PLACE IN RETURN BOX to remove this Checkout from your "cord.
TO AVOID F INES mum on or More data duo.

DATE DUE DATE DUE DATE DUE

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mun-9.1

————

SYNTHESIS AND CHARACTERIZATION OF POLYMERIC PSEUDOCROWN
ETHERS AND REVERSIBLE BLOCK/CRAFT COPOLYMERIC EMULSIFIERS
BASED UPON POLYMER COMPLEXATION
By

Arvind Mohan Mathur

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment Of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemical Engineering

1996

ABSTRACT

SYNTHESIS AND CHARACTERIZATION OF POLYMERIC PSEUDOCROWN
ETHERS AND REVERSIBLE BLOCK/GRAFT COPOLYMERIC EMULSIFIERS
BASED UPON POLYMER COMPLEXATION
By

Arvind Mohan Mathur

The complexation properties of poly(ethylene glycol) (PEG), arising due to its
Lewis base character, have been exploited for the development of two materials. First, a
scheme for the synthesis of polymeric pseudocrown ethers has been devised based upon
ion-induced templatization of oligomeric ethylene glycol diacrylates (PEGDA) in which
the oligomer assumes a circular conformation with the end groups in close proximity. A
combination of experimental and molecular modeling studies was used to characterize the
templatization process and design a one-step free-radical copolymerization of templated
PEGDA oligomers with a hydrophilic comonomer to produce crosslinked pseudocrown
ether (PCE) resins. Experimental evidence of templatization included the PEGDA-
induced solubilization of heavy metal salts into organics, proton NMR peak shifts, and
excimer fluorescence of pyrene end-labeled PEG oligomers. Molecular dynamics
simulations provided visualization of the templated conformations, and demonstrated a
decrease in the mean end-to-end distance in the presence of a cation for both the PEGDA
and the pyrene end-labeled oligomers. Ion-binding studies revealed the effects of the

monomer to solvent ratio. and the presence of a template cation (N i”) during synthesis on

the ion-binding capacity of the resins. The experimentally observed capacities of 0.3 mg
of Ni2+ per gram of polymer are comparable to that of previously reported polymer-bound
crown ethers.

The polymer complexation properties of PEG were also exploited in the design of
polymeric block/graft copolymeric emulsifiers comprising of a poly(methacrylic acid)
(PMAA) backbone and PEG grafts. The swelling properties of crosslinked PMAA-g-
PEG hydrogels revealed that changes in pH or solvent composition may be used to
disrupt the complex. The reversible oil emulsification and interfacial tension properties
of uncrosslinked PMAA-g—PEG copolymers were studied as a function of pH and
molecular architecture. Under complex-promoting (acidic) conditions these copolymers
exhibited surface active behavior (as evidenced by their ability to lower the oil-water
interfacial tension and form oil-in-water emulsions) while under basic conditions the
complex was broken and the surface activity was lost. The polarity sensitive fluorescence
of solubilized pyrene was used to study aggregate formation as a function of pH and to

determine a critical aggregate concentration as a function of molecular architecture.

To my parents and my wife

iv

ACKNOWLEDGMENTS

I would like to express my gratitude to several persons who have contributed to
the successful completion of this dissertation and have made my stay at Michigan State
University enjoyable.

My major professor and mentor, Dr. Alec Scranton, has contributed in more than
one way towards the accomplishment of the objectives of this research and my evolution
as a scientist. I will be forever grateful for his guidance, encouragement and enthusiasm.
I would like to thank the members of my committee DrS. Steven Boyd, Daina Briedis and
Bob Ofoli for their help and use of equipment in their laboratories. Dr. John Klier at
Dow Central Research, deserves special thanks for his interest, input and collaboration.
Thanks are also due to the Department of Chemical Engineering, specially to Julie
Caywood, Candy McMaster, and Faith Peterson, for being such good friends.

I would like to acknowledge financial support by a grant (# CTS-9110163) from
the National Science Foundation; the Research Excellence Fund of the State of Michigan;
and the Environment Enhancement Grant from the Dow Chemical Company.

Several members of my research group have contributed towards a gratifying and
healthy research environment. For this, I wish to express my sincere thanks to Bharath,

Dave, Eric, Jeff, Julie, Katy, Scott, Shail and Vijay. I would also like to thank several

undergraduates who I had the pleasure of working with and in particular, Brenda Becker
and Amanda Meeks, for their valuable help in the laboratory.

I am grateful to Dilum, Himanshu, Irma, Murty, Ram, Sanjay and their significant
others, for their friendship and all the good times we shared. I will always remember the
trips to the Upper Peninsula, the Broadway shows and other such activities that made my
stay at Michigan State much more complete.

I cannot even begin to express my gratitude to my wife Shalini, whose love,
support and understanding made this dissertation possible. I am grateful for her patience
and words of encouragement when I needed them the most. Finally, this dissertation
could not have been completed without the blessings and good wishes of our families, for

which I am eternally grateful.

vi

TABLE OF CONTENTS

 

 

 

LIST OF TABLES xiii

LIST OF FIGURES xiv
CHAPTER 1

BACKGROUND AND MOTIVATION l

1.1. INTRODUCTION .......................................................................................................... 1

1.2. POLYMERIC PSEUDOCROWN ETHERS - MOTIVATION ................................................. 2

1.2.1. Heavy Metal Ions in Aqueous Industrial Wastes ............................................. 2

1.2.2. Current Methods for Removing Heavy Metals from Water ............................ 4

1.2.3. Need for New Separation Methods for Heavy Metal Ions ............................... 7

1.3. POLYMERIC PSEUDOCROWN ETHERS - BACKGROUND ............................................... 8

1.4. MOLECULAR MODELING OF MACROMOLECULES AND LIGANDS .............................. l 1

1.4.1. Molecular Mechanics ..................................................................................... 12

1.4.2. Molecular Dynamics ...................................................................................... 13

1.5. REVERsIBLE COPOLYMERIC EMULSIFIERS - BACKGROUND AND MOTIVATION ........ 15

1.5.1. Need for the Development of Multiblock Reversible Emulsifiers ................ 19

1.5.2. Polymer Complexation Through Hydrogen Bonding .................................... 22

1.6. REFERENCES ............................................................................................................ 25

Vii

CHAPTER 2

 

 

OBJECTIVES OF RESEARCH 29
CHAPTER 3
ION-SOLUBILIZATION AND ION INDUCED TEMPLATIZATION BY
OLIGO(ETHYLENE GLYCOL) AND ITS DIACRYLATES 33
3.1. INTRODUCTION ........................................................................................................ 33
3.2. EXPERIMENTAL METHODS AND TECHNIQUES .......................................................... 38
3.2.1. Ion—Binding Studies of Oligomeric PEG and PEGDA .................................. 38
3.2.2. NMR Studies .................................................................................................. 39
3.2.3. Molecular Modeling ....................................................................................... 39
3.3. RESULTS AND DISCUSSION ...................................................................................... 40
3.3.1. Ion-Binding Studies with Oligomeric PEG ................................................... 40
3.3.2. NMR Studies .................................................................................................. 41
3.3.3. Molecular Dynamics Simulations .................................................................. 43
3.4. CONCLUSIONS .......................................................................................................... 51
3.5. REFERENCES ............................................................................................................ 53
CHAPTER 4

EFFECT OF ION-INDUCED TEMPLATIZATION ON THE END-TO-END
DISTANCE OF PYRENE END-LABELED TETRA- AND PENTA-ETHYLENE

 

GLYCOL 54
4.1. INTRODUCTION ........................................................................................................ 54
4.2. EXPERIMENTAL METHODS AND TECHNIQUES .......................................................... 57

4.2.1. Synthesis of Pyrene End-Labeled Oligomeric Ethylene Glycols .................. 57
4.2.2. Fluorescence Studies ...................................................................................... 59
4.2.3. Molecular Dynamics Simulations .................................................................. 6O

viii

4.3. RESULTS AND DISCUSSION ...................................................................................... 61

 

4.3.1. Fluorescence Studies ...................................................................................... 61
4.3.2. Molecular Dynamic Simulations ................................................................... 65
4.4. CONCLUSIONS .......................................................................................................... 73
4.5. REFERENCES ...................................................................... ~ ...................................... 77
CHAPTER 5
SYNTHESIS AND ION-BINDING PROPERTIES OF POLYMERIC
PSEUDOCROWN ETHERS 79
5.1. INTRODUCTION ........................................................................................................ 79
5.2. EXPERIMENTAL METHODS AND TECHNIQUES .......................................................... 83
5.2.1. Synthesis of Polymeric Pseudocrown Ethers ................................................. 83
5.2.2. Removal of Template Cation And Conditioning Of Resin ............................ 85
5.2.3. Ion-Binding Studies ....................................................................................... 85
5.2.4. Column Regeneration and Determination of Concentration of Eluted
Solution ......................................................................................................... 87
5.3. RESULTS AND DISCUSSION ...................................................................................... 88
5.3.1. Polymeric Pseudocrown Ether Synthesis ..................................................... 88
5.3.2. Ion-binding Studies ...................................................................................... 89
5.3.3. Polymer Regeneration .................................................................................. 93
5.4. CONCLUSIONS .......................................................................................................... 94
5.5. REFERENCES ............................................................................................................ 96
CHAPTER 6
SWELLING PROPERTIES OF POLY(METHACRYLIC ACID-G-ETHYLENE
GLYCOL) HYDROGELS - EFFECT OF SOLVENT 97

 

ix

6.1. INTRODUCTION ........................................................................................................ 97

6.2. EXPERIMENTAL METHODS AND TECHNIQUES ........................................................ 101
6.2.1. Hydrogel Synthesis ...................................................................................... 101
6.2.2. Equilibrium Swelling Studies ...................................................................... 103
6.2.3. Molecular Mechanics Studies ...................................................................... 104

6.3. RESULTS AND DISCUSSION .................................................................................... 105
6.3.1. Hydrogel Swelling Studies .......................................................................... 105
6.3.2. Molecular Modeling of PMAA-PEG Complexation ................................... 113

6.4. CONCLUSIONS ........................................................................................................ 116

6.5. REFERENCES .......................................................................................................... 119

CHAPTER 7

SYNTHESIS AND EMULSIFICATION PROPERTIES OF REVERSIBLE
BLOCK/CRAFT POLYMERIC EMULSIFIERS BASED UPON POLYMER

 

COMPLEXATION 120
7.1. INTRODUCTION ...................................................................................................... 120
7.2. EXPERIMENTAL METHODS AND TECHNIQUES ........................................................ 126

7.2.1. Copolymer Synthesis ................................................................................... 126
7.2.2. Emulsification Studies ................................................................................. 128
7.2.3. Interfacial Tension Measurements ............................................................... 128
7.3. RESULTS AND DISCUSSION .................................................................................... 130
7.3.1. Emulsification Studies ................................................................................. 130
7.3.2. Interfacial Tension Studies .......................................................................... 130
7.4. CONCLUSIONS ........................................................................................................ 137

7.5. REFERENCES .......................................................................................................... 139

CHAPTER 8

A FLUORESCENCE SPECTROSCOPY STUDY OF THE AGGREGATE
FORMATION BEHAVIOR OF REVERSIBLE COPOLYMERIC BLOCK/GRAF T

 

EMULSIFIERS 140
8.1. INTRODUCTION ...................................................................................................... 140
8.2. EXPERIMENTAL METHODS AND TECHNIQUES ........................................................ 144
8.3. RESULTS AND DISCUSSION .................................................................................... 146

8.3.1. Excitation Spectra ........................................................................................ 146

8.3.2. Estimation of the Critical Aggregate Concentration from the Excitation

 

Spectra ......................................................................................................... 150

8.3.3. Emission Spectra .......................................................................................... 154

8.3.4. Estimation of the CAC from Emission Spectra ........................................... 157

8.4. CONCLUSIONS ........................................................................................................ 160

8.5. REFERENCES .......................................................................................................... 161
CHAPTER 9

CONCLUSIONS - 163

9.1. SUMMARY OF RESULTS .......................................................................................... 163

9.1.1. Synthesis and Characterization of Polymeric Pseudocrown Ethers ............. 163

9.1.2. Synthesis and Characterization of Reversible Block/Graft Copolymeric

Emulsifiers Based Upon Polymer Complexation ....................................... 167
CHAPTER 10
RECOMMENDATIONS FOR FUTURE WORK - -170

 

10.1. SYNTHESIS AND CHARACTERIZATION OF POLYMERIC PSEUDOCROWN ETHERS 1 70

10.1.1. Ion-induced Templatization Studies ......................................................... 171

xi

10.1.2. Molecular Modeling ................................................................................. 171

10.1.3. Synthesis and Ion-Binding Studies ........................................................... 172
10.2. SYNTHESIS AND EMULSIFICATION PROPERTIES OF REVERSIBLE BLOCK/GRAFT
COPOLYMERIC EMULSIFIERS BASED UPON POLYMER COMPLEXATION ............... 174
10.2.1. Study of Reversible Complexation ........................................................... 174
10.2.2. Copolymeric Emulsifier Synthesis ........................................................... 175
10.2.3. Inclusion of Comonomers ........................................................................ 176
10.2.4. Copolymer Characterization and Emulsification Studies ......................... 176

10.3. ALKALl-SWELLABLE THICKENERS COMPRISING POLY(METHACRYLIC ACID-G-
ETHYLENE GLYCOL) COPOLYMERS BASED UPON POLYMER COMPLEXATION ...... 178

10.4. REFERENCES ........................................................................................................ 183

xii

LIST OF TABLES

Chapter 1

Table 1.1. Classification of surfactants based upon chemical structure. .......................... 17
Chapter 3

Table 3.1. Solubilities of salts upon the addition of PEG 300 .......................................... 41

xiii

Figure 1.1.

Figure 1.2.

Figure 1.3.

Figure 3.1.

Figure 3.2.

Figure 3.3.

Figure 3.4.

Figure 3.5.

Figure 3.6.

Figure 3.7.

Figure 3.8.

LIST OF FIGURES

Chapter 1
a) Structure of 18-crown-6 and b) Structure of cryptand-222 .......................... 9
Crown ethers attached as pendent groups and as part of a polymer
backbone. .......................................................................................................... 9
Potential application for reversible emulsifiers as an aqueous cleaner useful
in the clean up of greased components in service stations. ............................ 22
Chapter 3

Scheme for the synthesis of polymeric pseudocrown ethers based upon a one-
step free-radical polymerization of templated polyethylene glycol diacrylate
with a comonomer such as hydroxy ethyl methacrylate ................................. 35

Schematic showing the cation complexation ability of oligomeric
polyethers. ...................................................................................................... 35

Synthesis scheme of Warshawsky et al. for the synthesis of polymeric
pseudocrown ethers. ....................................................................................... 36

‘H NMR spectra of a) PEG 300 and b) PEG 300 in the presence of chromium

chloride.(hydroxyl) protons. ........................................................................... 42
MD 50 ps simulation results for the mean end-to-end distance of PEGDA
with different number of EG repeating units. ................................................. 44

Molecular representation of a PEGDA (n = 4) conformation templated
around a Na“ ion showing charged surface of the molecule. Shown here is

one conformation extracted from a 50 ps run. ................................................ 46
Effect of different simulation times on the mean end-to-end distance for
PEGDA with Na+ with various numbers of EG repeating units. ................... 48

Normalized frequency distribution of end-to-end distance for PEGDA (n = 4)
with and without Na+. .................................................................................... 49

xiv

Figure 3.9.

Figure 4.1.

Figure 4.2.

Figure 4.3.

Figure 4.4.

Figure 4.5.

Figure 4.6.

Figure 4.7.

Figure 4.8.

Figure 4.9.

Cumulative frequency distribution of end-to-end distance for PEGDA (n = 4)
with and without a Na+ ion. ........................................................................... 50

Chapter 4

Scheme for synthesis of pyrene end labeled oligoethylene glycols. .............. 58

Fluorescence emission spectra of (A) pyrene end-labeled pentaethylene
glycol and (B) l-pyrenebutyric acid. The excitation was at 322 nm and both
samples were at a concentration of 10‘5 M. .................................................... 62

Fluorescence profiles of pyrene end-labeled tetraethylene glycol in THF
solvent (10'5 M) in the (A) absence and (B) the presence of nickel chloride.
The excitation frequency is 322 nm. .............................................................. 64

Fluorescence emission spectra of pyrene end-labeled pentaethylene glycol
(10'5 M) in chloroform in the (A) absence and (B) presence of tin chloride.
Excitation is at 322 nm. .................................................................................. 66

Comparison of excimer-to-monomer ratio for pyrene end-labeled
pentaethylene glycol (10'5 M) in chloroform and THF in the presence and

absence of nickel chloride. ............................................................................. 67
Starting conformation of end-labeled pentaethylene glycol in the presence of
a sodium cation. .............................................................................................. 69

Pyrene end-labeled pentaethylene glycol conformation with templating cation
showing pyrene excimer and the end-to-end distance (ETED). ..................... 70

Comparison of the normalized frequency distribution of end-tO-end distance
for TEGDA and Py-TEG-Py demonstrating the effect of pyrene end-
labeling. .......................................................................................................... 71

Normalized frequency distribution of end-to-end distance for pyrene end-
labeled pentaethylene glycol in the presence and absence of a templating
sodium cation .................................................................................................. 72

Figure 4.10. Comparison of normalized frequency distribution of end-to-end distance

Figure 5.1.

obtained from molecular dynamics simulations for Py-TEG-Py and

TEGDA, both with a templating Na+ cation ................................................ 74
Chapter 5
Schematic of free-radical polymerization scheme for the synthesis of
polymeric pseudocrown ethers. ...................................................................... 81

XV

Figure 5.2.

Figure 5.3.

Figure 6.1.

Figure 6.2.

Figure 6.3.

Figure 6.4.

Figure 6.5.

Figure 6.6.

Figure 6.7.

Figure 7.1.

Figure 7.2.

Figure 7.3.

Experimental setup for ion-binding and regeneration studies. ....................... 86

Experimental ion-binding capacities for polymeric pseudocrown ethers
(PCES) and “control” p(HEMA) hydrogels as a function of the monomer to
solvent ratio during synthesis. ........................................................................ 90

Chapter 6

Synthesis scheme for the preparation of poly(methacrylic acid-g-ethylene
glycol) self-associating hydrogels. ............................................................... 102

Effect of solvent on the equilibrium degree of swelling for a PMAA-g-PEG
gel synthesized in 50% solvent (50 % monomers). ...................................... 106

Equilibrium swelling profiles showing the effect of polymerization
conditions for swelling in ethanol. ............................................................... 108

Equilibrium swelling profiles of crosslinked PEG-g-PMAA (50/50) and
PMAA homopolymer (50/50) in methanol. ................................................. 1 10

Equilibrium Swelling profiles of crosslinked PMAA-g-PEG (50/50) and
PMAA (50/50) in acetone. ........................................................................... 1 12

A molecular representation of a PMAA segment with four repeating units
complexed with a PEG segment of the same length showing 1:1 hydrogen

bonding between the acid protons and the ether oxygens. ........................... 115

A POLYGRAF molecular simulation depicting a 1:1 hydrogen bonding

between a PMAA and PEG segment both with 14 repeating units. ............. 117
Chapter 7

(A) Reduction of the viscosities of heavy oils by the formation of oil-in-water
emulsions, and (B) A schematic showing a treatment process that may be
used for the pumping of heavy oils by forming oil-in-water emulsions. ..... 121

Schematic and mechanistic depiction of the block/graft copolymer, consisting
of a poly(methacrylic acid) backbone and poly(ethylene glycol) grafts, under

acidic or complex promoting conditions and under basic or complex breaking
conditions. .................................................................................................... 123

Hydrogen bonded complexes formed between poly(methacrylic acid) and
poly(ethylene glycol) units in (A) non-grafted and, (B) grafted
configurations. .............................................................................................. l 25

xvi

Figure 7.4.

Figure 7.5.

Figure 7.6.

Figure 7.7.

Figure 7.8.

Figure 8.1.

Figure 8.2.

Figure 8.3.

Figure 8.4.

Figure 8.5.

Figure 8.6.

Figure 8.7.

Schematic showing the free-radical polymerization scheme for the synthesis
of reversible block/graft emulsifiers. ............................................................ 127

Schematic of the methodology adopted for the oil emulsification studies... 129

Emulsification behavior of two different block/ graft copolymers (10:1 and
20: 1) as a function of pH showing the percent oil emulsified (V/v) as a
function of pH ............................................................................................... 131

Interfacial tension between the oil (methyl laurate) phase and the aqueous
phase containing the emulsifiers, as a function of pH. Data is presented for
the 10:1 and the 20:1 copolymeric emulsifiers ............................................. 133

Schematic representation of the possible emulsification mechanism adopted
by the reversible emulsifiers (A) under acidic and basic conditions, and (B)
the steric stabilization that may result. ......................................................... 135

Chapter 8

Pyrene excitation spectra for 20:] PMAA-g-PEGI 000 copolymer solutions of
various concentrations with constant pyrene concentration (6 x 10'7 M).
Emission was monitored at 397 nm .............................................................. 147

Plot of the pyrene excitation ratio (1342/1337) as a function of pH for 20:1
PMAA-g-PEGIOOO copolymer solutions with 6 x 10'7 M pyrene. The
emission was monitored at 397 nm. ............................................................. 149

Plot of percent pyrene emulsified versus pH for a 20:1 PMAA-g-PEGIOOO
copolymer solutions calculated from excitation ratio data. .......................... 151

A plot of the pyrene excitation ratio as a function of copolymer concentration
for 10:1 and 20:1 PMAA-g-PEGlOOO copolymers at pH 2. ........................ 153

Pyrene emission spectra for 20:1 PMAA-g-PEG1000 copolymer solutions at
pH 2. Excitation was at 322 nm and pyrene concentration in each sample
was6x 10'7 M. ............................................................................................. 155

A plot of the pyrene emission ratio 1373/1339 (II/I3) as a function of pH for 20:1
copolymer solutions with a pyrene concentration of 6 x 10'7 M. Excitation
was at 322 nm. .............................................................................................. 156

A plot of the pyrene emission ratio (excitation at 322 nm) versus
concentration for 10:1 and 20:1 PMAA-g-PEGIOOO copolymers. .............. 158

xvii

Chapter 10

Figure 10.1. Conformational changes resulting from going from an acidic (A) to a basic
(B) environment causing the polymeric chain to expand. .......................... 180

Figure 10.2. A plot of Brookefield viscosity as a function of pH for 1:1 PMAA-g-
PEG1000 (with a MAA to EG ratio of 1 :1) copolymer solutions at two

concentrations. ............................................................................................ 1 82

xviii

CHAPTER 1

BACKGROUND AND MOTIVATION

1.1. Introduction

This chapter presents the background and motivation behind this research. The
motivation behind development of novel polymeric pseudocrown ethers, which have
potential applications in the separations industry, will be presented first. Since molecular
simulations have been utilized to elucidate phenomenon at the molecular level, an
overview of molecular modeling techniques and a brief review of their applications in
macrocyclic ligand simulation will be discussed next. Following this a brief overview of
potential applications which provided the motivation for the development of novel
reversible block/graft emulsifiers will be presented. This will be followed by a short
introduction into the phenomenon of hydrogen-bonded complexation between polymeric
acids and bases.

The unifying thread in all this work is the utilization of the complexation
properties of poly(ethylene glycol). Other than its commonly knownl uses (such as, in
cosmetics and toiletries, detergents and cleaners, inks, paints and coatings, etc.)
poly(ethylene glycol) has the ability to form complexes with electron accepting species
because of its Lewis base character. The cation complexation ability of poly(ethylene

glycol) (PEG) has been used to design a novel technique for the synthesis of polymeric

pseudocrown ethers. The ability of PEG to form reversible hydrogen bonded complexes
with species possessing a Lewis acid character (such as poly(methacrylic acid)) has been
utilized to design reversible block/graft emulsifiers. This chapter will lead into the

specific objectives of this research which are addressed in the following chapter.

1.2. Polymeric Pseudocrown Ethers - Motivation

1.2.1. Heavy Metal Ions in Aqueous Industrial Wastes

The control of heavy metal discharges to the environment promises to be a
significant industrial challenge in the coming years. Although heavy metals such as lead,
chromium, mercury, nickel, tin, copper, zinc, and cadmium are commonly present in
ionic form in aqueous waste streams, their toxicity to plants, animals, and humansz'5 has
led to increasingly stringent industrial emission standards.5 Heavy metals accumulate in
living organisms and can cause chronic diseases and even death. Awareness of the
potential dangers of heavy metals in industrial wastes was aroused over twenty-five years
ago when aqueous industrial eflluents containing mercury and cadmium ions poisoned
the food chain in Japan, causing the deaths of thousands of animals and over one hundred
people.2 More recently, there has been mounting concern over the possibility of adverse
health effects from exposure to heavy metal ions at relatively low concentrations
previously thought safe. For example, lead is suspected of causing subtle, irreversible
brain damage in growing children,5 while heavy metal ions in general are under

investigation for their possible gene-mutation effect in living organisms.6

Heavy metal ions appear in industrial waste streams from a variety of sources.
Relatively large concentrations of heavy metal ions occur in the waste streams from
plants in which a heavy metal is an integral part of the manufacturing process. For
example, waste streams from battery manufacturers and electroplating facilities typically
contain heavy metals in concentrations of over one hundred parts per million.5 Other
industries which produce waste water with relatively large concentrations of heavy metal
ions include mining,4 steelworks,s and a variety of specialty chemical plants.6 Many
processes which do not directly involve heavy metals may produce waste streams which
contain significant concentrations of heavy metal ions. For example, zinc and chromium
are often used as corrosion inhibitors in cooling towers, and are present in concentrations
approaching ten parts per million in the blowdown streams? Furthermore, corrosion of
catalysts or other metals may lead to Significant amounts of heavy metal ions in the
aqueous waste streams from any chemical plant.

The United States Environmental Protection Agency has established standards for
industrial discharges to inland waters. For heavy metals, the discharge limits are
typically in the range of fifty parts per billion.2 The heavy metal ions which enter natural
bodies of water often concentrate in the sediments of lakes and oceans.5 Ground water
often exhibits a very low concentration of heavy metal ions because they are immobilized
in the top layers of the soil. This fact has led to increasing concern over the accumulation
of heavy metals in the soils.7 In fact, at the Sapp Battery Superfund site in northern
Florida, the upper soil contains lead in concentrations approaching 135,000 parts per
million.8 Extensive ground water contamination has been observed in areas in which the

soil is very sandy9 and unable to immobilize the metal ions. Finally, if heavy metal laden

water is routed to a municipal water treatment plant, the metal ions concentrate in the
sewage Sludge. The subsequent use of the sludge for agricultural fertilizer further
contributes to the accumulation of heavy metals in the soil.7 Furthermore, heavy metals
have been observed to inhibit biological sewage treatment processes if present in

concentrations above a few parts per million.5

1.2.2. Current Methods for Removing Heavy Metals fiom Water
The most common technique for removing heavy metal ions from aqueous

streams is precipitation.'°"'

This method is based upon the fact that the aqueous
solubility of the heavy metal ions depends strongly upon pH. Under basic conditions, the
metals typically precipitate as the corresponding hydroxide salts. The minimum
attainable concentration of a heavy metal ion is determined by the magnitude of the
solubility minimum, typically about 500 parts per billion.” If the waste stream contains
two or more ions which exhibit solubility minima at different values of pH, the process
must be run at a compromise pH value, thereby undermining the efficiency of the
purification.'0 The sedimentation of the heavy metal precipitates usually takes several
hours to complete, even with the addition of flocculating agents.5 Often precipitation is
used only for an initial treatment of a waste stream, with fiIrther treatment by other
methods being required to decrease the heavy metal ion concentration to desirable
levels.5'IO

A variety of other methods have been suggested for removing heavy metal ions

from aqueous wastes, including activated carbon adsorption}10 ion exchange,”"'2 reverse

osmosis?“ liquid-liquid extraction}l5 evaporation,5"6 freezing,” foam fractionation or

18.19 I

flotation, and micellar enhanced ultrafiltration?“2 The latter two methods use
surfactants in the separation process. In foam or bubble fractionation, positive heavy
metal ions are attracted to anionic surfactant molecules at an air-liquid interface.
Typically air is bubbled through the solution, allowing the ion laden surfactants to adsorb
on the interface to be carried to the surface foam. In micellar enhanced ultrafiltration the
surfactant forms micelles around the contaminant, rendering it too large to pass through
an ultrafiltration membrane. Although this technique has been used primarily for

removal of organic contaminants,”2|

with an appropriate surfactant it could possibly be
used for removing ions as well.

All of the purification techniques developed to date have significant limitations
when considering the removal of heavy metal ions from aqueous industrial wastes.
Because heavy metal ions are typically present in relatively small concentrations,
methods such as evaporation and freezing which inherently remove the water rather than
the ions from the water, are relatively unattractive. Similarly, because the distribution
coefficient of the metal ion is essentially insensitive to its concentration, liquid-liquid
extraction is not well suited for removing heavy metals from waste streams.22 A major
shortcoming of all of the techniques developed to date is a lack of selectivity, which is
probably most pronounced for activated carbon. Although activated carbon will absorb
most metals which are organically complexed,5 it will also absorb any other organic
molecules present in the water. This necessitates frequent regeneration, and undermines
the cost effectiveness of the method. For the methods which involve surfactants, the lack
of selectivity results in a Significant decrease in the separation efficiency with increasing

inert salt concentration.'8"9

Another shortcoming of the present purification techniques is the fact that they do
not decrease the total quantity of heavy metal wastes, but merely concentrate the ions into
a smaller volume of liquid or solid.2'5 Therefore, under the current technology, the
problem of disposal of the concentrated wastes still remains. In many cases, the heavy
metal ions themselves have some value, and ideally they would be recycled rather than
discarded. However, since most industrial wastes contain several different types of ions,
the concentrated wastes will contain a broth of many types of ions, and a separations
problem still exists. For this reason, recycling of heavy metal ions has been employed
only in a few special cases, such as the recycling of chromium ions in cooling tower
blowdown.6

Ion exchange is the technique which is perhaps best suited to the removal of small

quantities of an ionic component from water,”23

and has been used extensively for the
removal of heavy metals from waste streams.5’lo A particularly attractive feature of ion
exchange is that the cost of the separation decreases with decreasing concentration of the
ion to be removed. This is important because the concentrations of heavy metals in waste
streams are typically relatively small; on the order of hundreds of parts per million.
Unfortunately, innocuous ions present in the water tend to reduce the capacity of the

resins as well as the efficiency of the separation.24

This fact has led to a great deal of
work in the area of selective chelating resins.”26 However, current chelating resins have
not met the selectivity and performance requirements”26 for aqueous waste treatment. A

second drawback of ion exchange is the problem of disposal of the regeneration wastes.

In the ideal case, regeneration results in a doubling of the number of ions which must be

discarded. However, in most cases an excess of regenerant is required,23 and the total

quantity of waste ions is much higher than the ideal value.

1.2. 3. Need for New Separation Methods for Heavy Metal Ions

The future promises to bring increasing concern over the accumulation of heavy
metals in the environment due to industrial wastes. Undoubtedly, the threat of heavy
metal pollution will lead to the adoption of more stringent effluent standards for industrial

processes. This realization has caused many authorsz‘3'5"°

to discuss the "zero discharge"
goal for heavy metal ions. While absolutely zero discharge is intrinsically impossible, the
separation methods in use today are not well suited to this goal. The current methods do
not decrease the total quantity of heavy metal wastes, but merely concentrate the wastes.
Although recovering and recycling of heavy metal ions has been practiced in a few
specialized cases, the current technology is not amenable to this approach because the
concentrated wastes that are generated typically contain a mixture of many ions. Further
separations would be required for recycling to be feasible.

There is a need for a heavy metal ion separation technique which would allow the
ions to be removed from the waste stream and recycled without an extra separation step.
Such a technique could enhance the economic attractiveness of recycling heavy metal
wastes rather than Simply disposing of them, especially in the face of increasing
regulatory pressures. A desirable technique should have the following properties: i) it
should be capable of selectively removing one distinct type of metal ion from an aqueous

mixture containing many ions, and ii) it Should allow the heavy metal to be reclaimed

without creating additional ionic wastes. Furthermore, the method Should be cost

effective and, ideally, not require existing waste treatment equipment and technology to

be completely abandoned.

1.3. Polymeric Pseudocrown Ethers - Background

Since the pioneering work of Pedersen”28 cyclic oligomers of ethylene glycol
known as crown ethers have been extensively studied for their unique ion-binding
properties. Pedersen demonstrated the ability of crown ethers to form stable complexes
with metal cations and to solubilize inorganic salts in nonpolar solvents. This ion-

binding ability arises from the electron lone pairs on the oxygen atoms of the cyclic

29.30

ether, with the most stable complexes formed with ions that fit snugly in the

macrocyclic cavity thereby interacting optimally with all donor Oxygens (see Figure 1.1).

Since Pedersen's work on crown ethers several related compounds have been

31-33

synthesized and studied, including cryptands, which are tricyclic ligands;

34.35

coronands, in which nitrogen, sulfur or phosphorous replace oxygen as the electron

donor; podands,”36

which are open-chain analogs of the cyclic ligands; and
podandocoronands34 which combine the structural properties of podands and coronands.
Several reviews on the synthesis and ion-binding of multidentate macrocyclic

compounds”39 and host-guest chemistry‘o‘“

provide an excellent overview of the area of
supramolecular chemistry that has burgeoned since Pedersen's efforts in 1967.
The unique ion-binding properties of macrocyclic ligands make them attractive

for a variety of applications, most notably phase transfer catalysis. However these

compounds are generally synthesized in low yields and through relatively long reaction

(“Do m
[0 O] @3637
ROJ k‘L/OJ

Figure 1.1. a) Structure of 18-crown-6 b) Structure of cryptand-222

o o _ (OI—tr? 1
[O O] {MILD/CHOW—

OH OH

 

 

Figure 1.2. Crown ethers attached as pendent groups and as part of a polymer backbone.

10

sequences,”36

making them rather expensive. For this reason, investigators developed
schemes for attaching these ligands to polymeric supports thereby allowing the ligands to
be efficiently retrieved and recycled (see Figure 1.2). Methods for incorporating
macrocyclic ligands either into the polymeric backbone or as pendant side groups36‘42’“
have been reported.

Synthetic methods which have been used to produce macrocycle-containing
polymers include condensation of di-functional macrocycles with appropriate
comonomers, free-radical polymerizations of Vinyl crown ethers, and substitutions of
macrocycles for leaving groups on existing polymer chains. Macrocycle-containing
polymers have been used in a variety of applications in analytical and preparative
chemistry, including anion and cation separations, column and thin layer
chromatography, salt conversions, and phase transfer catalysis.3"3("“2“““"5“47 Macrocyclic
ligands attached to polymeric chains may exhibit different ion-binding characteristics
than the corresponding monomeric species. Interactions which affect the affinity and
selectivity of polymer-bound macrocycles can arise from the polymer backbone, the point
of attachment, neighboring ligands, or other pendant groups.

Some of the limitations of crown ethers, such as toxicity (since they readily
absorb through the skin) and ease of recovery, could be overcome by immobilization on
polymeric chains. However, these materials were also synthesized in low yields with
long reaction sequences resulting in a potentially expensive resin. One alternative to
immobilizing crown ethers onto polymeric supports is to form crown ethers in-situ during
a polymerization reaction. In such a scheme the crown ethers would be produced only

during the polymerization process and would thus overcome limitations associated with

ll

toxicity and recovery. In addition, this scheme would be more attractive if it would
involve a one-step and easy to perform reaction which would result in high yield and low
cost. Materials, called pseudocrown ethers, offer all of the above advantages. A more
detailed description of work related to the development of polymeric pseudocrown ethers

and the scheme adopted in this research will be discussed in Chapter 3.

1.4. Molecular Modeling of Macromolecules and Ligands

Molecular modeling has emerged as an invaluable tool in understanding the
micro-environment of various systems. It has been extensively used in conformational
analysis of proteins and for other bio-molecules. For example, conformational analysis of
macromolecules have been performed and energies of various conformations calculated

using semi-empirical potential energy functions.48

Earlier, involved computer programs
were written in order to calculate and minimize the energy of a certain molecule and the
visualization of its conformation was done by interpreting radial distribution fimctions in
the case of molecular dynamics calculations, or based upon bond lengths and distances in
the case of molecular mechanics calculations. With the advent of excellent computing
techniques it is not only possible to decrease computation time and thus perform
simulations previously considered tedious, but also to visualize the resulting
conformations graphically on a computer terminal. Molecular modeling, however, can
also have its own limitations49 and is limited by the size of the molecule under study.

Molecular modeling may be useful as a tool for the scientific visualization of

molecules under a certain set of conditions. It has been broadly divided into the

12

following categories: i) ab initio calculations on atoms and molecules using quantum
mechanical fundamentals, ii) semi-empirical and empirical calculations which are
extensions of ab initio calculations performed on larger molecules after making certain
assumptions while performing the quantum mechanical calculations, iii) molecular
mechanics calculations which are designed for larger molecules and are based on force
fields, and, iv) molecular dynamics calculations, which are concerned with determining
the trajectories of atoms within a molecule over a period of time by solving Newton's
equations of motion subject to certain potential energy functions and constraints. While
the theoretical chemist uses the first two methods of molecular modeling quite
extensively to predict charge densities, extract molecular orbital information, and
determine energies of various molecules, the latter two methods are most widely used for
macromolecules like polymers and proteins for conformation analysis and energy

calculations.

1 . 4. I . Molecular Mechanics

Molecular mechanics (MM), which is often referred to as the force field method,
is one of the most commonly used methods available to predict geometry. It requires
limited computer resources and its results have the same quality with regard to geometry
as those obtained by sophisticated quantum methods. The molecule is viewed as a
collection of points (atoms) connected by springs (bonds) with different elasticities (force
constants). The force holding the atoms together can be described by potential energy
functions of structural features like bond lengths, bond angles, non bonded interactions

and so on. The combination of these potential energy functions is called a force field.

13

There exists "natural" lengths and angles for bonds and a molecule will assume a
geometry with those values in the simple cases. In general, steric, electrostatic and other
strain forces must be included too. Because of the intimate connection between structure
and energy, molecular mechanics involves both. To find the structure, one necessarily
has to examine the energy to find where the energy minima occur. The energy, E, of a
molecule in a force field arises from deviations from ideal structural features, and can be
approximated by a sum of energy contributions due to bond stretching, bond bending, out
of plane bending, torsional energy due to twisting about bonds, van der Waals non
bonded interactions, electrostatic interactions and other user specified constraints. The
value E is the difference in energy between the real molecule and a hypothetical molecule
where all the structural features such as, bond angles and bond lengths, are exactly at their
ideal or natural values. These equations are then solved using a minimization routine
such as, the steepest descent method, the conjugate gradient method or the Broydon
Fletcher steepest gradient method, in order to attain an energy minimum. Various force
fields are currently available to describe different sets of molecules. AMBER50 and
MM25| were two of the first models or "force fields" to be described and translated into
an efficient computer program. Since then various other force fields have been

developed, including CHARMM,52 DRIEDING,53 and MM3.54

I . 4. 2. Molecular Dynamics
As the name suggests, molecular dynamics simulations are used to perform
conformational analysis by solving Newton's equations of motion for the whole molecule.

The force between two atoms is the gradient of the potential between them and, knowing

14

various potential energy expressions, one can integrate Newton's equation twice to
calculate the new coordinates for an atom. The potential is assumed to be continuous and
no hard Sphere forces are considered. Several algorithms, like predictor corrector
methods, can be used to solve the resulting differential equations by finite difference
techniques. Typical simulations are performed on the order of pico (10“) seconds due to
the immense number of calculations involved. Periodic boundary conditions are typically
used in molecular modeling with solvents in order to simulate a system having a constant
volume and constant number of molecules. Periodic boundary conditions imply that if
any molecule moves out of a face of the cube containing the system of molecules, another
moves in from the opposite face to keep the number of molecules in the periodic
boundary constant. Thus, molecular dynamics Simulations may provide information
about molecular behavior and is widely used to estimate energies of molecules.

Computer software such as BIOGRAF and POLYGRAF55 (Molecular Simulations
Inc.; these programs now have been combined and the currently marketed software is
known as CERIUS2) have been developed thereby making it easy to perform MM and
MD calculations on a host of biological, organic and polymeric molecules. These
packages also possess the ability to graphically display, in real time, changes in
conformations of the molecules under study. This consolidation of a graphical interface
with algorithms that perform MM and MD calculations has redefined molecular
modeling, thereby making it a user-friendly tool to pursue investigations at the molecular
level. Various studies have utilized the ability of this software to perform molecular

investigations in an easy and visual manner. For example, the binding of phosphines to

15

Cr(CO)5 was studied using MM calculations performed56 on BIOGRAF. MD

calculations using SYBYL (Version 5.3, Tripos Associates) have been used to predict
equilibrium statistics and study the conformational dynamics of poly(dialkylsiloxanes).57
Molecular Simulation techniques are now being used to characterize the structure and

properties 0f POIYmeric systems,”59

thereby making this technique viable for predicting
mechanical and thermal properties which may lend a new dimension to the design and
synthesis of new materials. POLYGRAF has been used to derive the phase diagrams of
binary mixtures59 and to test the F lory-Huggins theory for a number of binary systems
and to characterize their miscibility behavior. Molecular dynamics Simulations have also
been used to obtain the diffusion coefficient of a penetrant gas in a polymeric material."0

Finally, molecular dynamics of aromatic polyesters performed using the AMBER code

have yielded experimentally accessible properties such as persistence length.6|

1.5. Reversible Copolymeric Emulsifiers - Background and Motivation

Surfactants, or surface active agents, are among the most versatile products of the
chemical industry. They are used as detergents and cleaners, and in pharmaceuticals,
drilling muds, motor oils, floatation agents, etc. The surface active nature of these
materials arises from their ability to lower the interfacial tension between two interfaces
thereby permitting miscibility by partitioning partially into both phases. The unique
chemical structure that permits such behavior is the presence of two distinct groups

within the same surfactant molecule, thereby imparting to them an amphipathic structure.

16

Usually, the two active segments in a surfactant which account for its surface activity are
a hydrophilic (or lyophobic) group and a hydrophobic (lyophilic) group. AS the terms
suggest, the hydrophilic nature is essential for allowing miscibility in the aqueous phase,
whereas the hydrophobic character is required for partitioning into the oil phase.
Typically above a certain concentration, known as the critical micelle concentration
(CMC), surfactants form aggregates of molecules due to the amphiphilic nature of each
molecule. These aggregates, or micelles as they are commonly referred to, may consist of
an inner hydrophobic core (in aqueous solutions) while the hydrophilic moieties are
extended into the continuous phase. Most surfactants can thus emulsify oil by entrapping
small oil droplets into the hydrophobic domains. The emulsions so formed are stabilized
by the presence of surfactant molecules on the interface which inhibit and often prevent
coalescence of two droplets. Numerous texts are available which describe various kinds
of surfactants commercially available. In general surfactants may be broadly classified as
follows: anionic, cationic, and non-ionic based upon the surface active portion of the
molecule (see Table 1.1). While most common types of surfactants fall into the first two
categories most research is currently focused on the development of novel non-ionic
surfactants. Non-ionic surfactants are usually either block or graft copolymers possessing
amphiphilic nature.

Block and graft copolymers are used as dispersing agents and stabilizers for
particles and emulsions in water. In these systems, one block of the dispersing agent is
designed to absorb onto the surface of the stabilized particle, and the other block is
designed to extend into the continuous phase. Therefore, each particle possesses a layer

of absorbed polymer chains which provides steric stabilization, ultimately arising from

17

Table 1.1. Classification of surfactants based upon chemical structure.

 

 

Type of Surfactant Structure/Example
a) Anionic O
//
R—C
Sulfates, sulfonates, carboxylates and \ - +
O ...Na
phosphates
Fatty Acid Soap
1i
R—O—Ir—O' ...Na
O....Na+

Alkyl Phosphate

O
R—O—isl—Q’...Na
(I;
Alkyl Sulfate

 

b) Cationic

Alkyl amines (Cg-Cm), Primary, secondary,
tertiary and quaternary amines

R
I +

R—N —R'...Cl
RI

Quaternary Amine

 

 

c) Non-ionic

Block(diblock, triblock), Graft
Hydrophilic block -oxyethylenes
Hydrophobic block - alkyl or aryl chains

 

Rfo —CH2-CHzi‘OH

n-Alkyl polyoxyethylene ether

 

 

18

the crowding of polymer chains which would occur if two particles approached one
another. Polymeric surfactants have gained considerable importance in recent years
resulting in the development of several di-and tri-block copolymers. Several polymeric
surfactants are now commercially available. Among these, di- and tri-block copolymeric
surfactants are the most widely used as emulsifiers and agrochemicals.”64 Diblock and
triblock copolymers of ethylene oxide (EO) and propylene oxide (PO) have received the
most attention due to their relatively simple architecture and the ability to “tailor-make”
the copolymers by controlling the lengths of the hydrophilic (EO) and hydrophobic (PO)
blocks.

Block copolymers are most commonly synthesized using complex anionic
polymerizations which are constrained by a variety of factors. For example, block
copolymers may be formed by sequential monomer addition to a living anionic
polymerization and while this technique is extensively utilized to produce a variety of di-
and triblock copolymers, it has several limitations. The anionic polymerizations are
sensitive to trace amounts of impurities such as oxygen, carbon dioxide, and water and
therefore must be carried out with rigorously clean and dry reagents and reactors. For
these reasons these reactions are typically carried out in small volumes and the resulting
polymers are expensive. Anionic polymerizations may also be applied to only a limited
group of monomers, and may not be used to polymerize vinyl monomers containing
electron donating substituents in the DI position or polar monomers containing

substituents that are reactive towards nucleophiles. Important classes of monomers such

19

as a-Olefins, vinyl esters, vinyl ethers, as well as some acrylates cannot be polymerized
anionicaly.

In contrast, free radical polymerizations are much more versatile and may be
applied to nearly any vinyl monomer. Graft or “comb” copolymer surfactants consisting
of a hydrophilic backbone and a hydrophobic graft or vice versa have also been
synthesized and evaluated for their surfactant properties. This area of non-ionic
surfactants consisting of block and comb or graft copolymers has been the focus of many

reviews.”67

1.5.]. Need for the Development of Multiblock Reversible Emulsifiers

Simple diblock copolymers have also been used as polymer compatibilizers and
dispersion stabilizers for many years. Theoretical work by Noolandi68 has suggested that
multi-block copolymers would be more efficient as compatibilizers than di- or triblock
copolymers. His results suggest that less of the multi-block copolymer would be lost into
the bulk phase as micelles or mesophases than the di- or triblock systems. Similarly,
multi-block dispersing agents and stabilizers should be more efficient than the common
di or triblock systems. This multiblock molecular design strategy is aimed at reducing
the ability of the surfactant to form a separate phase, thereby resulting in an increase in
the interfacial activity in a non-equilibrium system with a large interfacial area. Noolandi
also suggests that multiblock systems will have less tendency to form micelles

particularly if randomness is allowed in the structure of the block. Also, if well defined

20

blocks are not necessary then the costs involved in synthesizing these systems will also
be considerably lower.

The above discussion provides the motivation for an investigation into the
feasibility of multiblock copolymers as emulsifiers. As mentioned before, most di-block
and tri-block copolymers contain ethylene oxide (glycol) as the hydrophilic repeat unit
and in many cases propylene oxide (glycol) as the hydrophobic repeat unit. The synthesis
of a multiblock copolymer can be easily achieved since it does not require the tedious and
multi-step synthesis needed to produce most di- and tri-block copolymeric emulsifiers.

Most surfactants emulsify oil and exhibit consistent properties over a wide range
of pH, temperature and other micro-environmental variables. Ionic surfactants are
however susceptible to changes in pH and their efficiency is affected with changes in
ionic strength. However, in some applications it may be desirable to expect the same
surfactant molecule to behave as an emulsifier under a set of conditions and not behave as
one under a different set of conditions. Some common applications that may benefit from
this feature include most applications requiring the use of de-emulsifiers at some point in
the processing scheme. A good example of such an application is the use of drilling
“muds” in oil drilling operations and the concerns involved with forming and reforming
emulsions during the entire oil recovery process. A stable emulsion (usually oil dispersed
in water) is usually used to lubricate the cutting bit during drilling and to carry cuttings
up to the surface. An emulsion may be desirable in one part of the oil production process
and undesirable at the next stage. In oil fields, an in situ emulsion that is purposely
created in a reservoir as a part of an oil recovery process may change to a different,

undesirable type of emulsion (water-in-oil) when produced at the well-head. This

21

emulsion may have to be broken and reformulated as a new emulsion suitable for
transportation by pipeline to a refinery. At the refinery the new emulsion will have to be
broken and water from the emulsion removed, otherwise the water would cause problems
during the processing of the crude oil.

Another related application that requires emulsification followed by de-emulsi-
fication after a series of operations have been performed is the use of surfactant solutions
in the formation of an oil-in-water emulsion for the pumping of heavy crude oils and
bitumens. Most heavy crudes and bitumens have viscosities of the order of thousands of
centipoise rendering them impossible to pump without the use of heated pipelines. An
emerging technology involves the use of emulsifiers to form low Viscosity Oil-in-water
(comprising 70 wt.% oil-in-water) emulsions which facilitate easy pumping at mostly
ambient temperatures. Once the emulsion has reached its destination the emulsion needs
to be broken and the water removed for further processing of the crude.

Other possible applications include the need for an aqueous cleaner which may be
used in clean up operations in most service stations (Figure 1.3). Most routine clean up
of greasy parts or service station floors is accomplished by either cleaning the greasy part
in organic solvents, using soap solutions, or by impingement by high velocity water jets.
In all of these methods the waste liquid that is generated contains the oil phase either
dispersed in an aqueous phase or suspended in a two-phase mixture. Either way the
problem of separation of the emulsified oil phase by a de-emulsification process still
poses a challenge. An aqueous cleaner that would allow the greased part to be efficiently
cleaned followed by easy separation of the resulting oil-in-water emulsion would be of

commercial interest.

22

Aqueous Recycle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Oil Disposal
I —>
ApH
e
. be _. o _. . .
e e AT e
' C7 6 e
Emulsification Phase
Separation

Figure 1.3. Potential application for reversible emulsifiers as an aqueous cleaner useful
in the clean up of greased components in service stations

Hence, there clearly exists great potential for the development of reversible

emulsifiers. In the next section, background on polymer complexation, which forms the

basis for the design of novel reversible block/graft copolymeric emulsifiers, will be

presented.

1.5.2. Polymer Complexation Through Hydrogen Bonding

Polymer complexation occurs due to the association of two or more
complementary polymers, and may arise from electrostatic forces, hydrophobic
interactions, hydrogen bonding, van der Waals forces or various combinations of these

interactions.”72

Once one pair of complementary repeating units associate to form a
segmental complex, many other units may readily associate without the loss of
translational degrees of freedom owing to the long-chain structure of the polymers.

Complexation between polymeric acids and polymeric bases formed as a result of

electrostatic forces may occur by proton transfer resulting in poly(cation):poly(anion)

23

pairs. However, complexes formed by hydrogen bonding between two complementary
polymeric Lewis acids and bases typically dissociate under a wide variety of conditions.
Hydrogen bonded complexes are formed between a Lewis acid containing an electron
deficient proton and a Lewis base with lone pairs of electrons. Hydrogen bonds are
Specific and directional and are more localized than any other type of weak
intermolecular interaction. Hence, the equilibrium between complexed and uncomplexed
segments for two hydrogen bonded species in solution may be highly sensitive to
conditions such as solution pH, temperature, solvent composition and polymer
composition. Such complexes may be easily broken and reformed by controlling the
surrounding conditions.

Hydrogen bonded complexes between poly(methacrylic acid) (PMAA) and
poly(ethylene glycol) (PEG) have been studied by several investigators (see Klier et al.,73
Scranton et al.,74 and references therein). Poly(ethylene glycol) is a Lewis base by virtue
of the lone pairs of electrons on the ether oxygen, whereas poly(methacrylic acid) is a
Lewis acid. While both PEG and PMAA are themselves soluble in water and are highly
hydrophilic in nature, the hydrogen-bonded complex is hydrophobic. If a solution of
PMAA is added to an aqueous solution of PEG an insoluble phase begins to form
instantly. However, the complex is formed when the acid is protonated and can be
disrupted by changes in surrounding conditions such as pH.”74 This interesting feature of
these reversible complexes between PMAA and PEG may be used to design novel
materials and polymeric systems.

In this chapter the motivation behind the development of two novel materials:

polymeric pseudocrown ethers and reversible block/ graft emulsifiers, has been presented.

24

Though the two proposed materials may have potential applications in two distinct areas,
their development is an outcome of the unique complexation properties exhibited by
poly(ethylene glycol). The next chapter will address the specific objectives of this
research and will provide a brief outline of the layout of this thesis with an overview of

each chapter.

10.

11.

12.

13.

14.

25

1.6. References

Powell G.M. in, Handbook of Water-Soluble Gums and Resins, R. L. Davidson
Ed., McGraw Hill, New York, Chapter 18, (1980).

Mance G., Pollution Threats of Heavy Metals in Aquatic Environments, Elsevier
Applied Science, London, (1987).

Fergusson J.E., The Heavy Elements: Chemistry, Environmental Impact and
Health Eflects, Pergamon Press, New York, (1990).

Kelly M., Mining and the Freshwater Environment, Elsevier Applied Science,
London, (1988).

Harrison RM. and Laxen D.P.H., Lead Pollution: Causes and Control, Chapman
and Hall, London (1981).

Anderson R.E., in Traces of Heavy Metals in Water Removal Processes and
Monitoring, J. E. Sabadell Ed., United States Environmental Protection Agency,
Washington DC, pp 43 (1973).

Alloway B.J., Heavy Metals in Soils, John Wiley & Sons Inc., New York (1990).

Tmovsky M., Oxter J.P., Rudy R.J., Hanchak M.J., and Hartsfield B., in
Hazardous Waste: Detection, Control, Treatment, R. Abbou Ed., Elsevier
Applied Science, New York, pp 1581 (1988).

Czupyma G., Levy R.D., MacLean AL, and Gold H., In-Situ Immobilization of
Heavy Metal Contaminated Soils, Noyes Data Corp, Park Ridge, NJ (1989).

Lanouette K.H., Chem. Eng, 73 (1977).

Holl W. and Horst J., in Recent Developments in Ion Exchange, P.A. Williams
and M.J. Hudson Eds., Elsevier Applied Science, London, pp 165 (1987).

Lee LA. and Davis H.J., in Natural Waters, P.C. Singer Ed., Ann Arbor Science,
pp 363 (1974).

Houle P.C., in Traces of Heavy Metals in Water Removal Processes and
Monitoring, J. E. Sabadell Ed., United States Environmental Protection Agency,
Washington DC, pp 159 (1973).

Furukawa D.H., in Traces of Heavy Metals in Water Removal Processes and
Monitoring, J. E. Sabadell Ed., United States Environmental Protection Agency,
Washington DC, pp 179 (1973).

15.

l6.

17.

18.
19.
20.
21.

22.

23.

24.

25.

26.

27.
28.

29.

30.

31.

26

Konda K., Tsuneyuki T., Hiramatsu K., and Nakashio F., Sep. Sci. Technol, 1213
(1990)

Probstein R.F., in Traces of Heavy Metals in Water Removal Processes and
Monitoring, J. E. Sabadell Ed., United States Environmental Protection Agency,
Washington DC, pp 71 (1973).

Campbell R.J., in Traces of Heavy Metals in Water Removal Processes and
Monitoring, J. E. Sabadell Ed., United States Environmental Protection Agency,
Washington DC, pp 97 (1973).

Lin J. and Huang S.D., Sep. Sci. Tech, 24, 1377 (1990).

Huang S.D., Fann C.F., and HSieh H.S., J. Colloid Interface Sci, 89, 504(1982).
Dunn RD. and Scamehom J .F., Sep. Sci. and Tech, 22, 763 (1987).

Kandori K. and Schechter R.S., Sep. Sci. and Tech, 25, 83 (1990).

Valdes-Krieg E., King C.J., and Sephton H.H., in Traces of Heavy Metals in
Water Removal Processes and Monitoring, J. E. Sabadell Ed., United States
Environmental Protection Agency, Washington DC, pp 189 (1973).

Millar J.R., in Recent Developments in Ion Exchange, P.A. Williams and M.J.
Hudson Eds., Elsevier Applied Science, London, pp 129 (1987).

Calmon C., in Traces of Heavy Metals in Water Removal Processes and
Monitoring, J. E. Sabadell Ed., United States Environmental Protection Agency,
Washington DC, pp 7 (1973).

Sahni SK. and Reedijk J., Coord. Chem. Revs., 59, l (1984).

Akelah A. and Moet A., Functionalized Polymers and Their Applications,
Chapman and Hall, London (1990).

Pedersen C. J., J. Am. Chem. Soc., 89, 2495 (1967).
Pedersen C. J ., J. Am. Chem. Soc. 89, 7017 (1967).

Liotta C. L., in Synthetic Multidentate Macrocyclic Compounds, R. M. Izatt and J.
J. Christensen Eds., Academic Press, New York, pp 111 (1978).

Weber W. P. and Gokel G. W., Phase Transfer Catalysis in Organic Synthesis,
Springer-Verlag, Berlin (1977).

Dietrich B., Lehn J. M., and Savage J. P., Tetrahedron Lett,. 2885 (1969).

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

27

Lehn J. M., Pure Appl. Chem, 49, 857 (1977).
Lehn J. M., Pure Appl. Chem, 50, 871 (1978).

Weber E. and Vogtle F., in Host-Guest Complex Chemistry: Macrocycles, F.
Vogtle and E. Weber, Eds., Springer-Verlag, New York, pp 1 (1985).

Hancock R. D. and Martell A. E., Comments Inorg. Chem, 6, 237 (1988).
Smid J., Ind. Eng. Chem. Prod. Res. Dev., 19, 364 (1980).

Christensen J. J., Hill J. O., and Izatt R. M., Science, 174(4008), 459 (1971).
Christensen J. J., Eatough D. J., and Izatt R. M., Chem. Rev., 74, 351 (1974).
Bradshaw J. S. and Stott P. E., Tetrahedron, 36, 461 (1980).

Cram D. J. and Cram J. M., Science, 183(4127), 803 (1974).

Cram D. J., Angew. Chem. Int. Ed. Engl., 27, 1009 (1988).

Blasius E. et al., J. Chromatogr., 147 (1980).

Blasius E. and Janzen K. P., Pure Appl. Chem, 54, 2115 (1982).

Gramain P., in Recent Developments in Ion-Exchange, P. A. Williams and M. J.
Hudson Eds., Elsevier Applied Science, London, pp 300 (1987).

Cinouuini M., Colonna S., Molinari H., and Montanari F., J. C. S. Chem. Comm,
394 (1976).

Smid J., Varma A. J., and Shah S. C., J. Am. Chem. Soc., 191, 5764 (1979)
Gramain P. and Frere Y., Ind. Eng. Chem. Prod. Res. Dev., 20, 524 (1981).
Scott RA. and Schearga H.A., J. Chem. Phy., 45(6), 2091 (1966).

Pensak D.A., Pure & Appl. Chem, 61(3), 601 (1989).

Weiner P.K. and Kollman P.A., J. Comput. Chem, 2 (3), 287 (1981).
Allinger N.L., J. Am. Chem. Soc., 99, 8127 (1977).

Brooks B.R., Bruccoleri R.E., Olafson B.D., States D.J., Swarninathan S. and
Karplus M., J. Comput. Chem, 4(2), 187 (1983).

Mayo S.L, Olafson B.D., Goddard W.A., J. Phys. Chem, 94, 8897 (1990).

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

28

Allinger N.L., Yuh Y.H. and Li J.H., J. Am. Chem. Soc., 111, 8551 (1989).

POLYGRAF, v 3.21, Molecular Simulations Inc., 16 New England Executive
park, Burlington, MA 01803.

Lee K.J. and. Brown T.L, Inorg. Chem, 31, 289 (1992).
Neuburger N., Bahar I. and Mattice W.L., Macromolecules, 25, 2447 (1992).
Fan CF. and Hsu S.L., Macromolecules, 25, 266 (1992).

Fan C.F., Olafson B.D., Blanco M. and Hsu S.L., Macromolecules, 25, 3667
(1992)

Muller-Plathe F ., Rogers SC. and van Gunsteren W.F., Macromolecules, 25,
6722 (1992).

Depner M. and Schurmann B.L., Polymer, 33(2), 398, (1992).
Schmolka I.R., J Am. Oil Chem. Soc., 54, 110-116 (1977).
Schmolka I.R., J. Am. Oil Chem. Soc., 59(7), 322-327 (1982).
Schmolka I.R., J. Am. Oil Chem. Soc., 68(3), 206-209 (1991).
Piirma 1., “Polymeric Surfactants”, Marcel Dekker (1992).

Schofield J.D., in Critical Reports in Applied Chemistry, M.R. Porter Ed., Elsevier
Applied Science, New York, pp 35 (1990).

Porter M.R., Handbook of Surfactants, Second Edition, Chapman & Hall, New
York (1994).

Noolandi J ., Makromol. Chem. Theory Simul., 1(5), 295-8 (1992).

Tsuchida E. and Abe K., Adv. Polym. Sci., 45, 1 (1982).

Bekturov E. and Bimendina L.A., Adv. Polym. Sci., 45, 100 (1980).

Osada Y., Adv. Polym. Sci., 82, 1 (1987).

Kabanov VA. and Zezin A.B., Vysokomol Soyed., A21, 243 (1979).

Klier J ., Scranton AB. and Peppas N.A., Macromolecules 23, 4944 (1990).

Scranton A.B., Klier J. and Aronson C.L., in Polyelectrolyte Gels, R. S. Harland
and R. K. Prud’homme Eds., ACS Symposium Series 480, Washington, DC, pp
171(1992)

CHAPTER 2

OBJECTIVES OF RESEARCH

It is clear that there exists a need for an investigation into the development of
polymeric pseudocrown ethers and reversible emulsifiers. The ability of poly(ethylene
glycol) to form complexes with cations through ion-dipole interactions and with
complementary Lewis acids through hydrogen bonding, will be utilized in the
development of these materials. Essentially, the objective behind the development of
both materials is the use of a Simple fundamental concept in the easy and inexpensive
design of a polymeric system with potential commercial applications. However, a
fundamental understanding of the complexation phenomenon in both cases is essential in
order to better design the resulting materials.

Broadly, this research may be divided into two sections: a) an investigation into
the cation binding ability of oligo(ethylene glycol) and its use in the design of a
polymeric pseudocrown ether through a one-step polymerization scheme, and b)
utilization of the polymer complexation ability of poly(ethylene glycol) in the design of a
novel reversible emulsifier. Molecular modeling techniques will be employed to guide
the synthesis in both projects and to provide a fundamental understanding of the
underlying complexation phenomenon and its effect on molecular conformations. These
methods will not only be utilized to visualize various complexation behavior but will also

be used to validate experimental phenomenon.

29

ii.

iii.

iv.

Vi.

30

The specific objectives of this research are as follows:

an experimental and theoretical study of the cation binding ability of oligo(ethylene
glycol) diacrylates;

characterization of the phenomenon of templatization using pyrene end-labeled tetra-
and penta-ethylene glycol;

synthesis of polymeric pseudocrown ethers based upon a template ion scheme and an
investigation of ion-binding properties;

synthesis of poly(methacrylic acid-g-ethylene glycol) hydrogels and an investigation
into the complexation phenomenon through the study of equilibrium swelling in
response to solvent type and composition and polymerization conditions;

synthesis of novel reversible block/graft emulsifiers and a study of their oil
emulsification and surface acitive behavior; and,

use of the polarity sensitive fluorescence of pyrene to study the aggregate formation

behavior of the reversible emulsifiers.

Presented next is a brief description of the scope of each chapter, its contents and

how the specific objectives of this research will be met.

Chapter 3 is the first of three chapters which describe the synthesis and

characterization of polymeric pseudocrown ethers. This chapter is devoted to the

understanding of the ion-solubilization phenomenon observed experimentally through

cation solubility studies in organic solvents. NMR spectroscopy studies along with ion-

solubilization studies will be utilized to demonstrate the cation binding properties of

31

poly(ethylene glycol) and poly(ethylene glycol)-diacrylate. Further, molecular dynamics
simulations will be employed to demonstrate the effect of the presence of a template ion
on the end-to-end distance of poly(ethylene glycol)-diacrylates of varying chain lengths.
These studies will provide a better understanding of the fundamental basis for the
synthesis scheme which relies on template ion-induced cyclization of oligo(ethylene
glycol) diacrylates.

Chapter 4 will further build upon the experimental and modeling results obtained
in Chapter 3. The cations for use in the design of the pseudocrown ethers will be
identified and the size of the corresponding oligo(ethylene glycol)-diacrylates will be
determined which will most effectively template the cation, thereby bringing the
unsaturated end-groups into close proximity. For this purpose the synthesis of pyrene
end-labeled tetra- and penta-ethylene glycol will be discussed and pyrene excimer
fluorescence will be used to identify the oligomer which is most effectively able to
template around a given cation. Once again, molecular dynamics Simulations will be
employed to corroborate experimental results and demonstrate the effect of chain length
on the decrease in the mean end-to-end distance observed in the presence of a template
cation.

Finally, in Chapter 5 the synthesis of polymeric pseudocrown ethers will be
discussed. Various issues which were identified in previous chapters as being important
in the design of polymeric pseudocrown ethers will be summarized and a synthesis
scheme will be formulated. An experimental methodology will be developed for the
study of the ion-binding capacities of the resins. The monomer to solvent ratio will be

varied and its effect on the ion-binding capacity will be discussed. Finally, the ion-

32

binding properties of a polymeric pseudocrown ether for the binding of nickel cations
from aqueous waste streams will be presented.

Chapter 6 is the first of three chapters devoted to the development of reversible
block/graft polymeric emulsifiers based upon polymer complexation. This chapter will
describe the effect of polymer complexation on the swelling behavior of crosslinked
poly(methacrylic acid-g-ethylene glycol) hydrogels in response to solvent type and
composition. This work will build upon previous studies which have demonstrated the
effect of pH on the complexation phenomenon exhibited by copolymers of
poly(methacrylic acid) with oligo(ethylene glycol) grafts. In addition, molecular
mechanics studies will be employed to investigate the stoichiometry of the complex.

Chapter 7 will primarily describe the synthesis and emulsification properties of
the reversible emulsifiers. Variables such as methacrylic acid to ethylene glycol repeat
unit ratio, and their effect on the oil emulsification ability of the resulting copolymers,
will be investigated. Interfacial tension measurements will be obtained as a function of
pH for the block/graft copolymeric emulsifiers.

In Chapter 8 the environment-sensitive fluorescence of pyrene will be employed
to study the aggregate formation behavior of the emulsifiers as a function of
concentration and pH. These studies will also be used to identify a “critical aggregate
concentration” for the emulsifiers based upon peak intensity ratios in the excitation and
emission spectrum of solubilized pyrene in aqueous copolymer solutions.

Finally, a chapter each will be devoted to the presentation of the conclusions and

recommendations for future work.

CHAPTER 3

lON-SOLUBILIZATION AND ION INDUCED TEMPLATIZATION BY
OLIGO(ETHYLENE GLYCOL) AND ITS DIACRYLATES'

3.1. Introduction

In this chapter, the basis for a novel method for producing inexpensive polymeric
pseudocrown ethers in situ during free-radical polymerizations will be investigated, both
experimentally and by molecular dynamics Simulations. The synthetic scheme is based
upon a template ion and exploits the tendency of oligomeric ethylene glycol diacrylates to
form intramolecular cycles during the polymerization. Oligo(ethylene glycol) diacrylates
undergo free radical polymerizations by virtue of their carbon double bonds. If a growing
radical chain reacts with both double bonds within a few propagation steps, a cyclic
"pseudocrown ether" will be formed. The probability of this occurrence depends upon
the conformation of the diacrylate oligomer as well as the concentration of double bonds
in the proximity of the free radical. This probability will be fairly small for bulk
polymerizations and somewhat higher for dilute solution polymerizations. However, if
the polymerization is carried out in a nonpolar solvent with a small concentration of a
solubilized ion, the ion will bind with the oligo(ethylene glycol) diacrylate, inducing it to

assume a cyclic conformation which brings the unsaturated end groups in close proximity

 

’ Adapted from: Mathur A. M. and Scranton A. 3., “Synthesis and Ion-Binding Properties of Polymeric
Pseudocrown Ethers: A Molecular Dynamics Study”, Sep. Sci. & Tech, 30 (7-9), 1071 (1994).

33

34

(Figure 3.1). This template effect brought on by the complexed ion Should dramatically
increase the probability of primary cyclization and should, therefore, increase the
pseudocrown ether yield. Compared to traditional schemes in which a polymeric chain is
functionalized with a preexisting crown ether, our scheme offers safety advantages (since
crown ethers are not produced until they are polymer-bound) and is less expensive.

This novel synthetic scheme for polymeric pseudocrown ethers builds upon
previous research on the ion-binding properties of noncyclic oligo(ethylene glycol)s and
polyglycol dimethyl ethers (glymes) (see Figure 3.2). The complex stability for alkali
cations with glymes exhibits an increase with increasing chain length for relatively small
chains' but levels off at a threshold value which depends upon the Size of the cation.
Ethylene oxide oligomers have been reported to form 1:1 and 4:1 complexes with
mercuric chloride,2 and the structures of the resulting complexes have been elucidated.
Various studies have Shown that poly(ethylene glycol) (PEG) and glymes of various
chain lengths may be used as a cosolvent phase in heterogeneous reaction systems, due to
their ability to bring into contact reagents from aqueous and organic phases, and as agents
for phase-transfer catalysts in organic syntheses.”

This research also builds upon previous work on polymeric pseudocrown ethers.
Warshawsky et al.7 first synthesized polymeric pseudocrown ethers by reacting
chloromethylated polystyrene resins with oligomeric ethylene glycols (Figure 3.3). In
their method, the alcohol moieties on the ethylene glycol reacted with the chloromethyl
groups by a nucleophilic substitution reaction to produce pseudocrown ethers with
varying ring sizes. The resulting polymeric pseudocrown ethers were found to bind

various alkali and transition metal ions.

35

o
I CHK \ __ +
CHz—C—C 0 CH2 C—C—C “2 + Metal
n Catio
PEGDA

 

CH3
_| CHZ=CH CH,=CH
Cit—if) lb=o (i=0
— O
(:D + < ‘6’?
CH2 0 Ozttiaoln
l
PCE HEMA Templated PEGDA

Figure 3.1. Scheme for the synthesis of polymeric pseudocrown ethers based upon a one-
step free-radical polymerization of templated polyethylene glycol diacrylate

with a comonomer such as hydroxy ethyl methacrylate.

OH
CHz/ "
CH2 Metal Cation =OH
\O. +
/ . CH2
CH2 " 6———-CH2
0
\CH2/ \ /

CH2 “—0 H2

Figure 3.2. Schematic showing the cation complexation ability of oligomeric polyethers.

36

CHZCI

+ OH—CHz—CHfEO—CHz—CHZEl—O—CHz—CHz—OH

l

O_C H2 _C H2\

HZCI

I
M
0—0

0—0—

CH2

Figure 3.3. Synthesis scheme of Warshawsky et al. for the synthesis of polymeric
pseudocrown ethers.

37

The conformations of macrocyclic ligands and their complexes with ions have
been studied using molecular modeling techniques. For example, a molecular mechanics
study of 18-crown-6 and its alkali complexes8 demonstrated that the lowest energy
conformers possessed the Ci symmetry, as verified by X-ray crystallography. Wipf et al.8
have used molecular mechanics simulations to explain the temperature-dependent dipole
moment of 18-crown-6 and to investigate the stability of its complexes with Na+ and K+.
A distance geometry approach was adopted by Weiner et al.9 to study the geometry of
ring systems including 18-crown-6 using both the MM2 and AMBER force fields.
Molecular mechanics has also been used to estimate complexation energies and to
provide insight into host-guest interactions of alkali cations with anisole spherands.l0
Computer simulations illustrated that for some ligands, steric strain involved in wrapping
the ligand around the ion is comparable to the free energy of complexation.‘| Kollman
and collaborators'2 have used a host of simulations methods, including distance geometry,
molecular mechanics, and molecular dynamics, to investigate complexes of spherands,
crown ethers and porphyrins. These studies revealed that the simulations yielded
satisfactory binding energies in good agreement with experimental values. Finally,
molecular mechanics and molecular dynamics studies of cation complexes of a cyclic
urea-anisole spherandl3 have yielded valuable information about host-guest
preorganization.

In this chapter we will describe a series of molecular dynamic simulations of
various oligomeric ethylene glycol diacrylates in the presence and in the absence of
cationic species. The molecular modeling studies were performed to corroborate

experimental investigations which demonstrated ion solubilization in organic solvents

38

using PEG. In addition, 1H NMR studies were performed to confirm the complexation of
the cations with the PEG. The goal of this modeling effort is to provide an underlying
molecular understanding which will allow the development of an efficient reaction
scheme which maximizes the yield of the polymeric pseudocrown ether. Clearly the
number of ethylene glycol repeating units in the diacrylate is important because the
optimal ligand must effectively envelop the cation in order to bring the unsaturated end-
groups in close proximity. Therefore, simulations were performed for ligands containing
between two and ten ethylene glycol repeating units, with and without the presence of a
cation. In addition, a series of simulations were performed to investigate the effects of

simulation time and boundary conditions.

3.2. Experimental Methods and Techniques

3.2.1. Ion-Binding Studies of Oligomeric PEG and PEGDA

A series of ion-binding and solubility studies were performed in various solvents.
The objective of the studies was to identify ion/ligand/solvent systems in which an
initially insoluble ion was solubilized by the PEG ligand. This would ensure that the
ligand and the cation complexed with one another. Metal salts that were investigated
include cuprous, nickel, lead, tin and chromium chlorides, while the solvents in which the
studies were carried out include chloroform, methylene chloride, carbon tetrachloride,
and dioxane. All salts were dried in an oven at 60°C for 4 days and stored in a desiccator
before use. PEG with an average molecular weight of 300 daltons (Polysciences Inc.,

Warrington, PA) was chosen as the representative oligomer for these solubility studies.

39

In these experiments, 1 wt % of salt was added to solvent in a 20-ml glass vial. The
contents of the Vial were then mixed thoroughly in a vortex mixer for about 10 min. If
the salt was not soluble in the solvent, then about 10 wt % PEG 300 was added to the vial
and the Vial agitated for about 15 min. The solubilization of the salts (if any) upon the

addition of the PEG 300 was noted.

3. 2.2. NMR Studies

1H NMR spectroscopy was used to characterize binding by the PEG 300 by
comparing PEG 300 spectra collected in the absence and in the presence of the cations.
These studies were performed in deuterated chloroform and deuterated methylene
chloride, and the salts under study were tin chloride and chromium chloride. The NMR
spectra were taken on a Varian 300 instrument at the Max T. Rogers NMR facility at
Michigan State University. All samples had a concentration of 0.1 wt % in the deuterated
solvent. The salt-containing samples were prepared by dissolving 0.1 wt % saturated
PEG 300 in the NMR solvent. All spectra were taken at 25°C and were obtained by
signal averaging 64 scans. The CHC13 residual peak at 7.24 ppm in CDCI3 was used as a

chemical shift reference.

3. 2. 3. Molecular Modeling

Molecular dynamics (MD) simulations were performed on a Silicon Graphics
IRIS 4D/220GTX workstation using the molecular modeling software POLYGRAF.l4
These MD simulations were performed on PEGDA containing between two and ten

ethylene glycol repeating units, with and without the presence of a cation (Na+ or Al3+).

40

The molecular structures were built using the polymer builder and were energy
minimized before further MD calculations. Molecular mechanics (energy minimization)
calculations were based on the DRIEDING force field parameters, and the steepest
descent algorithm was used to perform the minimization. The Gasteiger algorithm was
used to charge equilibrate the structures. Canonical TVN MD Simulations were
performed on the oligomers for a period ranging from 50 to 200 ps and the molecular
trajectories were stored over the entire run. The trajectory file was written at every 0.1
ps, and the temperature for all the runs was kept at 300 K. The 200-ps simulations were
performed on a single PEGDA molecule within a periodic cell with an external pressure
of 1 atm at 300 K with and without the cation (sodium). Plots of total energy versus time
and of end-to-end distance (ETED) versus time were generated from the Simulations. In

addition, the time-averaged values of the above two parameters were calculated.
3.3. Results And Discussion

3. 3. 1. Ion-Binding Studies with Oligomeric PEG

Results from the ion solubility studies are Shown in Table 3.1. In each case
shown in the table, the ion was insoluble before the addition of the PEG. A table entry of
"Y" indicates that the addition of the PEG solubilized the metal salts. This solubilization
indicates that the ligand and cation are complexed with one another. Similar ion-binding
properties were observed for the diacrylates and dimethacrylates of PEG. Since the

pseudocrown ether synthesis should be carried out in a solvent in which the PEGDA is

41

soluble but the salt is insoluble unless it is complexed with the PEGDA, chloroform and

methylene chloride are good choices for the synthesis solvent.

Table 3.1. Solubilities of Salts Upon the Addition of PEG 300

 

 

 

 

 

 

Solvents/ CHCl3 CHzClz CCI4 Dioxane
Salts
CuCl Y Y Y ---
NiClz Y Y m ---
PbC12 --- m m ---
SnClz Y Y Y Y
CTCI3 Y Y partial ---

 

3.3.2. NMR Studies

1H NMR spectra of PEG 300 in deuterated chloroform are shown in Figure 3.4.
The spectrum in Figure 3.4a was collected in the absence of any metal salts, while that in
Figure 3.4b was obtained in the presence of chromium chloride. In these spectra, the
singlet peak located at 3.6 ppm corresponds to the PEG backbone methylene protons
(CH2) while the end group hydroxyl protons appear at 2.4 ppm in Figure 3.4a and at 2.7
ppm in Figure 3.4b. Therefore, the presence of chromium chloride bound by PEG 300
induces a shift in the location of the peak corresponding to the end-group hydroxyl
protons. A similar Shift was observed in deuterated methylene chloride and when tin
chloride was substituted for chromium chloride. These shifts may be attributed to
complexation of the metal ion with the PEG ligand. The ion may withdraw electron

density from the donor oxygen atoms, resulting in inductive deshielding of local

42

(a)

 

 

 

‘1" V Y r V Y Y Y I

(b)

 

 

 

(I
.3
A
P) a
O.
O
O
3

Figure 3.4. 1H NMR Spectra of a) PEG 300 and b) PEG 300 in the presence of chromium
chloride.(hydroxy1) protons.

43

Similar shifts in NMR peak position induced by metal salts have been reported in the
literature. For example, "B and 13C NMR Spectroscopy studies on crown ethers revealed
that the chemical shift of a ligand was shifted to a higher field in the presence of an ion as
compared to that of the uncomplexed host.l5 Proton NMR chemical shifts induced by
ionic association on a poly(ethylene oxide) chain have been observed, and the effect of

counter ions on the magnitude of the downfield shift has been studied.”17

3. 3. 3. Molecular Dynamics Simulations

A series of simulations were performed to elucidate the effect of the presence of a
cation on the conformation of a PEGDA ligand. To determine the effect of the ligand
size, simulations were performed on PEGDA containing between two and ten ethylene
glycol repeating units with and without the presence of a cation (Na+ or Al3+). The first
set of experiments was carried out for a duration of 50 pS. For these simulations, the total
system energy as a function of time fluctuated about the mean value typically with a
standard deviation less than 5%. Therefore, all of the individual conformations had
nearly the same energy and are equally probable. The PEGDA end-to-end distance
(distance between unsaturated end groups), which is a parameter useful in determining
whether the ion causes any templatization, was also plotted as a function of time for all
the simulations. This value also fluctuated with a standard deviation as large as 20% of
the mean.

Figure 3.5 is a plot of simulation results for the mean end-to-end distance (ETED)
as a function of the number of EG repeating units in the PEGDA chain. The figure

illustrates that the addition of a sodium ion results in a decrease in the mean ETED

44

18

 

16 4-

 

Mean End-to-End Distance (A)

 

 

 

4 ~1-
+PEGDA
2 " +PEGDA with Na‘
0 I a I I I I I
2 3 4 5 6 7 8 9 10

Number of Ethylene Glycol Repeat Units "n"

Figure 3.5. MD 50 ps simulation results for the mean end-to-end distance of PEGDA
with different number of EG repeating units.

45

for all of the PEGDA chain lengths. Similar trends were observed when the sodium ion
was replaced with an aluminum ion. This suggests that there is indeed a propensity for
cyclization owing to the presence of a cation since the oligomer tends to template around
the cation thereby decreasing the mean ETED. The effect of the ion is illustrated in
Figure 3.6, which contains a pictorial representation of one of the conformations adopted
in the presence of a sodium ion. This figure indicates that the PEGDA assumes a crown
ether-like conformation around the ion with the unsaturated end groups in close
proximity. The results Shown in Figure 3.5 also suggest that there exists an optimum
PEGDA chain length since the PEGDA ligand containing four EG repeating units
exhibits the lowest ETED and therefore the maximum templatization. This result is in
agreement with studies on Size-match selectivity reported in the literature. For crown
ethers, it has been reported that maximum complexation is achieved when the size of the
cavity is comparable to the ionic diameter. For Na“, the crown ether tert-butyl-
cyclohexyl-lS-crown-S with 5 donor oxygens has been found to provide the required Size
match selectivity.‘8 For PEGDA, our simulations indicate that the end group acrylate
carbonyl oxygens participate in the ion binding, making the PEGDA with n = 4 the
optimal ligand to bind with sodium.

To investigate the effect of Simulation time, MD runs were performed for 50, 100,
and 200 pS. In general, MD simulations of greater duration allow a larger number of
conformations to be sampled and therefore provide more reliable averaging. However,
the inevitable cost of longer simulations is greater computational requirements (in our
studies a 50-ps simulation takes a few hours, whereas the 200-ps runs require days to

complete). Therefore, it is important to determine the minimum simulation time which

46

 

Figure 3.6. Molecular representation of a PEGDA (n = 4) conformation templated
around a Na“ ion showing charged surface of the molecule. Shown here is
one conformation extracted from a 50 ps run.

47

provides reliable results. The effect of simulation time on the mean ETED of PEGDA in
the presence of a sodium ion is illustrated in Figure 3.7. The figure illustrates that for
PEGDA ligands containing seven or fewer EG units, the simulation results are very
repeatable; however, the Simulations exhibit greater variations for larger ligands. This
interesting behavior suggests that the simulation time has an effect only for relatively
long ligands in which the length of the oligomer becomes considerably greater than that
required to encompass the ion. For shorter ligands, the ion imposes an overwhelming
constraint on the conformations adopted by the ligand due to binding interactions. For
the longer ligands, the ion interacts only with a portion of the chain, placing milder
constraints on the end-to-end distance.

Differential and cumulative distributions for the PEGDA end-to-end distance
from the 200-ps simulations are plotted in Figures 3.8 and 3.9, respectively. The points
in Figure 3.8 correspond to the simulation data, while the curves represent Gaussian fits
to the data. Both figures clearly illustrate that the presence of the sodium ion shifts the
end-to-end distance to lower values while increasing the dispersion slightly. The
cumulative distributions shown in Figure 3.9 may be used to estimate the fraction of the
(equally probable) conformations with an end-to-end distance below a give value. By the
ergodic theorem, this will correspond to the fraction of PEGDA ligands in the system
with an end-to-end distance below this threshold value. Therefore, the Simulations can
provide information about the fraction of "templated" ligands which could lead to the

formation of pseudocrown ethers upon reaction.

48

18

16,

Mean end-to-end distance (A)

L If, , I, ,, LI L, m .1— L

2 3 4 5 6 7 8 9

Number of ethylene glycol repeat units "n"

Figure 3.7. Effect of different simulation times on the mean end-to-end distance for
PEGDA with Na+ with various numbers of EG repeating units.

 

49

0.045 -

0,040: .. A PEGDA (n=4) with Na“
PEGDA(n=4)

0.035 -
0.030 -
0.025 ;
0.020 —
0.015 -

Normalized Frequency

0.010 —
0.005 -
0.000 -

 

 

End-tO-End Distance (A)

Figure 3.8. Normalized frequency distribution of end-to-end distance for PEGDA (n = 4)
with and without Na+.

50

0.9 ,_
0.8 «F
0.7 4-

.0
w

     

+ PEGDA (n=4)
-o— PEGDA (n=4) with Na“

Normalized Cumulative Frequency
O
01

 

A I L l
T I I I

25 30 35 40
End-to-End Distance (A)

Figure 3.9. Cumulative frequency distribution of end-to-end distance for PEGDA (n = 4)
with and without a Na+ ion.

51

3.4. Conclusions

A novel scheme to synthesize polymeric pseudocrown ethers in situ during a free-
radical polymerization has been presented. This scheme is based upon a template ion and
exploits the tendency of oligomeric ethylene glycol diacrylates to form intramolecular
cycles during polymerization. Compared to traditional schemes in which a polymeric
chain is functionalized with preexisting crown ethers, our scheme offers safety
advantages and is less expensive. It will thus be possible to economically synthesize
polymers with unique ion-binding capabilities appropriate for applications such as phase-
transfer catalysis, ion chromatography, and metal ion removal from waste streams.
Preliminary experimental studies demonstrated that certain salts that were not soluble in
nonpolar solvents, such as chloroform and methylene chloride, were made soluble upon
the addition of oligomeric PEG 300. Further evidence of cation binding by oligomeric
PEG was obtained by 1H NMR studies of PEG and its complexes with metal salts.

In an effort to optimize the template ion synthesis approach, molecular modeling
studies were performed on oligomeric PEGDA containing between two and ten ethylene
glycol repeating units, with and without the presence of cations. Simulation results
indicated that the presence of the templating cation significantly decreased the mean end-
to-end distance, thereby bringing the unsaturated end-groups into close proximity.
Although the presence of the cation decreased the ETED for all PEGDA chain lengths,
the PEGDA ligand that resulted in the most effective templatization for Na+ contained
four ethylene glycol repeating units. The simulation time had little effect on the results

for relatively short PEGDA ligands containing seven or fewer ethylene glycol repeating

52

units. These ligands are highly constrained due to binding interactions with the ion. A
higher degree of variation with Simulation time was observed for longer ligands in which
only a portion of the chain interacts with the ion. These Simulation results have affirmed
the role of a templating ion in the synthetic scheme. Moreover, the results provide insight
into the selection of the templating ion and the PEGDA oligomer that will maximize

templatization by bringing the unsaturated end groups into proximity.

10.

ll.

12.

13.

14.

15.

16.

17.

18.

53

3.5. References

Chan L. L., Wong K. H., and Smid J., J. Am. Chem. Soc., 92, 1955 (1970).
Iwamoto R., Bull. Chem. Soc. Jpn., 46, 1114 (1973).

Lee D. J. and Chang V. S., J. Org. Chem, 43, 1532 (1978).

Regen S. L., Besse J. J., and McLick J., J. Am. Chem. Soc., 101, 116 (1979).
Zupancic B. and Kakalj M., Synthesis, 913 (1981).

Sukata K., Bull. Chem. Soc. Jpn., 56, 280 (1983).

Warshawsky A., Kalir R., Deshe A., Berkovitz H., and Patchomik A., J. Am.
Chem. Soc., 101, 4249 (1979).

Wipf G., Weiner P. and Kollman P., J. Am. Chem. Soc., 92, 666 (1982).
Weiner P. K. et al., Tetrahedron, 39, 1113 (1983).

Kollman P. A., Wipr., and Singh U. C., J. Am. Chem. Soc., 107, 2212 (1985).
Hancock R. D. and Martell A. E., Comments Inorg. Chem, 6, 237 (1988).

Kollman P. A., Grootenhuis P. D. J., and Lopez M. A., Pure Appl. Chem, 61, 593
(1989)

Maye P. V. and Venanzi C. A., J. Comput Chem, 12, 994 (1991).

POLYGRAF, v 3.21, Molecular Simulations Inc., 16 New England Executive
park, Burlington, MA 01803.

Live D. and Chan S. I., J. Am. Chem. Soc., 98, 3769 (1976).
Ono K. and Honda H., Macromolecules, 25, 6368 (1992).

Yanagida S., Takahashi K., and Okahara M., Bull. Chem. Soc. Jpn., 51, 1294
(1978).

Christensen J. J., Hill J. O., and Izatt R. M., Science, 174 (4008), 459 (1971).

CHAPTER 4

EFFECT OF ION-INDUCED TEMPLATIZATION ON THE END-TO-
END DISTANCE OF PYRENE END-LABELED TETRA- AND
PENTA-ETHYLENE GLYCOL‘

4.1. Introduction

In the previous chapter a novel schemel was presented for the synthesis of
polymeric pseudocrown ethers which have potential applications in cation binding,
separations, and waste management. This scheme is based upon the tendency of
oligomeric ethylene glycol diacrylates to assume circular conformations in the presence
of templating cations. This templatization phenomenon may induce the diacrylate
unsaturated end-groups to be in proximity. If a free-radical polymerization is initiated
with a comonomer such as hydroxy ethyl methacrylate, both double bonds of such a
templating oligomer may be incorporated into the same kinetic chain resulting in the
formation of pendent loops. The probability of cyclization is increased if the
polymerization is carried out in non-polar solvents in which the cation itself is insoluble
thus ensuring maximum templatization and hence primary loop formation. Furthermore,

if the free-radical polymerization is performed in dilute solution, the probability of

 

' Adapted from: Mathur A. M. and Scranton A. 8., “Synthesis and Ion-Binding Properties of Polymeric
Pseudocrown Ethers II: Ion-Induced Templatization of Oligomeric Ethylene Glycol Diacrylates”, Sep. Sci.
& Tech, (in Press).

54

55

cyclization is enhanced and would result in reduced crosslinking and more cyclization as
shown in Chapter 3, Figure 3.1.

Our previous solubility and 'H NMR studies demonstrated that various cations
were solubilized by oligomeric ethylene glycols' by virtue of ion-dipole interactions
between the cation and the electron lone pairs of the ether linkages. Further, molecular
dynamics simulations were performed on oligomeric ethylene glycol diacrylates both in
the presence and absence of templating cations. A reduction in the mean end-to-end
distance in the presence of a cation was observedl due to ion-induced constraints on the
possible number and type of conformations adopted by the oligomer. These Simulations
affirmed the role of the templating ion in the synthetic scheme. Experimental
corroboration of the theoretical results suggesting that templatization brings the chain
ends into proximity is presented in this chapter.

Pyrene has been extensively used as a molecular probe due to its excellent
fluorescence properties and its ability to form bi-molecular complexes (excimers) which
fluoresce in a different spectral region than the individual “monomer” pyrene molecules.
Cuniberti and Perico2 were among the first to study polymer cyclization using steady-
state pyrene excimer fluorescence. These investigators synthesized pyrene end-labeled
poly(ethylene oxide) and studied the influence of molecular weight on the extent of
intramolecular excimer formation. They observed an increase in the intramolecular
excimer fluorescence intensity with decreasing molecular weight. Winnik et al.3'4 studied
the cyclization dynamics of pyrene end-labeled polystyrene and demonstrated that
intramolecular excimer formation could be observed only for chains containing less than

1000 to 2000 bonds. For higher molecular weight chains the rate constant for cyclization

56

decreases rapidly. Winnik’s group has also studied the cyclization dynamics of pyrene
end-capped polystyrenes using transient fluorescence experiments. For example,
researchers from Winnik’s group reported the determination of cyclization rate constants
from fluorescence decay data,5‘6 asymmetric end-labeling of polystyrene and
determination of cyclization rate constants from intramolecular exciplex emission decay
data6 and fluorescence quenching in pyrene/benzil end-labeled poly(tetrarnethylene
oxide).7 A review of their earlier work8 on the cyclization of polymer chains in solution
and luminescence techniques to study the morphology of prototype industrial materials, is
a good source of literature in this area. Recently, Martinho et al.9 studied the effect of
hydrostatic pressure on the cyclization of pyrene end-labeled polystyrene in various
solvents and observed that the cyclization rate decreases monotonically with pressure.
Frank et al. '0'” studied the complexation of poly(acrylic acid) with pyrene-labeled
poly(ethylene glycol) by monitoring the excimer-to-monomer fluorescence intensity
ratio. These authors used this technique to study the hydrophobic attraction between
pyrene-end-labeled polyethylene glycols in water and water-methanol mixtures, and the
effect of the hydrophobic interaction in the poly(methacrylic acid)/pyrene end-labeled
poly(ethylene glycol) complexes. They also used the monomer/excimer ratio to
demonstrate the effect of pH on complexes of poly(acrylic) or poly(methacrylic acid)
with pyrene end-labeled poly(ethylene glycol).'3'” Frank’s group has also developed a
new anionic synthesis schemel5 for the synthesis of pyrene end-labeled polystyrene
having no ester linkages. The resulting polymers demonstrated enhanced thermal and

hydrolytic stability as compared to those synthesized by previous methods.

57

The preceding two paragraphs illustrate the utility of pyrene excimer fluorescence
experiments for conformation studies in cyclic or complexing polymer systems. In this
study, this technique will be used to investigate ion-induced cyclization of relatively
short-chain oligo(ethylene glycol) molecules. The monomer to excimer fluorescence
ratio will be utilized to characterize the extent of oligomer cyclization. Comparison of
the fluorescence ratio obtained in the presence of the templating ion to that observed in
the absence of the ion, will provide insight into the conformation of the templated ligand.
Therefore, these studies serve as an experimental test of the previously reported
molecular modeling results.l In addition, we present new molecular dynamics
Simulations of pyrene-labeled ligands to investigate the effect of the fluorescent
chromophore on the simulation results. These simulations will provide molecular insight
into the fluorescence results by elucidating the effects of the templating ion and/or the

pyrene end groups on the conformations adopted by the oligo(ethylene glycol) chains.

4.2. Experimental Methods and Techniques

4. 2. 1. Synthesis of Pyrene End-Labeled Oligomeric Ethylene Glycols

Pyrene end-labeled tetraethylene glycol (TEG) and pentaethylene glycol (PEG)
were prepared based on the procedure reported by Cuniberti and Perico2 (Figure 4.1).
One-pyrenebutyric acid (3.11 g), was mixed with tetraethylene glycol (0.899 g) or
pentaethylene glycol (1.013 g) and para-toluenesulphonic acid monohydrate (0.268 g) in
a 500 ml round bottom flask equipped with a Dean Stark trap as well as a condenser.

These chemicals were used as received from Aldrich. To this mixture was added 250 ml

58

CHZCHZCHZCOOOH
O Toluene
PTSA
+ OH<CH2-C Hz-OfiH ——>
I -H20

l-pyrenebutyric acid Oligo-ethylene glycol (n = 4,5)

I (CH2)3—C—O<-CH2-CH2-Oirfl) —-(CH2)3 I
06 End- to-end- distance (ETED) a0

Figure 4.1. Scheme for synthesis of pyrene end labeled oligoethylene glycols.

59

of dry toluene. The mixture was stirred and allowed to reflux overnight. The next day,
the reaction was stopped and TLC (silica gel) revealed one strong spot at Rf = 0.13
(hexane: ethyl acetate 1:1). The starting material, l-pyrenebutyric acid, appears at Rf =
0.43. The product was isolated as follows. The contents of the round-bottom flask were
emptied into a separatory funnel and extracted with 3 x 100 ml saturated sodium
bicarbonate solution to remove the catalyst para—toluenesulphonic acid and reactants. The
toluene layer was extracted once with distilled water (100 ml) and once with brine (100
ml), then stirred over anhydrous sodium sulfate for 20 minutes and filtered. Finally the
solvent was evaporated on a rotary evaporator at 60°C. The dried product was placed on
a high vacuum line for one hour and pumped at room temperature. The product yield was
96% and the proton and carbon-13 NMR spectra were consistent with the desired
products: tetraethylene glycol and pentaethylene glycol which are pyrene-labeled on both

ends (henceforth abbreviated Py-TEG-Py and Py-PEG-Py, respectively).

4. 2. 2. Fluorescence Studies

The fluorescence studies of pyrene end-labeled oligomers were performed using
an Aminco-Bowman Series AB2 Luminescence Spectrometer. An excitation wavelength
of 322 nm was chosen based upon absorption spectrum of pyrene, and the steady-state
fluorescence emission was collected for wavelengths between 330 and 630 nm. The
solvent to be used for the fluorescence studies had to meet several criteria. First, it should
be a solvent in which the templating ions themselves are insoluble, but are solubilized

upon addition of the oligo(ethylene glycol) chains.l This ensures that all ions in solution

60

are bound by the oligomers, and therefore ensures maximum templatization. In addition,
the solvent must meet the optical requirements of low absorption and emission in the
Spectral window between 300 and 630 nm. With these criteria in mind, chloroform and
THF were selected for the studies. Dilute solutions of the pyrene end-labeled oligomers
(Py-PEG-Py and Py-TEG-Py) were prepared in HPLC grade chloroform and THF at
concentrations varying from 10“ to 10'7 M. For studies performed in the presence of a
templating ion, the cation (tin, chromium or nickel) was introduced into the polymer
solution as its hydrated chloride salt.'6 These salts were added in excess and the resulting
solutions were agitated for 2-3 days to ensure maximum templatization and equilibration.

The solutions were then filtered prior to the fluorescence studies.

4. 2. 3. Molecular Dynamics Simulations

Molecular dynamics (MD) simulations were performed on pyrene end-labeled
tetraethylene glycol and pentaethylene glycol in the presence and in the absence of the
templating ion. The pyrene end-labeled tetra- and pentaethylene glycols were modeled
using molecular modeling software POLYGRAF.l7 The molecular structures were
energy minimized using the DRIEDING force field and were charge equilibrated using
the Gastieger algorithm prior to the MD simulations. For Simulations in the presence of
the templating ion, sodium (N ai) was introduced in proximity to the ether oxygens before
energy minimization and charge equilibration. TVN MD simulations were performed in
the absence of any solvent18 at 300K for either 100 or 200 ps and their trajectories were
recorded at every 0.1 ps. Our previous simulations' indicated that for oligomers of this

relatively short chain length, there is little or no effect of simulation time (above 50 pS)

61

on the mean end-to-end distance. However, in the case of tetra- and pentaethylene glycol
simulations were performed for 100 and 200 ps respectively, to allow for the bulky
pyrene end-groups to sample more conformations and hence provide a better average.
Plots of end-to-end distance versus time were generated and plots of energy versus time
yielded almost a horizontal line thereby ensuring that all the assumed conformations were

equally probable and possessed almost the same energy.

4.3. Results And Discussion

4. 3. 1. Fluorescence Studies

Figure 4.2 contains fluorescence profiles of two distinct systems: a dilute solution
(10’5 M) of 1-pyrenebutyric acid (curve A) and a similarly dilute solution (10'5 M) of the
pyrene-end labeled oligomer Py-PEG-Py (curve B), both in chloroform. The figure
illustrates that these two systems have markedly different characteristic fluorescence
profiles. While both spectra exhibit Sharp peaks at approximately 382 and 402 nm
(characteristic of the uncomplexed, monomeric pyrene), the Py-PEG-Py exhibits a second
broad peak centered at around 480 nm (characteristic of the pyrene excimer). Therefore,
Figure 4.2 illustrates the effect of placing two pyrene molecules in proximity on the two
ends of an oligomeric ethylene glycol chain. At this low concentration, the l-
pyrenebutyric acid exhibits essentially no excimer formation, indicating that the pyrene
moieties all exist as uncomplexed “monomers.” However, at the same low concentration,
the Py-PEG-Py system exhibits significant excimer formation, indicating that pyrene

moieties have formed sandwich complexes with one another. A series of studies were

62

 

.._A Py-PEG-Py
__B l-PBA

 

 

Emission Intensity (arb. units)

 

 

 

...2 °

330 380 430 480 530 580 630

Emission Wavelength (nm)

Figure 4.2. Fluorescence emission spectra of (A) pyrene end-labeled pentaethylene
glycol and (B) l-pyrenebutyric acid. The excitation was at 322 nm and both
samples were at a concentration of 10'5 M.

63

performed as a function of concentration which suggested that, at these low
concentrations, the ratio of the fluorescence intensity of the excimer peak to that of the
monomer peak (the excimer-to-monomer ratio) is essentially constant, while at higher
concentrations (10“ M and above) this ratio exhibits an upward Slope as the concentration
is increased. This suggests that at low concentrations (10‘5 M and below) the excimer
formation is primarily intramolecular, with the pyrene moieties on either end of the
oligomeric chain coming together to form an excimer, while at higher concentrations the
intermolecular excimer formation between pyrene moieties on two different oligomers
becomes Significant (hence the dependence on concentration). Therefore, these studies
suggest that some oligomer cyclization occurs in the pyrene end-labeled PEG even in the
absence of a templating ion. This is presumably due to the fact that the oligomeric chains
are rather small, leading to a locally high concentration of the two pyrene moieties
attached to a Single chain.

Figure 4.3 illustrates the effect of a templating cation on the fluorescence profile
of Py-TEG-Py. The figure illustrates that in the presence of the templating ion (Nizi in
this case) the fluorescence spectrum of Py-TEG-Py exhibits a significantly enhanced
pyrene excimer-to-monomer ratio relative to spectrum obtained in the absence of any ion.
Similar results were obtained using Sn2+ or Cr” as the templating ion and with
chloroform as the solvent. We attribute the observed enhancement of the excimer peak
upon addition of a templating cation to ion-induced cyclization of the oligo(ethylene
glycol) chain. In agreement with our previous molecular modeling studiesl these
fluorescence studies indicate that the templating ion binds with the oligo(ethylene glycol)

ligand, and facilitates the formation of cyclic conformations which bring the two chain

64

 

10
9
_A Py-TEG- Py
8 _B Py-TEG-Py with Ni

 

Emission Intensity (arb. units)
LII

 

 

 

 

 

4

3

2

‘ J

0 .
330 380 430 480 530 580 630

Emission Wavelength (nm)

Figure 4.3. Fluorescence profiles of pyrene end-labeled tetraethylene glycol in THF
solvent (10.5 M) in the (A) absence and (B) the presence of nickel chloride.
The excitation frequency is 322 nm.

65

ends in proximity. Similar trends were observed for the fluorescence profiles of the
longer oligomeric ligand Py-PEG-Py, as shown in Figure 4.4.

Experimental results for the pyrene excimer-tO-monomer fluorescence ratio for
Py-PEG-Py in two different solvents (chloroform and THF) are illustrated in Figure 4.5.
The figure contains data Obtained both in the presence and in the absence of Ni” as a
templating ion for both solvents, and illustrates some interesting results. For both
solvents, in the absence of the templating ion, the excimer-to-monomer ratio (Iex/Im) was
found to be 0.8, suggesting the probability of cyclization was essentially the same.
However, in the presence of the nickel ions the excimer-to-monomer ratio was
considerably higher in chloroform than in THF. This suggests that more oligomer
cyclization occurs in chloroform than THF. This result, combined with the fact that the
oligo(ethylene glycol) ligands more effectively solubilize the nickel ions in chloroform,‘
provides further indirect evidence for ion-induced cyclization. In addition, these studies
indicate that the pyrene excimer-to-monomer fluorescence ratio provides a sensitive
measure of the ability of the oligo(ethylene glycol) to solubilize the ions into an organic

solvent.

4. 3. 2. Molecular Dynamic Simulations

In the previous chapter,l we reported molecular dynamic Simulations of
oligo(ethylene glycol) diacrylates both in the presence and in the absence of a templating
cation. These studies revealed that the addition of the templating cation leads to a
significant decrease in the average end-to-end distance (in the case of tetra(ethylene

glycol) diacrylate (TEGDA), the shift is from ~ 19 A in the absence of the ion to ~ 7 A in

66

 

8 B —A Py-PEG-Py
_B Py-PEG-Py with Sn

 

Emission Intensity (arh. units)

 

 

 

 

I J
0 I I
330 380 430 480 530 580 630

Emission Wavelength (nm)

 

Figure 4.4. Fluorescence emission spectra of pyrene end-labeled pentaethylene glycol
(10'5 M) in chloroform in the (A) absence and (B) presence of tin chloride.
Excitation is at 322 nm.

67

'0

 

Cl Py-PEG-Py
. Py-PEG-Py with Ni

Excimer to monomer peak ratio
.9 .O O :- .-
45 O5 00 -- N A

.O
N

 

 

 

 

 

 

 

 

Chloroform
Solvent

Figure 4.5. Comparison of excimer-to—monomer ratio for pyrene end-labeled penta-
ethylene glycol (10'5 M) in chloroform and THF in the presence and absence
of nickel chloride.

68

the presence Of the ion). In the present study, molecular dynamic simulations were
performed on pyrene end-labeled ligands to elucidate the effect of the large fluorescent
chromophores on the chain conformations (Figure 4.6 and 4.7). These new simulations
are compared to our previous simulation results in the following discussion.

Figure 4.8 contains simulation results for the normalized frequency distribution of
the end-to-end distance for two distinct systems: pyrene end-labeled tetra(ethylene
glycol), and the analogous unlabeled diacrylate (TEGDA). These results indicate that
replacing the acrylate group with the pyrene chromophore results in a marked decrease in
the mean end-to-end distance (ETED) while leaving the dispersion relatively unchanged
(the ETED is ~19 A for the TEGDA, and ~8.5 A for Py-TEG-Py). These trends may be
attributed to the associative interactions between two pyrene moieties (these interactions
are responsible for the formation of the sandwich complex). These interactions tend to
bring the chain ends in proximity, resulting in the decreased mean end-to-end distance
(Figure 4.7). These simulation results suggest that even in the absence of a templating
cation, the pyrene end-labeled oligomers exhibit an increased tendency to form cycles
compared to the TEGDA. This result is consistent with the significant excimer formation
even in the absence of the templating ion exhibited in Figures 4.2-4.4. Figure 4.9
demonstrates the further reduction in the mean end-tO-end distance upon addition of the
templating cation to Py-TEG-Py. The figure contains simulation results for Py-TEG-Py
in the absence and in the presence of a templating ion. These molecular dynamics
simulations demonstrate that the mean end-to-end distance decreases to ~56 A from
~8.5 A in the presence of the cation. Again the simulation results are consistent with the

ion-induced enhancement in the excimer fluorescence shown in Figures 4.3 and 4.4.

69

  
  

   

.‘K‘cf‘ua -Z . '

Figure 4.6. Starting conformation of end-labeled pentaethylene glycol in the presence of
a sodium cation.

70

 

Figure 4.7. Pyrene end-labeled pentaethylene glycol conformation with templating cation
showing pyrene excimer and the end-tO-end distance (ETED).

71

 

0.06 -

0.05 -

0.04 -

0.03 -l

0.02 -

Normalized Frequency

0.01 —

 

0.00 ~

 

 

 

25

04
or
A
o
_l
or
N
o

End-to—End Distance (A)

Figure 4.8. Comparison of the normalized frequency distribution of end-to-end distance
for TEGDA and Py-TEG-Py demonstrating the effect of pyrene end-

labeling.

72

 

 

 

 

 

0.08 - o D Py'TEGPy
o o 0 Py-TEG-Py Wllh Na+
g“ 0.06 -
a
E
'8 0.04 -
£1
'3'
E 0.02 -
Z
0.03 -
I 1 I I I fit I
0 5 10 15 20
FJld-tO-FJld Distance (A)

Figure 4.9. Normalized frequency distribution of end-to-end distance for pyrene end-
labeled pentaethylene glycol in the presence and absence of a templating
sodium cation.

73

Figure 4.10 demonstrates the effect of the pyrene end-labeling on the possible
conformations of the templated tetra(ethylene glycol). The figure contains simulation
results for both Py-TEG-Py and TEGDA in the presence of the templating ion. Figure
4.10 once again illustrates that the templating ion greatly restricts the conformations
adopted by the chain as illustrated by the reduced mean end-to-end distance compared to
the ion-free simulation results shown in Figure 4.8. It is interesting that the Simulation
results suggest that, while the presence of the pyrene end groups has only a slight effect
on the mean value of the end-to-end distance, it has a marked impact on the dispersion.
This suggests that the combined presence of the templating ion and the pyrene end groups

places considerable restrictions on the conformations adopted by the oligomer.

4.4. Conclusions

In this chapter we have demonstrated the use of pyrene excimer fluorescence
experiments to investigate the ion-induced cyclization of relatively short-chain, pyrene
end-labeled oligo(ethylene glycol) molecules. The excimer fluorescence results
demonstrated that even in the absence of any templating cation, a significant excimer
emission was observed, indicating that the locally high concentration of the two pyrene
moieties attached to the same oligomer leads to intramolecular cyclization. The addition
of templating cations such as Ni2+ and Sn2+ leads to further enhancements in the excimer
fluorescence. These results suggest that the templating ion binds with the oligo(ethylene
glycol) chain and facilitates the formation of cyclic conformations which bring the two

chain ends in proximity. For pyrene end-labeled penta(ethylene glycol), the excimer

74

 

 

 

 

 

 

0.08
D O
.. o Py-TEGPy with Na+
a TEGDA with Na+
0.06 -
>:
2
3
8'
a 0.04 -
E
,g
E
g 0.02 -
z
0.00 -
l ' l ' l ' l ' l ' l
0 5 10 15 20 25
End-to-End Distance (A)

Figure 4.10. Comparison of normalized frequency distribution of end-to-end distance
Obtained from molecular dynamics simulations for Py-TEG-Py and
TEGDA, both with a templating Na+ cation.

75

fluorescence enhancement induced by nickel ions was more pronounced in chloroform
than in tetrahydrofuran. Therefore the pyrene excimer-to-monomer fluorescence ratio
was found to provide a sensitive measure of the ability of the oligo(ethylene glycol)
mixture to solubilize the ions into an organic solvent.

Molecular dynamics Simulations of pyrene-labeled oligo(ethylene glycol) ligands
were performed to provide molecular insight into the fluorescence results by elucidating
the effects of the templating ion and/or the pyrene end groups on the conformations
adopted by the oligo(ethylene glycol) chains. Simulation results indicated that even in
the absence of a templating cation, the pyrene end-labeled oligomers exhibit an enhanced
tendency to form cycles compared to the analogous unlabeled ligands. This result is
consistent with the fluorescence results in which significant excimer formation was
observed even in the absence of any cation. Simulation results further indicated that the
addition of a templating cation leads to a further reduction in the mean end-to-end
distance for pyrene end labeled tetra(ethylene glycol) in a manner consistent with the
observed enhanced excimer fluorescence. Finally, for two systems containing the
templating ion, the presence of the pyrene end groups on the ethylene glycol oligomer has
only a slight effect on the mean end-to-end distance, but leads to a marked reduction in
the dispersion of the distribution. Hence, the combined presence of the templating ion
and the pyrene end groups appears to place considerable restrictions on the conformations
adopted by the oligomer.

The studies reported in this contribution provide considerable insight into the ion-
induced cyclization of oligomeric ethylene glycols, and therefore enhance our

.understanding of the synthesis of polymeric pseudocrown ethers by the template ion

76

method. These studies have contributed to the selection of a synthesis solvent and the
template ion-PEGDA oligomer length that may be used for the synthesis of polymeric
pseudocrown ethers. In the next chapter these results will be used to develop a synthetic

scheme for the synthesis of polymeric pseudocrown ethers.

10.

ll.

12.

13.

14.

15.

16.

77

4.5. References

Mathur A. M. and Scranton A. B., Sep. Sci. & T ech., 30 (7-9), 1071 (1994).
Cuniberti C. and Perico A., Eur. Polym. J., 13, 369 (1977).

Winnik M. A., Redpath A. E. C. and Richards D. H., Macromolecules, 13, 535
(1980)

Redpath A. E. C. and Winnik M. A., Ann. N. Y. Acad. Sci., 75, 366 (1981).

Winnik M. A., Redpath A. E. C., Paton K. and Danhelka J., Polymer, 25, 91
(1984)

Winnik M. A. and Sinclair A. M., Can. J. Chem, 63, 1300 (1985).
Slomkowski S. and Winnik M. A., Macromolecules, 19(2), 500 (1986).

Winnik M. A., in Photophysical and Photochemical Tools in Polymer Science:
Conformation, Dynamics and Morphology, M. A. Winnik Ed., NATO ASI Series
C: Mathematical and Physical Sciences 182, pp 15, 293 and 611 (1986).

Martinho J. M. G., Catanheira E. M. S., Reis e Sousa A. T., Saghbini 8., Andre J.
C. and Winnik M. A., Macromolecules, 28, 1167 (1995).

Oyama H. T., Tang W. T. and Frank C. W., Macromolecules, 20, 474 (1987).

Char K., Frank C. W., Gast A. P. and Tang W. T., Macromolecules, 20, 1833
(1987)

Oyama H. T., Tang W. T. and Frank C. W., Macromolecules, 20, 1839 (1987).
Oyama H. T., Hemker D. J. and Frank C. W., Macromolecules, 22, 1255 (1989).
Hemker D. J ., Garza V. and Frank C. W., Macromolecules, 23, 4411 (1990).

Tang W. T., Hadziioannou G., Smith B. A. and Frank C. W., Polymer, 29, 1718
(1988)

Several authors have studied the extraction of alkali metal and alkaline earth metal
cations from aqueous media by macrocyclic and polymer bound macrocyclic
compounds. For overviews see (i) Sahni S. K. and Reedijk J ., Coordination
Chemistry Reviews, 59, 1 (1984) (ii) Christensen J. J ., Eatough D. J. and Izatt R.
M., Chem. Rev., 74 (3), 351(1974). Specifically, the binding of hydrated salts of
nickel and chromium by macrocyclic ligands has been reported by (iii) Prabhakar

17.

18.

78

L. D. and Umarani C., J.M.S.-Pure Appl. Chem, A 32(1), 129 (1995), and (iv)
Blasius E. and Janzen K. P., Pure & Appl. Chem, 54(11), 2115 (1982).

POLYGRAF, v 3.21, Molecular Simulations Inc., 16 New England Executive
Park, Burlington, MA 01803.

The primary reason for not including any solvent in our simulations was the fact
that the solvent-cation interactions would be minimal since the cation is itself
insoluble in the solvents used (chloroform and THF). Also, since in our
laboratory we have solubilized salts in bulk oligoethylene glycols, the modeling
may be interpreted to reflect those observations.

CHAPTER 5

SYNTHESIS AND ION-BINDING PROPERTIES OF POLYMERIC
PSEUDOCROWN ETHERS

5.1. Introduction

In the previous chapters, the cation binding properties of oligomeric ethylene
glycol diacrylates has been studied, both experimentally and by molecular modeling. It
has been demonstrated1 that oligomeric ethylene glycol diacrylates solubilize metal salts,
such as nickel chloride and chromium chloride, in organic solvents such as chloroform.
Ion-solubilization studies along with NMR Spectroscopy provided evidence for the cation
binding properties of these oligomeric polyethers. Further, molecular modeling studies
have corroborated these experimental findings by demonstrating that there is a decrease
in oligomeric ethylene glycol diacrylate mean end-to-end distance in the presence of a
templating cation. In order to investigate the template effect further, pyrene end-labeled
tetra— and penta-ethylene glycol were synthesized and cation induced templatization was
observed2 by studying excimer formation in the presence and absence of cations.
Molecular modeling studies demonstrated that there is a decrease in the mean end-to-end
distance in the presence of cations such as Na” and Al”. Studies with tetra- and penta-
ethylene glycol diacrylate suggested that the templating efficiency is maximum with
cations such as nickel and chromium as evidenced by a large increase in pyrene excimer

fluorescence. These studies form the basis of a template ion scheme for the synthesis of

79

80

polymeric pseudocrown ethers. In this method, the templatization ability of short chain
oligomeric ethylene glycol diacrylates in the presence of a metal cation, is exploited to
synthesize polymeric pseudocrown ethers via a single step polymerization. As described
earlier,l if a free-radical polymerization is performed with a hydrophilic comonomer such
as hydroxy ethyl methacrylate, the probability of cyclization is greatly enhanced due to
ion-induced templatization. Moreover, if the polymerization is carried out in an organic
solvent in which the template effect is maximized (a solvent in which the metal salt is
itself insoluble) and under dilute conditions, the probability of cyclization is maximized.
Based upon these studies a model cation-oligo(ethylene glycol)-diacrylate system can
now be studied for investigating the cation binding properties of polymeric pseudocrown
ethers synthesized by this scheme.

In this chapter, the optimization of this synthesis scheme will be discussed. Tetra-
ethylene glycol diacrylate with nickel as the templating cation will be the system under
investigation, and polymeric pseudocrown ethers (PCES) synthesized from this model
system will be investigated for their cation binding properties. Moreover, issues such as
PCE regeneration and effect of monomer to solvent ratio on the binding capacity, will be
addressed. The ion-binding properties of “control” hydrogels synthesized in the absence
of a templating cation have been compared to the binding capacity of the PCES.

A schematic for the PCE synthesis by the free-radical polymerization of hydroxy
ethyl methacrylate with templated oligo(ethylene glycol)-diacrylate is shown in Figure
5.1. Synthesis of the PCE hydrogels was achieved by performing a photopolymerization
with chloroform as the solvent. The choice of solvent was based upon previous studies

which demonstrated that salts such as nickel chloride were solubilized by oligo(ethylene

81

a; 3‘,“
é‘w'im‘i ,-,., ,‘F‘ 4".
3 Solvent . “N ,
“- ”-t V:- r..' ' ., 1.8,,- ' ‘ i~ "57.". ' that .
”a I 't' t 5‘29‘”:QED55‘£E&3$' “v war-“a
”' 'a 0' 4? 0:2;
.1. . -., . " ___> 35.5», :1;
$.‘1‘2fir‘fifi a ‘ .a-p
. 53% .v- 9', -
ogfi’g’fii nght

 

 

CH3 \

CH2=C
es
1;“ 0

CH2
8:4“,

I
C H—CH—o ’ €le
0” fC 2 2 2ft

\ PEGDA / k HEMA /

 

 

 

 

Figure 5.1.

Schematic of free-radical polymerization scheme for the synthesis of
polymeric pseudocrown ethers.

82

glycol)-diacrylate. Further tetra- and penta-ethylene glycol were most efficient in
templating nickel and chromium as determined from the decrease in end-to-end distance
observed in both experimental and modeling results. For the purposes of this model
system under study, poly(ethylene glycol)-diacrylate (PEGDA 200) was chosen as the
templating oligomer and nickel chloride or nitrate as the salt. (The PEGDA 200 contains
on an average 4.5 ethylene glycol (EG) repeat units since the molecular weight of one EG
repeat unit is 44 and the number 200 represents the average molecular weight of the EG
segment in the PEGDA oligomer). The comonomer chosen was hydroxyethyl
methacrylate (HEMA) due to its hydrophilic nature. Both the HEMA and the PEGDA
undergo free-radical polymerizations. The free-radical polymerization was initiated by
light instead of thermal means because of the low boiling point of chloroform and the
unavailability of efficient thermal initiators with a low half life at room temperature. For
purposes of comparison and to demonstrate the enhancement in cation-binding,
crosslinked homopolymers of HEMA containing the same amount of PEGDA (in the
absence of any salt) were synthesized. Their ion-binding properties have been compared
to those exhibited by the PCES. Further, various regeneration schemes have been studied
for the removal of the template cation after synthesis and for recovery of bound cations

during ion-binding studies.

83

5.2. Experimental Methods and Techniques

5. 2. I . Synthesis of Polymeric Pseudocrown Ethers

Polymeric pseudocrown ethers were synthesized based upon a template ion
scheme as described earlier in Chapter 3 (Figure 3.1). All chemicals and materials were
used as received. All organic solvents were HPLC grade and water used for making salt
solutions was also HPLC grade.

First, salt containing the template cation was added to PEGDA 200 (Polysciences,
Warrington, PA) with constant stirring until saturation was reached and no more salt
could be added to the PEGDA without precipitation. A saturated PEGDA-nickel solution
was formed upon adding upto 0.7 g of Ni(NO3)2.6HZO to 5 g of PEGDA 200. This
mixture was calculated to contain 0.4814 moles of salt per gram of PEGDA 200 or
0.09715 moles of Ni2+ per gram of oligomer.

Chloroform was the solvent in which the synthesis was performed and the ratio of
monomers to solvent was varied from 20/80 to 50/50 (by weight). In the case of the PCE
hydrogels the PEGDA ZOO-nickel solution was used while for the poly(HEMA)
hydrogels the pure PEGDA 200 was used in the synthesis. The PEGDA 200 was selected
to be 5 wt.% of the monomers. This was based upon several studies in which the amount
of PEGDA 200 was varied from 1% of monomers (by weight) to 10% of monomers
which showed that 5% of monomers was the optimum amount. Values lower than 5%
resulted in minimal ion-binding while an amount greater than 5% may result in higher
crosslinking density and hence a decrease in the ion-binding capacity. The following

procedure was optimized for the synthesis of the PCE hydrogels and the “control”

84

crosslinked poly(HEMA) hydrogels. In a 4 02. square glass bottle, 0.2148 g of PEGDA
200 was added to 3.8472 g of HEMA. Next, 0.0419 g of Irgacure 651 (Ciba-Giegy) was
added to the monomer mixture as the free-radical photoinitiator. The photoinitiator was
present at 1 wt.% of the monomers. For a dilute synthesis (20/80 monomer/solvent ratio),
16.0388 g of chloroform was added to the reaction mixture. The solution was mixed well
for about 5 minutes in the sealed glass bottle using a vortex mixer. A Black-Ray UV
lamp (200 W) was used to initiate the photopolymerization by exposing one face of the
horizontally placed glass bottle to the lamp. The reaction mixture was exposed to light
for 30 minutes in which time the clear reaction mixture turns into an opaque whitish solid
mass.

In order to remove unreacted monomer, initiator molecules, and residual solvent
trapped within the gel, 40 ml of methanol was added to the polymer in the glass bottle.
This would facilitate the swelling of the gel and the removal of unreacted monomer,
initiator, and chloroform from the gel into the bulk solvent phase. The methanol solvent
was replaced with fresh methanol every day for 3 days. To facilitate the removal of
methanol from within the gel, the gel was swollen in deionized water next. The water
was replaced everyday for 4 days in order to extract all the methanol and any remaining
monomer. The gel was then removed from the bottle and placed in an oven at 80°C for
drying. The drying process ensured complete removal of any trapped solvent in the gel.
After drying the gel for about 4-5 days the polymer was then ground in a mortar and
pestle, and for further particle size reduction, in a coffee grinder. The polymer was now
weighed and the yield recorded. This polymer was used for further regeneration and ion-

binding studies.

85

5.2.2. Removal of Template Cation And Conditioning Of Resin

Before using the polymer for ion-binding studies, complete removal of the
template cation was essential. The ground polymer particles were soaked in 0.1M HCI
solution for 2 days after which the solution was replaced with fresh acid solution. The
polymer was then filtered over a vacuum filter and stored in deionized water for further

1186.

5. 2. 3. Ion-Binding Studies

An experimental setup that was used for ion-binding studies is Shown in Figure
5.2. A 25 cm tall jacketed glass column with an internal diameter of about 10 mm was
used for the ion-binding studies. The ion-binding experiments were carried out at 25°C
by circulating water from a controlled temperature water bath. The polymer particles
swollen in water were filtered and loaded into the column. The column was gently
compacted so as to decrease the void volume and channeling. To condition the column
before ion-binding studies were performed, 50 ml of 1M HCI was pumped through the
column at a rate of 2 ml/min followed by 50 ml of deionized water. The metering pump
was capable of delivering flow rates from 1 ml/min to 9 ml/min. For ion-binding studies,
atomic absorption spectroscopy was employed for the determination of ion
concentrations. A Perkin Elmer AA-20 atomic absorption spectrometer was used for this
purpose. For the determination of concentrations of various aqueous nickel solutions, the
352.7 nm wavelength from a nickel lamp (Varian Scientific) with a slit width of 0.5 was
used. A 100 ppm (based upon Ni”) stock solution was prepared in HPLC grade water

using nickelous nitrate hexahydrate as the stock solution. Various dilutions of this

Jacketed
Column

 

 

 

Temperature controlled water bath

86

 

       
  

 

 

 

 

 

Atomic Absorption Spectrophotometer

 

 

 

 

 

 

 

 

Metering pump

Figure 5.2. Experimental setup for ion-binding and regeneration studies.

87

solution were used as standards for generating calibration curves for the AA
spectrometer.

Ion-binding studies were performed by pumping 100 ml of the 100 ppm Ni”
stock solution through the column at a rate of 2 ml/min. The concentration of the
solution after passing through the column was determined for each 5 ml of solution
collected and a breakthrough curve was generated for each run. A similar experiment
was performed in which case the concentration of the total volume of solution collected at
the column outlet was determined. The difference between the inlet concentration and
outlet concentration was used to calculate the amount of Ni” bound by the column. This
value was represented as mg of Ni” bound per gram of polymer. However, the ion-
binding capacity was calculated from the concentration of the eluted solution as described

below.

5. 2. 4. Column Regeneration and Determination of Concentration of Eluted Solution
Column regeneration was achieved by eluting HCl solutions with concentrations
varying from 0.1M to IM at temperatures ranging from 40°C to 60°C. It was observed
that complete elution of the bound cations is achieved by a minimal volume of an HCI
solution at a concentration of 1M and at 60°C. The water bath temperature was set to
60°C and once the temperature equilibrated in the bath, a graduated cylinder with 20 ml
of 1M HCl was placed in the bath to raise the temperature of the eluting acid solution to
60°C. The elution was then performed by pumping the acid solution through the column

at 2 ml/min. The eluent was collected at the column exit and its concentration was

88

determined by AA spectroscopy. From this calculation an accurate determination of the

ion-binding capacity of the polymeric resin could be made.

5.3. Results and Discussion

5.3.1. Polymeric Pseudocrown Ether Synthesis

The synthesis scheme described earlier and pictorially represented in Figure 5.1
was optimized through a series of syntheses. First, based upon previous studies,2 it was
demonstrated that Ni” and Cr3+ can be optimally bound by tetra- and penta—ethylene
glycol thereby bringing the end-groups in close proximity. Since the synthesis scheme
relies on efficient templatization, it was decided to select nickel and chromium as the
cations for ion-induced templatization, and oligomeric poly(ethylene glycol)-diacrylate
with 4-5 ethylene glycol repeat units as the templating oligomers. PEGDA 200, with an
average ethylene glycol molecular weight of 200 daltons representing an average 4.5
number of ethylene glycol repeat units, was readily available from Polysciences,
Warrington, PA. Further, in order to develop a methodology for the study of the
synthesis and ion-binding properties of polymeric pseudocrown ethers by the ion-induced
templatization scheme, nickel was chosen as the candidate cation.

Nickel is one of the many heavy metal ions found in many waste streams in
inorganic form. Several methods for the binding of nickel from aqueous waste streams
have been employed. Precipitation through chemicals such as soda ash,3 use of chelating
agents such as EDTA,4 and separation by ion-exchange resins5 are examples of some

techniques used conventionally. More recently, liquid surfactant and emulsion liquid

89

membranes employing poly[poly(ethylene glycol) phosphate] as a macromolecular,
multisite carrier of nickel cations has been investigated.6 This polymer, dissolved in 1,2-
dichloroethane, has been shown to be an efficient carrier of Ni” ions in bulk liquid
membranes with a pH gradient as the driving force. From the above discussion it is
apparent that nickel is an important heavy metal cation for which an efficient separation
technology would be desirable.

The synthesis of the pseudocrown ethers was achieved after several trial runs to
determine conditions that would maximize PCE yield and ion-binding capacity. One of
the variables investigated was the amount of PEGDA 200 to be incorporated into the
PCE. As discussed before, an increase in the templated PEGDA concentration may
increase the final ion-binding capacity but will also result in an increase in the fraction of
PEGDA lost towards crosslinking. Our previous modeling studies2 have indicated that
the probability of cyclization even in the absence of a templating cation is increased as
the length of the oligomeric ethylene glycol chain decreases. This may be attributed to
the short chain length of the oligomer and the limited number of equilibrium
conformations that may be adopted resulting in a low entropy. Based upon the
optimization studies a PEGDA concentration of 5 wt.% of the monomers was chosen for

this study.

5. 3. 2. Ion-binding Studies
The results of the ion-binding properties exhibited by the PCE synthesized with
nickel as a templating cation are shown in Figure 5.3. The data presented is an average of

two experimental runs for each polymer. This figure illustrates the ion-binding capacity

90

0.35

 

a PCE
CI P(HEMA)

    
      
     
   
     
     
   

0.3 _-

   

l
Ii

.lil‘

iim‘
hi

i
I .

    
       
       
     
     
     
   

  

     
       
       
       
 

  

C 1542?; T T ‘ ”T
a, E _—L_

g 0.25 ._.. ;:—=:_—::: E% L L

:5” : _‘~—:—_——._——

5 0 2 :=—_—:

2 - * -‘ * ELL ' E E
or E: :—

V ————=:-—'L_—__LL"" :————*‘—“—_.__

b 0.15 :%—f '—'——— —_—‘ — —

ID :~__—_—.=—:T: EE-z _____

g, ; E

U) 0'1 Li 7? ‘_"_" T ........... ’
.5 ~—- _ ' F __ _ __

'o E: —:___L_

5 TE ‘

m 0'05 T :53? — -- _:_~ — — L

  

i

   
   

.m—F
...._..__.__._
...___.__—
_____.—.

 
 

l

l

ll
lllllllllll
.,.

 

l

 

 

 

 

 

1‘
l

r
l

 

 

20/80 35/65 50/50
Monomer/solvent ratio during synthesis

 

 

 

Figure 5.3. Experimental ion-binding capacities for polymeric pseudocrown ethers
(PCES) and “control” p(HEMA) hydrogels as a function of the monomer to
solvent ratio during synthesis.

91

of the PCB as a function of the monomer to solvent ratio during synthesis. Also shown in
the figure is a comparison between the ion-binding ability of the PCB and the p(HEMA)
hydrogel synthesized with the same amount of PEGDA 200 but in the absence of any
template cation.

There are two primary observations to be made from this data. First, as the ratio
of monomer to solvent increases the average ion-binding capacity decreases. A PCE gel
synthesized in a 20/80 monomer to solvent ratio exhibited an ion-binding capacity of
about 0.30 mg/g polymer while that synthesized in monomer to solvent ratio of 50/50 had
an average ion-binding capacity of 0.17 mg/g polymer. This is consistent with our
hypothesis that the probability of cyclization would be enhanced under dilute solution
polymerization conditions since interactions between two templated oligomers would be
minimal thereby preserving the cyclic conformation. The data also suggests that a
plateau may be reached on the maximum attainable ion-binding capacity as the monomer
to solvent ratio is decreased. Also, experiments demonstrated that if the monomer to
solvent concentration is decreased to 10/90, the mechanical integrity of the polymeric
material may also be compromised since a highly swellable hydrogel is produced. Since
strength is of importance when considering loading and repeated ion-binding operations
both in the lab and potentially in industry, a lower ratio than 20/80 was not investigated
for ion-binding properties.

Since the synthesis of the first crown ether almost three decades ago, most cation
binding studies reported in the literature have dealt with the alkali or alkaline earth
metals. Fewer studies have focused on transition metal binding from aqueous streams.

However it is mostly transition metals that present a challenge in the separations industry

92

due to their toxicity. While conventional ion-exchange resins have ion-binding capacities
of upto a few hundred milligrams per gram of resin, most polymeric crown ethers and
other polymer bound cyclic ligands have binding capacities at least an order of magnitude
lower. For example, Kahana et 01.7 have reported the ion-binding capacities of various
polymeric pseudocrown ethers ranging from as low as 0.035 mg/g to 0.245 mg/ g polymer
for Li” and from 1.32 mg/g to 13.2 mg/g for Csi. By comparison, in a study of the ion-
binding capacities of several chelating ion-exchange polymers for Ni”, capacities ranging
from 0.23 mg/g for picolinic type resins8 to 31.7 mg/g for ethyleneimine type resins,9
have been reported. The PCEs synthesized and investigated in this study exhibit ion-
binding capacities comparable to the polymeric pseudocrown ethers mentioned above.

The second important result of these studies was the observed enhancement in the
ion-binding properties demonstrated by the PCB as compared to the p(HEMA) hydrogel.
The PCE gel synthesized at ratio of 20/80 exhibited an increase in ion-binding capacity
by a factor of 7.5 while that synthesized under a monomer to solvent ratio of 50/50
possessed about double the ion-binding capacity of its p(HEMA) counterpart. This
enhancement may be attributed to the ion-induced templatization of PEGDA in the case
of the PCB, which results in an increase in cyclization as compared to crosslinking.
Another trend that is observed 'is the increase in ion-binding capacity of the p(HEMA)
polymers as the monomer to solvent ratio is increased. This may be attributed to the
increased cyclization of the PEGDA as the monomer to solvent ratio is increased possibly
resulting from chain interactions and dynamics.

Templatization increases the fraction of PEGDA 200 oligomer chains that have a

low mean end-to-end distance thereby bringing the unsaturated acrylate end-groups in

93

close proximity. This may facilitate the incorporation of the PEGDA into a growing
polymer chain, resulting in the formation of pendent loops in a crown ether type
conformation. This method not only results in the incorporation of pendent cyclic oligo-
ethers onto a hydrophilic polymeric support by a one-step inexpensive free-radical
polymerization, but also ensures the formation of a crown ether like structure only during
synthesis when such a structure is immobilized onto the polymeric backbone. This
overcomes the health hazards associated with crown ethers and crown ether derivatives
(since they ma readily absorb through the skin). This, along with a high ion-binding
capacity, makes PCEs suitable for use as an easy to synthesize and inexpensive ion-

binding resin.

5. 3. 3. Polymer Regeneration

7,10

It was observed by Kahana and Warshawsky that the ion-complexation
properties of polymeric crown ethers are temperature dependent. They observed that, for
ions bound by crown groups by a salt-coordination mechanism, a temperature increase
from 20°C to 60°C caused enhanced spontaneous elution of the bound salt. These
investigators have also studied the binding of protic and Lewis acids to polymeric

pseudocrown ethers.ll

They observed that complexation of the pseudocrown ether with
acids such as HCl in water starts at HCl concentrations exceeding 1M. Further, other
researchers have also demonstrated that a combination of acidic conditions and high

temperature'z"3 is most efficient in eluting cations bound by polymeric crown ethers. Our

regeneration scheme was developed based upon all these considerations. The first

94

occasion when regeneration was employed was after synthesis, for the removal of the
template cation. Measurement of the concentration of the eluted solution revealed that
upto 95% of regeneration of the PCB could be achieved. Further, during ion-binding
studies, regeneration was performed in order to reuse the column and to determine the ion
uptake. Values obtained for the ion-binding capacity based upon the difference between
the initial and final concentrations of the solution passed through the bed were in good
agreement with those calculated from the eluent concentration after regeneration. In most
cases, upto 100% of the bound nickel was eluted from the column upon regeneration.
This thermo-regulated decomplexation ability of the PCB makes it attractive for repeated

loading and re-use.

5.4. Conclusions

In this chapter the synthesis and ion-binding properties of polymeric pseudocrown
ethers has been presented. Specifically, the synthesis of a pseudocrown ether hydrogel
with hydroxy ethyl methacrylate as the comonomer and nickel as the template cation has
been discussed. Based upon earlier studies, poly(ethylene glycol)-diacrylate 200, with an
average of 4.5 ethylene glycol repeat units, was chosen as the templating oligomer.
Hydrogels with hydroxy ethyl methacrylate as the comonomer and PEGDA 200 as the
crosslinking agent were also synthesized, but in the absence of any templating cation.
The conditioning, ion-binding properties and regeneration of the PCEs was discussed.
The monomer to solvent ratio during synthesis were varied from 20/80 to 50/50 and its

effect on the ion-binding capacity was investigated.

95

The PCE synthesized with a monomer to solvent ratio of 20/80 was observed to
possess the highest ion-binding capacity (0.3 mg Nip/g polymer) which is comparable to
capacities of polymer bound crown ethers reported in the literature. The binding capacity
of the PCEs was observed to decrease with an increase in the monomer to solvent ratio
during synthesis. This was attributed to a possible decrease in templatization efficiency
and inter-oligomer interactions resulting in greater crosslinking at the expense of pendent
crown ethers. Also, the effect of the template ion on the ion-binding capacity of the PCE
was demonstrated by comparing the binding capacity of the PCE with that exhibited by a
p(HEMA) hydrogel synthesized similarly but without any templating cation. The PCE
exhibited an enhanced ion-binding capacity for nickel. Finally, an efficient regeneration
scheme was optimized which resulted in complete regeneration of the PCB.

These experiments have demonstrated the promise of the template ion induced
cyclization scheme for the synthesis of polymeric pseudocrown ethers. With an easy one-
step synthesis scheme and comparable binding properties for Ni”, this research could
result in the development of an inexpensive resin for the binding of nickel from aqueous
waste streams. Further, this scheme could be extended to synthesize PCEs for the
binding of other toxic heavy metal ions of interest such as chromium, copper, lead, zinc

among others.

10.

11.

12.

13.

96

5.5. References

Mathur A. M. and Scranton A. B., Sep. Sci. & T ech., 30 (7-9), 1071 (1994).

Mathur A. M. and Scranton A. B., “Synthesis and Ion-Binding Properties of
Polymeric Pseudocrown Ethers II: Ion-Induced Templatization of Oligomeric
Ethylene Glycol Diacrylates,” Sep. Sci. & Tech., (in press).

Lanouette K.H., Chem. Eng., Oct. 17 (1977).

Lee LA. and Davis H.J., in Trace Metals and Metal-Organic Interactions in
Natural Waters, P. C. Singer Ed., Ann Arbor Science, pp 363 (1974).

Hudson M., in Ion Exchange: Science and Technology, A.E. Rodrigues Ed.,
NATO ASI Series E, 107, pp 463 (1986).

Wodzki R., Wyszynska A. and Narebska A., Sep. Sci. & Tech., 25(11 &12), 1175-
1187(1990)

Kahana N. and Warshawsky A., J. Polym. Sci. .' Polym. Chem. Ed., 23, 231-253
(1985)

Warshawsky A., in [an Exchange: Science and Technology, A.E. Rodrigues Ed.,
NATO ASI Series E, 107, pp 67 (1986).

Warshawsky A., Die Angew. Makromol. Chem, 109/110, 171-196 (1981).
Warshawsky A. and Kahana N., J. Am. Chem. Soc., 104, 2663-2664 (1982).
Warshawsky A. and Borer B., Hydrometallurgy, 4, 83-92 (1979).

Koshima H. and Onishi H., Analyst, 114, 615-617 (1989).

Warshawsky A., Deshe A., Shechtman L., and Kedem O., React. Polym., 12, 261-
268 (1990).

CHAPTER 6

SWELLING PROPERTIES OF POLY(METHACRYLIC ACID-g-
ETHYLENE GLYCOL) HYDROGELS - EFFECT OF SOLVENT

6.1. Introduction

In this chapter, a brief discussion of hydrogels and their potential applications will
be followed by a study of the effect of solvent in the disruption of the complex formed
between poly(methacrylic acid) and poly(ethylene glycol). For this purpose, crosslinked
hydrogels with a poly(methacrylic acid) backbone with poly(ethylene glycol) grafts were
synthesized and their swelling behavior in response to several conditions was
investigated. These studies further our knowledge of reversible complexation and the
ability of a solvent to disrupt the complex.

Hydrogels are macromolecular networks which swell, but do not dissolve, in
water. The ability of hydrogels to absorb water arises from hydrophilic functional groups
attached to the polymeric backbone, while their resistance to dissolution arises from
crosslinks between network chains. Many materials, both naturally occurring and
synthetic, fit the definition of hydrogels. Crosslinked dextrans and collagens are
examples of natural polymers which are modified to produce hydrogels. Classes of
synthetic hydrogels include poly(hydroxyalkyl methacrylates), poly(acrylamide), poly(N-

vinyl pyrrolidone), poly(acrylic acid), and poly(vinyl alcohol).

97

98

Synthetic hydrogel networks are useful for applications which require a material
that has good compatibility with aqueous solvents, yet will not dissolve. Such
applications include biomaterials, controlled release devices, chromatographic packings,
and electrophoresis gels. Many properties of synthetic hydrogels make them suitable for
biomedical applications which require contact with living tissue. The ability to absorb
and retain aqueous media not only gives hydrogels a strong superficial resemblance to
living tissue, but also makes them permeable to small molecules such as oxygen,
nutrients, and metabolites. The soft, rubbery consistency of swollen hydrogels minimizes
fiictional irritation felt by surrounding cells and tissue, while the low interfacial tension
with aqueous solvents reduces protein adsorption and denaturation. Furthermore,
synthetic hydrogel networks can be swollen and washed of undesirable byproducts,
residual initiators and monomers, and may be fabricated in a variety of shapes and
geometries.

Wichterle and Liml first suggested that a hydrogel based on poly(2-hydroxyethyl
methacrylate) (PHEMA) could be a biocompatible synthetic material. Since this initial
work, hydrogels have been investigated for a wide variety of biomedical applications.
Sofi contact lenses are an example of a biomedical application in which PHEMA
hydrogels have found extensive use. The requirements of a polymer to be used in this
application are quite severe. A suitable material must show optical clarity, strength, and
dimensional stability under a variety of conditions. Furthermore, the material must be
compatible with the cornea and must be permeable to oxygen and carbon dioxide.
PHEMA hydrogels meet these requirements. The same properties have rendered these

hydrogels useful for many other biomedical applications. Hydrogels have been suggested

99

for use as coatings for catheters and sutures, membranes for hemodialysis, burn dressings,
artificial cartilage, heart valves, and vitreous humor, as well as many other applications.

The high solute permeability of hydrogels has led to their use in devices for
controlled release of drugs or other active agents. One mechanism of controlled release
involves first dispersing a water-soluble drug into the hydrogel network, then drying the
device. In the dry polymer matrix, the drug is essentially immobile. When the device is
placed in an aqueous environment, the hydrogel swells, and the drug diffuses out of the
network. The rate of drug release depends on two simultaneous processes, water
migration into the device and drug diffusion out of the swollen network. The net effect is
a gradual release of the drug over the course of a well defined time period, thereby
maintaining close to optimum therapeutic medication levels while avoiding intermittent
and massive dose effects. Furthermore, controlled release from hydrogel microspheres
allows the possibility of targeting the drug to specific sites in the body where the drug is
needed. This targeting may make optimum use of expensive drugs while minimizing
system-wide toxic effects.

Hydrogels exhibiting extremely high swelling capacities may be synthesized by
free radical crosslinking polymerizations of ionogenic monomers such as acrylic and
methacrylic acid (or their sodium salts). For example, poly(acrylic acid) hydrogels may
exhibit a maximum swelling of more than 99% water, and are used in applications which
require “superabsorbing” polymers. These superabsorbing polymers have received
considerable attention in recent years, and have been extensively reviewedz‘4

Polymeric hydrogels that exhibit swelling transitions in response to external

stimuli such as changes in pH, temperature, or composition have been investigated as

100

candidate materials for controlled release, separations, control of enzyme-substrate
reactions, and sensor applications. Poly(methacrylic acid) and poly(ethylene glycol) form
hydrophobic hydrogen bonded complexes when the acid is protonated. This hydrogen-
bonded complex can be reversibly formed by species that may compete favorably for the
formation of hydrogen bonds with each of the constituents. Self-associating networks of
poly(methacrylic acid-g-ethylene glycol) have been investigated by Klier et al.5 and
Scranton et al.6 These researchers studied the synthesis and environmentally sensitive
swelling characteristics of poly(methacrylic acid-g-ethylene glycol) networks. The
complex-forming constituents were covalently linked to one another and swelling was
largely regulated by control of complexation. The oligomeric ethylene glycol grafts were
capable of forming hydrogen bonded complexes with the PMAA backbone. The
complex was formed when the acid was protonated (low pH) and could be broken by
neutralizing the acid (high pH). The dependence of equilibrium network swelling on
external pH, temperature, composition and network structure was examined.5 The
P(MAA-g-EG) hydrogels exhibited a sharp swelling transition in aqueous solutions at an
external pH of 4. These swelling characteristics were attributed to the complexation and
decomplexation induced by changes in the surrounding conditions. Nuclear Overhauser
Effect (N OE) studies were used to demonstrate complexation between dilute solutions of
poly(methacrylic acid) and poly(ethylene glycol) (PEG) at lower PEG molecular weights
than previously reported.6

This work builds upon previous work on the study of self-associating networks of
poly(methacrylic acid-g-ethylene glycol) which exhibit reversible swelling transitions in

response to pH and temperature. The objective of this study is to extend the

101

understanding of the reversible complexation phenomenon in response to solvent type
and composition. Synthesis conditions were also varied in order to study their effect on
the maximum equilibrium degree of swelling. In addition, crosslinked homopolymers of
poly(methacrylic acid) have been synthesized and their swelling behavior in various
solvents has been compared to the graft copolymers. Molecular modeling (molecular

mechanics) techniques were also used to investigate the stoichiometry of the complex.

6.2. Experimental Methods and Techniques

6. 2. 1. Hydrogel Synthesis

Poly(methacrylic acid-g-ethylene glycol) hydrogels, henceforth referred to as
P(MAA-g—EG), were synthesized by copolymerization of methacrylic acid (MAA,
Aldrich) with methoxy poly(ethylene glycol) methacrylate 400 (MPEGMA 400,
Polysciences, Warrington, PA), in the presence of ethylene glycol dimethacrylate
(EGDMA, Polysciences, Warrington, PA). Homopolymers of poly(methacrylic acid),
henceforth known as PMAA, were synthesized by polymerizing MAA in the presence of
EGDMA. All monomers and solvents were high purity and were used as received.

The graft copolymers were synthesized by the free-radical solution
polymerization of MAA and MPEGMA 400 (with about 9 EG repeat units) in the
presence ethylene glycol dimethacrylate (EGDMA) as crosslinking agent (Figure 6.1).
The molar ratio of ethylene glycol repeating unit to methacrylic acid was chosen to be 1:1
for all copolymers synthesized. The crosslinking agent (EGDMA) was 0.2 wt % based

on monomer weight for all syntheses. The reaction was initiated by a mixture (1 :1 by

102

 

(EH3 (EH3 (EH3
CH2:IC CH2=C Csz'C
:0 I _ C =0
3 ‘F‘° cs
. + 0 + A
H Ah CH2
9 2 6H
Methacrylic Acid (MAA) CH2 . 2
l O
9, Y"
n " c = 0
CH3 l
Methoxypoly(ethylene GlycoI)-methacrylate (MPEGMMA) CH2: (,3
11:9 CH3
Ethylene glycol di-methacrylate (EGDMA)
n=1
lnitiators Free Radical
Solvent Polymerization
40 C
V
Backbone PMAA

 
  

0000

{‘9 EGDMA Crosslink

£76)

Complexed MPEGMMA Grafts

Figure 6.1.

Synthesis scheme for the preparation of poly(methacrylic acid-g-ethylene

glycol) self-associating hydrogels.

103

weight) of sodium bisulfite and ammonium persulfate (0.01 wt % of monomer). The
polymerization was carried out in a solution of a 50/50 (w/w) mixture of distilled water
and ethanol (200 proof) to ensure proper miscibility of the initiators with the water
soluble monomers. The ratio of monomers to solvent was varied between 20/80 to 80/20
(w/w) to study the effect of solvent content during synthesis upon the swelling properties
of the resulting hydrogels. The reaction mixture was placed in 6 ml polyethylene vials
and were thoroughly mixed using a vortex mixer for 10 minutes. The vials were sealed
tight and placed vertically in a circulating water bath at 45°C for 24 hr. The resulting
gels were removed from the vials and cut into small cylindrical pieces about 1 cm. long.
Each of the gels was then placed in a container of distilled water for 24 hr. In order to
remove any residual initiator, the gels were swollen in acidic water (pH=4) which was

periodically changed for 3 days.

6. 2. 2. Equilibrium Swelling Studies

Polymer gels were first weighed to record their equilibrium mass after being
swollen in distilled water. Next, the hydrogels were placed in containers with various
solvent-water mixtures ranging from pure water to pure solvent. The solvents studied
included methanol, ethanol, acetone, n-propanol and iso-propanol. The swollen mass of
the hydrogels was determined by periodically removing them from the container and
weighing the drip-free sample until equilibrium had been achieved. Usually it took about
a week to achieve an equilibrium degree of swelling. The hydrogel samples were then

dried in an oven maintained at 65°C for 7-8 days until there was no change in mass

104

recorded. The equilibrium solvent weight fraction was determined by dividing the

difference between the swollen mass and the dried mass by the swollen mass.

6. 2. 3. Molecular Mechanics Studies

To study the stoichiometry of the PMAA-PEG complex, molecular mechanics
simulations were performed using the molecular modeling software POLYGRAF.7 The
polymer builder module was used to build syndiotactic PMAA and PEG segments (each
of 4, 8 and 14 repeating units). The structures were energy minimized and charge
equilibrated to facilitate the assumption of realistic minimum energy conformations with
realistic molecular geometries. The two polymer segments were then placed in close
proximity and energy minimization calculations (molecular mechanics, MM) were
performed until an energy minimum was reached and the calculations converged. In
order to ensure that the minimum attained was not a local minimum, molecular dynamics
(MD) calculations were performed on the system for a period of 1 ps, followed by further
energy minimization calculations. (For a brief description of both MM and MD
techniques, please refer to Chapter 1). Hydrogen bonds between the ether oxygens and
the acid protons were said to be formed by setting a cutoff distance of 3.5 A and

highlighting the oxygen-proton pairs that satisfied the criteria.

105

6.3. Results and Discussion

6. 3. 1. Hydrogel Swelling Studies

The equilibrium swelling profiles for P(MAA-g-EG) gels synthesized in a 50/50
monomer/solvent ratio are shown in Figure 6.2. The three curves represent data for
swelling in aqueous mixtures of solvent, from pure water to pure solvent. A homologous
series of alcohols was chosen for this study and the data shown in Figure 6.2 illustrates
the equilibrium swelling behavior of the P(MAA-g-EG) hydrogels in methanol, ethanol
and n-Propanol. The data are represented by the symbols, and lines drawn through the
data points represent trends. As shown in Figure 6.2, an equilibrium solvent weight
fraction just over 0.3 is observed in the limit of infinite dilution (pure water or 0 weight
fraction). In addition, the equilibrium degree of swelling increases as the weight fraction
of solvent in the water-solvent mixture increases. This trend is consistent with the idea
that solvents with Lewis base character may compete with PEG for hydrogen bonding
sites on the backbone PMAA, resulting in an increased degree of swelling due to
decomplexation of the covalently bonded grafts with the acid backbone. The
effectiveness of a solvent in breaking the complexes is probably related to its ability to
disrupt the hydrophobic interactions as well as compete for hydrogen bonding sites.
Also, other than in methanol, in which an equilibrium solvent weight fraction of about 0.8
is observed in pure methanol, both ethanol and n-propanol curves go through a maximum
and decrease to a lower degree of swelling as the amount of solvent in the aqueous

mixtures increases.

106

 

1.0
0.9.3
0.8-I A -. .. -_..-._._..........,
0.71
0.6:
0.5;

0.3 -

 

Eq. Solvent Wt. Fraction

0.2 a
~——l-——- Methanol

-—O—-— Ethanol

0.1 -
—V— n-Propanol

 

 

 

0.0 l l T l I T l T
0.0 0.2 0.4 0.6 0.8 1.0

Wt. Fraction Solvent in Water

Figure 6.2. Effect of solvent on the equilibrium degree of swelling for a PMAA-g-PEG
gel synthesized in 50% solvent (50 % monomers).

107

The swelling results at low concentrations of solvent demonstrate that n-propanol
is most effective in breaking the complex (as indicated by the slope of the trendline at
lower concentrations). However, gels swollen in ethanol or methanol achieve a higher
equilibrium degree of swelling (between 0.3 and 0.7 weight fraction solvent in water) as
compared to n-propanol. This trend may be explained by the ability of n-propanol to
disrupt the complex at lower concentrations by better stabilizing the hydrophobic
interactions due to the propyl group. Swelling transitions of P(MAA-g-EG) gels in
alcohol solvent mixtures may be attributed to two competing phenomena: i) the ability of
the solvent to disrupt hydrogen bonding and stabilize hydrophobic interactions, and ii) the
dissociation/ionization of the polymeric acid in the solvent. As shown in Figure 6.2,
swelling in the homologous series of alcohols with increasing hydrophobic character
results in a decrease in the equilibrium degree of swelling at high solvent concentrations.
This downward trend may be explained by the re-association of the PMAA in
environments that are more non-polar and inhibit ionization. The pK, of PMAA may be
affected by the ease with which the acid proton can exchange with the solvent molecules.
For example, the pK, of PMAA in water varies between 4 and 4.8 and allows for rapid
exchange between the acid proton and the dissociated water protons. At higher solvent
concentrations PMAA ionization may be the controlling factor which affects
complexation and thus swelling behavior. In environments such as high concentrations
of n-propanol or ethanol, the complexes may be re-formed due to the re-association of the
acid, resulting in a lower equilibrium degree of swelling.

Figure 6.3 illustrates the effect of polymerization conditions on the maximum

equilibrium degree of swelling achieved by a P(MAA-g-EG) hydrogel. In this figure,

108

1.0

 

0.81
0.7-
0.6:
0.53
0.4 -

0.3-

Eq. Solvent Wt. Fraction
b

0.2_ —‘-— 20% Monomer
. —O— 50% Monomer
0.1 - + 80% Monomer

 

 

 

0.0 I I I I I T I I
0.0 0.2 0.4 0.6 0.8 1.0

Wt. Fraction Ethanol in Water

Figure 6.3. Equilibrium swelling profiles showing the effect of polymerization
conditions for swelling in ethanol.

109

data are shown for three different (20/80, 50/50 and 80/20) monomer to solvent ratios
during synthesis. The swelling profiles are for swelling in various concentrations of
aqueous ethanol mixtures. For all three cases, the equilibrium degree of swelling
increases as the concentration of ethanol in the ethanol-water mixture increases and
reaches a maximum at around 0.6 wt. fraction ethanol in water. As shown in Figure 6.3,
gels synthesized at a lower monomer to solvent ratio (20% monomer) are capable of
achieving an equilibrium solvent fraction of 0.9 at intermediate solvent concentrations,
while those synthesized at a high monomer to solvent ratio (80% monomer) reach an
equilibrium solvent weight fraction of only about 0.7 at similar solvent concentrations.
These trends demonstrate that the gels synthesized in more solvent exhibit a
correspondingly higher equilibrium degree of swelling as compared to gels synthesized in
a higher monomer to solvent ratio. This may be explained by considering the
conformations of the oligomers and growing polymer chains during polymerization. In a
dilute system, the monomers and crosslinking agents may be relatively far apart resulting
in a decrease in the crosslinking density (and entanglement) and an increase in the sol
fraction. However, at higher monomer to solvent ratios, close proximity of growing
polymer chains to crosslinking agents may result in a higher degree of crosslinking (and
entanglement) and incorporation of more monomer units into the hydrogel network,
thereby decreasing the sol fraction.

To demonstrate the effect of complexation on the equilibrium degree of swelling
of a P(MAA-g-EG) hydrogel, comparisons were made with the swelling characteristics of
a crosslinked PMAA gel synthesized under the same conditions. Figure 6.4 shows the

swelling behavior of a PMAA hydrogel compared to a P(MAA-g-EG) hydrogel

110

 

 

 

 

 

c
.9.
*6
9
LL
E
E
(D
2
O
(I)
IB- 4
0.2: —l-— PMAA
Q1 _ “~0— 50/50 PMAA-g-PEG
0.0 I I I I I I I I I
0.0 0.2 0.4 0.6 0.8 1.0

Wt. Fraction Methanol in Water

Figure 6.4. Equilibrium swelling profiles of crosslinked PEG-g-PMAA (SO/50) and
PMAA homopolymer (SO/50) in methanol.

lll

(synthesized in a 50/50 monomer solvent ratio). The swelling profiles shown are for
swelling in increasing concentrations of aqueous methanol solutions. The swelling
characteristics of the P(MAA-g-EG) gel follow a sigmoidal trend with a low equilibrium
solvent uptake at infinite dilution (0 weight fraction methanol) and ~0.8 equilibrium
solvent weight fraction at higher concentrations of methanol. On the other hand the
swelling characteristics of PMAA show little variation with increasing methanol
concentration. The slight increase (from a value of 0.7 solvent weight fraction) followed
by a nominal decrease to an equilibrium solvent weight fraction of 0.75 may be explained
by the varying degree of dissociation of the polyacid in different concentrations of
aqueous methanol solutions. The effect of complexation can be easily demonstrated by
the difference in the equilibrium degree of swelling between the PMAA and the P(MAA-
g-EG) gels at low methanol concentrations. This difference can be attributed to the
compact configuration (low degree of swelling) adopted by the complexed hydrogel at
low methanol concentrations. However, as the concentration of methanol increases it is
able to compete effectively for hydrogen bonding sites on the backbone PMAA and
disrupt the hydrogen bonding between the grafted PEG segments and the PMAA. In the
absence of any grafts the equilibrium degree of swelling of the PMAA hydrogel is a
function of the degree of dissociation in a particular aqueous methanol solution, and does
not exhibit the lower degree of swelling demonstrated by the P(MAA-g-EG) hydrogel.
Figure 6.5 illustrates the difference in swelling behavior in acetone between a
PMAA gel and a P(MAA-g-EG) gel synthesized in a 50/50 monomer solvent ratio. The
two curves once again demonstrate the effect of complexation on the equilibrium

swelling behavior of these hydrogels. However, since acetone is a comparatively more

112

 

1.0

 

 

 

 

I:
.9
‘6
‘3
LL
3
E
(D
.2
O
(0 .
U" 0.24
LU 4 —l_ PMAA
0.1 - -O- PMAA-g-PEG
0.0 I I I I I I I 1 I
0.0 0.2 0.4 0.6 0.8 1.0

Wt. Fraction Acetone in Water

Figure 6.5. Equilibrium Swelling profiles of crosslinked PMAA-g—PEG (50/50) and
PMAA (50/50) in acetone.

113

hydrophobic solvent (with two methyl groups) compared to methanol, at high acetone
concentrations the pKa of the PMAA may decrease and the acid may undergo re—
association resulting in a more compact configuration evidenced by a lower equilibrium
solvent weight fraction. Figure 6.5 brings out the two effects of complexation and
dissociation quite clearly as seen by, i) the difference in equilibrium swelling at low
concentrations between the two hydrogels and ii) a lowering in the equilibrium degree of
swelling at high acetone concentrations shown by both the PMAA and the P(MAA-g—EG)

hydrogels.

6. 3. 2. Molecular Modeling of PMAA-PEG Complexation

It has been demonstrated by numerous investigators (see reference 5 and
references therein) that PMAA and PEG form a 1:1 complex based on a molar ratio of
methacrylic acid and ethylene glycol repeat units. Klier et al.5 demonstrated that
P(MAA-g-EG) hydrogels synthesized with a 1:1 molar ratio exhibited the least
equilibrium degree of swelling at a pH of 4 (complex promoting conditions). It has also
been proposed that deviations from this stoichiometry may lead to free polymer chains or
increased hydration of the complex (due to the presence of uncomplexed hydrophilic
units within a complexed chain). Experimental work by various researchers provides
evidence of the formation of the complex in a 1:1 stoichiometry. However, these
observations leave significant questions about the fundamental nature of the complexes,
and do not conclusively establish that the complexes form in a strict 1:1 stoichiometry

with each acidic moiety bound with each successive ether linkage.

114

In this study, we have attempted to demonstrate the stoichiometry of the PMAA-
PEG complex (under complex promoting conditions when the acid is protonated) by
performing molecular simulations on segments of PMAA and PEG of varying lengths.
Syndiotactic PMAA was built using the POLYGRAF polymer builder, and segments
consisting of 4, 8 and 14 repeat units were studied for complexation with segments of
PEG with exactly the same number of repeat units.

Molecular mechanics calculations were first performed on a tetrad of PMAA in
close proximity to a tetrad of PEG. The objective of these calculations was to
demonstrate that sterically it is possible to form stable 1:1 complexes between four
repeating units of PMAA and PEG. Molecular mechanics calculations demonstrated the
formation of such as complex by satisfying the hydrogen bonding criteria. The ratio of
energy contribution due to hydrogen bonding to the total energy of the minimized
structure was higher, compared to the starting conformation. Figure 6.6 shows the
PMAA tetrad complexed with a PEG segment with 4 EG repeat units thereby validating
the 1:1 complex stoichiometry. This conformation was the lowest energy conformation
hence most probable and favorable. In addition, the same complexed state was attained
after perturbing the system by molecular dynamics calculations at 25°C followed by
energy minimization. A molecular dynamics run ensured that the conformation observed
earlier was not at a local minimum and was indeed the favored conformation. It can also
be observed that, though the PEG repeat unit contains three atoms (-C-C-O) in the
backbone and the PMAA repeat unit contains two atoms (-C-C-) in the backbone, it is

possible to form a 1:1 complex. To further extend this study to longer segments,

115

(

    
   

     

-bonds

it \

I.

, II
PEI}

Figure 6.6. A molecular representation of a PMAA segment with four repeating units
complexed with a PEG segment of the same length showing 1:1 hydrogen
bonding between the acid protons and the ether oxygens.

116

calculations with 8 and 14 repeating units yielded similar results. Figure 6.7 represents
an equilibrium conformation of a segment consisting of 14 repeat units of PMAA
complexed with a PEG segment with the same number of repeating units. The curved
shape of the complex points to the possibility of a helical structure when the number of
repeat units are higher. This would be consistent with the fact that the PEG chain may
have to wrap around the PMAA chain due to an extra atom present in its repeat unit, in

order to form a 1:1 complex.

6.4. Conclusions

The swelling characteristics of graft copolymers of poly(methacrylic acid) with
poly(ethylene glycol) grafts were investigated. The swelling studies were performed in
increasing concentrations of water-solvent mixtures and the equilibrium solvent weight
fraction was recorded. Swelling behavior in a homologous series of alcohols (methanol,
ethanol and n-propanol) demonstrated the effectiveness of the solvent in disrupting the
complex and stabilizing the hydrophobic interactions. The ability of a solvent to break
the complex (at low solvent concentrations) may be related to the ease with which it
could compete in the formation of hydrogen bonds with the PMAA backbone. However
at high solvent concentrations the degree of dissociation of the PMAA could possibly
govern the equilibrium degree of swelling. For alcohols with longer alkyl groups (n-
propanol) the degree of swelling decreased gradually at higher solvent concentrations.
The ratio of monomer to solvent during synthesis affected the maximum equilibrium

degree of swelling attained by a hydrogel. Hydrogels synthesized in a 20/80 monomer

117

 

Figure 6.7. A POLYGRAF molecular simulation depicting a 1:1 hydrogen bonding
between a PMAA and PEG segment both with 14 repeating units.

118

solvent ratio achieved a higher degree of swelling which may be attributed to a lower
degree of crosslinking and entanglement.

Comparisons between the swelling behavior of a crosslinked homopolymer of
PMAA with a P(MAA-g-EG) gel in methanol and acetone demonstrated the effect of
complexation as evidenced by the difference in the equilibrium degree of swelling at low
solvent concentrations. A lowering in the degree of swelling at high concentrations of
acetone or n-propanol was attributed to a lower degree of dissociation of the polyacid
(PMAA).

Molecular simulations strengthened the evidence of the existence of a 1:1
complex stoichiometry, by demonstrating 1:1 complexation at the molecular level.
Molecular mechanics simulations on PMAA and PEG segments with 4, 8 and 14 repeat

units exhibited the formation of stable conformations with a 1:1 complex stoichiometry.

119

6.5. References

Wichterle o. and Lim 1)., Nature, 185, 117 (1960).

Gross J. R., in Absorbent Polymer Technology, L. Brannon-Peppas and R. S.
Harland Eds., Elsevier Science Publishing Company Inc., New York, pp 3-22
(1990)

Buchholz F. L., in Absorbent Polymer Technology, L. Brannon-Peppas and R. S.
Harland Eds., Elsevier Science Publishing Company Inc., New York, NY, pp 23-
44 (1990).

Buchholz F. L., Trends in Polymer Science, 2(8), 277-281 (1994).

Klier J., Scranton A. B., and Peppas N. A., Macromolecules, 23, 4944-4949
(1990)

Scranton A. B., Klier J., and Aronson C. L., in Polyelectrolyte Gels, R. S.
Harland, R. K. Prud’homme Eds., ACS Symposium Series 480, Washington DC,
pp 170-189 (1992).

POLYGRAF, v 3.21, Molecular Simulations Inc., 16 New England Executive
park, Burlington, MA 01803.

CHAPTER 7

SYNTHESIS AND EMULSIFICATION PROPERTIES OF
REVERSIBLE BLOCK/CRAFT POLYMERIC EMULSIFIERS
BASED UPON POLYMER COMPLEXATION

7.1. Introduction

In this chapter, the synthesis and emulsification properties of block/graft
polymeric emulsifiers based upon polymer complexation, is presented. The motivation
behind the development of reversible block/graft emulsifiers has been presented in
Chapter 1. These emulsifiers could potentially impact enhanced oil recovery1 and
pipeline emulsion transportation of heavy oils.2 The reversible emulsification ability of
these polymers may be useful in both applications. For example, oil-in-water emulsions
provide a cost-effective alternative to heated pipelines or the addition of diluents for the
pumping of heavy oils. Typical transport emulsions are composed of 70 wt.% oil, 30
wt.% aqueous phase and 500-2000 ppm of a stabilizing surfactant formulation.2 The
resulting emulsion has a viscosity in the range of 50-200 cP at the pipeline operating
conditions as compared to heavy oil viscosities that may range from 10,000 to 100,000 cP
(see Figure 7.1A). Nonionic surfactants may be used for this purpose due to their relative
insensitivity to the salt content of the aqueous phase. However, once pumping has been
completed the oil-in-water emulsion needs to be broken for the recovery of oil (Figure

7.1 B). This may be achieved by either the addition of de-emulsifiers3 or by boiling off

120

121

 

   
 

100,000 ‘ ~ - ~ ~
Heavy Crude Oils

and Bitumens

10,000 ,

1,000 ‘ * ~ ‘ ‘
Operating Limit (Crude)

 

Viscosity (cP)

 

100

 

Oil-in-water emulsions

 

10 l 1 g I l l l
40 60 80100120140160
T(°F)

(A)

      
 
   
  
 

Heavy Crude Mixer Emulsion Storage

 

Pipeline System

Water

0»

Surfactant

Emulsion Treatment

  

 

Dry Crude

‘ ............
.........

 

—> Water
(B)

Figure 7.1. (A) Reduction of the viscosities of heavy oils by the formation of oil-in-water
emulsions, and (B) A schematic showing a treatment process that may be
used for the pumping of heavy oils by forming oil-in-water emulsions.

122

the water and in some cases, a combination of both techniques. Either way, this step
represents enormous energy costs and the addition of a de-emulsification agent. If the
emulsification and de—emulsification can be performed by the same emulsifier without the
encumbrance of high energy costs, it would result in substantial savings and relatively
simple operating conditions. The block/graft emulsifiers presented in this chapter could
have potential applications in such situations requiring emulsification and de-
emulsification under different sets of conditions.

Simple diblock copolymers have also been used as polymer compatibilizers and
dispersion stabilizers for many years. Theoretical work by Noolandi4 has suggested that
multi-block copolymers would be more efficient as compatibilizers than di- or triblock
copolymers. His results suggest that less of the multi-block copolymer would be lost into
the bulk phase as micelles or mesophases than the di- or triblock systems. Similarly,
multi-block dispersing agents and stabilizers should be more efficient than the common
di or triblock systems.

In this research, a potentially new kind of block copolymer, the block/graft
copolymer, has been investigated. Block/graft copolymers consist of a polymer backbone
that has side chain grafts extending linearly off the polymer backbone at regular intervals.
The graft side chains are capable of complexing with the backbone under certain
conditions to form blocks of hydrophilic and relatively hydrophobic polymer as shown in
Figure 7.2. The complexation occurs by hydrogen bonding under acidic conditions.
Because of the regular side chains, block/graft copolymers are often referred to as “comb”

copolymers in their uncomplexed state.

123

Basic Conditions

 

/ Poly(ethylene glycol) grafts

Hydrophilic methacrylic acid backbone

Acidic Conditions

 

Hydrophobic chai sections due to omplexation of PEG gr fts with backbone

 

Hydrophilic sections of backbone not complexed with PEG grafts

Figure 7.2. Schematic and mechanistic depiction of the block/ graft copolymer, consisting
of a poly(methacrylic acid) backbone and poly(ethylene glycol) grafts, under
acidic or complex promoting conditions and under basic or complex breaking
conditions.

124

In this chapter, a simple one-step method for producing reversible block/graft
copolymers for applications as dispersing agents and stabilizers, is presented. These novel
polymers have been formed by free radical polymerization of a vinyl monomer with small
amounts of a macromonomer. The polymers consist of a methacrylic acid backbone and
oligo(ethylene glycol) grafts. The oligo(ethylene glycol) grafis are incorporated into the
chain during the polymerization of the methacrylic acid, eliminating the need for a multi-
step synthesis. The backbone Lewis acid repeat units are capable of forming hydrogen
bonded complexes with the Lewis base ether oxygens of the oligomeric grafis (Figure
7.3B). These complexes are formed when the acid is protonated or under acidic
conditions and result in the complexed regions of the backbone becoming hydrophobic.
By choosing the ratio of methacrylic acid to ethylene glycol repeat units it is possible to
form pseudo-multiblock copolymers consisting of a hydrophilic acid backbone
interspersed with hydrophobic segments due to complexation. Such a polymer behaves
like a multiblock polymeric emulsifier under acidic conditions and as a hydrophilic graft
copolymer under basic conditions. Thus this reversible emulsification behavior can be
turned “on” or “off’ by controlling the pH or other variables that affect complexation.

The emulsification behavior of these novel reversible emulsifiers has been
investigated as a function of pH. The effect of hydrophilic segment length has also been
studied by varying the ratio of the methacrylic acid to ethylene oxide repeat units.
Variables such as emulsifier concentration and chemical structure and their effect on the
emulsification capacity, have been investigated. Interfacial tension measurements were
performed to elucidate the surface active behavior of these polymers as function of

solution pH.

125

—> rrrrvyw
I I l

o b (5 o b b o
WWW/W _/\_/\_/\_/\_/\_/\_/\_
(A)
CH3

   

CH3 CH3 CH3 CH3 CH3 CH3

I 10“

CH3\. / \ / \ / \“m
o o o 0 ..
2 : : : 9
i i i i *i l
o o o o o

O
I I I I I I
C=O C=O C=O C=O C=O C=O

          

CH3 CH3 CH3 CH3 CH3 CH3

(B)

Figure 7.3. Hydrogen bonded complexes formed between poly(methacrylic acid) and
' poly(ethylene glycol) units in (A) non-grafted and, (B) grafted
configurations.

126

7.2. Experimental Methods and Techniques

7. 2. 1. Copolymer Synthesis

The graft copolymers were synthesized via a free-radical solution polymerization of
methacrylic acid (MAA) with methoxy poly(ethylene glycol) methacrylate (MPEGMA)
using hydrogen peroxide (30% solution) as initiator and a 50/50 mixture of ethanol/water as
the solvent (see Figure 7.4). The macromonomer employed was MPEGMA 1000
consisting of an average 22.5 number of ethylene glycol (EG) repeating units. Polymers
with different molar ratios of MAA to EG repeat units (such as 5:1, 10:1 and 20:1) were
synthesized. All chemicals were high purity and were used as received. For a repeat unit
ratio of 20:1 MAAzEO the synthesis is described below.

MAA (Polysciences, Warrington, PA) (1.945 g) was added to 0.055 g of MPEGMA
1000 (Polysciences) in a clean glass bottle. To this mixture, 8 grams of solvent was added
comprising a 50/50 mixture of ethanol (200 proof) and water (by weight). The reaction was
initiated by adding 40 uL of a 30% solution of hydrogen peroxide as a thermal initiator.
The sample was placed in a controlled temperature water bath at 80°C and polymerized for
36 hours, after which it was dried in an oven at 80°C. The resulting solid polymer was
ground using a mortar and pestle. Polymer solutions were made by dissolving required
amounts of dry polymer in HPLC grade water with constant stirring. The solutions were
stirred overnight or until a clear solution was obtained. The solutions were filtered using a
coarse #4 qualitative (Whatrnan) filter paper under vacuum in order to remove any

suspended impurities.

127

  

 

 

 

 

 

.
. tawny” .
C O . .
' g ‘ . . 50/50 EtOHNVater
37,; O .‘r.¢"s,% ’
we fit
, . 0 ' H202
i; ’ . ' é ' Heat
. O I‘D Q
W \
.
CH3
HzCi
0
OH
\ MAA / kMPEGMA1000J

 

 

 

 

Figure 7.4. Schematic showing the free-radical polymerization scheme for the synthesis
of reversible block/graft emulsifiers.

128

7. 2. 2. Emulsification Studies

Emulsification studies were performed to determine qualitatively and quantitatively
the emulsification ability of the copolymers. For these studies, a 0.1 wt.% solution of the
copolymer was prepared in HPLC grade water under constant stin'ing. Methyl laurate
(Aldrich) was chosen as the “oil” phase. Samples, each consisting of 10 ml of the polymer
solution, were adjusted to different values of pH by the addition of small volumes of either
lM HCl or IM NaOH solution. In 10 ml stoppered graduated cylinders, 2 m1 of methyl
laurate was added to 8 ml of the pH-adjusted polymer solutions (Figure 7.5). The mixture
was agitated vigorously for about 2 minutes and then allowed to stand for 4 days or until no
change in the emulsion quality and volume was noted. The percent oil emulsified was
calculated by measuring the volume of un-emulsified clear oil in the oil phase and the

volume of the cream (stable emulsion).

7. 2. 3. Interfacial Tension Measurements

All interfacial tension measurements were performed on a Kriiss Processor
Tensiometer K12 (Krfiss, USA) equipped with a controlled temperature water bath. The
Wilhelmy plate method was used to determine the interfacial tension between the oil
phase (methyl laurate) and the aqueous phase. Measurements were performed in a 49 ml
(small) cup at 25°C as per the instrument procedure for interfacial tension measurements.
Three readings per sample were recorded and the average of the three was reported as the
interfacial tension. Polymer solutions at various pH for the 10:1 and 20:1 (MAA to EG

repeat unit ratio) polymers were prepared from a 0.1 wt.% stock solution. As before the

129

 

Oil

  
 

 

Emulsion
“Cream”

Agitate

Emulsifier we“ —>
solution

N

\ Microemulsion
or water

 

 

 

 

 

 

Emulsion

Methyl laurate - oil phase

Figure 7.5. Schematic of the methodology adopted for the oil emulsification studies.

130

pH of the samples was adjusted by addition of small quantities of 1M HCl or IM NaOH
solutions. Interfacial tension measurements were performed for solution pH ranging

between 2 and 12 for both the 10:1 and 20:1 copolymer solutions.

7.3. Results And Discussion

7. 3. I . Emulsification Studies

The emulsification properties of two different block/ graft copolymeric emulsifiers is
illustrated in Figure 7.6. The figure represents results from tests performed with 20 % (v/v)
methyl laurate dispersed in a 0.1 wt.% polymer solution. The figure demonstrates that even
at fairly low concentrations under acidic conditions, the copolymers are able to emulsify 20
vol.% oil by forming a stable emulsion. In addition, the emulsions may be readily broken
by raising the pH, with the percentage emulsified dropping to zero under basic conditions.
It is interesting to note that these properties are reversible, and that a system cycled from
acidic to basic conditions will reform a stable emulsion if the pH is returned to acidic
conditions.

The data in Figure 7.6 also illustrates that molecular architecture plays an important
role in the performance of the emulsifiers. The figure shows that the emulsification ability,
as characterized by the percent oil emulsified, depends upon the methacrylic acid to
ethylene glycol repeat unit ratio. The 20:1 MAA to EG polymer exhibits a decrease in the
emulsification capacity at a considerably lower pH than the 10:1 system. The lines in
Figure 7.6 are drawn to illustrate the emulsification trends for the two repeat unit ratios

under investigation.

131

 

1001
80*
60-

40‘

Percent Oil Emulsified

 

 

 

 

Figure 7.6. Emulsification behavior of two different block/graft copolymers (10:1 and
20:1) as a function of pH showing the percent oil emulsified (v/v) as a
function of pH.

132

7.3. 2. Interfacial Tension Studies

The data presented in Figure 7.7 illustrates the trends observed for the interfacial
tension between the oil (methyl laurate) phase and the aqueous phase containing the
polymeric emulsifiers. Once again, data for both the 10:1 and the 20:1 copolymers is
represented by the symbols whereas the lines show trends. Under acidic conditions, it
can be observed that the interfacial tension is lowered to a value of about 10 mN/m as
compared to a value of about 21 mN/m under neutral conditions. The interfacial tension
between methyl laurate and pure water, in the absence of any polymer, was determined to
be 21.5 mN/m at 25°C. Both copolymeric emulsifiers under investigation exhibit similar
trends, however, the 20:1 copolymer exhibits a marginally lower interfacial tension under
acidic conditions.

We attribute the observed emulsification and lowering of interfacial tension
properties exhibited by the reversible block/graft copolymers to a self associating
phenomenon in which the grafts form complexes with the backbone. In this system, the
self-complexation of the block copolymer occurs because of hydrogen bonding between
the ether linkages of the poly(ethylene glycol) (PEG) and the acid moieties of the
poly(methacrylic acid) (PMAA). In the complex, the hydrophilic moieties are bound
with one another as illustrated in Figure 7.2. Therefore the complex is significantly more
hydrophobic than its constituent polymers, and the complexed structure mimics a pseudo-
multiblock copolymer architechture. The fact that the complex is significantly more
hydrophobic than the uncomplexed polymers has been established by the study of the

swelling properties of crosslinked poly(methacrylic acid-g-ethylene glycol) networks

133

 

 

 

 

 

 

22‘ - 10:1
19‘
2
E,
C
.9
(I)
C
m
I...
33
8
‘l:
9.
5 10‘ ' Wilhelmy Plate Method
‘ (Methyl Laurate - Water)
8 I I I I l I 1 l I l r
O 2 4 6 8 10 12
pH

Figure 7.7. Interfacial tension between the oil (methyl laurate) phase and the aqueous
phase containing the emulsifiers, as a function of pH. Data is presented for
the 10:1 and the 20:1 copolymeric emulsifiers.

134

reported by Klier and collaboratorss'6 For example, these authors concluded that the
aqueous swelling minimum observed under acidic conditions for gels with 1:1 MAAzEG
repeat unit ratios can be attributed to the formation of hydrogen bonded hydrophobically
stabilized complexes. They also observed a reduction in the equilibrium degree of swelling
with temperature for non—stoichiometn'c gels due to the increase in strength of the
hydrophobic interactions between the complexing polymers.‘5 In addition, the hydrogen-
bonded hydrophobic complex formed between hydrophilic Lewis acid carboxylic groups
and hydrophilic Lewis base groups can be disrupted by a change in pH, solvent type or
temperature. Due to this phenomenon the block/graft copolymers described in this chapter
can be used as reversible emulsifiers which can be turned “on” or “off” by triggering a
“switch” such as pH, solvent or temperature.

The emulsification trends can be explained by the formation and disruption of the
complex between the hydrophilic backbone and the ethylene glycol grafts. For the
MAAzEG molar ratios being discussed here, the long hydrophilic sections of the
complexed polymer would sterically stabilize oil-in-water emulsions as shown in Figure
7.8. The hydrophobic sections of the pseudo-block copolymer would preferentially
partition onto the surface of an oil droplet while the hydrophilic segments would impart
solubility in the bulk aqueous phase. This would prevent droplet coalescence and thus
result in a stable emulsion (Figure 7.8). Since the complexes are formed under acidic
conditions and can be disrupted under basic conditions, the copolymers are able to
emulsify oil only under complex promoting conditions. For the 20:1 copolymer, a
decrease in the percent oil emulsified with an increase in pH may be attributed to an

increase in the hydrophilic carboxylic acid segments, resulting in inefficient

135

Hydrophilic Sequences

       
  

‘—

Water Insoluble
Complex Sequence

 

Basic conditions

Acidic conditions

(A)

KEV

 

 

 

 

Figure 7.8. Schematic representation of the possible emulsification mechanism adopted

by the reversible emulsifiers (A) under acidic and basic conditions, and (B)
the steric stabilization that may result.

136

emulsification. A decrease in the MAA to EG ratio may result in shorter hydrophilic
segments per chain, thereby facilitating better partitioning into the oil phase under acidic
conditions. However, both copolymers display a trend in which they emulsify oil under
acidic conditions and act as poor emulsifiers under basic conditions. When base was
added to the acidic samples, the emulsion was broken instantly upon agitation resulting in
a two-phase system. Upon lowering the pH and vigorous mixing thereafter, the emulsion
was re-formed thereby elucidating their reversible nature. Hence pseudo-multi
block/graft copolymers may be “tailor made” to work as reversible emulsifiers in the pH
range of choice by controlling the MAA to EG repeat unit ratio. The emulsification
behavior may also be controlled by addition of hydrophilic or hydrophobic oligomers as
grafts to impart additional hydrophilicity or hydrophobicity.

The lowering of interfacial tension may also be explained by the formation of a
pseudo-multi block structure under complex promoting conditions. With an alternating
hydrophobic-hydrophilic architecture, the polymeric emulsifiers may form aggregates
much like micelles. However, these aggregates may be formed either by a single chain if
it can assume a suitable conformation or by the cooperative effort of several chains. The
aggregates may consist of a hydrophobic interior and a hydrophilic domain extending
into the bulk aqueous phase. This picture is consistent with the emulsification studies,
since oil droplets may be preferentially incorporated into the interior of such aggregates.
Similarly, the lowering of interfacial tension may be attributed to the preferential
partitioning of the hydrophobic segments into the oil phase all along the oil-water

interface. In the following chapter, a fluorescence spectroscopy study will be performed

137

with pyrene as a fluorescent probe to study the aggregates formed by these copolymers as
a function of pH and concentration.

The block/graft copolymers may be useful in a variety of applications which may
require emulsification under certain conditions, while phase separation may be desired
under other conditions. One such example is an aqueous cleaning system which offers all
of the separation advantages of an emulsifying system but also allows (by breaking the
emulsion) the oil and water to be readily separated from one another. After the emulsion
is broken, the oil phase may be readily skimmed from the water, and the aqueous phase
(including the surfactant) may be recycled. These types of emulsifiers would ultimately
affect a variety of industries ranging from petroleum processing (for wastewater treatment

and separations, and enhanced oil recovery) to automotive service stations.

7.4. Conclusions

The synthesis and emulsification behavior of novel block/graft copolymeric
emulsifiers based upon a reversible hydrophobic architecture has been presented. In these
systems, the graft copolymer assumes a multiblock copolymer configuration under
certain conditions. The copolymers exhibit reversible emulsification behavior and may
be useful in a variety of applications. Emulsification studies demonstrated that the
copolymers can emulsify oil under acidic conditions in a reversible fashion. As the pH of
the emulsifier solution is raised towards neutral, the percent oil emulsified decreases, and
under basic conditions no emulsification is observed. Interfacial tension studies

illustrated the surface active nature of the copolymers under acidic conditions. A

138

lowering of the interfacial tension was observed under acidic conditions for the
copolymeric emulsifiers under investigation. These properties exhibited by the
block/graft copolymers can be attributed to the formation of hydrogen bonded
hydrophobic complexes between the methacrylic acid backbone and the ethylene glycol
repeat units on the graft. This may result in an alternating hydrophilic-hydrophobic
pseudo-multi block structure which may be responsible for the surface active properties

of these copolymers.

139

7.5. References

Taylor KC. and Hawkins BE, in Emulsions: Fundamental Applications in the
Petroleum Industry, L.L. Schramm Ed., ACS Advances in Chemistry Series 231, pp
263 (1992).

Rimmer D.P, Gregoli A.A., Hamshar J.A. and Yildirim E., in Emulsions:
Fundamental Applications in the Petroleum Industry, L.L. Schramm Ed., ACS
Advances in Chemistry Series 231, pp 295 (1992).

Grace L., in Emulsions: Fundamental Applications in the Petroleum Industry, L.L.
Schramm Ed., ACS Advances in Chemistry Series 231, pp 313 (1992).

Noolandi, J., Makromol. Chem. Theory Simul. 1(5), 295-8 (1992).

Scranton A.B., Klier J ., Aronson C.L., in Polyelectrolyte Gels, R. S. Harland and
R. K. Prud’homme Eds., ACS Symposium Series 480, Washington, DC, pp 171
(1992)

Klier J., Scranton AB, and Peppas N.A., Macromolecules 23, 4944 (1990).

CHAPTER 8

A FLUORESCENCE SPECTROSCOPY STUDY OF THE
AGGREGATE FORMATION BEHAVIOR OF REVERSIBLE
COPOLYMERIC BLOCKIGRAFT EMULSIFIERS

8.1. Introduction

In the previous chapter, the synthesis and emulsification properties of block/graft
copolymeric emulsifiers were presented. The chemical nature of the reversible hydrogen
bonded complexes between backbone methacrylic acid repeat units and the oligomeric
ethylene glycol grafts, was also discussed. The oil emulsification properties of the
copolymers were presented as a function of pH. The copolymers emulsified oil under
acidic conditions since the grafts form hydrophobic complexes with the backbone
resulting in an alternating hydrophobic-hydrophilic structure. It was also demonstrated
that these complexes were reversible and that, upon raising the pH, the emulsion could be
broken, and reformed thereafter upon agitation if the pH was lowered. Interfacial tension
measurements between a model oil (methyl laurate) phase and the aqueous polymer
solution phase exhibited a lowering of the interfacial tension for the samples at acidic pH
as compared to those under neutral or basic pH.

The studies mentioned above, have established the surface active nature of the
block/graft copolymers under acidic pH. In this chapter, the aggregate formation

behavior of the copolymeric emulsifiers under acidic conditions, has been investigated. It

140

141

is hypothesized that the aggregates may consist of a hydrophobic interior which may be
responsible for the emulsification properties observed in the previous chapter. The
environment-sensitive fluorescence of pyrene has been utilized to study the formation of
the aggregates with hydrophobic interiors. Pyrene not only would serve as a fluorescent
probe but also as the model “oil” phase for these studies.

Pyrene has been used as a polarity sensitive probe due to the effects of iosmeric
solvents on the vibronic band intensities in the fluorescence spectrum of pyrene' (also
known as the Ham2 effect). This environmental effect on the vibronic band intensity in
pyrene monomer fluorescence was utilized by Kalyanasundaram and Thomas3 in studies
of micellar systems. The strong perturbation of the vibronic band intensities was used as
a probe to accurately determine critical micelle concentrations and also to investigate the
extent of water penetration in micellar systems. An excellent review article by Turro et

al.4

discusses the use of luminescence probes as sensors for probing the
microenvironment of micellar systems. More recently, both solubilized pyrene as well as
pyrene end-labeling has been used to study micellar solutions formed by block copolymer
surfactants. For example, the use of pyrene end labeled poly(ethylene oxide) to study its
interactions with sodium dodecyl sulfate (SDS) in aqueous solution, has been reported.’
These investigators observed an increase in the excimer/monomer emission intensity with
an increase in SDS concentration which was explained by the incorporation of the pyrene
end—labeled poly(ethylene oxide) into aggregates comprising two or more end-labeled
chains.5 Similarly, Prochazka et al.° and Kiserow et al.7 have utilized the time-resolved

fluorescence of naphthalene to study the chain dynamics of micelles formed by

napthalene-labeled diblock and triblock copolymers in aqueous media. It was observed

142

that intramolecular excimer formation (which is controlled by the mobility of the pendant
fluorescent groups and the polymer chain dynamics) is sterically hindered in the micellar
cores as they become more compact.° Bakeev et al. have used l-pyrenecarboxylaldehyde
as a fluorescence probe to study the complexation of amphiphilic polyelectrolytes with
surfactants of the same charge in aqueous solutions.8 In addition, the pH and electrolyte
related chain dynamics of a water soluble terpolymer based on acrylamide and the surface
active monomer sodium 11 acrylamidoundecanoate has been investigated by
incorporation of 2-(l-pyrenylsulfonamido) ethyl acrylamide into the monomer.9 The
steady state fluorescence studies conducted by these investigators revealed significant
constrictions of the polymer chains as pH decreases or electrolyte concentration
increases.

Maltesh et al. studied the effect of binding of cations to pyrene end-labeled
polyethylene glycol (PEG) on its interactions with sodium dodecyl sulfate (SDS).'0 The
conformational changes of pyrene end-labeled PEG during its association with SDS has
also been investigated by monitoring the excimer formation between the pyrene end-

Il,l2

groups. The applications of fluorescence spectroscopy to the study of polymer-

surfactant interactions and polymer association in water have been well presented in
recent reviews.'3'”

Other aromatic luminescent probes solubilized in aqueous media have also been
used as probes for the study of micellar solutions.” Intrinsic excimer fluorescence of
p0lystyrene-block-poly(ethylene propylene) has also been used to study the block

copolymer micelles and to determine the critical micelle concentration (CMC).'° The

determination of the critical micelle concentration using pyrene as a solubilized

143

fluorescent probe for diblock copolymers,”l8 block electrolyte solutions,”20 and triblock
copolymers,” has been recently investigated. Investigators have also used other aromatic

22.23

fluorescent probes such as 1,6-Diphenyl-1,3,5-hexatriene (DPH) and dypyme24 for the
investigation of micelle formation by the Pluronic (BASF) class of triblock copolymeric
surfactants. These investigators have utilized the environment-sensitive fluorescence of
the solubilized fluorescent probe to obtain estimates for the CMC and to study micelle
formation in the copolymer systems.

The CMC of many of the block copolymer systems is too small to be determined
by light-scattering techniques.” In such cases a method based on the use of pyrene as a
fluorescent probe has been employed to determine the CMC.”“‘9 In the presence of
micelles, the pyrene probe partitions between the aqueous and the micellar phases. This
partitioning leads to a number of interesting changes in the (0,0) band in both the
excitation and the emission spectra of pyrene. There is a red shift in the excitation
spectrum, change in the vibrational fine structure of pyrene fluorescence (the ratio Il/I3
decreases) and an increase in the fluorescence decay time (manifested by an increase in
intensity), accompanying the transfer of pyrene from the aqueous to the hydrophobic
micellar phase. The excitation peak maximum for the broad peak centered at around 332
nm was observed to shift to about 338 nm in hydrophobic environments.”“9 This shift
was also reported to be accompanied by an increase in the emission intensity attributed to
an increase in the fluorescence lifetime of the excited pyrene. In the case of the emission
spectrum of pyrene, the ratio of the peak at ~378 nm (1,) to the peak at ~389 nm (13) is
very sensitive to the polarity of the environment.'8"9'2| The Il peak, which arises from the

(0,0) transition from the lowest excited electronic state, is a “symmetry forbidden”

144

transition that can be enhanced by the distortion of the n-electron cloud. Whereas, the
electronic transition that results in the I3 peak is not forbidden and thus is relatively
solvent-insensitive.‘9 Thus, the ratio (I,/I3) can serve as a measure of the polarity of the
environment experienced by pyrene. The CMC values estimated using this ratio have
been reported3"8"9'21 to be in agreement with those obtained from conventional techniques
such as surface tension measurements.

In this chapter, we will utilize the environment-sensitive fluorescence of
solubilized pyrene to study the aggregate formation behavior of block/graft copolymers
of poly(methacrylic acid-g-ethylene glycol) as a function of pH and concentration. The
ability of the copolymers to emulsify oil under acidic conditions points to the formation
of hydrophobic domains. Steady state fluorescence measurements of the excitation and
emission spectra of solubilized pyrene, as it partitions between the hydrophobic domains
and the bulk aqueous phase, will be analyzed. In addition, possible methods to determine

the critical aggregate concentration will be presented.

8.2. Experimental Methods and Techniques

The synthesis of reversible block/graft copolymeric emulsifiers of
poly(methacrylic acid) (PMAA) with poly(ethylene glycol) (PEG) grafts (referred to as
PMAA-g-PEGIOOO) was discussed in detail in the previous chapter. Two PMAA-g-
PEG1000 copolymers investigated in this study contained MAA to EG ratios of 10:1 and

20:1. Fluorescence spectroscopy was performed on 00polymer samples at pH values

145

ranging from <2 to 7 and concentrations ranging from 0.0 wt.% (no polymer) to 0.5
wt.%. The studies were performed for two sets of copolymeric emulsifiers, those
containing 10:1 and 20:1 ratio of methacrylic acid to ethylene glycol repeat units.
Polymer solutions were prepared by dissolving known amount of the dry polymer into
HPLC grade water. The solutions were filtered by a coarse #4 filter paper (Whatman) in
order to remove any suspended impurities. Typically a 100 ml sample with a polymer
concentration of 0.5 wt.% was prepared and successive dilutions were made from this
stock solution. Pyrene was used as the polarity sensitive fluorescent probe for all the
experiments.

The solubility limit of pyrene in water is reported at a concentration of 6x107 M.
To prepare a saturated pyrene solution in water, 0.010 g of pyrene was added to 100 ml of
HPLC grade water and stirred for 4 days. The solution was then filtered using a medium
#2 filter paper (Whatrnan) to remove any undissolved pyrene. The concentration of this
pyrene solution was assumed to be 6x10'7 M, which is the saturation concentration of
pyrene in water. For all the studies reported in this chapter, the pyrene concentration was
kept constant at a value of 6x10'8 M for each sample. This was to ensure that no excimer
formation would occur since the concentration of pyrene was 10% of its saturation
concentration. Sample concentrations varied from 0.0 wt.% (no polymer) to 0.5 wt.%
and each 10 ml sample contained 6x10'8 M pyrene. The pH of the samples was varied
from <2 to about 7 by adding a few microliters of either a 1M HCl or IM NaOH solution.

The fluorescence spectroscopy was performed on an Aminco-Bowman Series 2

Luminescence Spectrometer. Both the excitation and emission spectra were recorded for

146

every sample. For the excitation spectra, emission was at 397 nm while the excitation
was varied between 290-360 nm. For emission spectra, excitation was at 322 nm and the
emission was collected between 370 and 450 nm. The slit width was kept at 0.2 nm and
the spectra were collected at intervals of 0.2 nm at a rate of 1 nm/s. The detector voltage

was kept constant for the excitation and emission scans over all concentrations.

8.3. Results and Discussion

8. 3. 1. Excitation Spectra

Representative excitation spectra of solubilized pyrene for a series of copolymer
samples are shown in Figure 8.1. This figure is a plot of the emission intensity (arbitrary
units) at 397 nm as the excitation intensity is varied from 290 nm to 360 nm. The plot
consists of four spectra representing aqueous solutions of a 20:1 (methacrylic acid
(MAA) to ethylene glycol (EG) molar repeat unit ratio) copolymer with concentrations
ranging from 0.005 wt.% to 0.1 wt.%, each at pH 2. This figure illustrates the red shift
associated with the partitioning of pyrene into hydrophobic domains. In addition, as the
copolymer concentration increases from 0.005 wt.% to 0.1 wt.% there is an increase in
emission intensity seen in the excitation spectra. The excitation peak maximum shifts
from 337 nm (corresponding to pyrene in bulk aqueous phase) to 342 nm (representing
pyrene in a hydrophobic phase).

As discussed in the previous chapter, PMAA-g—PEGIOOO block/graft copolymers
form a pseudo-block copolymer structure under acidic conditions. This occurs possibly

due to the formation of hydrophobic segments arising from hydrogen bonded MAA-EG

147

1.8

 

 

 

 

 

 

 

. 337 342
1.6 -_ Polymer concentration
-0.005 wt. % (A)
1.4-— —0.01 wt. % (B)
—0.05 wt. % (C)
1.2-- —0.1 wt. % (D)
D
a 1 ‘"
2 0 3.-
o .
E C
_ 0.6--
o 4 "3
0.2-- A
H, - A“... ., —”
0 :

290 300 310 320 330 340 350 360
Wavelength (nm)

Figure 8.1. Pyrene excitation spectra for 20:1 PMAA-g-PEGIOOO copolymer solutions of
various concentrations with constant pyrene concentration (6 x 10‘7 M).
Emission was monitored at 397 nm.

148

units along the copolymer backbone (see Figures 7.2 and 7.3). Since the ratio of MAA to
EG repeat units is 20:1, under complex promoting conditions this may result in the
formation of alternating hydrophobic and hydrophilic segments within the same chain.
One or more chains may then associate to form aggregates with hydrophobic cores in the
interior with the hydrophilic MAA groups providing solubility in the bulk aqueous phase
(Figure 7.8).

Most researchers have used the 11/13 ratio in pyrene emission spectra to study

micellization and estimate the CMC of block copolymer solutions.3"°'2‘

However, the
excitation spectrum of pyrene may also be used to study aggregate formation and to
estimate the critical aggregate concentration (CAC). Figure 8.2 illustrates the pH
dependence of the ratio of the peak at 337 nm to that at 342 nm (1337/1342). The data
presented in this figure is for a 20:1 MAzEG copolymer for concentrations ranging from
0.005 wt.% to 0.1 wt.%. The data are represented by the symbols while the lines drawn
through them illustrate trends. There are two important observations that can be made
from this figure. First, as the pH is raised from 2 to 7 the excitation ratio 1337/1342,
decreases and approaches a value of ~0.4 at pH 7. Second, this decrease is more
pronounced as the concentration of the copolymer is increased. In addition, under acidic
conditions the excitation ratio 1337/1342 increases from about 0.6 for a concentration. of
0.005 wt.% to about 1.8 when the copolymer concentration is raised to 0.1 wt.%. It was
also observed that for a pH greater than 7 the ratio remained constant at about 0.4. This

was also the value observed from the excitation spectrum of pyrene in water without any

copolymer present.

149

 

 

 

 

 

 

 

 

2.0
1 8 2 . Polymer concentration
' ~ . 0.005 wt. %
’7‘ 1-6 ‘ A l 0.01 wt. °/o
5 . ‘ A 0.05 wt. %
: 1-4 ‘ o 0.1 wt. %
N .
:3 1.2 -
.9 l
E 1.0 -
c 0.8 - I A
.9 .
E 0.6 - o--———-ILN\\
'6 , \\\
g 0.4 4 :MM
0.2 -
0.0 . , , I . I .
pH

Figure 8.2. Plot of the pyrene excitation ratio (1342/1337) as a function of pH for 20:]
PMAA-g-PEGIOOO copolymer solutions with 6 x 10'7 M pyrene. The
emission was monitored at 397 nm.

150

This data may be used to plot the percent pyrene “emulsified” as a function of pH
by considering a simple two-phase model for the partitioning of pyrene in the
hydrophobic and hydrophilic domains. In our studies, pyrene may be considered as the
“oil” phase due to its hydrophobic nature. If an excitation ratio of 0.4 is considered to
represent pyrene in the aqueous phase and a value to 2.0 represents pyrene in the
hydrophobic phase (based upon experiments with a polymer concentration of 0.5 wt.%),
then a simple linear relationship may be used to calculate the percent pyrene “emulsified”
by the copolymer. Figure 8.3 is plot of the percent pyrene emulsified calculated in the
above manner versus pH for the data presented in Figure 8.2. This figure demonstrates
aggregate formation exhibited by the 20:1 copolymer surfactant under acidic conditions.
An increase in the polymer concentration results in the formation of an increased number
of polymeric aggregates under acidic conditions. This may be concluded from the
increase in the excitation ratio resulting from pyrene molecules experiencing a
hydrophobic environment. This figure also illustrates the surface inactive behavior of the
block/graft copolymer under neutral and basic pH due to the decomplexation of the PEG

graft from the PMAA backbone.

8. 3. 2. Estimation of the Critical Aggregate Concentration fiom the Excitation Spectra
The PMAA-g-PEGIOOO copolymers employed in this study have shown evidence
of aggregate formation which was first demonstrated by oil emulsification studies
discussed in the previous chapter. While most ionic and di-and tri-block copolymeric
emulsifiers form micelles above the critical micelle concentration, the PMAA-g-

PEG1000 copolymers may not form micelles in the strict sense. Due to the proposed

151

 

 

 

 

 

 

 

 

100
90 o Polymer concentration
‘ 0 0.005 wt. %
5c: 807, ‘ I 0.01 wt. %
93 A A 0.05 wt. %
5': 7O -
g . e 0.1 wt. %
E 60 -
I5” ..
g 50 1 .
S 40 ‘ I
0. I
*5 30 -
8 20 -
d.’ ‘ I “We “\
10 - \\-\\
0 I I . l ' r T l . TS“ I . I '
0 1 2 3 4 5 6 7 8
pH

Figure 8.3. Plot of percent pyrene emulsified versus pH for a 20:1 PMAA-g-PEGIOOO
copolymer solutions calculated from excitation ratio data.

152

pseudo-multiblock architecture under acidic conditions, aggregates may be formed by
one or more chains depending on the length of the polymeric chain and the size of the
aggregate. In this study, a critical aggregate concentration will be identified for the
PMAA—g—PEGIOOO copolymers. The CAC may be defined as the concentration at which
the first copolymer aggregate is formed analogous to the definition of the CMC.

While most researchers have used the pyrene emission spectrum to estimate the
CMC of copolymeric surfactants, a few investigators have used the pyrene excitation
spectrum to estimate the CMC of block copolymer surfactants.‘8"9 In general, a ratio of
peaks in the excitation spectrum has been employed. For example, the peak ratio 1333/1333
has been employed to estimate the CMCs of polystyrene-b-poly(ethylene oxide) diblock
copolymers'g"9 In this study, the peak ratio 1337/1342 has been employed to estimate the
CMC of the reversible block/graft emulsifiers. Figure 8.4 demonstrates the effect of
copolymer concentration on the excitation ratio (1337/1342) at a pH of 2 for the PMAA-g-
PEGlOOO copolymeric emulsifiers. This figure illustrates the trends exhibited by two
different copolymers with MAA to EG ratios of 10:1 and 20:1. The data are represented
by the symbols whereas the lines drawn through them illustrate trends. The sigmoidal
nature of the trends in this figure is similar to plots of properties such as surface tension
plotted as a function of surfactant concentration.'7'22'25‘26 For example, most surfactants
exhibit a decrease in surface tension with an increase in the aqueous concentration.
However, a leveling-off is observed after a certain concentration and the CMC may be
estimated from the intersection of tangents drawn along the two slopes.‘7'22'25'26 A similar
procedure can be adopted to estimate the critical aggregate concentration (CAC) from this

data for the two copolymeric emulsifiers. From Figure 8.4 the estimated values for the

153

 

 

 

 

 

 

I 1021
2'0‘ . 20:1
7,, 1.5-
.36
.9
g 1.0-
C
.2 4
iii
2 0.5—
LlJ I:
I I l
I
l i
0.0 IIIHII I I IIIIIII r Trnlnl I r IIIIII' I rrrnrfl r rrrrnr] n rrrrrul
1E—6 1E-5 1E-4 1E-3 0.01 0.1 1

Polymer Concentration (wt. %)

Figure 8.4. A plot of the pyrene excitation ratio as a function of copolymer concentration
for 10:1 and 20:1 PMAA-g-PEGIOOO copolymers at pH 2.

154

CAC for the 10:1 and the 20:1 copolymeric emulsifiers were 8 x 10‘4 wt.% and 2 x 10'3
wt.% respectively. For di-block and tri-block copolymeric surfactants, the CMCs
reported in literature range from 4 x 10“1 wt.% to 6 x 10'4 wt.% for poly(ethylene oxide)-
polystyrene tri- and diblock copolymers respectively.” The values obtained from the
excitation spectrum for the PMAA-g—PEGIOOO block/graft copolymers under acidic
conditions are of the same order of magnitude as those reported in literature for diblock

and triblock copolymeric surfactants.

8. 3. 3. Emission Spectra

The steady state pyrene emission spectra for a 20:1 PMAA-g-PEGIOOO
copolymeric emulsifier at a pH of 2 is shown in Figure 8.5. The figure contains spectra
for copolymer solutions from 0.005 wt.% to 0.5 wt.% with the same pyrene concentration
(6 x 10‘7 M). The two trends that can be readily observed are an increase in overall
emission intensity as the copolymer concentration increases, and the change in the peak
intensity ratio (1373/1339) commonly referred to as the II/I3 ratio. As in the case of the
excitation spectrum, these trends may be explained by the partitioning of the pyrene
between the aqueous and hydrophobic domains created by the copolymer aggregates
under acidic conditions. The intensity increase may be attributed to an increase in the
lifetime of the excited state pyrene, while the increase in the ratio Il/I3 arises from
forbidden vibronic band transitions which exhibit marked intensity enhancement under
the influence of solvent polarity. Figure 8.6 illustrates the dependence of the Il/I3 ratio

(ratio of peak at 378 nm and 389 nm respectively) for a 20:1 copolymer as a function of

155

 

 

 

 

 

 

1 378 389
0,9 -_ Polymer concentration
-- 0.005 wt. “/0 (A)
0.8 -— —0.01 wt. % (B)
0.7 .L . — 0.05 wt. % (C)
D — 0.1 wt. % (o)
0.6 -«-—
3. -_ .
.5.) 0.5 I C
c
,9 0.4 ar- ,
E .
0.3 +- '
A
0.2 -— -
0 1 -.~ “N .K
0 I i i ‘ i ‘ i i

l I I
370 380 390 400 410 420 430 440 450
Wavelength (nm)

Figure 8.5. Pyrene emission spectra for 20:1 PMAA-g-PEGIOOO copolymer solutions at
pH 2. Excitation was at 322 nm and pyrene concentration in each sample
was 6 x 10'7 M.

156

 

 

 

 

 

 

 

 

 

1.8-
A 1.6-1
g
:0, 1.4-
I?) .
E 1.2-
E
c 1.0— .
.9 Polymer concentration
<0 o
-‘-’-’ 0.8- . 0.005wt.%
UEJ . - 0.01wt.%
06— A 0.05 wt.%
' 0 0.1wt.%
0.4 I I I I r I I
0 2 4 6 8

Figure 8.6. A plot of the pyrene emission ratio 1373/1339 (I,/I3) as a function of pH for 20:1
copolymer solutions with a pyrene concentration of 6 x 10'7 M. Excitation
was at 322 nm.

157

pH. The figure contains data for copolymer solutions with concentrations ranging from
0.005 wt.% to 0.1 wt.%. The data are represented by the symbols and the lines drawn
through them illustrate trends. From this figure it can be noted that, under acidic
conditions, the emission ratio (II/I3) decreases as the copolymer concentration is
increased. From experiments performed on solutions without any copolymer, the II/I3
ratio for pyrene was found to be ~1.8 and independent of pH. In addition, the pyrene
emission spectrum for a solution containing 0.5 wt.% of 20:1 copolymeric emulsifier
yielded a Il/I3 ratio of ~0.8. Values for the emission ratio I,/I3 for pyrene in water
reported in literature range from 1.58 to 1.9.3'17’” The I,/I3 values for pyrene in a
hydrophobic environment range from 0.55-0.6 in various hydrocarbon solvents,3 0.95 for
polystyrene films,l7 1.15-1.20 in polystyrene-b—poly(ethylene oxide) solutions,‘8 and

between 1.05-1.12 in polystyrene-b-poly(sodium acrylate) solutions.19

By considering a
Il/I3 value of 1.8 to represent pyrene in the aqueous phase and a value of 0.8 to represent
pyrene in a predominantly hydrophobic phase, a plot of percent pyrene emulsified as a

function of pH may be constructed in a manner similar to that described in the pyrene

excitation spectrum studies.

8. 3. 4. Estimation of the CA C from Emission Spectra

A plot of the emission ratio (II/I3) as a function of concentration at a pH of 2 is
presented in Figure 8.7. The figure contains data for two PMAA-g-PEGIOOO copolymers
with MAA to E6 repeat units ratios of 10:1 and 20:1. The data is represented by the
symbols and lines drawn through them represent trends. Both the copolymers exhibit

similar sigmoidal trends and the technique used to estimate the CAC from excitation data

158

 

 

 

 

 

 

1.6- . _,
I \\ _
“a
_<:> 1.4-
}; 1.2~
76
(I
‘5
.2!
E
LlJ
0.8- I ‘
I l
l I I lWIFl 1 r1 rlllll fiT 1 Illlll l l l llllll I l T llllll l l l lllll
1E-6 1E-5 1E-4 1E-3 0.01 0.1 1

Polymer Concentration (wt. %)

Figure 8.7. A plot of the pyrene emission ratio (excitation at 322 nm) versus
concentration for 10:1 and 20:1 PMAA-g-PEGIOOO copolymers.

159

can be similarly applied to determine the CAC from the data in Figure 8.7. This may be
achieved by drawing tangents to the two faces of the sigmoidal curve as shown in the
figure and determining the CAC as the concentration corresponding to the point where
the tangents intersect. Using this method the CAC for the 10:1 and 20:1 copolymeric
emulsifiers was found to be 8 x 10“ wt.% and 2 x 10'3 wt.% respectively. These values
are in perfect agreement with those obtained from the excitation spectra for the
copolymers.

The differences in aggregate formation and the estimated CAC between the two
copolymeric emulsifiers are clearly evident in Figures 8.4 and 8.7. The 10:1 PMAA-g-
PEGIOOO copolymer contains 10 times the number of hydrophobic sites under complexed
conditions as compared to the 20:1 copolymer. It is therefore correct to presume that on
an average the 10:1 copolymer will contain twice the number of hydrophobic segments
per chain as compared to the 20:1 copolymer. The lower critical aggregate concentration
exhibited by the 10:1 copolymer may be attributed to the higher number of hydrophobic
units per chain formed under complex promoting conditions. This may facilitate the
formation of hydrophobic aggregates at a lower concentration than the 20:1 copolymer.
However, a ratio lower than 10:1 and closer to 1:] may not exhibit a lower CAC since the
excess of hydrophilic methacrylic acid units is essential for the solubilization of the

aggregates into the bulk aqueous phase.

160

8.4. Conclusions

In this chapter, the aggregate formation behavior of reversible block/graft
copolymers has been presented. Fluorescence spectroscopy has been used to study the
formation of hydrophobic aggregates. The polarity sensitive fluorescence of solubilized
pyrene was used to study the aggregate formation as a function of pH and concentration.
Both the excitation and emission spectra of pyrene have been used to estimate the critical
aggregate concentration for 10:1 and 20:1 copolymers. Specifically, the 1337/1342 ratio in
the excitation spectrum and the 1373/1339 ratio in the emission spectrum were plotted as a
function of pH for various copolymer concentrations. As the copolymer concentration
was increased, the excitation and emission ratios indicated increased partitioning of the
pyrene into hydrophobic domains. This may be interpreted as direct evidence of the
increase in the number of aggregates formed as a result of an increase in copolymer
concentration. Further, the critical aggregate concentration for the 10:1 and 20:1
copolymers was estimated from both the excitation and emission spectra to be 8 x 10“

wt.% and 2 x 10'3 wt.% respectively.

10.

11.

12.

13.

14.

15.

161

8.5. References

Nakajima A., J. Mol. Spec, 61, 467-469 (1976).
Ham J. S., J. Chem. Phys., 21, 756-758 (1953).
Kalyanasundaram K. and Thomas J. K., J. Am. Chem. Soc., 99, 2039 (1977).

Turro N. J., Gratzel M., and Braun A. M., Angew. Chem. Int. Ed. Engl., 19, 675-
696 (1980).

Quina F., Abuin E. and Lissi E., Macromolecules, 23, 5173-5175 (1990).

Prochazka K., Kiserow D., Ramireddy C., Tuzar Z., Munk P., and Weber S. E.,
Macromolecules, 25, 454-460 (1992).

Kiserow D., Chan J ., Ramireddy C., Munk P., and Weber S. E., Macromolecules,
25, 5338-5344 (1992).

Bakeev K. N., Ponomarenko E. A., Shishkanova T. V., Tirrell D. A., Zezin A. B.,
and Kabanov V. A., Macromolecules, 28, 2886-2892 (1995).

Kramer M. C., Steger J. R., and McCormick C. L., in Multidimensional
Spectroscopy of Polymers: Vibrational, NMR and Fluorescence Techniques, M.
W. Urban and T. Provder Eds., ACS Symposium Series, 598, Washington DC, pp
379 (1995).

Maltesh C. and Somasundaran P., Langmuir, 8, 1926-1930 (1992).

Maltesh C., Somasundaran P., and Ramachandran R., J. Appl. Polym. Sci. .' Appl.
Polym. Sym., 45, 329-338 ( 1990).

Hu Y.-Z., Zhao C.-L., and Winnik M. A., Langmuir, 6, 880-883 (1990).

Winnik F. M., in Interactions of Surfactants with Polymers and Proteins, E. D.
Goddard and K. P. Anantpadmanabhan Eds., CRC Press, Ann Arbor, pp 367
(1992)

Winnik M. A. and Winnik F. M., in Structure Property Relations in Polymers:
Spectroscopy and Performance, M. W. Urban and C. D. Craver Eds., ACS
Advances in Chemistry Series, Washington DC, 236, pp 485 (1993).

Almgren M., Grieser F., and Thomas J. K., J. Am. Chem. Soc., 101(2), 279-291
(1979)

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

162

Yeung A. S. and Frank C. W., Polymer, 31, 2101-2111 (1990).
Zhao C.-L. and Winnik M. A., Langmuir, 6, 514-516 (1990).

Wilhelm M., Zhao C.-L., Wang Y., Xu R., and Winnik M. A., Macromolecules,
24, 1033-1040 (1991).

Astafieva I., Zhong X. F., and Eisenberg A., Macromolecules, 26, 7339-7352
(1993).

Astafieva 1., Khougaz K., and Eisenberg A., Macromolecules, 28, 7127-7134
(1995)

Nivaggioli T., Alexandridis P., and Hatton T. A., Langmuir, 11, 730-737 (1995).

Alexandridis P., Athanassiou V., Fukuda S., and Hatton T. A., Langmuir, 10,
2604-2612 (1994).

Alexandridis P., Holzwarth J. F ., and Hatton T. A., Macromolecules, 27, 2414-
2424 (1994).

Nivaggioli T., Tsao B., Alexandridis P., and Hatton T. A., Langmuir, 11, 119-126
(1995)

Rosen M. J ., Surfactants and Interfacial Phenomenon, Second Ed., John Wiley &
Sons, New York, 1989.

Clint J. H., Surfactant Aggregation, Chapman Hall, New York, 1992.

CHAPTER 9

CONCLUSIONS

In this research, two polymeric materials have been designed based upon the
complexation properties of poly(ethylene glycol), and their properties have been
investigated. These materials have several potential applications and may be employed
commercially to solve current engineering problems. During the course of this research,
several fundamental issues have also been addressed in order to understand phenomena at
the molecular level. Molecular modeling techniques have been employed for purposes of
visualization and the study of polymeric conformations. These studies have provided
validation of proposed hypotheses and have been used to guide the synthesis efforts.
Finally, this research has accomplished its objectives of developing strategies for the
synthesis and characterization of polymeric pseudocrown ethers and reversible

copolymeric emulsifiers based upon polymer complexation.

9.1. Summary of Results

9.1.]. Synthesis and Characterization of Polymeric Pseudocrown Ethers
Ion-induced Templatization of Oligomeric Ethylene Glycol Diacrylates. The
scheme for the synthesis of polymeric pseudocrown ethers is based upon the ability of

oligomeric ethylene glycol diacrylates to assume circular conformations when associated

163

164

with metal cations of appropriate size. Experimental studies demonstrated that certain
salts that were not soluble in nonpolar solvents, such as chloroform and methylene
chloride, became soluble upon the addition of oligomeric PEG 300. Further evidence of
cation binding by oligomeric PEG was obtained by 'H NMR studies of PEG and its
complexes with metal salts. Pyrene excimer fluorescence was also employed to
investigate the ion-induced cyclization of relatively short-chain, pyrene end-labeled
oligo(ethylene glycol) molecules. The excimer fluorescence results demonstrated that,
even in the absence of any templating cation, a significant excimer emission was
observed, indicating that the locally high concentration of the two pyrene moieties
attached to the same oligomer leads to intramolecular cyclization. The addition of
templating cations such as Ni” and Sn2+ lead to further enhancements in the excimer
fluorescence. These results suggest that the templating ion binds with the oligo(ethylene
glycol) chain and facilitates the formation of cyclic conformations which bring the two
chain ends into proximity. For pyrene end-labeled penta-ethylene glycol, the excimer
fluorescence enhancement induced by nickel ions was more pronounced in chloroform
than in tetrahydrofuran. Therefore the pyrene excimer-to-monomer fluorescence ratio
was found to provide a sensitive measure of the ability of the oligo(ethylene glycol)
mixture to solubilize the ions into an organic solvent.

Molecular Modeling Studies. In an effort to optimize the template ion synthesis
approach, molecular modeling studies were performed on oligomeric PEGDA containing
between two and ten ethylene glycol repeating units, with and without cations.
Simulation results indicated that the templating cation significantly decreased the mean

end-to-end distance, thereby bringing the unsaturated end-groups into close proximity.

165

Although the presence of the cation decreased the ETED for all PEGDA chain lengths,

the PEGDA ligand that resulted in the most effective templatization for Na+ contained
four ethylene glycol repeating units. The simulation time had little effect on the results
for relatively short PEGDA ligands containing seven or fewer ethylene glycol repeating
units. These ligands are highly constrained due to binding interactions with the ion. A
higher degree of variation with simulation time was observed for longer ligands in which
only a portion of the chain interacts with the ion. These simulation results have affirmed
the role of a templating ion in the synthetic scheme. Moreover, the results provide insight
into the selection of the templating ion and the PEGDA oligomer that will maximize
templatization by bringing the unsaturated end groups into proximity.

In addition, molecular dynamics simulations of pyrene-labeled oligo(ethylene
glycol) ligands were performed to provide molecular insight into the fluorescence results
by elucidating the effects of the templating ion and/or the pyrene end groups on the
conformations adopted by the oligo(ethylene glycol) chains. Simulation results indicated
that, even in the absence of a templating cation, the pyrene end-labeled oligomers exhibit
an enhanced tendency to form cycles compared to the analogous unlabeled ligands. This
result is consistent with the fluorescence results in which significant excimer formation
was observed even in the absence of any cation. Simulation results further indicated that
the addition of a templating cation leads to a further reduction in the mean end-to-end
distance for pyrene end labeled tetra-ethylene glycol in a manner consistent with the

observed enhanced excimer fluorescence.

166

Ion-binding Studies. The synthesis of a pseudocrown ether hydrogel with
hydroxy ethyl methacrylate as the comonomer and nickel as the template cation was
discussed. Based upon earlier studies, poly(ethylene glycol)-diacrylate 200 with an
average of 4.5 ethylene glycol repeat units, was chosen as the templating oligomer.
Hydrogels with hydroxy ethyl methacrylate as the comonomer and PEGDA 200 in the
absence of any templating cation were also synthesized as a “control” for comparison
purposes. The conditioning, ion-binding properties and regeneration of the PCEs was
discussed. The monomer to solvent ratio during synthesis were varied from 20/80 to
50/ 50 and its effect on the ion-binding capacity was investigated.

The PCE synthesized with a monomer to solvent ratio of 20/80 was observed to
possess the highest ion-binding capacity (0.3 mg Nip/g polymer) which is comparable to
capacities of other polymer bound crown ethers reported in the literature. The binding
capacity of the PCEs was observed to decrease with an increase in the monomer to
solvent ratio during synthesis. This was attributed to a possible decrease in
templatization efficiency and inter-oligomer interactions resulting in greater crosslinking
at the expense of pendant crown ethers. Also, the effect of the template ion on the ion-
binding capacity of the PCB was demonstrated by comparing the binding capacity of the
PCB with that exhibited by a p(HEMA) “control” hydrogel synthesized without any
templating cation. The PCE exhibited an enhanced ion-binding capacity for nickel.
Finally, an efficient regeneration scheme was optimized which resulted in complete
regeneration of the PCB.

These studies have demonstrated the promise of the template ion induced

cyclization scheme for the synthesis of polymeric pseudocrown ethers. With an easy one

 

167

step-synthesis scheme and comparable binding properties for Ni“, this research could
result in the development of an inexpensive resin for the binding of nickel from aqueous
waste streams. Further, this scheme could be extended to synthesize PCEs for the
binding of other toxic heavy metal ions of interest such as chromium, copper, lead and

zinc, among others.

9.1.2. Synthesis and Characterization of Reversible Block/Graft Copolymeric Emulsifiers
Based Upon Polymer C omplexation

Swelling Properties of Poly(methacrylic acid-g—ethylene glycol) Hydrogels. To
further the understanding of reversible complexation between poly(methacrylic acid) and
poly(ethylene glycol), the equilibrium swelling of crosslinked PMAA-g-PEG hydrogels
was investigated. The swelling studies were performed in increasing aqueous
concentrations of the solvents and the equilibrium solvent weight fraction was calculated.
Swelling behavior in a homologous series of alcohols (methanol, ethanol and n-propanol)
demonstrated the effectiveness of a solvent in disrupting complex and stabilizing the
hydrophobic interactions. The ability of a solvent to break the complex (at low solvent
concentrations) was related to the ease with which it could compete in the formation of
hydrogen bonds with the PMAA backbone. However, at high solvent concentrations, the
degree of dissociation of the PMAA may govern the equilibrium degree of swelling. For
alcohols with longer alkyl groups (n-propanol) the degree of swelling decreased gradually
at higher solvent concentrations. The ratio of monomer to solvent during synthesis
affected the maximum equilibrium degree of swelling attained by a hydrogel. Hydrogels

synthesized in a 20/80 monomer solvent ratio achieved a higher degree of swelling which

168

may be attributed to a lower degree of crosslinking and entanglement. Comparisons
between the swelling behavior of a crosslinked homopolymer of PMAA with a P(MAA-
g—EG) gel in methanol and acetone demonstrated the effect of complexation as evidenced
by the difference in the equilibrium degree of swelling at low solvent concentrations.
These studies also illustrated the effect of dissociation of the PMAA on its swelling
behavior at high solvent concentrations. A lowering in the degree of swelling at high
concentrations of acetone or n-propanol may be attributed to a lower degree of
dissociation of the polyacid (PMAA).

Molecular simulations demonstrated the existence of a 1:1 complex stoichiometry
between complexing PMAA and PEG segments. Molecular mechanics simulations on
PMAA and PEG segments with 4, 8 and 14 repeat units exhibited the formation of stable
conformations with a 1:1 complex stoichiometry.

Synthesis and Emulsification Properties of Reversible Block/Grafi Copolymers
Based Upon Polymer Complexation. The synthesis and emulsification behavior of novel
block/graft copolymeric emulsifiers based upon a reversible hydrophobic architecture has
been presented. In these systems the graft copolymer assumes a multiblock copolymer
configuration under certain conditions of pH. These copolymers exhibit reversible
emulsification behavior and may be useful in a variety of applications. Emulsification
studies demonstrated that these copolymers can emulsify oil under acidic conditions in a
reversible fashion. As the pH of the emulsifier solution is raised towards neutral the
percent oil emulsified decreases, and under basic conditions no emulsification is
observed. Interfacial tension studies illustrated the surface active nature of the

copolymers under acidic conditions. A lowering of the interfacial tension was observed

169

under acidic conditions for the copolymeric emulsifiers under investigation. These
properties exhibited by the block/graft copolymers can be attributed to the formation of
hydrogen bonded hydrophobic complexes between the methacrylic acid backbone and the
ethylene glycol repeat units on the graft. This may result in an alternating hydrophilic-
hydrophobic pseudo-multi block structure which may be responsible for the surface
active properties of the copolymers.

Aggregate Formation by Reversible Copolymeric Block/Graft Emulsifiers.
Fluorescence spectroscopy has been used to study the formation of hydrophobic
aggregates. The polarity sensitive fluorescence of solubilized pyrene was used to study
the aggregate formation as a function of pH and concentration. Both the excitation and
emission spectra of pyrene has been used to estimate the critical aggregate concentration
for 10:1 and 20:1 copolymers. Specifically, the 1337/1342, ratio in the excitation spectrum
and the 1373/1389 ratio in the emission spectrum was plotted as a function of pH for various
copolymer concentrations. As the copolymer concentration was increased, the excitation
and emission ratios indicated increased partitioning of the pyrene into a hydrophobic
domains. This may be interpreted as direct evidence of the increase in the number of
aggregates formed as a result of an increase in copolymer concentration. Further, the
critical aggregate concentration for the 10:1 and 20:1 copolymers was estimated from
both the excitation and emission spectra to be 8 x 10“ wt.% and 2 x 10'3 wt.%

respectively.

CHAPTER 1 O

RECOMMENDATIONS FOR FUTURE WORK

Two novel materials have been developed based upon polymer complexation.
While a significant contribution has been made in the understanding of fundamental
concepts which led to the design of these materials, there is considerable potential for
further investigations. Future work that is appropriate may be divided into two broad
categories; further investigations into the synthesis and characterization of the materials
and theoretical modeling studies for visualization of molecular behavior and
conformations. In addition, other potential applications of these materials should be
addressed based upon their unique properties. In this chapter, some recommendations for
future work will be made for both the polymeric pseudocrown ethers and the reversible

block/graft copolymer projects.

10.1. Synthesis and Characterization of Polymeric Pseudocrown Ethers

The synthesis and ion-binding properties of polymeric pseudocrown ethers
(PCES) have been investigated. The ion-induced templatization of oligomeric ethylene
glycol diacrylates has been studied by ion-solubilization, NMR spectroscopy, molecular
modeling and pyrene end-labeling studies. In addition, a synthetic scheme has been

developed based on a free-radical photo-polymerization for the synthesis of the

170

171

pseudocrown ethers. However, further investigations need to be carried out in each of the

above mentioned aspects of this research.

10.1.1. Ion-induced Templatization Studies

Studies performed resulted in the selection of nickel and chromium as the two
heavy metal candidates for the template ion-induced synthesis of PCES. This selection
was based on ion-solubilization studies of salts of nickel and chromium including
chlorides and nitrates of nickel(II), chromium (III and VI). Studies performed focused on
PEGDA 200 as the oligomer of choice due to theoretical considerations based upon the
optimum number of ethylene glycol units that may be required for templatization which
resulted in the end-groups being in close proximity. Other salts and cations that should
be investigated include lead, copper, tin and other platinum group metals. In addition,
metal cations such as technetium and ruthenium may be good candidates in an attempt to
apply this methodology to study the binding of radioactive metals such as plutonium and
uranium. Experiments to be performed may include the determination of the optimum
cation-oligomer combination that results in maximum templatization. The experiments
should include ion-solubilization and NMR spectroscopy to determine the extent and
nature of cation binding. Also, the effect of the counterion on templatization has not been

addressed in this study and needs to be investigated.

10.1.2. Molecular Modeling
Molecular simulations may be performed on each of the systems under

investigation to visualize the configurations adopted by the oligomer in the presence of a

172

cation. End-to-end distance studies performed using molecular dynamics (MD)
simulations will yield valuable information on the optimum oligomer chain length
required for templatization. However, when performing MD simulations, the inclusion of
solvent would account for oligomer solvent interactions. As a result the simulation time
would have to be increased to a number which allows for the equilibration of the system.
Simulations time of 200ps+ would be suitable for these studies. However, a final
assessment can be made only after examining a plot of energy versus time.

The estimation of binding energies is possible from the molecular dynamics
simulations. The binding energy per mole may be estimated as the difference between
the average total energy for the oligomer and the energy of the oligomer with the cation.
These values may be compared to those reported in the literature or with the binding
energy values obtained from experimental binding constant data.

MD simulations performed have primarily employed sodium, aluminum,
ruthenium, technetium and copper. However, other metal cations such as chromium and
nickel that were used in the synthesis of PCEs were not available on POLYGRAF for
modeling purposes. The new version of POLYGRAF currently available (CERIUS2,
Molecular Simulations Inc.) should be used in further simulations since it may include

several other metal cations and contains an efficient and faster MD module.

10.1.3. Synthesis and Ion-Binding Studies
Based upon the ion-solubilization studies, nickel was chosen as the template
cation and PEGDA 200 (with an average 4.5 ethylene glycol repeat units) was chosen as

the oligomer that would optimally template the cation. Synthesis and ion-binding studies

173

were performed on PCEs based upon this model system. In addition, the ratio of PEGDA
200 to the comonomer hydroxy ethyl methacrylate (HEMA) was chosen to be 5 wt.%.
This was based upon a 20/80 monomer to solvent ratio during synthesis. It may be
possible to increase the PEGDA 200 concentration to 10 wt.% or more if a 10/90
monomer to solvent ratio (or lower) is chosen. That may result in PCEs with higher
binding capacities.

Other possible investigations include the use of other hydrophilic comonomers
such as hydroxy ethyl acrylate, carboxylic monomers such as acrylic and methacrylic
acid, and other hydrophilic monomers that may be readily polymerized with the
templated PEG diacrylates. Requirements for the comonomers include hydrophilicity,
strength in the polymerized form, and non-interference with the templatization process. It
would be interesting to observe the characteristics of a PCE with an ionogenic backbone
comprising acrylic or methacrylic acid. That may lead To thyevelopment of conductive
polymers and enhanced cation uptake.

Ion-binding studies were performed using atomic absorption (AA) spectroscopy
and a difference technique was used in order to determine the ion uptake. While AA
spectroscopy is a very sensitive technique, it requires calibration before each run and is
sensitive to impurities. Further, since the ion-uptake per gram of polymer being
investigated is very small, other techniques such as ICP mass spectroscopy may prove to
be more accurate for the determination of cation concentrations in aqueous solutions.

The present study investigated the binding properties of PCB synthesized with
nickel as the templating cation. The ion-binding studies performed considered only the

binding of nickel to the PCE. Further studies should include PCEs synthesized with other

174

cations as template ions based upon the strategy presented in this research. In addition,
binding studies should include several cations in order to determine the selectivity of
these materials. The hypothesis that the PCE will be selective for the template cation

should be tested conclusively.

10.2. Synthesis and Emulsification Properties of Reversible Block/Graft
Copolymeric Emulsifiers Based Upon Polymer Complexation

Reversible block/graft copolymers were synthesized based upon the ability of
poly(ethylene glycol) (PEG) to form hydrophobic complexes with hydrophilic polymers
such as poly(methacrylic acid) (PMAA). Graft copolymers with a PMAA backbone and
PEG 1000 (containing on an average 23 EG repeat units) grafts were synthesized by a
one-step free-radical polymerization. The copolymers behaved as emulsifiers under
complex promoting (acidic) conditions while under conditions when the complex was
broken, the surface active behavior was also lost. Oil emulsification and interfacial
tension measurements demonstrated their surface active behavior. The polarity sensitive
fluorescence of pyrene was used to study aggregate formation and to estimate the critical
aggregate concentration (CAC). Studies that may be useful in further understanding the
emulsification behavior and aggregate formation by these copolymers are addressed

below.

10.2.1. Study of Reversible Complexation
The fundamental basis behind the reversible nature of these materials is the

formation of a hydrophobic complex by two constituent hydrophilic polymers.

175

Complexation of PMAA with PEG has been investigated by several researchers and the
variables that affect complexation have also been discussed. However, experimental
evidence that points to the formation of 1:1 complexes between MAA and EG repeat
units has not been supported conclusively via a molecular picture. Molecular dynamics
simulations may be performed on PMAA-g—PEG grafi copolymers in the presence of
solvent to study the complex stoichiometry. Molecular Simulations Inc. software
(POLARIS or CERIUSZ) may be used to model polymers in solvent. These studies will
provide a better understanding of the complex and the effect of solvent and pK, of the
acid on complexation. Further, two-dimensional lH or l3C NMR studies will be useful in
studying complexation of a PMAA-g-PEG copolymer. By performing 2-D 'H NMR on a
sample in water under complex promoting conditions using a suitable pulse sequence and
water-suppression, it may be possible to investigate the complex by evaluating the

backbone acid proton peaks.

1 0. 2. 2. Copolymeric Emulsifier Synthesis

The PMAA-g-PEGIOOO graft copolymer was synthesized by a free-radical
solution polymerization. Variables that may be further investigated should include the
composition of the solvent, choice and concentration of initiator, MAA to EG repeat unit
ratio, and the length of the PEG graft. All of these variables will have an effect on the oil
emulsification and aggregate formation behavior of the copolymer. For example, the
effect of MAA to EG ratio has been investigated in this study and has been found to

impact the CAC of the copolymer solutions under acidic pH. Finally, the synthesis

176

should be attempted under complex promoting conditions to determine the effect on the

structure of the graft copolymer that is obtained.

10. 2. 3. Inclusion of Comonomers

Studies dealing with the inclusion of comonomers to impart additional
hydrophilic or hydrophobic properties to the PMAA-g—PEG copolymer, should be
conducted. For example, a copolymer synthesized with a MAAzEG ratio of 1:1 and
containing a few mole percent of a hydrophobic oligomer such as lauryl methacrylate
would result in a graft copolymer that would be primarily hydrophobic under complex
promoting conditions and would result in a graft copolymer with a hydrophilic backbone
and both hydrophilic and hydrophobic grafts under basic conditions. Such a copolymer
would behave as a graft copolymeric emulsifier under basic conditions and its surface
active nature would be lost under acidic or complex promoting conditions. Preliminary
studies in our laboratory have demonstrated the viability of such a scheme in which the
emulsification behavior of the copolymer would be exactly opposite to that of the

PMAA-g-PEG copolymers synthesized with MAA to EG ratios greater than unity.

10. 2. 4. C opolymer Characterization and Emulsification Studies

The PMAA-g-PEG copolymers investigated in this research need to be
characterized. Determination of the molecular weight distribution of these materials
could lead to an understanding of the emulsification behavior of these materials. The
universal calibration technique may be used to determine the molecular weight

distribution of the graft copolymers by gel permeation chromatography (GPC). In

177

addition, molecular weight determination is also possible by first hydrolyzing the grafts
followed by determination of the backbone molecular weight since the MW of the grafts
would already be known. Proton NMR spectroscopy along with 13C NMR would be
useful in characterizing the extent of grafting and determining the number of grafts per
chain in conjunction with the molecular weight distribution.

The oil emulsification studies presented in this thesis are adequate in determining
qualitatively the emulsification ability of a particular copolymer solution under acidic
conditions and for quantitatively estimating the amount of oil emulsified as the pH is
raised. However, studies to be performed in the future should include emulsification
studies for the determination of: i) the maximum amount of oil emulsified under acidic
conditions as a function of copolymer concentration, ii) stability of the emulsion, iii)
compositions of the aqueous and emulsion phases, and iv) nature of the emulsion phase
under all conditions (whether oil or water continuous). Light scattering studies may be
useful in determining the size of the polymeric aggregates under various conditions of
pH. Centrifugation studies may be used to study the stability of the emulsions.

Another important aspect that needs to be addressed is the surface active behavior
of PMAA alone as a “control.” While PEG does not exhibit any appreciable surface
activity, PMAA has been known to form hydrophobic aggregates under low pH
conditions. This is due to the tight coiled conformations which arise from protonation of
the acid. However, when the acid is neutralized (the pH is raised) the PMAA coils
expand due to charge-charge repulsions. The contribution of the coiling and uncoiling of
the PMAA as a function of pH and its oil emulsification properties are currently being

investigated. The formation of hydrophobic aggregates by PMAA may be responsible for

178

its ability to behave as a surface active material. Hence, a thorough investigation needs to

be conducted to study the properties exhibited by PMAA alone.

10.3. Alkali-Swellable Thickeners Comprising Poly(methacrylic acid-g-ethylene
glycol) Copolymers Based Upon Polymer Complexation

Potential new applications for block/graft copolymers comprising PMAA
backbones and PEG grafts include alkali-swellable thickeners. Alkali-swellable or alkali
soluble thickeners (ASTs) have found wide application for thickening paint, coating,
textile, consumer product, and adhesive formulations. In general, these thickeners exhibit
low viscosity under acidic conditions, but high viscosity under basic conditions making
them easy to manufacture and blend into formulations while providing excellent
thickening properties. As described in several reviews and monographs“2 these
thickeners are typically produced by emulsion polymerizations of acrylic or methacrylic
acid with a hydrophobic monomer such as ethyl acrylate. The acid groups may be
positioned on the surface of the beads by semi-batch or multi-stage addition of the
hydrophilic monomer in the latter stages of an emulsion polymerization.3'4 In general,
these polymers contain more than 40 wt.% of the hydrophobic monomer. For example,
Murakami et al.2 used compositions containing 48-51 wt.% of the hydrophobic monomer;
Rodriguez and Wolfe5 report systems containing 66 wt.% of the hydrophobic ethyl
acrylate monomer or 33 wt.% of the hydrophilic methylmethacrylate monomer, while
Duprey6 suggests that at least 30% alkyl methacrylates with one to four carbon atoms
must be used. Shay7 recommends compositions with 15 to 50% carboxy monomer, 10%

surfactant monomer, and the balance a hydrophobic (ethyl acrylate) monomer. With high

179

amounts of the hydrophilic acidic monomer, it is often necessary to include crosslinking
agents to maintain insolubility and high thickening efficiency.2

Water-insolubility in the application described above is provided via hydrophobic
comonomers or functional groups (such as aliphatic esters of acrylic or methacrylic acid).
Thickeners contain hydrophobic comonomers to provide water insolubility at acidic pH.
This class of materials have never been synthesized without hydrophobic comonomers or
hydrophobic functional groups. Thickeners which exhibit very low solution viscosities
under acidic conditions while exhibiting an order of magnitude increase in viscosity
under basic conditions are been currently investigated. These thickeners are graft
copolymers of acrylic or methacrylic acid with oligomeric ethylene glycol grafts.
Carboxylic acids such as acrylic or methacrylic acid (Lewis acids) form strong
hydrophobic hydrogen-bonded complexes with Lewis base functionalities such as the
ether oxygens of polyethers on polyethylene glycol (or oxide). The mechanism of these
interactions and the nature of the complex are well understood (Scranton et al.“‘9 and
references therein). Under complex promoting conditions, the acid is protonated and the
grafts align along the backbone due to hydrogen bonding resulting in a coiled
conformation as shown in Figure 10.1. However, when the pH is raised the acid groups
are neutralized and the complex is broken resulting in a net volume expansion of the
coiled polymer due to charge-charge repulsion and mobility of the graft. The class of
polymers discussed above can be readily synthesized by a single step free radical
polymerization reaction. However, alkali swellable microparticles consisting of a similar
system can also be produced by a novel emulsion polymerization reaction of the two

hydrophilic monomers.

180

 

 

(A) __ Hydrophilic Backbone (B)
.................. Hydrophilic Graft

Figure 10.1. Conformational changes resulting from going from an acidic (A) to a basic
(B) environment causing the polymeric chain to expand.

181

Viscosity studies performed on a Brookfield viscometer for PMAA-g-PEGIOOO
solutions for two solution concentrations as a function of pH are shown in Figure 10.2.
The data shown in this figure is for a 1:1 MAA to EG copolymer of PMAA-g-PEGIOOO.
The symbols represent data while the line drawn through them is to illustrate the trend.
These studies demonstrate a large increase in viscosity as a function of pH for both the
0.05 wt.% and the 0.1 wt.% copolymers. This increase is however much more significant
for the 0.1 wt.% copolymer in which the viscosity increases from about 1-2 cP under
acidic conditions by over two orders of magnitude to ~ 700 cP at around neutral pH.

Clearly, great potential exists in the development and further investigations into
the synthesis and viscosity studies of PMAA-g-PEGIOOO copolymers with a 1:1 MAA to
EG ratio. These copolymers may also be synthesized with other comonomers to impart
hydrophobic or hydrophilic character to the thickeners. A host of applications exist for
alkali-swellable thickeners and further studies into the synthesis and characterization of

these materials should be undertaken.

182

 

 

800-
—-— 0.05 wt.% 1:1.2 PMAA-g-PEG1000
. —o—O.1wt.% .
E
9, 600-
E
U)
0
8
'5 400‘
2
(D
h:
.92
o 200‘
8
m
______._.———I
0___.=__,__.—qz—L , . , -.—=+—4
2 4 6 8 10 12

pH

Figure 10.2. A plot of Brookefield viscosity as a function of pH for 1:1 PMAA-g-
PEG1000 (with a MAA to EG ratio of 1:1) copolymer solutions at two
concentrations.

183

10.4. References

Shay G.D., Advances in Chemistry, 223, 457 (1989).

Murakami T., Fernando R.H., Glass J.E., Surface Coating International, 76, 8
(1993).

Eisenhart E.K. et al., US. patent 5266646 (1995)

Eisenhart E.K. et al., US. patent 5451641 (1995).

Rodriguez BE, and Wolfe M.S., Macromolecules, 27, 6642 (1994).
Duprey J., US. patent 4351754 (1982).

Shay G.D. et al., US. patent 4801671 (1989).

Scranton A.B., Klier J., Aronson C.L., in Polyelectrolyte Gels, R. S. Harland and
R. K. Prud’homme, Eds., ACS Symposium Series 480, American Chemical
Society, Washington, DC, pp 171 (1992).

Klier J ., Scranton A.B., and Peppas N.A., Macromolecules 23, 4944 (1990).