PEGMEMA from initiators on alumina yields the
desired composite membranes. Figure 2.9 shows a SEM image of a composite membrane
77
 

containing a poly(PEGMEMA-475-co-PEGMEMA-1100) skin on a porous alumina
support. The film is thin (130 nm) and clearly covers the pores without filling them.
However, selective gas permeation studies are needed to demonstrate that the films are
defect-free and non-crystalline.

 

Figure 2.9. SEM image of the cross section of a gold-coated porous alumina membrane
modified with a poly(PEGMEMA-475-co-PEGMEMA-1100) film. The alumina exhibits
unfilled vertical pores covered by a 130 nm-thick film that was grown from an aqueous
solution containing 25 mole% PEGMEMA-475 and 75 mole% PEGMEMA-1100.
2.3.5

Gas Permeation through Poly(PEGMEMA) Membranes
Figure 2.10 shows the CO2 and H2 fluxes through porous alumina membranes

coated with either a poly(PEGMEMA-1100) film or a copolymer film grown from a
solution containing a 1:1 mole ratio of PEGMEMA-475 and PEGMEMA-1100. For

78
 

poly(PEGMEMA-1100), the initial CO2/H2 selectivity is 12 ± 1. The highest roomtemperature CO2/H2 selectivity of PEG-containing membranes in the literature is also
12.

8, 22-24

However, the CO2/H2 selectivity of the poly(PEGMEMA-1100) film drops to

less than 1 within a day because the CO2 flux decreases 12-fold. This drop in CO2 flux
is almost certainly due to crystallization of the PEG side chains, which as noted above
occurs within one day of synthesis of poly(PEGMEMA-1100) films.

79
 

Figure 2.10. Single-gas fluxes through gold-coated porous alumina membranes capped
with (A) poly(PEGMEMA-1100) or (B) poly(PEGMEMA-475-co-PEGMEMA-1100)
grown from a solution containing 50 mole% PEGMEMA-475. (Fluxes are normalized
by multiplying by the SEM thickness of the membrane skin.) The circles represent CO2
data, and the triangles represent H2 data; filled symbols represent measurements taken
immediately after synthesis, and open symbols represent data obtained 24 h later. The
inset in Figure A shows the data on an expanded ordinate for clarity.
 

 

80
 

Figure 2.10 (cont’d)

 
 

By comparison, poly(PEGMEMA-475) films exhibit an initial CO2/H2 selectivity
of 9 ± 2. On silylated supports, poly(PEGMEMA-1100) and poly(PEGMEMA-475)
films show selectivities of 13 ± 1 and 5.8 ± 0.5, respectively, demonstrating that there is a
81
 

significant difference between the gas selectivities of these two polymers.

The

poly(PEGMEMA-475) films contain a smaller overall fraction of PEG than
poly(PEGMEMA-1100), which likely explains their lower selectivity. However, the
selectivity of the poly(PEGMEMA-475) does not decrease with time because the shorter
chains do not crystallize.
The goal of using copolymers as membranes is to achieve the initial selectivity of
poly(PEGMEMA-1100) films while avoiding decreases in selectivity and flux due to film
crystallization. Figure 2.10B shows that a poly(PEGMEMA-475-co-PEGMEMA-1100)
film (prepared on gold-coated alumina using a solution containing 50 mole%
PEGMEMA-475) has an initial CO2/H2 selectivity of 12-13, which is essentially the
same as that of the pure poly(PEGMEMA-1100).

Moreover, the selectivity of the

copolymer membranes is constant over at least several weeks at room temperature, again
presumably because the side chains of PEGMEMA-475 inhibit crystallization. Figure
2.11 shows the dependence of CO2/H2 selectivity on the mole fraction of PEGMEMA475 in poly(PEGMEMA-475-co-PEGMEMA-1100) films. As long as there is 75 mole%
or less PEGMEMA-475 in the polymerization solution, membranes formed on goldcoated alumina maintain average CO2/H2 selectivities of 12 at room temperature. The
films grown from silylated alumina show a maximum CO2/H2 selectivity of 17 ± 2 with
copolymers generated from a 50 mole% PEGMEMA-475 polymerization solution, but a
selectivity of only 7 when using 75 mole% PEGMEMA-475. Silanization is more likely
to produce initiators within the pores of the alumina, which might somewhat alter film
structure at the support surface and change selectivity relative to films on gold-coated
82
 

alumina. Moreover, the initiator packing densities might be different for the gold-coated
and silylated surfaces.

 

Figure 2.11. Single-gas CO2/H2 selectivity vs. mole fraction of PEGMEMA-475 in the
monomer solution used to create amorphous poly(PEGMEMA-475-co-PEGMEMA-1100)
films on porous alumina membranes. The triangles represent films grown from silylated
alumina, and the squares represent films grown from initiators on gold-coated alumina.
For the poly(PEGMEMA-1100) homopolymer on silylated alumina, the open circle near
the origin represents selectivity after crystallization. Error bars are one standard
deviation.
 

Determination of the permeability, P, of the films on porous alumina is
challenging because of the difficulty in obtaining accurate values for film thickness (see
equation 2.1). Film thicknesses in multiple cross-sectional SEM images of a given
membrane typically show a relative standard deviation (rsd) of 10-30%, although
sometimes we see even larger variations. Based on SEM images of membranes with a
83
 

rsd in thickness <25%, the H2 permeability of poly(PEGMEMA-475-co-PEGMEMA1100) films ranges from 2 to 6 Barrers, and the CO2 permeability ranges from 20 to 60
Barrers.

(Given the uncertainty in thickness, we could not differentiate among

permeabilities of films with different compositions.) Typical CO2 permeabilities of
PEG-containing films range from 12 Barrers in pure PEG to 570 Barrers in cross-linked
PEG-based acrylates.

50, 52

This wide range of permeabilities in different PEG-

containing polymers stems from differences in fractional free volume and the presence of
24, 53, 54

different polar groups that alter the solubility of gases in the membrane.

Although the poly(PEGMEMA-475-co-PEGMEMA-1100) films do not exhibit as high a
permeability as some cross-linked films, the ultrathin skin maximizes permeance for a
given permeability.
Selectivities in gas mixtures can differ significantly from those with single gases,
particularly when absorption of a gas such as CO2 plasticizes the membranes.

55

In

mixed-gas experiments, plasticization may increase the flux of H2 and decrease CO2/H2
selectivity relative to single-gas measurements. With feed gas streams of 79.7 mole% H2
and 20.3 mole% CO2, the CO2/H2 selectivity was only 13% lower on average than the
pure gas CO2/H2 selectivities.

84
 

2.4 Summary
In their amorphous state, poly(PEGMEMA-1100) membranes provide high CO2
permeabilities and CO2/H2 selectivities, but crystallization can decrease flux by an order
of magnitude and reduce CO2/H2 selectivity to below 1. Copolymerization of two
PEGMEMA monomers, one containing 8-9 and the other 23-24 ethylene oxide units per
side chain, avoids the problem of crystallization and yields films that maintain an
amorphous state for more than a year. Reflectance FTIR spectra indicate that the ratio of
the PEGMEMA-475 and PEGMEMA-1100 incorporated into copolymer films is within 8%
of the ratio of the monomers in the polymerization solution. Membranes consisting of
porous alumina supports coated with copolymer films exhibit stable single-gas CO2/H2
selectivities around 12 at room temperature, which is attractive for CO2 removal from H2
streams.
2.5 Acknowledgements & Copyright
We thank the U.S. Department of Energy Office of Basic Energy Sciences for
funding and S. Koszalka and Dr. Kathryn Severin for their assistance.
Reprinted (adapted) with permission from (Grajales, S.; Dong, X.; Zheng, Y.;
Baker, G.; Bruening. M.; Chem. Mater., 2010, 22, 4026–4033.). Copyright (2011)
American Chemical Society. See Appendice I for copyright permission.

85
 

REFERENCES

86
 

References
1. Grajales, S.; Dong, X.; Zheng, Y.; Baker, G.; Bruening. M.; Chem. Mater., 2010, 22,
4026–4033.
2. Scholes, C.; Kentish, S.; Stevens, G. Sep. Purif. Rev. 2009, 38, 1-44.
3. Prakash, S.; Kohl, P. A. J. Power Sources 2009, 192, 429-434.
4. Shao, L.; Low, B. T.; Chung, T. S.; Greenberg, A. R. J. Membr. Sci. 2009, 327, 18-31.
5. Department of Energy. U.S. Energy Information Administration: Annual Energy
Review. http://www.eia.doe.gov/emeu/aer/pdf/pages/sec1_13.pdf (accessed Sep 29,
2009).
6. Molburg, J. C.; Doctor, R. D. Proc. - Annu. Int. Pitt. Coal Conf. 2003, 20, 1388-1408.
7. Favre, E. J. Membr. Sci. 2007, 294, 50-59.
8. Lin, H. Q.; Van Wagner, E.; Freeman, B. D.; Toy, L. G.; Gupta, R. P. Science 2006,
311, 639-642.
9. Ockwig, N. W.; Nenoff, T. M. Chem. Rev. 2007, 107, 4078-4110.
10. Perry, E. Process for the Recovery of Hydrogen from Ammonia Purge Gases. U.S.
Patent 4,172,885, Oct 30, 1979.
11. Kesting, R. E.; Fritzsche, A. K. Polymeric Gas Separation Membranes. John Wiley
and Sons, Inc.: New York City, NY, 1993.
12. Matsuyama, H.; Teramoto, M.; Matsui, K.; Kitaura, Y. J. Appl. Polym. Sci. 2001, 81,
936-942.
13. Matsuyama, H.; Teramoto, M.; Sakakura, H.; Iwai, K. J. Membr. Sci. 1996, 117, 251260.

87
 

14. Kim, M.; Park, Y.; Youm, K.; Lee, K. J. Membr. Sci. 2004, 245, 79-86.
15. Noble, R. D.; Pellegrino, J. J.; Grosgogeat, E.; Sperry, D.; Way, J. D. Sep. Sci.
Technol. 1988, 23, 1595-1609.
16. Way, J. D.; Noble, R. D. Facilitated Transport. In Membrane Handbook; Ho, W. S.
W.; Sirkar, K. K., Eds.; Van Nostrand Reinhold: New York, 2001; pp 833-866.
17. Koros, W. J. J. Polym. Sci., Part B: Polym. Phys. 1985, 23, 1611-1628.
18. Zhong, J.; Lin, G.; Wen, W.-Y.; Jones, A. A.; Kelman, S.; Freeman, B. D.
Macromolecules 2005, 38, 3754-3764.
19. Bondar, V. I.; Freeman, B. D.; Pinnau, I. J. Polym. Sci., Part B: Polym. Phys. 2000,
38, 2051-2062.
20. Kusuma, V. A.; Matteucci, S.; Freeman, B. D.; Danquah, M. K.; Kalika, D. S. J.
Membr. Sci. 2009, 341, 84-95.
21. Patel, N. P.; Miller, A. C.; Spontak, R. J. Adv. Funct. Mater. 2004, 14, 699-707.
22. Kalakkunnath, S.; Kalika, D. S.; Lin, H.; Freeman, B. D. Macromolecules 2005, 38,
9679-9687.
23. Priola, A.; Gozzelino, G.; Ferrero, F.; Malucelli, G. Polymer 1993, 34, 3653-3657.
24. Lin, H.; Freeman, B. D.; Kalakkunnath, S.; Kalika, D. S. J. Membr. Sci. 2007, 291,
131-139.
25. Balachandra, A. M.; Baker, G. L.; Bruening, M. L. J. Membr. Sci. 2003, 227, 1-14.
26. Sun, L.; Baker, G. L.; Bruening, M. L. Macromolecules 2005, 38, 2307-2314.
27. Shah, R. R.; Merreceyes, D.; Husemann, M.; Rees, I.; Abbott, N. L.; Hawker, C. J.;
Hedrick, J. L. Macromolecules 2000, 33, 597-605.

88
 

28. Zheng, Y.; Bruening, M. L.; Baker, G. L. Macromolecules 2007, 40, 8212-8219.
29. Matyjaszewski, K.; Miller, P. J.; Shukla, N.; Immaraporn, B.; Gelman, A.; Luokala,
B. B.; Siclovan, T. M.; Kickelbick, G.; Vallant, T.; Hoffmann, H.; Pakula, T.
Macromolecules 1999, 32, 8716-8724.
30. Bao, Z.; Bruening, M. L.; Baker, G. L. Macromolecules 2006, 39, 5251-5258.
31. Kim, J.-B.; Huang, W.; Miller, M. D.; Baker, G. L.; Bruening, M. L. J. Polym. Sci.,
Part A: Polym. Chem. 2003, 41, 386-394.
32. Cheng, N.; Azzaroni, O.; Moya, S.; Huck, W. T. S. Macromol. Rapid Commun. 2006,
27, 1632-1636.
33. Zhou, F.; Zheng, Z.; Yu, B.; Liu, W.; Huck, W. T. S. J. Am. Chem. Soc. 2006, 128,
16253-16258.
34. Brown, A. A.; Khan, N. S.; Steinbock, L.; Huck, W. T. S. Eur. Polym. J. 2005, 41,
1757-1765.
35. Lad, J.; Harrisson, S.; Haddleton, D. M. Mechanistic Aspects of Copper-mediated
Living Radical Polymerization. In Advances in Controlled/Living Radical
Polymerization; Matyjaszewski, K., Ed.; American Chemical Society: Washington, DC,
2003; Vol. 854; pp 148-160.
36. Huang, W.; Kim, J.-B.; Bruening, M. L.; Baker, G. L. Macromolecules 2002, 35,
1175-1179.
37. Wang, X. S.; Armes, S. P. Macromolecules 2000, 33, 6640-6647.
38. Wang, X. S.; Lascelles, S. F.; Jackson, R. A.; Armes, S. P. Chem. Commun. 1999,
1817-1818.
39. Jain, P.; Dai, J.; Grajales, S.; Saha, S.; Baker, G. L.; Bruening, M. L. Langmuir 2007,
23, 11360-11365.
40. Tsarevsky, N. V.; Matyjaszewski, K. Chem. Rev. 2007, 107, 2270-2299.
89
 

41. Miyazawa, T.; Ideguchi, Y.; Fukushima, K. J. Chem. Phys. 1962, 37, 2764.
42. Yoshihara, T.; Murahashi, S.; Tadokoro, H. J. Chem. Phys. 1964, 41, 2902.
43. Takahashi, Y.; Tadokoro, H. Macromolecules 1973, 6, 672-675.
44. Li, X.; Hsu, S. L. J. Polym. Sci., Part B: Polym. Phys. 1984, 22, 1331-1342.
45. Schönherr, H.; Frank, C. W. Macromolecules 2003, 36, 1188-1198.
46. Schönherr, H.; Frank, C. W. Macromolecules 2003, 36, 1199-1208.
47. Wang, H. P.; Keum, J. K.; Hiltner, A.; Baer, E.; Freeman, B.; Rozanski, A.; Galeski,
A. Science 2009, 323, 757-760.
48. Bitter, J. G. A. Desalination 1984, 51, 19-35.
49. Preston, W. E.; Barlow, J. W.; Paul, D. R. J. Appl. Polym. Sci. 1984, 29, 845-852.
50. Lin, H.; Kai, T.; Freeman, B. D.; Kalakkunnath, S.; Kalika, D. S. Macromolecules
2005, 38, 8381-8393.
51. Lin, H.; Freeman, B. D. J. Membr. Sci. 2004, 239, 105-117.
52. Lin, H.; Freeman, B. D. Macromolecules 2005, 38, 8394-8407.
53. Lin, H.; Freeman, B. D. J. Mol. Struct. 2005, 739, 57-74.
54. Kelman, S. D.; Raharjo, R. D.; Bielawski, C. W.; Freeman, B. D. Polymer 2008, 49,
3029-3041.
55. White, L. S.; Blinka, T. A.; Kloczewski, H. A.; Wang, I. f. J. Membr. Sci. 1995, 103,
73-82.

90
 

Chapter 3 Coating of Regenerated Cellulose Membranes for
CO2/H2 Separations
3.1 Introduction
Over 90% of US hydrogen production employs steam reforming of natural gas
followed by a water gas shift reaction. The raw product of this process contains over 20%
CO2, which is removed from the H2 stream by pressure swing adsorption (PSA).

1

Compared to PSA, membrane separations of CO2 and H2 offer the potential for low
2

capital investment, a small footprint, simple scale-up, and low energy consumption.

Reverse-selective separations that remove CO2 and retain H2 at the high pressure feed
side of the membrane are especially attractive to avoid re-pressurizing the H2. The
challenges in applying reverse selective CO2/H2 separation membranes at a large scale
include (1) development of thin polymer films on porous substrates to maximize flux and
separation efficiency; and (2) selection of substrates that can pack into spiral-wound
3

modules for industrial scale applications. W.R. Grace developed a spiral-wound module
4

for CO2/CH4 separations, but spiral-wound modules and flat-sheet membranes for
CO2/H2 separation remain in research labs or small scales. The alumina substrates
described in chapter 2, while interesting for fundamental studies, are simply too
expensive and fragile for practical separations. Polymeric membranes are more robust

91
 

and less inexpensive than flat-sheet ceramic membranes, and thus are the most likely
substrates for composite, reverse-selective membranes.
Among the materials used to form polymeric membranes, cellulose is natural,
5

abundant and inexhaustible. Cotton, for example, is over 90% cellulose. Regenerated
cellulose (RC, also termed cellulose II, rayon or cellophane) results from chemical or
mechanical treatment of native cellulose (or cellulose I), which was traditionally one of
6

the materials in membranes for dialysis/hemodialysis processes. Mechanical treatment
7

of cellulose is usually performed through compression under high pressure. Scheme 3.1
illustrates the chemical treatment that creates RC. Regenerated cellulose often has less
8

exposed –OH groups but a better defined structure than cellulose.

Scheme 3.1. Production of regenerated cellulose from through chemical treatment of
cellulose.

 

RC membranes are commonly used for ultrafiltration and microfiltration and are
available in a wide range of molecular weight cut-offs (MWCOs). Liu et al. used RC
92
 

membranes (MWCO=100 kDa) to separate immunoglobulin (MWCO=180 kDa) from
chicken egg yolk based on the size difference of the proteins, and the immunoglobulin
purity was 85% after a single-stage separation and 91% after two consecutive
9

separations.

Surface-initiated grafting of polymer brushes to RC membranes should be
10

relatively straightforward using literature procedures for modifying cellulose.

In the

presence of triethylamine (TEA), exposed –OH groups will react with 2-bromoisobutyryl
bromide to immobilize an ATRP initiator from which polymerization can occur. Several
groups examined modified RC membranes because synthetic polymer coatings on the
membranes offer control over wettability, pore size, roughness, and propensity for
fouling. Chen and coworkers grafted poly(glydicyl methacrylate) (PGMA) on a RC
membrane via ATRP and opened the expoxide ring by reaction with N-methylglucamine.
The ring opening leads to a high density of hydroxyl groups, and the resulting membrane
exhibited high affinity for boron and arsenic, which might be useful for removal of boron
from water.

11

Husson and co-workers grew poly(acrylic acid) (PAA) brushes in the 1 μm

pores of RC, and the resulting membranes exhibited a high binding capability for
lysozyme.

12

Bhut

et

al.

also

grew

poly(2-dimethylaminoethyl

methacrylate)

(poly(DMAEMA)) brushes from RC membranes. The polymer brush thickness increased
linearly with polymerization times of 0-15 hr. BSA binding to the brushes increases
linearly with brush thickness before reaching a plateau of 66.3 mg BSA/mL of
membrane.

13

Lin and coworkers grafted zwitterionic sulfobetaine from cellulose

93
 

membranes via surface-initiated ATRP. Decreased nonspecific protein adsorption and
platelet adhesion to these membranes may increase their biocompatibility.14
To the best of my knowledge, no reverse-selective gas separation membranes
were developed from cellulose based substrates. Eckelt and coworkers prepared cellulose
membranes that were size selective  with a H2/CO2 selectivity of 6.2±1.1, and the
selectivity dropped to 1.4 after regeneration.15 This chapter describes my work in
growing

poly(PEGMEMA-1100-co-PEGMEMA-475)

brushes

from

regenerated

cellulose to create composite membranes for CO2/H2 separations. The results in chapter
2 show that similar films on porous alumina exhibit a CO2/H2 selectivity of 12, and the
copolymer film shows no sign of crystallization after over a year of preparation. The H2
permeability of poly(PEGMEMA-475-co-PEGMEMA-1100) films ranges from 2 to 6
3

Barrers, and the CO2 permeability ranges from 20 to 60 Barrers.

Freeman and

coworkers prepared a cross-linked copolymer from PEGMEMA and poly(ethylene glycol
diacrylate) PEGDA in 2005, and the CO2 permeability decreased dramatically from 580
Barrers to 100 Barrers when the PEGDA weight percentage increased from 0 to 100 %.
The maximum reported CO2/H2 selectivity of 13 for this system occurs for minimally
cross-linked networks containing 99 wt % PEGMEMA.

Increasing the fraction of

PEGDA brings the selectivity down, presumably because crosslinking minimizes the
16

fraction of free volume in the polymer.

A later report with 30 wt% PEGDA in the

poly(PEGDA-co-PEGMEMA-475) showed a CO2 permeability of 160 Barrers and a

94
 

17

CO2/H2 selectivity of 14 at 1 atm CO2 partial pressure and 10 °C.

In 2007, Freeman

and coworkers reported another study of poly(PEGDA-co-PEGMEMA-475), also
indicating that increasing the percentage of cross-linker has a negative effect on both
18

permeability and selectivity.

In my studies of CO2/H2 separations by polymer films on RC substrates, I
examined the effect of cross-linking on poly(PEGMEMA-475-co-PEGMEMA-1100)
films by introducing a 25% molar fraction of PEGDA during polymerization. The study
aims to (1) grow polymer membranes on RC substrate and (2) observe the effect of crosslinking on the transport properties of poly(PEGMEMA-475-co-PEGMEMA-1100-coPEGDA) films. In mixed-gas separations the cross-linked films on RC substrates show
CO2/H2 selectivities ranging from 6.5 to 19.9, but the CO2 permeability is only 5-15
Barrers.

3.2 Experimental
3.2.1

Chemicals and Materials
PEGMEMA monomers with Mn = 475 Da (PEGMEMA-475) and Mn = 1,100 Da

(PEGMEMA-1100) were obtained from Aldrich. The cross-linker, PEGDA Mn =700 Da,
was purchased from Aldrich as well. All monomers contain both 100 ppm 4methoxyphenol and 300 ppm 2,6-di-tert-butyl-4-methylphenol as inhibitors.

The

monomers were used as received without removal of the inhibitors. Proton NMR spectra
show that the average numbers of ethylene oxide units in PEGMEMA-475 and
95
 

PEGMEMA-1100 are 8.8 and 23.1, respectively (see Figure 2.1).3 The 2bromoisobutyryl bromide (BiBB) (98%), CuCl (99.995%), CuBr2 (99.999%), and
1,1,4,7,10,10-hexamethyltriethylenetetramine (HMTETA) (97%) were purchased from
Aldrich, and deionized water was obtained using a Millipore system (Milli-Q, 18.2
MΩ•cm). Triethylamine (TEA) and dimethylformamide DMF (99.8%) were purchased
from Jade Scientific and used as received. THF was distilled over sodium metal with
benzophenone. The ultrafiltration regenerated cellulose membranes (PLGC02510, 25 mm
disks with 10 kDa MWCO) were purchased from Milliopore, and the RC layer was
supported by ultra-high molecular weight polyethylene. Silicon(100) wafers were
obtained from NOVA Electronic Materials and sputter-coated with 20 nm of chromium
followed by 200 nm of gold by LGA Thin Films (Santa Clara, CA). The [Br-C(CH3)219

COO(CH2)11S]2 initiator was synthesized as described previously.

A cylinder

containing 20.3% CO2 and 79.7% H2 was purchased from Airgas.

3.2.2

Polymerization

3.2.2.1 Initiator Attachment
Scheme 3.2 illustrates the initiator attachment and ATRP procedures. The RC
membrane was immersed in 20 mL deionized water for 1 h with the support side up. To
ensure complete glycerin removal, the water was replaced every 20 minutes. After
soaking in water, the RC membrane was rinsed with 5 mL deionized water twice and
dried under flowing N2. The dried membrane was then placed in a Teflon membrane
holder that exposed the RC surface and transferred into a glove bag. Under a N2 blanket,
96
 

the RC membranes were rinsed with 2 mL anhydrous THF twice to remove any residual
water. An initiator solution with 10 mM TEA and 10 mM of BiBB was made by adding
28 μL of TEA and 25 μL of BiBB to 20 mL anhydrous THF. The membrane was then
coated with 2 mL of this solution for 2 min and rinsed with 2 mL anhydrous THF three
times and 2 mL deionized water once. Finally, the RC membrane was transferred outside
of the glove bag, and dried with flowing N2. Both rinsing and drying occurred with the
membrane remain in the holder. The initiator-modified RC membrane was stored in a N2
filled glove bag and used for polymerization within 48 hr.
For Au-coated silicon wafers, the initiator attachment occurred by soaking UVozone-cleaned (10 min) wafers in 1 mM ethanolic [Br-C(CH3)2-COO(CH2)11S]2
overnight. The wafers were then rinsed with ethanol and dried with a N2 stream.

97
 

Scheme 3.2. Growth of a cross-linked copolymer brush from a RC membrane.

3.2.2.2 Catalyst Solution
A catalyst stock solution was prepared by first degassing 30.0 mL deionized H2O
via three freeze, pump, thaw cycles in a round bottom flask. At the third freezing period,

98
 

0.0596 g (0.6 mmol) of CuCl and 0.0450 g (0.2 mmol) of CuBr2 were added to the round
bottom flask prior to finishing the pump thaw cycle. No CuCl or CuBr2 is lost during this
pumping step. Two more freeze, pump, thaw cycles were performed to enhance oxygen
removal. In a N2-filled glove bag, 490 µL (1.8 mmol) of HMTETA was added to this
solution. The solution turned dark blue upon HMTETA addition. The catalyst solution
was stored in a glove bag with constant stirring for future use. In some rare cases,
precipitation is visible in the catalyst solution after over two weeks of preparation. With
constant stirring, the precipitation does not appear, and the catalyst solution remains
+

2+

uniform and clear. The mole ratio of Cu :Cu :HMTETA was 3:1:9.

3.2.2.3 ATRP
The monomer solution, which contained a total of 0.75 M monomer in water, was
degassed via four freeze, pump, thaw cycles. Unless specifically stated, the molar ratio of
PEGMEMA-1100:PEGMEMA-475:PEGDA is 3:3:2. Monomer solution and initiatormodified RC membranes were transferred into the glove bag with the catalyst solution.
The polymerization mixture was prepared by combining degassed monomer and premade catalyst solutions in a 9:1 volume ratio to produce a light blue solution containing
0.67 M monomer, 2.0 mM CuCl, 0.60 mM CuBr2, and 6.0 mM HMTETA. The initiatormodified membranes or gold coated wafers were immersed in the polymerization
solution for designated periods of time, and the resulting polymer films were removed
from the glove bag, and rinsed with 5 mL DMF three times followed by rinsing with 5
mL of ethanol three times and drying under a N2 stream.
99
 

3.2.3

Characterization methods
Reflectance FTIR spectroscopy was performed using a Nicolet Magna-IR 560

spectrometer with a Pike grazing angle (80°) attachment. A UV-Ozone-cleaned, goldcoated Si wafer was used to obtain the background spectrum. An air background was
obtained immediately prior to measurement of the RC samples.

Attenuated Total

Reflectance (ATR) FTIR spectra were obtained with a Perkin Elmer Spectrum one FTIR
spectrometer using an air background. This spectrophotometer contains a diamond,
single-reflection ATR crystal.
Scanning electron microscopy (SEM) was performed mainly with a JEOL semiin-lens cold cathode field-emission scanning electron microscope, model JSM-7500F,
operating with an r-filter in signal maximum mode. Samples were sputter coated with 10
nm of osmium prior to imaging. A few SEM images were obtained with a JEOL 6400V
microscopy with a LaB6 emitter (Noran EDS). The samples for JEOL 6400V were
coated with 8 nm of gold before of imaging.
3.2.4

Gas Permeation
Mixed-gas experiments were performed by loading the membrane into a custom-

made membrane holder (all connections utilized Swagelok fittings and were tested to
ensure that they maintained pressure over a time scale longer than the experiment) that
allowed cross-flow of the feed gas as well as a sweep gas on the permeant side. A
backpressure valve was employed to sustain the feed gas pressure, and the feed flow rate
was high enough to maintain constant composition at the face of the membrane (the
100
 

stage-cut, or ratio of permeant flow to feed flow, was <1%). The N2 sweep gas/permeant
stream was connected to an automated six-port injector valve on a Hewlett-Packard 6890
GC equipped with a thermal conductivity detector (TCD) and an Agilent GSCarbonPLOT capillary column (i.d. = 0.53 mm, length = 30 m, 3 µm coating). Figure
3.1 illustrates the entire apparatus.

Gas Selection Valve
Pressure
Controller
N2

Pressure
Gauge
CO2

Two Channel
Mass Flow
Controler

br
em
M

H2

e
an

CO2
+ H2

Vent

carrier gas Detector
GC System

Column
 

Figure 3.2. Apparatus for testing mixed gas permeabilities.
The mixed-gas and pure-gas permeances were calculated from GC results using
equation 2.3,

101
 

Permeance 

B   N

2

 N  A  Δp

(2.3)

2

where χB is the mole fraction of a particular permeate gas, B, in the gas mixture injected
into the GC column, ΦN2 is the sweep gas flow rate, χN2 is the mole fraction of N2 in the
gas injected into the GC column, Δp is the transmembrane partial pressure difference for
the gas of interest, and A is the membrane area. (ΦN2/χN2 is the sum of the permeate and
sweep gas flow rate, and χN2 was determined by subtracting the mole fractions of the
permeates from unity. This assumes that the amount of N2 passing from permeate to feed
was negligible.) The sweep gas flow rate was programmed into a mass flow controller,
pressure and area were measured, and the mole fractions were determined from the
integrated gas chromatograms and a calibration curve.

102
 

O-C-O

C=O
C=C

Absorbance

Cross-linked copolymer

PEGMEMA-1100

PEGMEMA-475

PEGDA-700
2200

2000

1800

1600

1400

1200

1000

800

-1

Wavenumbers cm

 

Figure 3.2. Reflectance FTIR spectrum of a cross-linked co-polymer brush (top, 25 mole%
crosslinker, 37.5 mole% PEGMEMA-1100, and 37.5 mole% PEGMEMA-475 in the
polymerization solution that contained 0.67 M total monomer), and spectra of spin-coated
films of the three monomers. The co-polymer film was grown by ATRP from intiators
immobilized on Au-coated Si, and the monomers were spin coated on the surface of Aucoated wafers. Compare the cross-linked copolymer absorbance with poly(PEGMEMA1100) and poly(PEGMEMA-475) (Chapter 2, Figure 2.3.) the position of C=O peak and
O-C-O peak was in match.
 

 

103
 

 

3.3 Results and Discussion
3.3.1

FTIR Studies of Cross-linked Copolymer Films

3.3.1.1 Reflectance FTIR Spectra of Cross-linked Copolymer Brushes on a Aucoated Silicon Wafer
In an attempt to examine crosslinking in poly(PEGMEMA-1100-co-PEGMEMA475-co-PEGDA) brushes, we obtained reflectance FTIR spectra of a poly(PEGMEMA1100-co-PEGMEMA-475-co-PEGDA) film grown from initiators on a Au-coated Silicon
wafer. Figure 3.2 shows the spectrum of the cross-linked co-polymer film as well as
spectra of the three monomers involved in polymerization. In the spectrum of
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA), the absence of the C=C peak
-1

at 1635 cm

suggests that the brush does not contain a significant amount of residual

C=C bonds. Assuming that the brush has the same composition as the monomer solution,
the films contain 25% cross-linking agent, and the absence of signal from C=C bond
indicates a highly cross-linked film. Nevertheless, we have no direct evidence for
incorporation of PEGDA.
3.3.1.2 ATR-FTIR Study of Cross-linked Copolymer Grafted from RC
Figure 3.4 compares the ATR-IR spectra of an RC membrane before (bottom) and
after (top) growth of a 1.5 μm-thick film of poly(PEGMEMA-1100-co-PEGMEMA-475co-PEGDA, 25 mole % PEGDA in the polymerization solution). The IR spectrum of the
-1

bare RC membrane contains signature peaks at 3331 and 2904 cm

representing –OH

stretching (from both cellulose and water) and –CH stretching, respectively.20 The 1635
104
 

-1

cm peak is likely due to absorbed H2O, this means water also contributes for 3331 cm-1
-1

peak. The 1470 cm
-1

1416 cm

peak is due to –OH in plane bending. Absorbances at 1375 and
21

stem from the CH and CH2 bending modes respectively,

and the peak at

-1

1100 cm peak is due to C-O stretching.
Equation 2.4 gives the expression for the penetration depth, d, in ATR,

(2.4)

where λ is the wavelength, n1 (2.417) is the refractive index of the diamond crystal, n2 is
the refractive index of the polymer film (~1.5), and θ is the angle of incidence of the IR
22

beam. For this particular system, the angle of incidence is 45°.

-1

At 3331 cm , the

penetration depth is 0.6 μm. The thickness of the cross-linked copolymer film is 1.5 μm
(estimated from SEM images), which is more than twice the penetration depth in this
region, so the polymer brush should attenuate most of the signal from the RC substrate.
The film may also adsorb less water than the bare RC membrane.
-1

For the cross-linked copolymer film, the absorbance at 1721 cm due to the C=O
bond confirms the presence of the copolymer on the surface. Husson and co-workers
polymerized poly[poly(ethylene glycol methacrylate)] (PPEGMA) from RC, and their
23

ATR-IR spectra indicated similar peaks for the polymer layer.
-1

cm

stems from O-C-O bond stretching, and the absence of peak splitting indicates that
105

 

The absorbance at 1096

there is no crystallization of PEO side chains in this copolymer (see chapter 2 page
70).

3,24

-1

However, the C=C peak at 1631 cm

suggests a significant amount of monomer

trapped in the cross-linked framework or possibly unreacted cross-linking agent.
However, this signal may also contain some absorbance from the RC membrane.
Although the IR spectrum shows the presence of the film, it is unfortunately not very
useful for determining film composition.

Bare RC Membrane
Crosslinked Copolymer on RC

Absorbance

0.1

4000

3500

3000

2500

2000

1500

1000

-1

Wavenumbers cm

 

Figure 3.3. ATR-FTIR of a bare RC membrane (before, bottom), and after (top)
modification by growth of a poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA)
film.

106
 

3.3.2. SEM Characterization of RC Membranes and Cross-linked Polymer Films
Figure 3.4 shows SEM images of bare RC membranes. The RC membrane has a
200 μm-thick support layer made from dense polyethylene fibers. Based on the is
estimated by molecular weight cut off (MWCO) of 10 kDa, the RC layer on top of the
polyethylene has pores with diameters of 2-4 nm. This pore size is lower than the SEM
resolution. The thickness of the RC layer is about 30 μm.

 

 

Figure 3.4 SEM images of cross-sections of an RC membrane. The black lines indicate
the RC layer, which resides on support of polyethylene fibers.
Figure 3.5 shows cross-sectional images of RC membranes after growth of crosslinked polymer brushes. For most membranes the boundary between the polymer brush
and RC is evident, but for some samples the contrast between RC and cross-linked
polymer brush is not easy to detect by eye. Both RC and polymer films contain C, H and
O, so energy-dispersive X-ray spectroscopy (EDS) cannot define the boundary between
RC and the cross-linked brush. In such cases, two methods were used to find he interface
of RC and polymer brush layer. (1) Application of an electron beam on a small area of
the sample for 20-60 seconds can differentiate between the two materials. The polymer
brush layer melts and slightly moves during this time, while the more stable RC layer
107
 

shows no change in shape. This slight movement of the sample is very obvious under
SEM. The RC is probably more stable due to the regeneration treatment. (2) The
thickness of the RC layer is 30 μm (Figure 3.4). The total thickness of RC and polymer
brush could be measured and the thickness of polymer brush is determined by total
thickness minus the thickness of the RC layer. Using these methods or the two methods
combined, polymer brush film thicknesses could be obtained from SEM images.

Figure 3.5. SEM images of cross-sections of cross-linked copolymer films grafted on RC
membranes. The black lines indicate the polymer film on top of the RC membrane.

108
 

3.3.3. Gas Separations
Using the apparatus illustrated in Figure 3.1, I determined the gas permeability of
RC membranes modified with cross-linked polymer films. The feed gas cylinder initially
contained 20.3% CO2 and 79.7% H2. Over the time frame of years, the CO2/H2 ratio of
the mixed gas cylinder may have changed slightly, so I monitored its composition with
GC to ensure an accurate CO2/H2 ratio. Pure CO2 and H2 served as standards. Figure 3.6
shows a sample gas chromatogram of the feed gas. The H2 peak (2.9 min) area is 8414,
whereas the CO2 peak (4.35 min) area is only 19.3. The huge difference between the CO2
and H2 signals is due in part to the higher concentration of H2 in the feed mixture, but the
main factor for the large differences in the peak areas is that the thermal conductivity of
the N2 carrier gas is much closer to CO2 than H2.

109
 

H2

0

Signal Intensity

-500
130

-1000

125

-1500

CO2

-2000

0

1

2

3

4

Retention Time (min)

5
 

Figure 3.6. Gas chromatogram of the feed mixed gas. The molar ratio of CO2 to H2 is
79.7: 20.3.
During the permeability test, the feed gas flowing past the membrane was
adjusted to a pressure of 45 psi (3.1 bar), and the flow rate was much larger than rates of
CO2 or H2 permeation through the membrane to ensure a constant feed composition. The
permeate side is swept with ultra-pure N2, at a flow rate of 0.5 mL/min or 0.75 mL/min,
depending on the permeability of the sample.

110
 

H2
120

CO2
128

Signal Intensity

100
126

80
60

124

40
122
4.0

20

2

3

4.2

4.4

4

4.6

4.8

5

Retention Time (min)

6
 

Figure 3.7. Gas chromatogram of the permeate of a mixed-gas separation with a RC
membrane modified with a poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA)
film. (The splitting of the H2 peak is probably due to a mechanical disturbance on sample
injection, but this does not affect the integral of peak area)
Figure 3.7 shows a chromatogram of the gas permeating through a RC membrane
modified with a poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA) film. The H2
peak area is 216.9, the CO2 peak area is 13.3, and the N2 sweep gas flow rate is 0.75
mL/min. Remarkably, the CO2/H2 selectivity is 18.7.

Based on SEM images, the

thickness of this particular film is 700 nm, so the CO2 permeability is 15 Barrers.
However, after exposure to CO2/H2 for over 30 min, the selectivity of the film dropped
111
 

from 19 to 8. Nevertheless, after sitting in air overnight, the selectivity of the membrane
returns to approximately its original value of 18.
More than 10 membranes prepared under the same conditions with 25 mole%
cross-linker in poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA) showed CO2/H2
selectivities from 6.5 to 19.9, and the permeability of CO2 ranged from 5 to 15 Barrers.
Comparing this permeability result with that for the poly(PEGMEMA-1100-co3

PEGMEMA-475) films on alumina (PCO2 = 3-6 Barrers, chapter 2), the permeability for
these cross-linked film (5-15 Barrers) may be slightly higher than for the non-crosslinked polymer. Freeman and coworkers claimed that the crosslinking would dramatically
16,18

decrease both selectivity and permeability.

However, for this particular case, 25

percent cross-linker did not affect the selectivity and permeability dramatically.

3.4. Conclusion
ATRP from initiators on RC yields poly(PEGMEMA-1100-co-PEGMEMA-475co-PEGDA) films. The cross-linker mole percentage was 25% in solution and

presumably the same on the surface.

FTIR spectra and SEM images confirm the

presence of the film.

In mixed-gas separation tests, the CO2/H2 selectivity of the modified RC membrane
ranged from 6.5 to 19.9, and the permeability of CO2 varied from 5 to 15 Barrers. This is
comparable to the poly(PEGMEMA-1100-co-PEGMEMA-475) films described in
Chapter 2, showing that polymerization for polymers is a feasible method for creating
 
membranes.  
112
 

REFRENCES
 

 

113
 

References
1. Molburg, J. C.; Doctor, R. D. P. Annu. Int. Pitt. Coal Conf. 2003, 1388-1408.
2. Hussain, A.; Hagg, M. J. Membr. Sci. 2010, 359, 140-148.
3. Grajales, S. T.; Dong, X.; Zheng, Y.; Baker, G. L.; Bruening, M. L. Chem. Mat.
2010, 22, 4026-4033.
4. Baker, R. W.; Lokhandwala, K. Ind. Eng. Chem. Res. 2008, 47, 2109–2121.
5. Esau, K. Plant Anatomy ; John Wiley: New York, 1953.
6. Gelde, L.; Vázquez, M. I.; Benavente, J. Polym. Test.2011, 30, 457–462.
7. Nakagaito, A. N.; Yano, H. Appl.Phys. A: Mater. 2004, 78, 555-559.
8. Lindqvist, J.; Malmstrom, E. J. Appl. Polym. Sci. 2006, 100, 4155-4162.
9. Liu, J.; Yang, J.; Xu, H.; Zhu, H.; Qu, J.; Lu, J.; Cui, Z. Int. J. Food Eng. 2011, 7,
1–8.
10. Carlmark, A.; Malmstrom, E. Biomacromolecules 2003, 4, 1740-1745.
11. Wei, Y.-T.; Zheng, Y.-M.; J., C. P. Langmuir 2011 2011, 27, 6018–6025.
12. Singh, N.; Wang, J.; Ulbricht, M.; Wickramasinghe, S.; Husson, S. J. Membr. Sci.
2008, 309, 64-72.
13. Bhut, B. V.; Wickramasinghe, S. R.; Husson, S. M. J. of Membr. Sci. 2008, 325,
176-183.
14. Liu, P. S.; Chen, Q.; Liu, X.; Yuan, B.; Wu, S. S.; Shen, J.; Lin, S. C.
114
 

Biomacromolecules 2009, 10, 2809–2816.

15. Eckelt, J.; Wolf, B. A. Macro. Chem. Phys. 2005, 206, 227–232.
16. Kalakkunnath, S.; Kalika, D. S.; Lin, H.; Freeman, B. D. Macromolecules 2005,
38, 9679-9687.
17. Lin, H.; Van Wagner, E.; Freeman, B. D.; Toy, L. G.; Gupta, R. P. Science 2006,
311, 639-642.
18. Lin, H.; Freeman, B. D.; Kalakkunnath, S.; Kalika, D. S. J. Membr. Sci. 2007,
291, 131-139.
19. Shah, R.; Merreceyes, D.; Husemann, M.; Rees, I.; Abbott, N.; Hawker, C.;
Hedrick, J. Macromolecules 2000, 33, 597-605.
20. Liang, C. Y.; H., M. R. J. of Polym. Sci. 1959, 37, 385–395.
21. Marchessault, R. H.; Liang, C. Y. J. of Polym. Sci.1960, 43, 78-84.
22. Ghaucha, A.; Deveaub, P.; Jacobb, V.; Baussand, P. Talanta 2006, 68, 12941302.
23. Singh, N.; Chen, Z.; Tomer, N.; Wickramasinghe, S. R.; Soice, N.; Husson, S. J.
Membr. Sci. 2008, 311, 225-234.
24. Zheng, Y.; Bruening, M. L.; Baker, G. L. Macromolecules 2007, 40, 8212-8219.

115
 

Chapter 4 Synthesis of Nanoparticle-containing Membranes
and Investigation of their CO2/H2 Separations
This chapter focuses on surface-initiated polymerization of poly(ethylene glycol methyl
ether methacrylate) (PEGMEMA) and poly(ethylene glycol) diacrylate (PEGDA) in the
presences of SiO2 nanoparticles modified with initiators. Mixed-gas permeation studies
examine whether incorporation of nanoparticles alters CO2/H2 separations with these
films.
4.1 Introduction
4.1.1

Impact of Fractional Free Volume on Selectivity and Permeability
For non-porous polymer membranes, gas permeability (P) depends on both

solubility (S) and diffusivity (D) according to equation (4.1). The solution diffusion
model suggests that the ideal selectivity of gas A over gas B, αA/B, depends on the
1

solubility coefficients and diffusivities of the two gases, equation (4.2).

≡

(4.1)

/

(4.2)

The solubility of a certain gas in a polymeric membrane depends not only on the physical
properties of the gas, but also on the similarity between the solubility parameters of the

116
 

2

polymer and the gas. As discussed in Chapter 1, CO2/H2 diffusivity selectivity is always
less than one because H2 is smaller and lighter than CO2.
Equation 4.3 often describes the diffusivity of a gas in a polymer, where A is a
constant that reflects the mean free path of the permeate. B is proportional to the diameter
of the permeate,

3,4

and FFV is the fractional free volume in the polymer. Importantly, an

increase in FFV will give higher permeability.

	

exp

(4.3)

To increase FFV, Freeman and coworkers blended fumed SiO2 nanoparticles in
poly(4-methyl-2-pentyne) (PMP), and the permeability of both N2 and CH4 increased
with the amount of nanoparticles added. PMP containing 40% fumed SiO2 nanoparticles
showed a >2-fold increase in CH4 permeability. Positron anihilation lifetime
spectroscopy (PALS) indicated that FFV correlates with the wt% of nanoparticles in the
polymer. This work also indicated that the diffusion coefficient increases more for large
5

gases than for smaller ones. For reverse-selective CO2/H2 separations, an increase of
FFV will positively impact both CO2 permeability and CO2/H2 selectivity, because
diffusivity selectivity, which favors H2, will decrease.

117
 

4.1.2

Mixed Matrix Membranes for Gas Separation
Robeson summarized the selectivity and permeability data for a number of light

gas combinations to determine upper bound in the performance of polymeric membranes
6

for gas separations. For a given permeability selectivity is limited. Simply put, with sizesieving membranes more permeable materials show less selectivity. Freeman similarly
summarized data for reverse selective polymers in CO2/H2 separations and also found an
upper bound effect, but in this case selectivity may increase with permeability.

5

With the acknowledgement of an upper bound of selectivity and permeability in
polymeric gas separation materials, researchers focused on developing new materials
with higher selectivity and/or permeability. Forming films containing nanoparticles or
porous zeolites could break this upper bound. Mixed Matrix Membranes (MMMs)
contain micro to nanometer sized inorganic particles blended with polymers. By
combining two materials with different selectivity and permeability merits, MMMs may
overcome the upper bound between selectivity and permeability. Inorganic materials
applied in synthesizing MMMs include: zeolites,
5,12

and nanoparticles.

7-11

metal-organic frameworks (MOFs)

9

After physically mixing fumed SiO2 nanoparticles with diameters

smaller than 10 nm into poly(4-methyl-2-pentyne), Freeman and coworkers observed a
surprisingly 3 fold increase in CO2 permeability when 30 vol% nanoparticles are blended
in. The size of voids between nanoparticles and the polymers enhanced the gas
5

permeability. Polysulfone/ SiO2 nanoparticle MMMs exhibit a similar permeability

118
 

12

enhancement compared to the pure polymer.

This work attempts to create similar

improvements through incorporation of nanoparticles into membranes containing
poly(PEGMEMA).
While more nanoparticles are blend in the polymer film, it is hard to obtain defect
free film for gas separations. Introducing cross-links into polymers could help stabilize
membrane under aggressive conditions.

13

Freeman and co-workers focused on cross-

linking poly(ethylene glycol) methyl ether acrylate (PEGMEA) films for CO2/H2
separations. On slowly increasing from 0% to 100% cross-linking monomer,
poly(ethylene

glycol)

diacrylate

(PEGDA),

in

polymerization

of

[methyl

bis(trimethylsiloxy) silyl] propyl glycerol methacrylate, however, the resulting polymer
14,15

shows a linear decrease in CO2 peremeability.

Poly(ethylene glycol) methyl ether

acrylate (PEGMEA) with 10 wt% of PEGDA is very robust in separation of H2S at
humid and high pressure condition. The permeability of both CO2 and H2 in 30 wt%
cross-linking increased with the respect of CO2 pressure, especially at low temperature (20 °C). At low temperature, CO2/H2 selectivity increased with CO2 partial pressure as
well.
 

16

 

119
 

4.2 Experimental
4.2.1

Chemicals and Materials
PEGMEMA monomers with Mn = 475 Da (PEGMEMA-475) and Mn = 1,100 Da

(PEGMEMA-1100) and a crosslinkable monomer, PEGDA Mn=700 Da, were purchased
from Aldrich. All monomers contain both 100 ppm 4-methoxyphenol and 300 ppm 2,6ditert-butyl-4-methylphenol as inhibitors and were used as received without removal of
the inhibitors. Proton NMR spectra show that the average numbers of ethylene oxide
units in PEGMEMA-475 and PEGMEMA-1100 are 8.8 and 23.1, respectively (see
17

Figure 2.1).

SiO2 nanoparticles (12 nm diameter) were purchased from Aldrich, and the

diameter of the nanoparticles measured in TEM images of 17 particles was 8.0 ± 3.5 nm
(Figure 4.1).

 

Figure 4.1. TEM images of SiO2 nanoparticles spotted on a carbon-coated Cu grid.

120
 

2-bromoisobutyryl bromide (BiBB) (98%), CuCl (99.995%), CuBr2 (99.999%),
and 1,1,4,7,10,10-hexamethyltriethylenetetramine (HMTETA) (97%) were purchased
from Aldrich, and deionized water was obtained using a Millipore system (Milli-Q, 18.2
MΩ•cm). Triethylamine (TEA) and DMF (99.8%) were purchased from Jade Scientific,
and used as received. THF was distilled over sodium metal with benzophenone. The
porous alumina substrates were Anodisc membrane filters (25 mm disks with 0.02 µm
surface pores) purchased from Whatman. Figure 4.2 shows the SEM images of the top
and cross section of an alumina membrane. Silicon (100) wafers were obtained from
NOVA Electronic Materials. A silane initiator, Cl(CH3)2Si(CH2)11OCOC(CH3)2Br, was
18

synthesized as described previously.

A mixed gas cylinder containing 20.3% CO2 and

79.7% H2 was purchased from Airgas, and a cylinder with 50% CO2 and 50% H2 was
obtained from Praxair. A Sargent-Welch S-15700 (5000 rpm) centrifuge was used to
separate nanoparticles from suspensions. The Sharpertek Stamina XP sonicator  40,000
Hz, 380 watt was used for dispersing the nanoparticles.

121
 

 

Figure 4.2. SEM images of the top (A) and cross-section (B) of porous alumina
membranes with nominal 20 nm surface pores.

122
 

4.2.2

Polymerization

4.2.2.1 Initiator Attachement
Scheme 4.1 illustrates the initiator attachment to alumina membranes and SiO2
nanoparticles. The SiO2 nanoparticles, 160 mg, were dispersed in 20 mL dry THF during
5 min of sonication. After transfer into a N2-filled glove box, 10 drops of
dimethylchlorosilane initiator, Cl(CH3)2Si(CH2)11OCOC(CH3)2Br (light yellow color
liquid) was added to the suspension to achieve a 1 mM initiator concentration. The
suspension was left in the glove box for the desired amount of time, and the nanoparticles
were separated from the suspension via 5 min of centrifugation (5000 rpm), and rinsed
with 10 mL dry THF three times. The resulting initiator-modified nanoparticles were
dried under vacuum. Due to some loss in decanting following centrifugation, the process
yielded about140 mg of initiator-modified nanoparticles.
For alumina membranes, the initiator attachment was performed in a glove box by
soaking a UV/ozone-cleaned (10 min) membrane overnight in THF containing 1 mM
Cl(CH3)2Si(CH2)11OCOC(CH3)2Br. The membranes were then rinsed with 5 mL of
THF three times and dried under flowing N2. In a few cases, the alumina membranes
were rinsed by flowing 5 mL of THF through the membrane using a peristaltic pump.
The two rinsing methods gave no significant difference in the resulting polymer brush
film thickness.

123
 

Scheme 4.1. Initiator attachment to SiO2 nanoparticles or alumina membranes.

 

4.2.2.2 Catalyst Solution
A catalyst stock solution was prepared by adding 0.0596 g (0.6 mmol) of CuCl
and 0.0450 g (0.2 mmol), CuBr2 and 490 µL (1.8 mmol) of HMTETA ligand into
degasses 30.0 mL deionized H2O. Details of this step is described in (Chapter 3, page 88).
+

2+

The nominal mole ratio in the solution was Cu :Cu :HMTETA was 3:1:9.

124
 

Scheme 4.2. Preparation of nanocomposite membranes by ATRP. The reaction is
performed an in oxygen-free glove bag. The pink dots represent nanoparticles, but they
are greatly expanded to allow visualization.

4.2.2.3 ATRP
Scheme 4.2 illustrates the polymerization strategy. The monomer solution is
prepared by dissolving PEGMEMA-1100, PEGMEMA-475 and PEGDA in water, and
adding initiator-modified SiO2 nanoparticles. The total monomer concentration is 0.75 M.
Unless specifically stated, the molar ratio of PEGMEMA-1100, PEGMEMA-475 and
PEGDA is 3:3:2. The monomer solution was degassed via four freeze, pump, thaw cycles.
Monomer solution and the initiator–modified alumina membrane were transferred into a
N2-filled glove bag containing the catalyst solution. The polymerization mixture was
prepared by combining degassed monomer and catalyst solutions in a 9:1 volume ratio to

125
 

produce a light blue solution containing 0.67 M monomer, 2.0 mM CuCl, 0.60 mM
CuBr2, and 6.0 mM HMTETA.

The tops of initiator-modified membranes were

immersed in the polymerization solution for designated periods of time, and the resulting
polymer films were rinsed with 5 mL DMF three times followed by rinsing with 5 mL of
ethanol three times and drying under a N2 stream. No polymerization occurred on the
bottom side of the alumina membrane because the monomer solution did not pass
through the membrane during polymerization.

4.2.3

Characterization Methods and Gas Separation Tests
Scanning electron microscopy (SEM) was performed mainly with a JEOL semi-

in-lens cold cathode field-emission scanning electron microscope (Model JSM-7500F).
This microscope operates with an r-filter in signal maximum mode. Some SEM images
were obtained using a JEOL 6400V microscope with a LaB6 emitter (Noran EDS). All
samples were cracked right after freezing in liquid nitrogen to ensure a clean exposure of
the cross-section. The cross-section was then sputter coated with 10 nm of osmium
(JEOL 7500F) or 8 nm of gold (JEOL 6400V) prior to imaging.
Nuclear Magnetic Resonance (NMR) spectra of the reaction mixtures were
obtained with a Varian Unity±500 superconducting NMR-Spectrometer (500 MHz).
Transmission Electron Microscopy (TEM) images were obtained at the Center for
Advanced Microscopy using a JEOL100 CXII.
Mixed-gas permeation tests were performed as described in chapter 2 Page 57.
126
 

4.3 Results and Discussion
4.3.1 SEM Characterization of Nanoparticle-containing Membranes
SEM images confirm growth of a thick polymer film on top of the alumina
membrane. Figure 4.7 shows a cross-sectional image of a nanoparticle-containing film
prepared with 0.18 mg/mL SiO2 nanoparticles in the polymerization solution and a 45min reaction. The thickness of the poly(PEGMEMA-1100-co-PEGMEMA-475-coPEGDA-700)/nanopartices layer is 77.0 ± 4.3 μm. An expanded view Figure 4.7 (B)
suggests that the nanoparticles are embedded in the polymer framework. The diameter of
the particles in the image is 200 nm to 2 μm, and this large size is presumably due to the
significant polymerization from the 8 nm-diameter, initiator-modified particles. Chen and
co-workers previously used nanoparticle size increases as proof of successful
19

polymerization.

Figure 4.7 (C) is the cross-sectional image of an alumina membrane

modified with a poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700) film
without embedded nanoparticles, and Figure 4.7 (D) is a higher magnification image of
the poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700). As expected, Figure
4.7 (D) does not reveal any particles.

127
 

 

Figure 4.3. (A) and (B) SEM images of cross-sections of polymer films prepared by
polymerization of a solution containing 0.25 M PEGMEMA-1100, 0.25 M PEGMEMA475, 0.17 M PEGDA, 2.0 mM CuCl, 0.60 mM CuBr2, 6.0 mM HMTETA and 0.18
mg/mL SiO2 nanoparticles. (C) and (D) are the cross-sectional images of similar films of
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700)
prepared
without
nanoparticles.

4.3.2

Gas Separations with Nanoparticle-containing Membranes

4.3.2.1 Effect of Nanoparticle Loading on CO2/H2 Selectivity and Permeability
The gas permeability of MMMs is traditionally interpreted via a Maxwell model,
which uses the theory for electrical conduction through a heterogeneous medium to
obtain the exact solution for the conductivity of randomly distributed and non-interacting
128
 

homogeneous solid spheres in a continuous matrix.

20

The Maxwell model (or its

refinements) satisfactorily describes MMMs with very permeable fillers such as
21

zeolites.

However, Freeman and coworkers discovered that mixing SiO2 particles in

glassy, amorphous poly(4-methyl-2-pentyne) membranes significantly enhance the
permeability and selectivity. The improvement is beyond predictions by the Maxwell
model, and the enhancement may stem from free volume generated between the
nanoparticles and the polymer.

22

Unlike most nanocomposite membranes, the nanoparticles in my membranes
should be anchored to the surrounding polymer due to polymerization from immobilized
initiators. The covalently modified nanoparticles may offer more stable MMMs than
simple blends, but the effect of covalent grafting on permeability is unknown. With the
monomer concentration fixed at 0.67 M, and the monomer composition of 37.5%
PEGMEMA-475, 37.5% PEGMEMA-1100 and 25% PEGDA, the amount of
nanoparticle loading should depend on the amount of initiator-modified SiO2
nanoparticles in the monomer solution. We performed a series of experiment that altered
the nanoparticle loading in poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700)
films to find the highest permeability and selectivity for gas separations. Figure 4.4
shows the SEM images of cross-sections of nanocomposite membranes prepared with
different amounts of initiator-modified particles. The film thickness measured from SEM
images is crucial for calculating gas permeability, which is equal to permeance multiplied
by film thickness. For every membrane, at least five thickness measurements from

129
 

different areas were performed, and the average value is used for permeability
calculations.

 

Figure 4.4. SEM images of cross-sections of nanoparticle-containing membranes. The
nanoparticle concentrations in monomer solutions were 0 mg/mL (A), 0.18 mg/mL (B)
and (C), 0.45 mg/mL (D), 0.9 mg/mL (E), and 2.7 mg/mL (F). The film in (B) was grown
130
 

from a RC membrane, and all other nanocomposite membranes were grafted from
alumina. The black lines indicate the beginning and ending of the grafted polymer film.
For samples with seemingly unclear polymer-substrate boundaries, the beginning of the
film is determined by checking the difference of layer stability in the electron beam.
Figure 4.5 illustrates the effect of nanoparticle concentration in the monomer
solution on the resulting membrane’s CO2/H2 selectivity. Both selectivity and
permeability data represent the average of values from at least two membranes that were
prepared under the same condition. Within the region of 0-2.7 mg/mL nanoparticle
concentration, CO2 permeability reached a maximum of 79.0 ± 1.2 Barrer at 0.45 mg/mL.
The membranes with 0.18 mg/mL nanoparticles in the monomer solution were
synthesized using RC substrates (Figure 4.4). The top of the RC membrane is not as flat
as porous alumina (SEM image of a bare RC membrane is in Chapter 3, page 99), and
this is perhaps the reason that the permeability varies more on RC than with alumina
substrates. Comparing the permeability of nanoparticle-containing membranes with the
PEGMEMA-1100-co-PEGMEMA-475-coated membranes (PCO2 = 20 to 60 Barrers,
Chapter 2) and cross-linked PEGMEMA-PEGDA 25% film (PCO2 = 5 to 15 Barrers,
Chapter 3), the permeability of nanocomposite membrane is higher when the
concentration of nanoparticles in the monomer solution is 0.45 mg/mL. The CO2/H2
selectivity may increase with 0.45-0.9 mg/mL nanoparticles during polymerization, but
the scatter in the data prevents a definite conclusion. The fact that mixed gas selectivity is
22

generally smaller than pure gas selectivity,

and the pure gas selectivity of PEGMEMA-

1100-co-PEGMEMA-475 films is about 13, shows that the maximum selectivity of

131
 

nanocomposite membrane is almost the same as PEGMEMA-1100-co-PEGMEMA-475
film if not larger.
20
18
16
14
12
10
8
6
4
2
0

120

80

α CO2/H2

Permeability  Barrer

100

60
40
20
0
‐20

0

2

0

1
2
3
SiO2 Nanoparticles in 
Monomer Suspension mg/mL

4

SiO2 Nanoparticles in Monomer 
Suspension mg/mL

  

  

Figure 4.5. The influence of nanoparticle concentration during polymerization on CO2
permeability (left) and CO2/H2 selectivity (right) of membranes coated with
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700). The monomer ratio of
PEGMEMA-1100, PEGMEMA-475 and PEGDA is 3:3:2.
Higher concentrations of nanoparticles in polymerization solutions gave decreases
in permeability for the nanocomposite membrane. Previous studies of MMMs indicated
5

an increase in permeability of permeate gas due to either an increase of FFV, and
12

transport through permeable particles such as zeolites.

In my case, the nanoparticles

embedded in the polymer films are tightly surrounded with covalently grafted polymer,
which may lead to different interfaces between nanoparticles and polymers compared to
simple blending. As Figure 4.6 (A) and (B) illustrate, the interface between nanoparticles
and polymers should contain less voids with grafting from nanoparticles than with
blended polymer/nanoparticle films. The size of the free volume around SiO2
132
 

nanoparticles mixed with poly(4-methyl-2-pentyne) ranges from 0.02 nm to 1 nm
5,23

according to the PALS.

Gas transport is very sensitive to the size of the free volume.
6

For example, the kinetic diameters of N2 and O2 are 3.64 Å, and 3.46 Å.

Despite a

difference in diameter of only 0.18 Å, a small difference in the size of free-volume
24

elements dramtically impacts diffusion.

Spontak and co-workers studied the effect of

embedding of SiO2 nanoparticles by blending and covalently bonding. The nanoparticles
modified with methacrylate on the surface were able to incorporate into surrounding
polymer chains. The methacrylate-coated nanoparticles covalently bonded to the polymer
decreased CO2 permeability compare to a film without nanoparticles. However,
25

hydroxyl-terminated nanoparticles did not reduce the permeability as much.

With

polymer covalently linked to the nanoparticles, the inclusion of the particles likely does
not increase free volume to enhance permeability. In such a case, the nanoparticle is
simply an obstacle for gas permeation. In my study, at high loading the nanoparticles in
the films likely acted as obstacles in the permeation pathway. However, there are cases
(0.45 mg/mL nanoparticle loading) where the permeability of CO2 in the nanocomposite
film compare is higher than in the corresponding poly(PEGMEMA-1100-coPEGMEMA-475-co-PEGDA-700)

film without nanoparticles (Figure 4.5).

loadings, the nanoparticles may create some free volume.

133
 

At low

 

Figure 4.6. Illustration of the space between nanoparticles and the polymeric matrix
embedding the particles. A) The nanoparticles contain grafted polymer prepared by
surface-initiated ATRP; B) the nanoparticles are blended in the polymer matrix, there is
no covalent bond between nanoparticle and polymer; C) the nanoparticles contain grafted
polymer grown by surface-initiated ATRP with lower initiator density.
The elasticity of PEGMEMA-co-PEGDA may also affect changes in free volume
upon incorporation of nanoparticles in the membrane. According to Pechar and coworkers,8 the rigid nature of glassy polymers surrounding nanoparticles creates free
volume. In contrast, embedding nanoparticles in rubbery, cross-linked co-polymers may
create much less free volume. Base on the Tg of -47 °C for poly(PEGMEMA-475-co26

PEGDA) wih 50% of each monomer,

I expect, poly(PEGMEMA-1100-co-

PEGMEMA-475-co-PEGDA-700) is also rubbery at room temperature. Thus, the
permeability and selectivity poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA700) may not change greatly upon nanoparticle incorporation.

134
 

4.3.2.2 Influence of Nanoparticle Initiator Density on Gas Permeation Through
Nanoparticle-containing Films
As Figure 4.6 (A) illustrates, nanoparticles coated with grafted polymer and
encapsulated in a cross-linked polymer film may not increase free volume as much
uncoated nanoparticles physically blended into a film. However, although the voids
between nanoparticles increase with particle loading, eventually the voids will create a
non-selective path. A number of recent studies indicated that nanoparticle blending in
5,12

membranes has this limit,

and the unstable nature of such MMMs may be an problem

for their application. Perhaps if the nanoparticles had a low density of initiator on the
surface, covalent bonds could still stabilize nanocomposite membranes without
decreasing the induced increases in free volume (Figure 4.6 (C)).

Moreover, the

covalently bonded nanoparticles may not create a non-selective path as easily as the
blended in nanoparticles.
According to kinetics studies of the reaction of chlorosilanes with SiO2
surfaces,

27

the rate is first order with respect to monochlorotrimethylsilane concentration

and hydroxyl group density on the SiO2 surface. Moreover, after 10 min only 53%
percent of the reaction has occurred. Assuming the dimethylmonochlorosilane initiator,
Cl(CH3)2Si(CH2)11OCOC(CH3)2Br, has a similar reaction rate, a 10 min time for
initiator attachment to nanoparticles would derivatize only half of the reactive surface
hydroxyl groups. The resulting nanoparticles embedded in a polymeric matrix may be
more like Figure 4.6 (C) than Figure 4.6 (A). Figure 4.7 shows the SEM images of cross135
 

sections of nanocomposite membranes prepared using nanoparticles derivatized with
initiator for only 10 min. The averaged film thicknesses measured from at least two
images (at least eight measurements) were used for the gas permeability calculations.

 

Figure 4.7. SEM images of cross-sections of alumina membranes coated with a
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700)
film
containing
nanoparticles. The polymerization solution contained 0.25 M PEGMEMA-1100, 0.25 M
PEGMEMA-475, 0.17 M PEGDA, 2.0 mM CuCl, 0.60 mM CuBr2, 6.0 mM HMTETA
and SiO2 nanoparticles. The concentrations of nanoparticles in the monomer suspension
were 0.18 mg/mL (A) and 1.35 mg/mL (B). The polymerization time was 30 min.
Figure 4.8 shows permeability and selectivity data for nanoparticle-containing
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700) membranes prepared with
SiO2 nanoparticles reacted with initiators for 6 h (squares) or only 10 min (triangles).
The similarity between the two types of membranes may occur because at high initiator
density, steric constraints prohibit ATRP from occurring at all initiation sites. Thus the
two types of nanoparticles may contain similar densities of chains grafted to the surface.
Data from Kim et al. suggest that ATRP uses only 10% of the initiators in monolayers on
gold.

28

136
 

120

20
18
16
14
12
10
8
6
4
2
0

CO2 Permeability Barrer

100

α CO2/H2

80
60
40
20
0

‐20

0

 

1

2

3

0

SiO2 Nanoparticles in Monomer 
Suspension mg/mL

1

2

3

SiO2 Nanoparticles in Monome
Suspension mg/mL
 

Figure 4.8. The effect of initiator attachment time on CO2 permeability (left) and
CO2/H2 selectivity (right) of porous alumina membranes coated with poly(PEGMEMA1100-co-PEGMEMA-475-co-PEGDA-700) nanoparticle-containing films. The open
circles indicate membranes prepared with SiO2 nanoparticles reacted with silane initiator
for over 6 hr, and the open triangles indicate membranes prepared with SiO2
nanoparticles reacted with silane initiator for only 10 min.
Figure 4.9 is the summary of CO2/H2 selectivity and CO2 permeability data from this
section. The black line indicates the upper bound of CO2/H2 separations given by
Freeman. Unfortunately, although the nanoparticle-containing membranes approach the
upper bound, they do not cross it.

137
 

αCO2/H2

100

10

1
0.01

0.1

1

10

100

1000

10000

CO2 Permeability Barrer
Figure 4.9. The upper bound line29 of CO2/H2 separations and data for membranes
composed of alumina coated with poly(PEGMEMA-1100-co-PEGMEMA-475-coPEGDA-700) nanoparticle-containing films. Filled diamonds represent nanocomposite
membranes synthesized with nanoparticles reacted with initiator for 6 h, and open
triangles represent corresponding membranes with nanoparticles reacted with initiator for
only 10 min.
 

4.4 Conclusions
Surface-initiated ATRP of PEGMEMA-1100, PEGMEMA-475 and PEGDA in
the presence of initiator-modified SiO2 nanoparticles gave a membrane with covalently
embedded nanoparticles. The nanoparticles introduced in the polymer film did not
promote the CO2/H2 selectivity or CO2 permeably significantly. Further studies of
grafting density on the nanoparticles suggests that the free volume is not affected by the

138
 

initiator density on the nanoparticles, at least at high density (>0.5). However, there may
be an effect at much lower initiator density.

139
 

REFERENCES

140
 

References
1. Kesting, R. E.; Fritzsche, A. K. Polymeric Gas Separation Membranes; John
Wiley and Sons, Inc.: New York, 1993.
2. Lin, H.; Freeman, B. D. J. of Mol. Struc 2005, 739, 57-74.
3. Cohen, M. H.; Turnbull, D. J. Chem. Phys. 1959, 31, 1164-1169.
4. Thran, A.; Kroll, G.; Faupel, F. A. Thran, G. Kroll, F. Faupel 1999, 37, 3344-3358.
5. Merkel, T. C.; Freeman, B. D.; Spontak, R. J.; He, Z.; I. Pinnau; Meakin, P.; Hill,
A. J. Science 2002, 296, 519-522.
6. Robeson, L. M. J. Membr. Sci. 1991, 62 (2), 165-185.
7. Li, Y.; Chunga, T.-S.; Caoa, C.; Kulprathipanja, S. Jouranl of Membrane Science
2005, 260, 45-55.
8. Pechar, T. W.; Kima, S.; Vaughan, B.; Maranda, E.; Baranauskas, V.; Riffle, J.;
Jeong, H. K.; Tsapatsis, M. J. Membr. Sci. 2006, 277, 210-218.
9. Zornoza, B.; Seoane, B.; Zamaro, J. M.; Tellez, C.; Coronas, J. ChemPhysChem
2011, 12, 2781-2785.
10. Jianga, L. Y.; Chunga, T. S.; Caoa, C.; Huanga, Z.; Kulprathipanja, S. J. Membr.
Sci. 2005, 89-100, 252.
11. Li, Y.; Guana, H.-M.; Chunga, T.-S.; Kulprathipanja, S. J. Membr. Sci. 2006,
275, 17-28.
12. Ahna, J.; Chunga, W.-J.; Pinnau, I.; Guiver, M. D. J. Membr. Sci. 2008, 314.
13. Bernardo, P.; Drioli, E.; Golemme, a. G. Ind & Eng. Chem. Res. 2009, 48, 4638141
 

4663.
14. Kusuma, V. A.; Gunawan, G.; Smith, Z. P.; Freeman, B. D. Polymer 2010, 51,
5734-5743.
15. Kusumaa, V. A.; Freemana, B. D.; Bornsb, M. A.; Kalikab, D. S. J. Membr. Sci.
2009, 327, 195-207.
16. Lin, H.; Wagner, E. V.; Freeman, B. D.; Toy, L. G.; Gupta, R. P. Science 2006,
311, 639-642.
17. Grajales, S. T.; Dong, X.; Zheng, Y.; Baker, G. L.; Bruening, M. L. Chem. Mat.
2010, 22, 4026-4033.
18. Matyjaszewski, K.; Miller, P. J.; Shukla, N.; Immaraporn, B.; Gelman, A.;
Luokala, B. B.; Siclovan, T. M.; Kickelbick, G.; Vallant, T.; Hoffmann, H.; Pakula,
T. Macromolecules 1999, 32, 8716-8724.
19. Du, J.; Chen, Y. Macromolecules 2004, 37, 3588–3594.
20. Maxwell, C. Treatise on Electricity and Magnetism; Oxford University Press:
London, 1973.
21. Gonzo, E. E.; Parentis, M. L.; Gottifredi, J. C. J. Membr. Sci. 2006, 277, 46-54.
22. Ribeiro, C. P.; Freeman, B. D. Polymer 2010, 51, 1156-1168.
23. Kobayashi, Y.; Haraya, K.; Hattorit, S.; Sasuga, T. Polymer 1994, 35, 925-928.
24. Kusumaa, V. A.; Freemana, B. D.; Bornsb, M. A.; Kalikab, D. S. J. Membr. Sci.
2009, 327, 195–207.
25. Patel, N. P.; Zielinski, J. M.; Samseth, J.; Spontak, R. J. Macromol. Chem. Phys

142
 

2004, 205, 2409-2419.
26. Kalakkunnath, S.; Kalika, D. S.; Lin, H.; Freeman, B. D. Macromolecules 2005,
38, 9679–9687.
27. Hair, M. L.; Hertl, W. J. Phys. Chem. 1969, 73, 2372-2378.
28. Kim, J.; Bruening, M.; Baker, G. J. Am, Chem, Soc. 2000, 122, 7616-7617.
29. Freeman, B. D. Macromolecules 1999, 32, 375-380.
 

143
 

Chapter 5 An Investigation of the Rapid Surface-initiated
Polymerization of PEGMEMA

In chapters 3 and 4, studies of poly(PEGMEMA-1100-co-PEGMEMA-475) and
poly(PEGMEMA-1100-co-PEGMEMA-475-co-PEGDA-700)

brushes

showed

an

unusually rapid growth of these films from porous alumina. A 30-min surface-initiated
polymerization often yields 100 m-thick films. This chapter examines possible reasons
for these remarkable growth rates. I simultaneously grow brushes on both sides of porous
alumina to prove that the fast polymerization is not due to precipitation. Film thickness
increases non-linearly with monomer concentration, showing that the process is not a
simple first-order polymerization. Kinetic studies of in-solution polymerization of
poly(PEGMEMA-1100-co-PEGMEMA-475) and poly(PEGMEMA-475) that the most
probable reason behind the rapid polymerization is that a high concentration of
PEGMEMA-1100 alters solvent polarity to enhance catalyst activity.

5.1 Introduction
5.1.1

Growth of Thick Polymer Films from Porous Alumina
Chapters 3 and 4 discussed the synthesis of cross-linked poly(PEGMEMA-co-

PEGDA) and nanoparticle-containing poly(PEGMEMA-co-PEGDA) films from porous
supports. Remarkably, when grown from porous alumina, these films are as thick as 200
μm (Figure 5.1) after just 30 – 45 min of polymerization. In many previous studies,
surface-initiated polymerization of other monomers yielded films with thicknesses of
144
 

1

only a few nm in a similar time. Huang and co-workers reported a rapid polymerization
of poly(tert-butyl acrylate) brushes, but the ellipsometric thickness of the film on SiO2
2

was still only 200 nm after 20 min of polymerization. A few studies reported “rapid”,
surface-initiated ATRP (SI-ATRP) of PEG-based monomers. Xu and coworkers
polymerized PEGMEMA-360 from oxidized silicon to achieve a film thickness of 38 nm
3

in 4 h. All of these thicknesses are several orders of magnitude lower than the
thicknesses of the poly(PEGMEMA-1100-co-PEGMEMA-475) films that I prepared on
porous alumina with a 30-min polymerization. An understanding of the exceptionally fast
PEGMEMA polymerization from porous alumina is vital for reducing the thickness of
membrane skins to improve membrane permeance.

145
 

 

Figure 5.1. SEM images of cross-linked poly(PEGMEMA)-based films grown from
porous alumina. (A) poly(PEGMEMA-co-PEGDA), PEGDA molar fraction of 25%. (B)
poly(PEGMEMA-co-PEGDA) film with SiO2 nanoparticles, PEGDA mole fraction of
10%. (C) poly(PEGMEMA-1100-co-PEGMEMA-475) The total monomer concentration
in the polymerization solution was 0.67 M, and 0.34 M for (C); catalyst concentrations
were CuCl- 2 mM, CuBr2- 0.67 mM and HMTETA- 6 mM. Polymerization times were
120 min for (A) and 30 min for (B) & (C).
At least three factors might contribute to the exceptional thickness of
poly(PEGMEMA) on alumina. (1) The high porosity of the alumina may lead to a much
higher amount of immobilized initiator than the formation of initiator monolayers on flat
surfaces such as oxidized silicon wafers. However, a recent study of poly(Nisopropylacrylamide) (PNIPAM) growth from initiators on flat and etched silicon wafers
suggested that the increased surface area of the etched wafer did not enhance the rate of
146
 

4

ATRP. (2) Formation of polymer in solution and subsequent precipitation on the face of
the membrane could yield thick films. (3) The monomer/water mixture may provide a
solvent that is especially favorable to ATRP. Armes showed that the presence of water
sometimes enhances the rate of ATRP, presumably because the solvent favors the Cu(II)
5

form of the catalyst to promote the formation of active chains. After briefly reviewing
aqueous ATRP below, this chapter examines whether precipitation or solvent effects
might account for the extremely rapid growth of poly(PEGMEMA-1100-co-PEGMEMA475) brushes from porous alumina.

5.1.2

Aqueous ATRP
5

ATRP can occur in neat monomer, in a solution, or in a heterogeneous system
6

(e.g., an emulsion). “Various solvents, such as benzene, toluene, anisole, diphenyl ether,
ethyl acetate, acetone, dimethyl formamide (DMF), ethylene carbonate, alcohol, water,
7

carbon dioxide, and others were used for different monomers.” Coca et al. reported the
first aqueous ATRP. Polymerization of 2-hydroxyethyl acrylate (HEA) at 90 °C for 12 hr
gave 87% conversion with Mn = 14,700 and a polydispersity index (PDI, Mw/Mn) of
1.34. Compared with neat polymerization, the reaction rate was slightly slower (90 °C
polymerization for 10 hr in neat monomer gave 92% conversion) and the PDI (1.2 for
8

neat polymerization) was higher. Armes later reported aqueous ATRP of a carboxylatecontaining monomer. After polymerization for 10 h at 90 °C, sodium methacrylate
147
 

9

conversion reached 70-80% with a PDI of 1.2-1.3. The Armes group also reported
aqueous ATRP of PEGMEMA-408 (also called methoxy-capped oligo(ethylene glycol)
methacrylate, OEGMA in the original reference). Compared to neat monomer, the
presence of water accelerated the ATRP significantly. With the same concentration of
catalyst, 2.2'-bipyridine and CuCl, the aqueous polymerization reached 90% monomer
conversion in 20 min, whereas the neat polymerization required 15 h for comparable
conversion. The PDI resulting from aqueous polymerization was also slightly lower than
5,10

for the neat monomer: 1.15-1.30, and 1.30 respectively.

According to Wang and

Armes, this accelerating effect of water may occur because a) the catalyst is very active
in polar solvent; and b) “since OEGMA is a sterically congested monomer, its
termination rate constant is several orders of magnitude lower than that of conventional
5

methacrylates, which is expected to lead to improved living character.” Ye and Narain
studied ATRP of N-isopropylacrylamide (NIPAM) in water and water-organic solvent
mixtures. At room temperature, NIPAM ATRP in organic solvents follows wellcontrolled first-order kinetics, but in both water and water-organic solvent mixtures
(organic:water 4:1 or 3:1) the polymerization is very rapid and shows significant
termination. However, when performed at 0 °C, the aqueous polymerization yields molar
11

masses that increase linearly with conversion and pseudo-first-order kinetics.

Aqueous ATRP is very convenient for synthesis of water-soluble polymers.
Perruchot et al. successfully polymerized poly(PEGMEMA) from the surface of SiO2
nanoparticles.

12

Later, Wang and co-workers performed similar SI-ATRP from Au

148
 

nanoparticles. They successfully grafted poly(PEGMEMA) on Au nanoparticles through
a DNA-modified initiator.

13

A few recent studies of (poly(2-dimethylamino)ethyl-

methacrylate) (PDMAEMA) and PEGMEMA block copolymer films binding with DNA
through polyanion-polyanion electrostatic forces. The binding of active DNA indicated
14,15

the block-copolymer has some potential of gene delivery applications.

5.2 Experimental
5.2.1

Chemicals and Materials
PEGMEMA monomers with Mn = 475 Da (PEGMEMA-475), and Mn = 1,100

Da (PEGMEMA-1100) as well as PEGDA Mn=700 Da were obtained from Aldrich. All
monomers contain both 100 ppm 4-methoxyphenol and 200 or 300 ppm 2,6-di-tert-butyl4-methylphenol as inhibitors and were used as received. Proton NMR spectra show that
the average numbers of ethylene oxide units in PEGMEMA-475, and PEGMEMA-1100
are 8.8, and 23.1, respectively (Figure 2.1).

16

2-Bromoisobutyryl bromide (BiBB) (98%),

CuCl (99.995%), CuBr2 (99.999%), and 1,1,4,7,10,10-hexamethyltriethylenetetramine
(HMTETA) (97%) were purchased from Aldrich, and deionized water was obtained
using a Millipore system (Milli-Q, 18.2 MΩ•cm).

Triethylamine (TEA) and DMF

(99.8%) were purchased from Jade Scientific and used as received. THF was distilled
over sodium metal with benzophenone. The porous alumina substrates were Anodisc
membrane filters (25 mm disks with 0.02 µm surface pores) purchased from Whatman.

149
 

Silicon (100) wafers were obtained from NOVA Electronic Materials. Silane initiator,
Cl(CH3)2Si(CH2)11OCOC(CH3)2Br,

17

and

a

water-soluble

initiator,

5

CH3(OCH2CH2)3OCOC(CH3)2Br, were prepared by literature methods. A gas cylinder
containing 50% CO2 and 50% H2 was purchased from Praxair.

5.2.2

Polymerization

5.2.2.1 Initiator Attachment
Initiator attachment to both alumina membranes and oxidized silicon wafers was
performed

by

soaking

UV-Ozone

cleaned

(15

min)

substrates

dimethylchlorosilane initiator, Cl(CH3)2Si(CH2)11OCOC(CH3)2Br,

in

1

mM

in THF overnight.

The membranes were then rinsed with 5 mL THF three times and dried with flowing N2.
In a few cases, the alumina membranes were rinsed by flowing 5 mL THF through the
membrane using a peristaltic pump. The two rinsing method made no significant
difference in the thickness of the resulting polymer brush.
5.2.2.2 Catalyst Solution
A catalyst stock solution was prepared by adding 0.0596 g (0.6 mmol) of CuCl
and 0.0450 g (0.2 mmol), CuBr2 and 490 µL (1.8 mmol) of HMTETA ligand into
degasse 30.0 mL deionized H2O. Detail of this step is described in (Chapter 3, page 88).
+

2+

The nominal mole ratio was Cu :Cu :HMTETA was 3:1:9.

150
 

5.2.2.3 SI-ATRP
The monomer solution was prepared by dissolving the desired amount of
monomer in water and degassing via four freeze, pump, thaw cycles. Monomer solution
and initiator-modified alumina membranes were transferred into the N2-filled glove bag
with the catalyst solution. The polymerization mixture was prepared by combining
degassed monomer and catalyst solutions in a 9:1 volume ratio to produce a light blue
solution containing 0.67 M monomer, 2.0 mM CuCl, 0.60 mM CuBr2, and 6.0 mM
HMTETA. The initiator-modified membranes were immersed in the polymerization
solution for designated periods of time, and the resulting polymer films were rinsed with
5 mL DMF three times followed by 5 mL of ethanol three times and drying under a N2
stream.
5.2.2.4 Solution ATRP
Monomer solutions were prepared similarly to solutions for SI-ATRP, except that
water-soluble initiator, CH3(OCH2CH2)3OCOC(CH3)2Br, was added to the monomer
solution before freeze, pump, thaw cycles. The catalyst solution was added to degassed
monomer solution to start the polymerization. The reaction mixture was vigorously
stirred. Catalyst concentrations were the same as in surface-initiated polymerization.

151
 

5.2.3

Characterization Methods
Scanning electron microscopy (SEM) was performed mainly with a JEOL semi-

in-lens cold cathode field-emission scanning electron microscope, model JSM-7500F,
operating with an r-filter in signal maximum mode. Samples were sputter coated with 10
nm of Osmium prior to imaging. A few SEM images were obtained with a JEOL 6400V
instrument with a LaB6 emitter (Noran EDS). The samples for the JEOL 6400V were
coated with 8 nm of gold before of imaging.
Nuclear Magnetic Resonance (NMR) spectra of the reaction mixture were
obtained with a Varian 500 MHz superconducting NMR-Spectrometer operating at
499.738 MHz and interfaced with a Sun Microsystems Ultra5 UNIX console.
Gel permeation chromatography (GPC) was used to determine the molecular
weights of polymers. GPC was performed at 35 °C in DMF using a PLgel 20 μm
MIXED-B column at a flow rate of 1 mL/min. Two detectors were used: a Waters R410
differential refractometer and a Waters 996 photodiode array. The concentration of the
polymer samples was approximately 1 mg/mL, and each solution was filtered through a
Whatman 0.2 μm PTFE filter before injection. Molecular weights are reported relative to
18

monodisperse polystyrene standards.

152
 

5.3 Results and Discussion
5.3.1

Does Precipitation Increase Film Thickness
Precipitation of polymer on the membrane might explain the exceptionally high

film thicknesses on porous alumina. However, NMR spectra of the monomer solution
after growth of the poly(PEGMEMA-1100-co-PEGMEMA-475) (PPEGMEMA) from
the surface show that little polymer is present in solution. (The only initiator is that
attached to the substrate.) In the spectrum in Figure 5.2, there is no signal at 1 ppm, so the
signature peak of CH3 from the polymerized methacrylate backbone is absent. Strong
signals from the alkene protons (5.62 and 6.0 ppm) indicate the presence of extensive
monomer in the solution. This experiment is not definitive proof, however, that there is
no precipitation because if polymer were extremely insoluble, it would not remain in
solution. Nevertheless, the extent of polymerization in solution is minimal.

153
 

   6                     5                      4                      3                     2                     1       ppm 

 

Figure 5.2. NMR spectrum of the monomer and catalyst mixture (contains CuCl, CuBr2
HMTETA, PEGMEMA-1100 and PEGMEMA-475) after surface-initiated
polymerization from a porous alumina membrane. The absence of a PPEGMEMA signal
in region b and the large peak a for PEGMEMA show that there is little polymer in
solution.
Typically, we grow poly(PEGMEMA-1100-co-PEGMEMA-475) brushes with
the alumina in a holder that exposes only the top surface of the membrane (Figure 5.3,
left). In this apparatus, precipitating polymer should settle on the membrane. To better
examine whether precipitation contributes to film thickness, we also mounted the
membrane in the middle of the reaction mixture (Figure 5.3, right) and compared the
films thicknesses on the two sides of the membrane. As Figure 5.4 shows, exposing only
the top face of the membrane to the polymerization solution yields a thick films only on
154
 

the top face of the membrane. In contrast for the membrane in the middle of the
polymerization solution, thick films form on both sides of the membrane. Precipitation
should not increase the thickness of the film on the bottom of the membrane because the
precipitate should settle on the bottom of the reaction vessel not on the membrane.
Nevertheless, these films are still 48 m thick after 15 min of reaction.

 

Figure 5.3. Illustration of different positions of an initiator-modified alumina substrate in
a monomer-catalyst mixture during polymerization. On the left only the top of the
membrane is exposed to monomer and catalyst, whereas on the right both sides are
exposed to the solution
 

Figure 5.4. SEM images of the cross sections of membranes modified with
poly(PEGMEMA-1100-co-PEGMEMA-475) brushes. The condition for polymerizations
are [CuCl]=2 mM, [CuBr2]=0.67 mM, [HMTETA]=6 mM, and PEGMEMA155
 

1100:PEGMEMA-475=1:1, monomer total concentration is 0.34 M. (A) Only the top
side of the membrane was exposed to the monomer solution (B) Both sides of the
membrane were exposed to the monomer solution.
Figure 5.5 compares the film thicknesses on the top and bottom faces of a number
of membranes modified by polymerization with times ranging from 10 to 30 min. The
poly(PEGMEMA-1100-co-PEGMEMA-475)

brush

thickness

increased

with

polymerization time, and the difference in the film thickness on the top and bottom face
of the membrane is typically less than a factor of 2. Note that the alumina membrane has
different pore sizes on each side. The pore size on the top is 0.02 μm, and that on the
bottom is 0.2 μm. This difference in pore size might explain the small differences in
thickness on the two faces of the membranes. Overall, the high thickness of films on the
bottom face of the membranes suggests that precipitation is not responsible for the
exceptionally high film thickness.

156
 

120

SEM Thickness (μm)

100
80
60
40
20
0
0

5

10

15

20

25

30

35

Time of Polymerization (min)
 

Figure 5.5. Evolution of poly(PEGMEMA-1100-co-PEGMEMA-475) thickness with
polymerization time for films grown from both the top and bottom faces of initiatormodified porous alumina. The diamonds indicate the side of the membrane facing up, and
the circles represent the side facing down. The error bars indicate the standard deviation
of multiple SEM thicknesses on at least two membranes prepared under identical
conditions. The conditions for polymerization were [CuCl]=2 mM, [CuBr2]=0.67 mM,
[HMTETA]=6 mM, and PEGMEMA-1100:PEGMEMA-475=1:1, total monomer
concentration is 0.17 M.

5.3.2

The Effect of Monomer Concentration on Film Thickness
In controlled polymerization (minimal termination), the reaction rate is

proportional to the monomer concentration.

19

To investigate the kinetics of PEGMEMA

polymerization, I grew poly(PEGMEMA-1100-co-PEGMEMA-475) brushes from
porous alumina using total monomer concentrations ranging from 0.04 to 0.34 M. The
relative molar ratio of PEGMEMA-1100 and PEGMEMA-475 was 1:1 and the
157
 

polymerization time was 30 min. Figure 5.2 shows some representative SEM images of
film cross-sections, and I determined film thicknesses using at least two images (5
measurements) for a given film.

 

Figure 5.6. SEM images of poly(PEGMEMA-1100-co-PEGMEMA-475) films grafted
from porous alumina. The monomer concentrations and film thicknesses are A) 0.34 M,
150.0 ± 9.2 μm, (B) 0.17 M, 93.7 ± 2.7 μm, (C) 0.084 M, 19.7 ± 1.2 μm and (D) 0.042 M,
775 ± 120 nm. The monomer concentration is the sum of PEGMEMA-1100 and
PEGMEMA-475. The monomers were present in a 1:1 ratio, and polymerization times
were 30 min.
Figure 5.3 shows film thickness as a function of the total monomer concentration.
The film thickness increases faster than expected as the monomer concentration increases.
In particular, when the monomer concentration doubles from 0.077 to 0.17 M, the film

158
 

increases 5 fold, from 19.7 to 100 μm. Higher concentrations of monomers likely alter
the polarity and viscosity of the reaction mixture to accelerate the reaction.

180

SEM Film Thickness (μm)

160
140
120
100
80
60
40
20
0
0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

Monomer Concentration (M) 
 

Figure 5.7. Thicknesses of poly(PEGMEMA-1100-co-PEGMEMA-475) brushes as a
function of monomer concentration. The brushes were grown for 30 min from initiators
immobilized on porous alumina. The monomer concentration is the sum of the two
monomers, which were present in a 1:1 ratio. The error bars indicate the standard
deviation of multiple thickness measurements on at least two membranes prepared under
identical conditions, [CuCl]=2 mM, [CuBr2]=0.67 mM, [HMTETA]=6 mM, and
PEGMEMA-1100:PEGMEMA-475=1:1. The polymerization time was 30 min in all
cases.

5.3.3 In-solution Polymerization of PEGMEMA
To help us understand the rapid,

surface-initiated

polymerization of

poly(PEGMEMA-1100-co-PEGMEMA-475) from porous alumina, we also examined
159
 

polymerization from initiators in solution. Table 5.1 summarizes results from in-solution
polymerizations that employed the same monomer and catalyst concentrations as the
surface-initiated polymerization. The only variation with respect to polymerization from
a surface was the monomer:initiator ratio, which of necessity was much lower than in
surface-initiated polymerization. All solution reactions were quenched by bubbling air
through the samples for 30 s.
Table 5.1. Characterization of the polymer resulting from copolymerization of
PEGMEMA-1100 and PEGMEMA-475 initiated by CH3(OCH2CH2)3OCOC(CH3)2Br
The reaction was catalyzed by [CuCl]- 2 mM, [CuBr2]- 0.67 mM and [HMTETA]- 6 mM,
in water at room temperature. Each monomer initially was present at 0.34 M. The
conversions are calculated via integration of the NMR signals from the –CH3 group on
the polymethacrylate back bone and the vinyl protons of the unreacted monomer.
Monomer: Initiator Reaction Time (min) Conversion %

Mn by GPC

PDI

300:1

30

99.0

203000

5.71

300:1

2

83.9

216000

2.61

1600:1

20

~0

1100

-

200:1

0.5

89.8

112000

2.31

200:1

3.5

93.0

166000

2.92

The conversion data show that the reaction is very rapid. For the 200:1 monomer:
initiator ratio, the reaction is 90% complete in 0.5 min. However, PDI values >2 show
that the resulting polymers have a broad range of molecular weights. This suggests a
relatively uncontrolled reaction, where either polymerization is much faster than
initiation and/or significant termination occurs. When the monomer:initiator ratio was

160
 

1600:1, the reaction did not yield significant amounts of polymer. This might occur in
part because we did not remove the inhibitors from the monomers. The monomer:
inhibitor ratio is 480:1, so the ratio of inhibitor to initiator was 3:1 in the solution with a
1600:1 monomer: initiator ratio.

Notably, the reaction was much faster when the

monomer: initiator ratio decreased from 300:1 to 200:1. A higher initiation efficiency
combined with a larger amount of initiator may increase the polymerization rate.
However, the Mn value did not drop drastically on going to from a monomer:initiator
ratio of 300:1 to 200:1, which suggests that the initiation efficiency did not change by
even a factor of 2.
In a study by Armes and co-workers, ATRP of PEGMEMA (contains repeating
PEO units) in water (catalyzed by CuCl and bpy) was much faster than polymerization of
neat monomer. We performed copolymerization of PEGMEMA-475 under the same
5

conditions (Table 5.2, reaction 1 lists the conditions). and compared the results with
polymerization of PEGMEMA-475 using the catalyst we employed in surface-initiated
polymerization (Table 5.2, reaction 2). Additionally, we copolymerized PEGMEMA-475
and PEGMEMA-1100 (Table 5.2, reaction 3).

We examined the kinetics of the

polymerization by periodically taking a small aliquot from the reaction mixture,
quenching the polymerization by bubbling with air for 30 s, and subsequently examining
the aliquot NMR spectrum.

161
 

Table 5.2. Different conditions for PEGMEMA polymerization initiated by
Reactions 1 and 2 employed
CH3(OCH2CH2)3OCOC(CH3)2Br in solution.
PEGMEMA-475 as the monomer, but reaction 3 used equimolar PEGMEMA-475 and
PEGMEMA-1100 with the total monomer concentration listed.
Monomer

Monomer:Initiator

CuCl

CuBr2

Ligand

(mM)*

Ratio

(mM)

(mM)

(mM)

1

1634

33:1

49

0

bpy, 90

2

1634

33:1

2

0.67

HMTETA, 6

3

336

200:1

2

0.67

HMTETA, 6

Reaction

* The monomer concentration for reaction 3 is limited because the maximum
PEGMEMA-1100 monomer concentration is 0.9 M in pure monomer.

120

% Conversion

100
80
60
40
20
0
0

5

10

15

Reaction Time (min)

20

25
 

Figure 5.8. Evolution of monomer conversion with time for polymerization of
PEGMEMA-475 (diamonds-bpy catalyst, reaction 1, and circles-HMTETA catalyst,
reaction 2) and a PEGMEMA-1100, PEGMEMA-475 mixture (triangles, reaction 3).
Table 5.2 lists the reaction conditions.
162
 

Figure 5.8 shows monomer conversion versus time for the reactions in Table 5.2.
Under similar conditions, Armes achieved 80% conversion in 24 min, Mn=12500,
5

PDI=1.2, and I obtained 95% conversion in 22 min, Mn=16100, PDI=1.36). Note that
the monomer PEGMEMA-475 contains 1.5 more PEO units than the PEGMEMA that
Aarmes reported. Comparing the rates of reactions 1 and 2, the HMTETA catalyst system
(circles) gives lower rates than the bpy catalyst for polymerization of PEGMEMA-475.
I

Note that the concentration of Cu for the HMETEA catalyst system is about 10 times
I

less than the Cu concentration in the bpy system. GPC gives PDI values of 1.34 and 1.40,
for polymerization of PEGMEMA-475 from the bpy and HMTETA catalysts,
respectively. This result suggests that the rapid polymerization of poly(PEGMEMA1100-co-PEGMEMA-475) from the surface was probably not simply due to the high
activity of the HMTETA catalyst. However, the presence of PEGMEMA-1100 in the
monomer mixture increases the initial polymerization rate by an order of magnitude or
more (triangles, Figure 5.8). PEGMEMA-1100 has a very long PEO side chain that
swells in aqueous solution, and the monomer itself might serve as a solvent that
accelerates

polymerization.

Future

studies

will

investigate

polymerization

of

PEGMEMA-475 in the presence of oligo ethylene glycol to see if this increases
polymerization rates. We suspect that changing the polarity of the solvent increases the
activity of the catalyst to give a rapid, uncontrolled polymerization. Chapter 2 described
surface-initiated growth of poly(PEGMEMA) films with three monomers that had
different PEO side chain lengths (PEGMEMA-300, PEGMEMA-475, and PEGMEMA1100) using a catalyst solution prepared in DMF. The resulting polymerization rate from

163
 

Au did not reveal a dramatic impact of monomer side chain length on film thickness. The
combination of water and PEGMEMA is likely necessary to see the dramatic increase in
polymerization rate.

5.4 Conclusion
This chapter discussed why the synthesis of poly(PEGMEMA-1100-coPEGMEMA-475) films from porous alumina yields unusually thick films (100 m in 30
min). This rapid polymer brush growth is several orders of magnitude faster than
previous results.2 NMR results indicate that the reaction mixture did not polymerize
greatly in solution, and precipitation does not appear to contribute to film growth. By
altering the monomer concentration and comparing in-solution polymerization with work
from other groups, we think the most probable reason for the extremely rapid surfaceinitiated polymerization is a change in the solvent characteristics for aqueous solutions
containing large amounts of PEGMEMA-1100. However, the details behind acceleration
of the polymerization need further study.
 

 

164
 

REFERENCES

165
 

References
1. Sankhe, A. Y.; Husson, S. M.; S. Michael Kilbey, I. Macromolecules 2006, 39,
1376–1383.
2. Huang, W.; Kim, J.; Bruening, M.; Baker, G. Macromolecules 2002, 35, 11751179.
3. Xu, F. J.; Zhong, S. P.; Yung, L. Y. L.; Kang, E. T.; Neoh, K. G.
Biomacromolecules 2004, 2392-2403, 5.
4. Vasani, R. B.; McInnes, S. J. P.; Cole, M. A.; Jani, A. M. M.; Ellis, A. V.; Voelcke,
N. H. Langmuir 2011, 27, 7843–7853.
5. Wang, X.-S.; Armes, S. P. Macromolecules 2000, 33, 6640–6647.
6. Qiu, J.; Gaynor, S. G.; Matyjaszewski, K. Macromolecules 1999, 32, 2872-2875.
7. Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921-2990.
8. Coca, S.; Jasieczek, C. B.; Beers, K. L.; Matyjaszewski, K. J. Polym. Sci. Part A:
Polym. Chem. 1998, 36, 1417-1424.
9. Ashford, E. J.; Naldi, V.; O’Dell, R.; Billingham, N. C.; Armes, S. P. Chem.
Commun. 1999, 1285-1286.
10. Wang, X.-S.; Lascelles, S. F.; Jackson, R. A.; Armes, S. P. Chem. Commun. 1999,
1817-1818.
11. Ye, J.; Narain, R. J. Phys. Chem. B 2009, 113, 676–681.
12. Perruchot, C.; Khan, M. A.; Kamitsi, A.; Armes, S. P. Langmuir 2001, 17, 4479–
4481.
13. Lou, X.; Wang, C.; He, L. Biomacromolecules 2007, 8, 1385-1390.
14. Rungsardthonga, U.; Deshpandea, M.; Baileyb, L.; Vamvakakib, M.; Armes, S.
P.; Garnetta, M. C.; Stolnik, S. J. Contr. Release 2001, 73, 359-380.
15. Deshpande, M. C.; Davies, M. C.; Garnett, M. C.; Williams, P. M.; Armitage, D.;
Bailey, L.; Vamvakaki, M.; Armes, S. P.; Stolnik, S. J. Contr. Release 2004, 97, 143156.

166
 

16. Grajales, S. T.; Dong, X.; Zheng, Y.; Baker, G. L.; Bruening, M. L. Chem. Mat.
2010, 22, 4026-4033.
17. Matyjaszewski, K.; Miller, P. J.; Shukla, N.; Immaraporn, B.; Gelman, A.;
Luokala, B. B.; Siclovan, T. M.; Kickelbick, G.; Vallant, T.; Hoffmann, H.; Pakula,
T. Macromolecules 1999, 32, 8716-8724.
18. Liu, T.; Simmons, T. L.; Bohnsack, D. A.; Mackay, M. E.; Milton R. Smith, I.;
Baker, G. L. Macromolecules 2007, 40, 6040-6047.
19. Shipp, D. A.; Matyjaszewski, K. Macromolecules 2000, 33, 1553-1559.
 

167
 

Chapter 6 Summary and Future Work

This thesis discussed a few topics centered CO2/H2 separation membranes
developed on porous substrates. Chapter 1 introduced the topic of hydrogen economy,
and the separation techniques for CO2 and H2. Additionally, this chapter also introduced
solution-diffusion model, which provided the guide of material selection for reverseselective CO2/H2 separation membranes. Atomic Transfer Radical Polymerization
(ATRP) is a robust method of developing a selective skin, and is used for all the studies
of this entire thesis. Thus, a summary of ATRP mechanisms and each component is
gathered in chapter 1 as well.
Chapter 2 is represented the development on grafting a poly(ethylene glycol
methyl ether methacrylate) PEGMEMA-1100 and PEGMEMA-475 copolymer film from
porous alumina surface. The copolymer film successfully stopped PEGMEMA-1100 side
chain from crystallization without compromising the selectivity. Reflectance FTIR data
was used to study the film composition, and the how the composition affects the gas
separation properties.
Chapter

3

experimented

the

development

of

poly(PEGEMEMA)

and

poly(PEGMEMA-1100-co- PEGMEMA-475-co-PEGDA-700) cross-linked co-polymers.
By cross-linking, the nanocomposite membranes are prepared by introducing surface-

168
 

initiated SiO2 nanoparticles. Chapter 4 focused on this nanocomposite membrane, how it
is prepared and the gas separation properties are tested.
Chapter 5 is an investigation of an ultra-rapid polymerization of PEGMEMA1100 and PEGMEMA-475 in aqueous media. The polymer film thickness is usually
about 100 μm from 30-45 min of polymerization. A few controlled experiments were
performed to approach the reason of this ultra-fast polymerization. With more
understanding of the polymerization kinetics, thinner poly(PEGMEMA) films could be
prepared in the future for fast CO2/H2 separation.
With the work presented in this thesis the reported reverse-selective polymeric
membranes in the literature, I now want to propose some future work in this area. The
first is the nanocomposite membranes with covalently bonded nanoparticles embedded.
As discussed in chapter 4, the main reason that the nanoparticles failed to enhance the
permeability and selectivity could be the elasticity of rubbery poly(PEGMEMA-1100-coPEGMEMA-475-co-PEGDA-700) film is rubbery. Kalakkunnath and co-workers studied
1

the impact of the cross-link segments on the polymer’s transition temperature. Thran
reported the fraction of free volume (FFV) impact on gas separation properties in glassy
polymer.Lin, H. & Freeman, B. D. J. Molec. Structr. 2005 739, 57-74. Freeman and Lin
provided a detailed material selection guide for reverse-selective gas separation
3

membranes. Glassy polymer that is suitable for CO2/H2 separation combined with SIATRP on nanoparticles is promising in increasing the FFV and increase in the
permeability and selectivity.

169
 

The other meaningful approach of study reverse-selective gas separation
membrane is find a better support that is of less cost and ready to be packed in large
spiral wound modules. At the same time, evaluation of membranes’ gas separation
properties in high temperature and high humidity conditions are very important.

170
 

REFERENCES

171
 

References
1. Kalakkunnath, S., Kalika, D. S., Lin, H. & Freeman, B. D., Macromolecules 2005 38,
9679-9687.
2. Thran, A., Kroll, G. & Faupel, F. Gassy Plymers. 1999 37, 3344-3358.
3. Lin, H. & Freeman, B. D. J. Molec. Structr. 2005 739, 57-74.
 

172