unriwumfi n. .53“;
7. .3 n»...
«m 54me

#Emiiagaé

 

 

n

”mg

1.3.“.

J:

 

Ir
8

IL
.3“

Mtg

I
9a?
5? , J's/‘6‘ 3 1

This is to certify that the
thesis entitled

SYNTHESIS ANDIMORPHOLOGY OF MALEIMIDE VINYL
ETHER ALTERNATING COPOLYMERS IN THE SOLUTION
PHASE

presented by
RICHARD M. BELL III

has been accepted towards fulfillment
of the requirements for the

MS. degree in Chemistry

 

 

 

Major Proféssor’s Signature

57/ 3/67

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DatoDuopssop. 1 5

 

SYNTHESIS AND MORPHOLOGY OF MALEIMIDE VINYL ETHER
ALTERNATING COPOLYMERS IN THE SOLUTION PHASE

By

Richard M. Bell III

A THESIS

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

Master of Science

Department of Chemistry

2004

ABSTRACT

SYNTHESIS AND MORPHOLOGY OF MALEIMIDE VINYL ETHER
ALTERNATING COPOLYMERS IN THE SOLUTION PHASE

By

Richard M. Bell |I|

Maleimide vinyl ether alternating copolymers were synthesized according
to procedures established by the Blanchard group previously. The monomer
reactivity of a 3 monomer system was studied by 1H NMR to establish if one
monomer was more reactive than the other. N-phenylmaleimide monomer was
preferentially incorporated into the polymer chain relative to N-pyrenylmaleimide
monomer. The resulting polymers were then studied using steady state
fluorescence spectroscopy, and the results of this work were compared to a
theoretical model to investigate the morphology of the polymers in various
solvents. These studies suggest the polymer is coiled in solvents in which it is
minimally soluble. As the solubility of the polymer in the solvent increased, it

adapted a more linear conformation.

for Jimmy Wicks (October 9, 1955 - November 19, 2003)

ACKNOWLEDGMENTS

I would like to thank Gary Blanchard for his assistance in my research and
the writing of this thesis. He has worked tirelessly in editing and advising on this
document and I could not have done it without him. He has been a constant
source of motivation in my research always encouraging me to pursue every
detail and leave no stone unturned.

I also must thank group members past and present. Times can be hard
when there are so many people in a lab but I always felt everyone was motivating
me to succeed even when the motivation was a bit aggressive. I would
especially like to thank Dr. Jaycoda Major whom started this project and provided
a great deal of preliminary research on the topic.

My family and friends have been a constant source of support for me.
They have always believed in me and knew I could do it when at times I felt I
couldn’t. I hope my example will motivate them to try things they never thought

they could do and succeed where success may seem impossible.

TABLE OF CONTENTS

List of Figures ....................................................................................................... vi
Chapter 1. Introduction .......................................................................... 1
Chapter 2. Experimental ....................................................................... 17
Chapter 3. Results and Discussion ......................................................... 25
Chapter 4. Conclusions and Future Work ................................................. 50

Figure 1.1

Figure 1.2

Figure 2.1

Figure 2.2

Figure 2.3

Figure 3.1

Figure 3.2

LIST OF FIGURES

Figure of polymer in linear conformation and coiled. In a coiled

conformation side groups may interact ............................................ 9

Representation of how polymers are distributed on a silicon

surface ............................................................................. 12

Synthesis of substituted amine. R can be pyrene or another

functionality we desire ......................................................... 18

Synthesis of maleimide vinyl ether polymer using a three monomer
system. The coefficients n and m can vary based on solution
stiochiometry and do not imply the stiochiometry as a

whole ............................................................................... 20

1H NMR of MVE polymer synthesized with N-Pyrenylmaleimide, N-

phenylmaleimide and 1,4-butanediol vinyl ether ........................ 21

1H NMR spectrum of polymer containing 3:1 phenylzpyrene at 3 hrs

of reaction. Numbers above peaks are integrations .................. 27

1H NMR spectrum of polymer containing 3:1 phenylzpyrene at 6 hrs

of reaction. Numbers above peaks are integrations .................. 28

vi

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6

Figure 3.7

Figure 3.8

Figure 3.9

Figure 3.10

Figure 3.11

Figure 4.1

1H NMR spectrum of polymer containing 3:1 phenylzpyrene at 9 hrs

of reaction. Numbers above peaks are integrations .................. 29

PhenylzPyrene ratios on polymer chain as analyzed by 1H NMR

over time .......................................................................... 32

PhenylzPyrene ratios on polymer chain as analyzed by 1H NMR

over time .......................................................................... 33

Pyrene monomer and excimer bands at high (0.05M) and low

(0.6uM) concentrations ........................................................ 38

Fluorescence spectra of pyrene polymers with various pyrene

loading density in various solvents ......................................... 39

Excimer intensity of pyrene containing polymers in cyclohexane..41

Excimer intensity of pyrene containing polymers in acetonitrile....42

Excimer probabilities as a function of pyrene loading ................. 45

Comparison of experimental and statistical results .................... 47

Beer’s Law plot for N-pyrenylmaleimide. Extinction coefficient is

equal to slope .................................................................... 52

vii

Figure 4.2 Beer’s Law plot for N-phenylmaleimide. Extinction coefficient is

equal to slope .................................................................... 53

Images in this thesis are presented in color

viii

LIST OF ABBREVIATIONS

AIBN 2-2’azobisisobutyronitrile

DMSO Dimethylsulfoxide

GPC Gel Permeation Chromatography

HPLC High Performance Liquid Chromatography
MVE Maleimide vinyl ether

QCM Quartz Crystalline Microbalance

ZP Zirconium Phosphonate

INTRODUCTION

Chemical separations are an essential part of academic and industrial
chemistry. Academically they provide fascinating and challenging intellectual
problems.1 In industry chemical separations are essential for the purification of
products ranging from petroleum fractions to pharmaceuticals and for the
removal of toxins from waste water systems or other environmental processes.
In the semiconductor device industry, purification of elemental silicon is a
prerequisite for all subsequent fabrication steps.

Species in separations are partitioned between two distinct phases. For
example, in High Performance Liquid Chromatography (HPLC) a separation
occurs when the fraction of time spent in the stationary phase is different for two
species. The chemical potential differences between these two phases are
distinct and discontinuous. The Blanchard group seeks to separate analytes by
creating a chemical potential gradient of molecular dimension normal to the
stationary phase surface. Using this chemical potential gradient, separation
would occur with small changes in chemical potential as compared to other
analytical separations. In essence, we seek to create layered interfacial
structures where separation occurs by a chemical potential gradient rather than a
simple step function.

This thesis investigates the synthesis of a family of polymers which can be
used to form chemical potential gradients on a variety of substrates. Chemical
potentials are different for each analyte and the identity of each polymer layer.

The methods by which these interfaces are synthesized afford layer-by-layer

means by which one can control the chemical potential profile at the resulting
interface(s). Chemical potential can be controlled by substitutions made during
polymer synthesis and deposition of these polymers on a surface by established
methods?"3 This work focuses on the synthesis and morphology of MVE
(maleimide vinyl ether) polymers in the solution phase. The research can be
applied to a more detailed understanding of the morphological properties of these
polymers and lead to an understanding of how these polymers may be adsorbed
onto a substrate surface.

It is important to understand what a chemical potential gradient is and how
is it possible to synthesize one. In practice we are trying to create a multilayer
system on a surface. These multilayers consist of polymers where the
concentration of side groups is adjusted to affect the chemical potential as an
analyte encounters each layer. Factors such as dipole moment or dielectric
constant of the polymers can influence the interactions of the analytes within the
layer. Steric issues can be examined as well through the appropriate choice of
polymer side groups and the use of cross-linking chemistry within and between
polymer layers.

A multilayer system can be constructed on a variety of substrates (Au and
SiOx) and polymer properties could be used to make a gradient normal to the
surface. Synthesis of MVE polymers and bonding to a substrate surface has
been documented in our group previouslyz'“5 and work has also been done to
control the adsorption and desorption of methanol and cyclohexane on a layered

polymer film.7 We build on this work by seeking a deeper understanding of the

synthesis of MVE polymers and their morphology in the solution phase. With this
knowledge, the bonding of the polymers to a substrate surface will be better
understood and this understanding will be used to construct chemical potential
gradients. To reach this goal we seek to create a chemical potential gradient on
a polymer surface and control this gradient based on the choice and order of the
polymer layers. An understanding of chemical potential is the first step.

The change in standard state chemical potential associated with the
partioning of an analyte between two phases can be given by:

U°A<1)- lJ°A(2) = -RT ln CA(1)/CA(2) = 'RT In K (1)
where (PM) and mm) are the chemical potentials of the analyte A in phases 1
and 2. The terms cm) and cm) are the concentrations of analyte A in phases 1
and 2. K is the equilibrium constant which can also be described by the rate of
adsorption, kg, and desorption, kd, of the analyte between phases 1 and 2.

In a chemical separation the equilibrium constant K' for partitioning
between two phases is determined by the ratio of the adsorption and desorption
rate constants, k12’ and k21'. Traditional systems such as gas and liquid
chromatography do not allow substantial control of these rate constants unless
the composition of one of the phases is changed. Changing the identity of one
phase can be less molecule specific than changing the interface between two
phases in a chemical potential gradient. A chemical potential gradient has the
potential to give more degrees of freedom in chemical separations.

Using a structural gradient at the phase interface can afford greater

control over the separation process. If we think of a multilayered system the

adsorption and desorption rate constants can be called kjj and kji for each
adjacent layer pair, i and j. Thus the equilibrium constant would be:

K = kijlkj.
for partitioning between the i and j layers of an interface. If we were to use a two
phase system the equilibrium constant would be:

K = k12/k21
or for a surface comprised of four layers:

K = k12k23k34k45k46/k21k32k43k54k65 = K1 K2K3K4K5
We intend to control the polymer layers independently and thereby control the
adsorption and desorption rate constants of each layer independently. Using this
system, we can influence the kinetics of separation. This method is
advantageous in that we can control the identity of each layer and thereby affect
the separation substantially. Therefore, we seek to construct a chemical
potential gradient on a surface where we can control the identity of the layers and
the separation.

Our group uses MVE alternating copolymers for this research. We have

garnered expertise with these polymers and the ability to bond them to a surface

either ionically‘I's'w“6

or covalently.”17 Polymerization has been accomplished
through photo, thermal and radical initiation. Photoinitiation has been
accomplished through UV irradiation of a polymer matrix containing di(4-
vionxloxybutyl) succinate and 1,5 bis(maleimide)-2-methyl pentane.5 The same

polymerization has been accomplished through thermal initiation when the

temperature was raised to 60°C in CHCI3 (b.p. 605°C) for 24 hours. During

these polymerizations addition across the maleimide occurs both cis and trans,
which is advantageous for the multilayer bonding scheme we wish to use in our
research. The vinyl ether can also have a terminal phosphonate that allows ionic
bonding to a substrate.6

This research has focused on covalent bonding because of the avoidance
of polarizability of the Zirconium-Phosphonate(ZP) layer which can occur with an
ionic bonding scheme. This polarizability can interfere with multilayer
measurements which makes covalent chemistry an interesting alternative. It is
also possible to use mixed layering of ionic and covalent chemistry‘ but the
dominant technique in our research has been covalent chemistry.

By reacting an amine with maleic anhydride, the amic acid can be formed
and then closed forming a substituted maleimide. The resulting substituted
maleimides can be polymerized and covalently bound to a surface.”3 Using a
substituent such as pyrene we can study the spectroscopy of the polymers in
solution and on a surface. We have also investigated the adsorption of gases
such as hexane and methanol using a Quartz Crystal Microbalance (QCM).7
Data from these measurements show that our multilayer assemblies can in fact
alter the adsorption behavior of surfaces.

My research has focused on studying the synthesis and morphology of a
MVE polymer in the solution phase. MVE polymers are synthesized by radical
polymerization‘T 1‘ using the radical initiator 2-2’-azobisisobutyronitrile (AIBN). In
this scheme the AIBN is heated and splits into two radical initiators. The radical

attacks one monomer at the double bond creating a monomer radical. The

monomer radical attacks another monomer and a polymer begins to form. This
oligomer (which has a radical at the end of it) will then attack other monomers
lengthening the polymer chain. This reaction continues, creating an alternating
copolymer.

To better understand the polymer, it is necessary to understand what
chemical reactions are contributing to the growth of the chain. The two main
variables in a polymerization reaction are monomer and initiator concentration.
The monomer concentration is important because there must be enough
monomer units activated to avoid chain termination effects. For example, if in a
solution 3 chains were activated then these may be terminated very quickly.
Termination occurs when a radical is in solution and reacts with a monomer
radical thereby making it unreactive in a radical polymerization or can occur if two
monomer radicals react with one another. If there are many chains being
activated these termination effects become negligible and polymers begin to
form. The other factor, initiator concentration, works in much the same way. It
activates the monomers for polymerization but it cannot be used in great excess
if a long polymer is desired. If the mixture is thought of as a finite solution of
monomers too much initiator will be detrimental to the polymerization process.
This excess initiator concentration creates too many starting chains and will
create polymers with weights smaller than if a lower concentration of initiator is
used. A lower concentration will create fewer chains and they will grow using the
stock monomers available to them. With the fewer chains the polymers can grow

to longer lengths in the same amount of monomer stock solution. A balance is

necessary in choosing a starting initiator concentration. With a low concentration
of initiator the time of polymerization will increase. Therefore the amount of
initiator used must be small enough to make polymers of sufficient length but not
so low the polymerization process cannot be completed in a reasonable time.

At this point the polymers have been created and it can now be thought of
how to bond them to the surface of a substrate. There are two possible methods

of bonding polymers to a surface; ionically“"5"5“6

or covalently.2'3"17 In an ionic
bonding scheme zirconium phosphonate (ZP) chemistry is utilized. This
chemistry, while beneficial, can lead to difficulties in the spectroscopic
measurements we intend to do because the ZP layer can be subject to
polarizability. This polarizability will interfere with interlayer excitation transport
thereby precluding their use in certain applications. Thus a covalent bonding
scheme is used.

The procedure for covalent bonding of our polymer layers onto a hydroxyl
terminated surface has been reported elsewhere3. The silanol functionalities on
a hydrolyzed silicon surface can be reacted with polymers combining selected
vinyl ether or maleimide side group reactive functionalities. The surface is
activated using a diacid chloride and the terminal acid chloride is reacted with the
vinyl ether side group functionality to bond the polymer covalently to the surface
which has been accomplished using ester and amide bonds.

This thesis does not report on measurements taken with polymer layers

bonded to a silica surface, such studies will be referred to in the future work

chapter. At this point we are attempting to understand the morphological and

compositional issues associated with a synthesis of MVE polymers using one
type of vinyl ether and two different maleimides. The issues of interest here are
1) relative reactivity of the maleimides and 2) morphology of the polymers made
in a series of solvents of varying polarity.

The long term goal of this work is to gain the ability to synthesize a wide
range of MVE polymers and to control the conformation of these polymers in
solution. To accomplish these goals we use a statistical model to understand
solution phase emission spectra of the chromophores. The morphology of
polymers in solution and on the surface is important to understanding interlayer
interactions, coating uniformity and free volume. Interlayer interactions will be
essential knowledge if we are to understand how to construct a multilayer system
with a pre-determined chemical potential for a given analyte.

Free volume will also be important for understanding the porosity of these
films. Understanding the morphology in solution is the first step in this process
and we use steady state spectroscopy to investigate this. The morphology of the
polymers in solution should be related to how they will be bound to the surface
(Figure 1.1). That is, if a polymer is linear in solution it can be bound to the
surface in a linear way which may not be the case if the polymer is in a coiled
conformation. It is also possible that a coiled conformation may not absorb to the

surface as well as a linear polymer may.

 

Figure 1.1 : Schematic drawing of a
polymer in a linear (left) conformation and
coiled (right) conformation. In a coiled
conformation side groups may interact.

Adsorption to the surface is related to the availability of vinyl ether side
groups to react with an activated surface. In a coiled conformation the
chemically reactive side group could be sterically hindered and not available to
react with the surface to the same extent as in a linear polymer. This information
about the conformation is important knowledge when an attempt is made to build
a structured chemical gradient system.

We probe polymer conformation in solution by monitoring the interactions
between maleimide monomer side groups. Pyrene is useful because it easily
forms excimers allowing facile examination of the interaction of our polymer
chains. The spectroscopy of pyrene is well understood“’29 and is easily
integrated into our polymer synthesis using published reactions.2 Pyrene readily
forms excimers which prove very useful to this research.

Using pyrene as a fluorescent tag we can study the excimer properties of
the polymer in solution. Excimers are useful because they describe the
conformation of the polymer in solution based on the amount of excimer formed.
Using pyrene as an example, an excimer is formed when a monomer of pyrene,
M1, is excited optically:

M1+ hv—> M'
where M' indicates a monomer in its first excited singlet state. The excited
monomer can interact with another monomer, M2, to form an excimer, D', as
follows:

M'+M2 —+ D'

10

The dimer signal is shifted strongly to the red portion of the spectrum. This shift
is due to the different potential energy surface resulting from the interaction of
the two monomers. This dimer then disassociates emitting a photon:

D* —> M + M2 +hv
The emitted photons are collected and the relative amounts of monomer and
excimer are detected using emission spectroscopy. A monomer can absorb a
photon and then emit the photon without forming a dimer. Thus, the
concentration of monomer can have a direct effect on the amount of excimer
formed in a fluorescence experiment.

Excimer formation is a diffusion controlled process. That is, a process by
which the excimer formation is dependent upon the interaction between excited
and ground state monomer units. Once a pyrene chromophore is excited it must
find a ground state pyrene molecule within its fluorescent lifetime for an excimer
to be formed. Thus, pyrene excimer formation is dependent on the concentration
of pyrene in solution on a polymer chain.

Two limits can be identified for polymer morphology: linear (elongated)
and coiled. This is pictured in Figure 1.2. If the polymer exists as a linear chain
then the excimer concentration is predictable based on a statistical model. If it is
a coiled system then the excimer concentration will be higher than predicted by
the sample model. Due to the interaction of pyrenes that are not only adjacent to
each other but nonadjacent pyrene as well because of the coiled conformation

and the polymer bending into itself. This conformation allows for interaction

11

 

 

 

 

 

 

 

Figure 1.2: Representation of how
polymers can be distributed on a silicon
surface.

12

between pyrene functionalities not adjacent and an overall increase in the
intensity of the excimer signal.

Using steady state fluorescence we can determine the amount of excimer
relative to monomer. Excimer formation comes from pyrene chromophores
interacting with each other. We would like to determine whether the polymers
are straight chains in which there is a predictable excimer formation or whether
they exist as coiled systems in which excimer formation is going to be higher
than predicted by our statistical model.

In solution the coiled polymer conformation gives rise to excimer formation
from interactions between adjacent pyrene side groups and non-adjacent pyrene
side groups on the polymer chain. In a solvent that promotes “coiled”
conformations excimer formation between non-adjacent pyrene side groups can
occur. Excimer concentrations higher than what are predicted statistically can be
attributed to coiled conformations of the polymer in solution.

This research allows us to understand the solution phase confonnation(s)
of our polymers. In the future we would like to compare solution phase and
substrate bound excimer formation and determine which model, a coiled or
straight chain, best fits the experimental results. If the data show a statistical
excimer concentration then it can be assumed there is very little self association
and the polymer exists as a straight chain in solution. If our data indicate a
higher than statistically predicted excimer concentration then there is self

association in solution and we have a coiled system.

13

Doing these experiments has led to a wealth of knowledge on the
functions that influence the relative reactivity of monomers and the functions of
MVE polymers. This picture rests on a number of assumptions and we seek to
evaluate them in this work.

We discuss in Chapter 2 of this thesis the synthesis of a series of
maleimide-vinyl ether alternating copolymers where the relative reactivity of two
maleimide monomers are determined from experimental conditions. In Chapter 3
the analysis of these polymers of variable pyrene loading is done by emission
spectroscopy and the results are analyzed. These results are then compared to
a theoretical model and conclusions are drawn. Chapter 4 discusses methods
that will be implemented to examine relative reactivity ratios and how all of the
data will be used to give an overall picture of the synthesis and morphology of

these MVE polymers of varying pyrene loading density in various solvents.

14

Literature Cited

1

1O

11

12

13

14

15

16

17

18

S. Morgenthaler, S. Lee, S. Zurcher and N. D. Spencer Langmuir 2003,
15,

Major, J.S. and Blanchard G.J. Chem. Mater. 2002, 14, 2567-2573
Major, J.S. and Blanchard G.J. Chem. Mater 2002, 14, 2574-2581
P. Kohli and G.J. Blanchard Langumuir 16, 8518-8524

P. Kohli, A.B. Scranton and G.J. Blanchard Macromolecules 1998, 31,
5681-5689

P. Kohli and G.J. Blanchard Langmuir1999, 15, 1418-1422

J.S. Major and G.J. Blanchard Langmuir 2003, 19 2267-2274

K. G. Olsen and GB. Butler Macromolecules 1984, 17, 2480-2486
K. G. Olsen and GB. Butler Macromolecules 1984, 17, 2486-2501

Bevington, J.C. Radical Polymerization Academic Press Inc., New York
1 961

Cowie, J.M.G. Alternating Ctmolvmers Plenum Press, New York 1985

Misra, G.S. Introductory Polvmer Chemistry Wiley Eastern Limited, New
Delhi, India 1993

Seymour, R. B. |ntr_oguction to Polvmer Chemistry Robert E. Kreiger
Publishing Company, Hunington, New York 1978

Takemoto Kiichi, Ottenbrite RC. and Kamachi M. Functional Monomers
and Polymers Marcel Dekker, Inc., New York 1997

X. Wang and M. Lieberman Langmuir 2003, 19, 7346-7353

H.E. Katz, G. Scheller, T.M. Putvinski, M.L. Schilling, W.L. Wilson and
C.E.D. Chidsey Science 1991, 254, 1485-1487

P. Kohli and G.J. Blanchard Langumuir16, 4655-4661

Berlman, Isadore Handbook of Fluorescence Spectra of Aromatic

Molecules Academic Press Inc., New York 1971

15

19

20

21

22

23

24

25

26

27

28

29

Birks, J. B. Photophvsics of Arorrfitic Moleculejs John Wiley & Sons Ltd.,
London 1970

Demas, J.N. Excited State Lifetifme Measurements Academic Press Inc.,
New York 1983

CW. Wu, L. R. Yorbrough and F. Ying-Hsiueh Wu Biochemistry 1976, 13,
2863-2868

K. Kalyanasundaram and J.K. Thomas Joumal of the American Chemical
Society 1977, 99, 2039-2044

F.M. Winnik Chem. Rev. 1993, 93, 587-614

S. Ohmorl, S. Ito and M. Yamamoto Macromolecules 1990, 23, 4047-4053
S. Hamai, N Tamai and H. Masuhara J. Phys. Chem. 1995, 99, 4980-4985
J. Matsui, M. Mitsuishi and T. Miyashita Macromolecles 1999, 32, 381-386
G. Jones, II and V. I. Vullev J. Phys. Chem. A 2001, 105, 6402-6406

C. Tedeschi, H. Mohwald and S. Kirstein Journal of the American
Chemical Society 2001 , 123, 954-960

J. Matsui, M. Mitsuishi and T. Miyashita J. Phys. Chem. B 2002, 106,
2468-2473

16

Chapter 2

EXPERIMENTAL

N-Pyrenemaleimide monomer was first synthesized by the procedure
given below."3" This compound was then polymerized with N-Phenylmaleimide
and 1,4-butanediol vinyl ether according to a literature procedure.2'5 To obtain
relative reactivities of the monomers the polymerization solution was sampled at
various time and analyzed by 1H NMR.

Synthesis of N-Pyrenema/eimide monomer:

This reaction is represented in Figure 2.1. To a 100 mL round bottom
flask 1-aminopyrene (Fluka, 98%) and maleic anhydride (Acros, 99%) were
added in a 1:1 mol ratio in chloroform (EM Science, 99.8%). The reactants were
sealed with a septum and stirred for 12 hours at room temperature to give the
insoluble amic acid derivative. The product was then collected by suction
filtration and added to a 50 mL round bottom flask with 30 mL of acetic anhydride
(CCI, 97%). To this vessel a 1:1 mole ratio of sodium acetate (J.T. Baker,
99.3%) and the previous product were added. The contents were heated to 70°C
in a oil bath for 2-3 hours. The resulting product solution was then added drop
wise to 300 mL of stirring distilled water (distilled in house). The solid precipitate
was then collected by suction filtration and allowed to dry in the air overnight.

The solid was then analyzed by 1H NMR, confirming the absence of a
carboxylic acid proton resonance. The resonances from 7.8- 8.5 ppm are

attributed to the protons on the pyrene rings. At 7.2 ppm are the maleimide C=C

17

5:55:23 notmon Locaocm .0 0:23
on coo m .mEEEfiE 85533 .6 w_mo£:>m H FN oSmE

._._
mSo: m-m
Doom
025L323
o$oo<
\m
I O
I z O O O
o o I + N I zIm

950: S
E
:8

18

allenyl protons and the large resonance at 3.25 ppm is water, a product of the
amic acid ring closure dehydration reaction. At 2.5 ppm is the resonance for the
reference solvent (DMSO)-d°. This monomer was stored for later use in MVE
polymer synthesis.

N-Pyrenemaleimide, AIBN (Aldrich, 98%) 1,4-butanediol vinyl ether
(Aldrich, 99%) and chloroform were obtained from Aldrich and used as received.
N-pyrenemaleimide was synthesized using the above procedure and used as is.
Polymerization of N-Pyrenemaleimide, N-Phenylmaleimide and 1-4-butanediol
vinyl ether:

An MVE polymer was created using the monomers listed above. The
reaction scheme is displayed in Figure 2.2. The monomer concentrations in
solution were modified to create varied fractions of N-pyrenylmaleimide in the
polymer.

To a 50 mL round bottom flask the monomers were added in 30 mL of
chloroform and 1-2 mol % of AIBN was added as a radical initiator. The
reactants were refluxed at 70°C for approximately 24 hours under an inert
atmosphere of argon. The solution was then concentrated by rotary evaporation
until approximately 5-10 mL of the solution remained. The resultant product was
then added dropwise to hexanes (Jade Scientific, 98.5%) or diethyl ether (CCI,
99%) and the precipitant was collected by suction filtration.

This solid was analyzed by 1H NMR (Figure 2.3) confirming the presence
of both pyrene and phenyl chemical shifts. The protons due to the pyrene

moiety. are found from 7.8 — 8.5 ppm. The phenyl resonances are

19

.EoEoEoozm c0328 :0
comma bm> 58 E 25 c mEoBEooo o5. .Emflmtam 5695:. 8.5
m 9.6: .6823 .656 _>c_> 3:529: a Go m_mo£c>m ”Nd 93$

0 z oo 2 o AIII.IO

c

O

E
O

0
[£2
E O
Q
c

IO IO

micfigmm Babom 85m 33> 3532232

20

 

 

 

 

  

UW‘L.
'Isméwz'i 3""?! 'I I)

     

 

W‘TTgT TWETTWT T l T

Figure 2.3: 1H NMR of a MVE polymer
synthesized with N-pyrenylmaleimide,
N-phenylmaleimide and 1,4-butanediol
vinyl ether

21

from 7.2 to 7.6 ppm. The large peak at 3.2 ppm is water from the solvent used in
analysis by 1H NMR From 1 to 2 ppm we see broad resonances due to the loss
of vinyl character of the maleimide protons and the protons on the vinyl ether
used.

Sampling of reaction mixture for NMR analysis:

During synthesis 3 polymerization reactions were sampled and analyzed
by solution phase 1H NMR spectroscopy. The measurements were performed
according to the following procedure. The round bottom flask was disconnected
from the reflux condenser column and a ~1 mL sample was removed using a
Pasteur pipette. The reaction was continued following replacement of the reflux
condenser column. The reaction aliquot was concentrated by rotary evaporation
to ~05 mL and the resultant solution was added drop wise to ~30 mL of di-ethyl
ether. This mixture was stirred for approximately 20 minutes and a portion, ~5
mL, was then added to a disposable glass test tube. The test tube was then
centrifuged and the supernatant was removed. This was done to separate the
polymer from the solution containing both monomer and polymer. The solid was
allowed to air dry and was analyzed by 1H NMR.

Gel Permeation Chromatography.

Gel Permeation Chromatography experiments were performed using a
Waters Breeze GPC system. This system consists of a Waters 2414 Refractive
Index Detector, Waters 2487 Dual wavelength absorbance detector, Waters
1515 lsocratic HPLC pump and Waters 717+ Autosampler. The wavelength of

detection was 254 nm and the flow rate was 1mL/min. The analysis was done

22

using Breeze software provided by Waters. The column was Stragel HR 4E THF
dimensions 7.8 X 300 mm (part# WATO44240). Samples were prepared by
dissolving 1mg per mL of polymer in the mobile solvent, THF with BHT inhibitor.
Fluorescence Spectroscopy.

Fluorescence data were acquired using a Spex—J4 Flurolog3 fluorescence
spectrophotometer with a slit width of 1 nm and an excitation beam of 317 nm.
The data were collected and analyzed using SpectraAcq software.

Nuclear Magnetic Resonance:

NMR data were obtained using a 300 MHz Varian NMR. All samples were
compared to a reference solvent, deuterated DMSO. All samples were scanned
64 times and a Fourier Transform was applied. Data were then transferred and

analyzed using the computer program Mestrec.

23

Literature Cited:

1

2

Major, J.S. and Blanchard, G.J. Chem. Mater. 2002, 14, 2567-2573
Major, J.S. and Blanchard, G.J. Chem. Mater 2002, 14, 2574-2581

C.W. Wu, L.R. Yarbrough and F .Y. Hisiueh Wu Biochemistry 1976, 15
2863-2868

Weltman, J.K., Szaro, R.P., Frackelton, A.R., Jr., Dowben R.M., Binting,
JR, and Cathou, R.E. Journal of Biological Chemistry 1973, 248, 3173-
3176

K. G. Olsen and G. B. Butler Macromolecules 1984, 17, 2480-2486

24

Chapter 3

RESULTS AND DISCUSSION

It is the intention of this research to demonstrate knowledge of the
reaction conditions and an understanding of the properties of the polymers being
synthesized. To achieve this goal, samples are analyzed by 1H NMR and GPC.
NMR will discern the relative concentrations of pyrenyl and phenyl side groups
present in our polymers and can also be used to indicate how much of the
polymer is forming. GPC will give an understanding of the polymer chain length.

The goal of this research is to examine the length and composition of the
polymers being synthesized during the reaction and after completion. The length
will give us an understanding of whether or not the polymer size fits our
theoretical model which assumes each chain consists of about 10 monomer
units. A value greater than 10 monomer units has little effect but a value smaller
than this would create a highly variable distribution because one monomer unit
would cause a large fractional change in the composition of the chain.
Composition is important because we are studying the excimer formation of the
polymer sidegroups in solution. We need an understanding of the order of the
monomer units and whether or not there is a statistical distribution of monomer
units on the polymer.

In an ideal situation the pyrenyl side group loading density would be
directly proportional to the fractional concentration of pyrene monomer in

solution. For example, if we added a 10% concentration of pyrene maleimide

25

relative to phenyl maleimide and we assume 10-unit chains the amount of pyrene
on each chain would be 1 out of 10. If we were to change it to 20% then there
would be 2 out of 10. Through 1H NMR we intend to measure the relative
concentrations of phenyl and pyrenyl side groups using procedures developed
for other systems."2

GPC can be used to analyze the molecular weight and distribution of the
polymers. The polymers should be of sufficient length so they can be compared
to the theoretical model that has been constructed. Polymers of at least 10
monomer units should be sufficient for this comparison.

Analysis of polymers by NMR:

It is important to examine the reactivity of N-pyrenylmaleimde and N-
phenylmaleimide during the reaction. This issue relates directly to the monomer
reactivity ratios for each monomer unit. Equimolar amounts of N-
pyrenylmaleimide and N-phenylmaleimide were reacted in solution with 1,4-
butanediol vinyl ether. We wish to synthesize polymers where the concentration
of each monomer unit on the polymer equals that which we add to solution. If
this is not possible then we wish to know to what degree this deviates from a
statistical distribution.

Polymers were analyzed by 1H NMR and Figures 3.1, 3.2 and 3.3 show
the pyrenyl and phenyl side groups present in the polymer. The signals were
integrated and from these data the relative amounts of each side group were
determined. The relative amounts of pyrene and phenyl signals are not equal to

the amounts found in solution. 1H NMR indicates there is more phenyl maleimide

26

 

 

 

 

 

 

 

Um»...

IIII IIII IIII IIII IIII IIII IIII TIP IUTFII TI III—II IIII IIII
9IO 8I5 8I0 7I5 7IO 6I5 6IO SISII IISIOII IIIIST 4I0 3I5 3I0II I2I5 T2I0II IIS lIO OIS

 

Figure 3.1: NMR spectrum of a
polymer containing 3:1 phenylzpyrene
at 3hrs of reaction. Numbers above
peaks are integrations

27

 

 

 

 

 

I——_Lfi
llllllllIIIIIIIIIIIIIIIIIIIIllllllllll IIIIITITIIllIIIIIIITTIIIIITITTIIIITI]IITTITTTTTTIIT
9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5

Figure 3.2: NMR spectrum of polymer
containing 3:1 phenylzpyrene at 6hrs of
reaction. Numbers above peaks are
integrations

28

 

 

 

 

 

 

 

VIM

[ITIIIIIIIIIIIIIIIII]IIIITIIIIIIIIIIIIIIITIIIIIIIIIWIIIIIWI—FIIII]IIIIIIIIIIIIIIIIIIIIIIII
30 25

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 2.0 1.5 [.0 0.5

 

Figure 3.3: NMR spectrum of polymer
containing 3:1 phenylzpyrene at 9hrs of
reaction. Numbers above peaks are
integrations

29

incorporated into the polymer than would be expected from the solution phase
stoichiometry. The N-phenylmaleimide monomer is more reactive toward vinyl
ethers than N-pyrenylmaleimide. Monomer reactivity is influenced by a number
of factors which may include differences in the polarity of the monomers or
sterics of the monomers in the reaction. We can make simple arguments to
analyze the significance of polarity and sterics.

These monomers would not differ significantly in polarity because the only
difference between the two is the substitution, pyrene vs. phenyl, on the
maleimide. One could argue the n electrons in the pyrene rings could cause a
difference in the dipole moment of the N-pyrenylmaleimide in solution but such a
difference would be negligible in importance. The position is also important in
that the attack on the maleimide ring is of substantial distance from the
substitution on the maleimide and would not be a large factor in the overall
reactivity of the maleimdes in solution.

Steric issues are more likely to play a discernible role. The pyrenyl and
phenyl maleimides are significantly different in size. The size of the maleimide
monomers matters in the reaction because if the sterics are restricting on one
monomer the other will be more reactive in solution. Using this argument one
can see the phenyl ring is less restricting than the pyrene ring structure when
compared with one another. Therefore the phenyl maleimide can be more

reactive when compared to the pyrene maleimide in solution.

30

As figures 3.4 and 3.5 show the steric argument seems to explain our
data. Throughout the reaction 1H NMR spectra were taken and analyzed for
pyrenyl and phenyl components of the polymer. The ratio of sidegroups
incorporated should be the same as the solution phase reactant concentration
ratio if the monomers are equally reactive. What we find is that the two
maleimide monomers are not equally reactive, with the phenyl maleimide being
the more reactive species. 1H NMR analysis indicates for a reaction starting
concentration ratio of 1:1 (phenylzpyrene) maleimide side groups, the resulting
polymer incorporation is ~ 1.5:1 (phenylzpyrene). This difference can be
attributed to the additional bulk of N-pyrenylmaleimide.

Based on this result we modified the reaction conditions in an attempt to
compensate for the added bulk of the pyrenyl side group by reducing the amount
of N-pyrenylmaleimide in solution and increasing the amount of phenyl
maleimide. When the solution ratio of phenylzpyrene was changed to 3:1 a ratio
of 4:1 (phenylzpyrene) on the polymer chain was found by 1H NMR analysis. This
ratio is more comparable to our initial concentrations of each monomer in
solution. It also shows the monomer reactivity has changed and the difference in
reactivity between the two monomers has been reduced. These results give us
control of how much pyrenyl maleimide is incorporated in the polymer during
polymerization.

The sterics of the reaction play a potentially important, if ill-defined role in
controlling the distribution of pyrenyl and phenyl maleimides on the polymer

chain. When equimolar amounts of pyrenyl and phenyl maleimide monomers are

31

we: .06 E22
3 36:95 mm £20 .0628 m_>_>_ co more. 0539355; ”in 9:9”.

as 2.5
2 m. E. or 2 o man

I. \

 

5533
E35353: fix a .5 «a: we E55: a 3 £2: 2.2.5 1:25

~I- co. co. 3:.
O o 0
one: auaquKuaqd

. N.—
. v...

. o.—

 

-mé

32

9:: 56 «:22
E 86:95 mm Emzo LoEzoa w>_>_ co ooze. ocoiaicocn. ”Wm 9:9...

2: as:
F F or m m N. o n v m

 

 

.l\\ ,.

 

.8328 E38353:
fin a he 2:: ha 552:; a we are. «5:3»5:

one: auaquKuoqd

33

added, the relative reactivity of the monomers is different. The mixture has been
changed to reduce these steric influences by increasing the amount of N-
phenylmaleimide in solution and this has had an effect on the selectivity of the
monomers during polymerization. Our data indicate the phenylzpyrenyl ratio
does not change significantly with time. Future research will use N-
pyrenylmaleimide in low concentrations to avoid these undesirable effects of
having a nonstatistical amount of pyrenyl maleimide.

To this point the diffusion coefficients of each monomer have yet to be
examined in this thesis. The hydrodynamic volume of each monomer can be
calculated according to a method developed by Edward.3 The hydrodynamic
values of N-phenylmaleimide and N-pyrenemaleimide were found to be 150 A3
and 250 A3, respectively. If we apply these values to the Stokes-Einstein
equaflon:

0° = kT/6nnr
where D0 is the diffusion coefficient, k is the Boltzmann constant, T is
temperature in K, n is the viscosity of the solvent and r is the Van der Waals
radius of a sphere. Since both maleimides are in the same solution T and n are
the same and the Do is proportional to 1lr. Using the r values found above we
find 0° is equal to 1.67x10'7 cmzls for N-phenylmaleimide and 1.0x10'7 cmzls for
N-pyrenylmaleimide. Thus the diffusion coefficient for each maleimide is different
from each other but not to the degree of what is found in the 1H NMR data. We

conclude that diffusion does play a role which may be comparable to the role

34

sterics play in the polymerization process. Future work will be dedicated to
determining the degree diffusion plays in the polymerization process.

While NMR has been useful in determining the concentrations of pyrenyl
and phenyl sidegroups on the polymers synthesized it can only be used to
describe average concentrations in these polymers. It does not describe the
position of these functionalities on the polymers. For example, NMR can be used
to analyze a polymer of 10 monomer units at a concentration of 40% pyrene.
NMR can indicate there are 4 pyrene units on the chain but it cannot tell us the
position of these units. The difference between 4 units being adjacent to each
other or being at other positions on the chain is indistinguishable by NMR. While
this is true, the fact that the phenyl groups are more reactive than the pyrenyl
precludes bonding or pre-aggregation of the monomers being incorporated and
thus their distribution on the polymer chain. This point is important in our
modeling of these polymers. To understand the excimer formation properties of
the system we must have an understanding of the position of the monomer units
on the polymer chain. It has been found the phenyl maleimide is incorporated
into the chain more readily than the pyrenyl maleimide. This does not interfere
with our statistical model in that we can adjust it to compensate for this.

At this point we are able to understand the relative concentrations of each
of our monomer units on the polymer chains but not able to discern the position
of these units. We consider the statistical model mentioned above following a
discussion of the polymer MW and Mn, because the results are relevant to

determining the applicability of a statistical model. This information is important

35

when we compare our results and fluorescence measurements to our statistical
model. If the polymer is of too few monomer units then the experimental data will
not hold to the assumption the polymers follow a statistical distribution of
monomer units.

GPC analysis of polymers:

GPC results indicate the polymers MW values vary between 3000 and
16000 Daltons. This variation results from the reaction conditions used.
Specifically, the concentration of monomer and the amount of initiator influence
the Mw and Mn of the resulting polymers in predictable ways. Polydispersity
Index (PDI) values have varied between 1.3 and 2 depending on reaction
conditions. It has been found that the longer the chain, the more polydispersity in
the polymer. For polymers with 16,000 (M...) the PDI equals 2.

Initiator concentration has a substantial effect on the MW of our polymers.
Our initial experiments used 10 mol% of initiator. This initiator concentration
resulted in polymers with a MW of ~3000. Reduction of initiator concentration to 1
mol% resulted in a MW of ~7000. To increase the MN further the concentration of
monomers was increased resulting in a MW of ~16000 which is the highest value
seen to date.

We have investigated the excimer properties of the polymers in various
solutions using fluorescence spectroscopy. The excimer properties of pyrene
containing polymers have been studied in the literature“ 7 and this method is
used in determining whether the polymers exist as a straight chain or coiled

conformation. We have examined our polymers in dilute solutions to reduce the

36

amount of intermolecular association between polymer chains and to compare
these fluorescence measurements to a statistical model. If excess excimer
fluorescence, above that predicted by a statistical model is seen, then it is
attributed to intramolecular association of the pyrene on the polymers. This
excess excimer formation leads to the conclusion the polymers exists in the
coiled conformation. This experiment is done in several solutions where the
solubility of the polymers varies to some degree. When the solubility of the
polymer is low it will coil and intramolecular interactions will increase.

Figure 3.6 shows the fluorescence spectrum of pyrene at two
concentrations, 0.06 uM and 0.05M. There are several features evident in these
spectra. There are discreet vibronic transitions due to pyrene from ~370 to 400
nm. There is a broad featureless band between 425 and 500 nm due to excimer
formation in solution.

Fluorescence measurements were made in three solutions, cyclohexane,
acetonitrile and Dimethylsulfoxide (DMSO). The polymer solubility varies
considerably in these solvents. The polymer is very soluble in DMSO, slightly
less soluble in acetonitrile, and only slightly soluble in cyclohexane. This varying
solubility gives us a model of what a coiled conformation may exhibit when
excited. Figure 3.7 is an emission spectra of the polymers synthesized with
differing concentrations of pyrene on the polymer. All spectra contain both
monomeric and excimeric bands. As the pyrene loading on the polymer chains is

increased, the amount of excimer observed in each spectrum also increases.

37

289588 $8.8 26. Em 38.8
:9; pm 393 So Ezbomaw 8:60.652... ”9m 9:9“.

A85 £95.55;

 

 

 

 

 

 

 

 

 

 

can con omv oov own
P p — p n p —
I o
I ooooom
I ooooov
I ooooom
2.1m; I So

.285 I 2;
I oooooo—
I ooooomw

ocoid mEmD :ozmbcoocoo LoE_oxw Co cosmEELoSoo

(he) Kuwaiti!

38

2528

30:? E bacon mcfimo. 6:93 80:? 53> mLoEzoa
m_>_2 mEEmEoo 9.qu Lo 96QO 8:88.02”. ”Em 9:9“.

@5332?

 

Kusualui pezummou

 

 

REE Ewan—gas;

 

 

 

 

 

  

Alrsuatui poziluuuou

8:06 com 9595 5%.?
8305 22m mcmia oomdm
8:05 86 ocmia Osma—

wnaoa 09m ocean {one

 

 

 

 

 

23385230

Kiisuoiui pozinzuuou

39

This is expected as an increase in the concentration of pyrene on the polymer
will lead to increase of the probability of adjacent pyrenes forming excimers. The
issue at hand is the comparison of the experimental data to our model.

Acetonitrile is a “good” solvent, meaning the polymer is soluble in it. A
“poor” solvent is one in which the polymer is insoluble. The solvent used to
determine an insoluble conformation is cyclohexane. Figure 3.8 exhibits the
polymer emission response in cyclohexane for X% pyrene loading. It can be
seen that the excimer band is more intense than the monomer bands in contrast
to the acetonitrile spectrum (Figure 3.9) where the monomer bands are more
intense than the excimer bands under conditions of similar pyrene loading
density and polymer concentration.

The intensification of the excimer band relative to the monomer band is
due to the polymer being in a coiled conformation because of the insolubility of
the polymer. The excimer band is more intense because there is more
interaction between the pyrenyl side groups. The interaction is not only from
adjacent side groups but also from nonadjacent pyrenyl groups on the same
polymer chains. With the polymer in a coiled conformation this interaction is
possible, if the polymer is in a more linear conformation this interaction is less
frequent. This difference is evident in the acetonitrile spectrum where the
excimer band is much less intense when compared with the monomer band. The
polymer is more soluble in acetonitrile so it will exist in a more predominately
linear conformation and there will be fewer interactions between nonadjacent

pyrenyl sidegroups giving rise to a less intense excimer band.

40

2982296 E
99:28 mcEfiEoo 29:8 Co 36:95 .9503 ”Qm 9:9”.

 

 

 

 

 

 

€559.13!
9 9 9 9 g 9 9 g 9 9 9 .7 v v v .7 v t .7 v v 8 8 c 8 Do 8
Ir 0 6 8 I. 9 9 8 7w Ir 0 6 8 I. 9 9 v 09 Z Ir 0 6 9 I. 9 C.— v
8 7w 7v Ir Ir 0 o 6 6 8 8 I. I. 9 9 9 C.- .' v 8 06 Z Z Ir Ir 0 o
il‘i‘i o
Tier}.
5) .
51/51/
1.2 . go—
/115 K
9.,
.J .
,3; ......w
R a: . 88m
1 \‘i... _
. M1? Lian... J 2.7.. 3.. I
*N VII 4.1.. 7,r ;,.\,.....sl— if. ‘1‘. _ w
. z. ,{ . ._ . .88» m.
8m NFI J... . 5
$8 .9 I
.3 u 1 3* “
$ommI ... ,2 I
I ,
9:82 ocoin. _ ..
I .-,L. I
\fi.
- 88“

35.2206 5 9.83. 2.23 .o 5:83 a 8 has: 2:. 3.59m

41

QEEBmom

E 29:28 mEEmEoo 95:8 3 33:25 5:.me ”as,” 959“.

2:5 593.263
9 C. .7 7 .7 by
lv 0 6 9 I. 9
8 9 I. I. 9 9

8 l9
209
269
l99
HS
099
099
689
629
99?
9”
V8?
VZV
9 W
80?
262
288

 

 

 

 

 

fiend I
$.an I
$08
$w.mm I

 

 

 

>=mcmn
9:82 9.93

2.32803 5 3.2.3 9523. 2.23 .0 530:2 a «a :26 9:8 .38me

 

fooooomw

, 888m

7 808mm

 

h oooooon

(nu) [usuoaul

42

These experimental results have been compared to a theoretical model.
The polymer has been assumed to be linear and the theoretical model predicts
the probability of excimer formation found as a function of loading density on the
polymer chain.

The theoretical model was obtained by assuming our polymers consist of
linear chains. We assume these are 10-monomer chains for convenience.
We further assume if pyrenes are adjacent to one another on the polymer chain
they will form an excimer upon excitation with a probability equal to unity. Finding
the distribution of pyrene on the polymer chain is calculated by:

Combinations = nllxl(n-x)!

where n the number of monomer units possible on the chain and x is the number
of pyrene units possible on the chain. For example, if we use a 10 monomer unit
chain and a concentration of 20% pyrene in solution then n equals 10 and x
equals 2. Combinations refer to the number of ways the phenyl and pyrenyl side
groups can be arranged. This calculation tells us the number of combinations a
pyrene can occupy based on concentration and length of chain. What must be
found is the degeneracy of the combinations because one pyrene is no different
from another. For example if pyrene A occupies position 3 and pyrene B
occupies position 6 for our experiment it is no different than if pyrene A occupied
position 6 and pyrene B occupied position 3. When this is taken into account we
can find the number of unique combinations of monomers adjacent to one

another and divide this by the total number of combinations to find the probability

43

of excimer forming. We can assume the probability upon excitation of a pyrene
molecule is either going to form a monomer or excimer:

P(m) + P(x) = 1
where P(m) is the probability of monomer and P(x) is the probability of excimer
upon excitation. We can find the probability of monomer using the above method
to find the probability of excimer formation, P(x). We can determine the
probability for a variety of pyrene loading densities and determine the amount of
excimer we would expect based on this statistical model (Figure 3.10). Note the
excimer probability is unity at 60% pyrene loading concentration since there is no
way to add 6 pyrenes to a 10 unit chain without at least two of the pyrenes being
next to each other.

Also displayed in Figure 3.10 is how :I: 1 pyrene per chain, relative to the
expected loading will affect the probability of excimer forming. For example, a
pyrene loading concentration of 10% predicts that one pyrene will be found on
each 10 unit chain. While this is the most likely possibility there will be some
chains that will have 2 pyrenes and some will have no pyrene. This probability
may be small but it can occur and if the 2 pyrenes are adjacent then excimer can
form. This uncertainty will affect the measurement and has been factored into
the model when comparing the experimental data to it. The uncertainty will allow
a small amount of excimer to be found for chain loading concentrations where it
may not be obviously possible. This uncertainty has been determined by
assuming the pyrene distribution on the polymer chain will follow a Gaussian

function. If this is the case the error will be :10. The deviation in a 20%

44

  

mcfimo. ocoia ho cozocé m mm Ezfimnoa 5:..wa 5.8 939“.

mango— ocoba “cocoon

 

 

 

 

ow om ow om cm 2 o
. . _ . . . _ . . q 0 .......... . Hun: o
.......... .
can _ h a . .
ops—o5 8 32695 8:03.. mo
8:35me Eoumufim 53> ....0... .......u. .. cm
@233on “Sampson 6866 295m ....0.... . m
- 9. m.
m
o. . m
0..
r 8 m..
. m.
o...... .
M0. ow
u
g ............. - o2

 

45

measurement would therefore contain 10% and 30% pyrene possibilities. This
has been factored into the curve and displayed in the figure. This model
assumes polymer forms linear chains and tells us any deviations that appear will
manifest as more excimer formation than the model predicts because of
intramolecular association of nonadjacent pyrenes.

The experimental data were determined by obtaining fluorescence spectra
of the polymers in various solvents. By knowing a particular pyrene
concentration will give a certain fluorescence we determined the amount of
excimer formed in each of these solvents. This was done for polymers of various
pyrene loading densities and compared to our statistical model.

A comparison of the experimental data to the theoretical model is shown
in Figure 3.11. Experiments have shown the polymers exhibit excimer intensity
above what is predicted statistically. This is due primarily to the conformation of
the polymers in these solutions. The polymer is least soluble in cyclohexane
and, as expected, the excimer to monomer intensity ratio is highest in this
solvent. The polymer exhibits less excimer formation for acetonitrile and DMSO.
This is also expected as it is more soluble in each of these solvents. Overall the
polymers seem to extend but some amount of coiling in all of these solvents due
to the much higher than statistically predicted excimer intensity. This comparison
of experimental data to the theoretical model leads us to conclude the polymers
exist in a coiled conformation and not the linear chain that was expected.

It is important to understand these polymers were not created using the

experimental condition as described above. The initiator level used was much

46

33mm. 62.26% ncm _mEoE:maxm co cowtmano ”id 239“.

 

  

 

 

manome— ocoba Hoodoo
ow om ow om cm 3 o
a . . . a . _ . . . _ q ......... o
...o ...........
Engage—gov 8% 3:08:35 ..0
AflmbEQ-ooav Saw Eggheads ....0 .o.... . om
A828 58 3885.6 2.. .
can. 5%... 338:8 ....o w
i ow m.
3
.. m
.o.. . m
... o.
8 m
.... .
O ..................... O l Om
. ......... . ................. ....
................... 0... .
......... O O. .0 5......
o ........... . ..... . 1. l 2:

 

47

higher than what we have used recently and this will affect the amount of pyrene
on the polymer to a large degree. It is expected that these polymers that were
formed early on vary greatly in the amount of pyrene on the polymer chain. This
variation is due to the polymers being less than 10 monomer units in length.
Pyrene addition to a polymer of small length will dramatically change the
concentration on the polymer as opposed to a longer chain polymer. These
polymers might vary a great deal when being compared to our model system and
could lead to erroneous results. Work is now being done to understand the
consumption of the monomers involved in the polymerization and to repeat the
experiments. These polymers may have a more statistical distribution of pyrene
on the chain and will exhibit less excimer which will affect the comparison. For
this reason the results here should be interpreted as illustrative rather than

indicative of a polymer created under more controlled conditions.

48

Literature Cited:

1

P.Koh|i, A.B. Scranton and G.J. Blanchard Macromolecules 1998, 31,
5681-5689

C. Soykan, M. Coskun and M. Ahmedzade Polymer lntemational 2000,
49, 479-484

J.T. Edward Journal of Chemical Education 1970, 47, 261 -270

J. Matsui, M. Mltsuishi and T. Miyashita J. Phys. Chem. B 2002, 106,
2468-2473

J. Matsui, M. Mltsuishi and T. Miyashita Macromolecules 1999, 32, 381-
386

S. Ohmorl, S. Ito and M. Yamamoto Macromolecules 1990, 23, 4047-4053

S. Tazuke, H. Ooki and K. Sato Macromolecules 1982, 15, 400-406

49

Chapter 4

CONCLUSIONS AND FUTURE WORK

This thesis has described the synthesis of MVE polymers and the
morphology of the polymers in various solvents. Using a published synthesis N-
pyrenylmaleimide was synthesized and then integrated into a radical-initiated
polymer synthesis. The polymers were then added to several different solvents
in which the solubility of the polymer varied. Solutions in which the polymer was
insoluble exhibit high excimer fluorescence. When the polymer was soluble in a
solvent the excimer intensity decreased. This was then compared to a
theoretical model in which two assumptions were made. Excimer intensity is due
to adjacent interactions of pyrene on the polymer chain and the chain is linear as
to not have any interaction between nonadjacent pyrene. Comparison of the
experimental results to the theoretical model indicates the polymer is in a coiled
conformation in solvents such as cyclohexane and intramolecular interactions are
occurring. ln solvents such as DMSO where the solubility is higher
intramolecular interactions are not as dominant and excimer formation follows a
more statistical or linear trend.

Future work is being focused on finding the monomer reactivity ratios to
find if the polymer is statistical or blocky in nature. This work is important to the
model because if we attribute excimer fluorescence to adjacent pyrenes on a
polymer there must be a statistical distribution of these pyrene units on the chain.

If the polymer is blocky then the pyrene will not follow a statistical distribution.

50

This will lead to excimer formation from effects that are not due to the polymer
morphology. Excimer will form due to polymer morphology and from the blocky
nature of the polymer. If these polymers form in a blocky nature comparison to a
statistical model will be difficult.

Work has begun to describe whether or not these polymers formed are
blocky or statistical. In order to do this we must study how the monomers are
consumed in the reaction as it proceeds. Thus, the reaction must be sampled
and analyzed at points in the reaction to understand how the monomers are
consumed during the polymerization process.

Absorbance spectroscopy is the method of monomer analysis we are
using presently. Absorbance spectroscopy could prove useful because each
monomer could be resolved using this technique. The N-phenylmaleimide can
be resolved from the N-pyrenylmaleimide based on the different absorbance of
the phenyl and pyrene functionalities. What would be needed are the extinction
coefficients of each monomer.

The extinction coefficients can be found using Beer's Law:

A=€bc
Where A is absorbance, C is extinction coefficient, b is the path length of the cell
in cm and c is concentration in mol/liter. A plot of absorbance as a function of
concentration will yield a slope equal to the extinction coefficient. The plots for
each monomer are shown in Figure 4.1 and 4.2. The extinction coefficient for N-

phenylmaleimide was found to be 17200 L/mol-cm at 220 nm, and the extinction

51

32m 0. Race m_ Eo_oEooo
cozoczxm .mn_E_m_mE 9.2.3-... ..o. 83 >22 m...mmm H; 939“.
53 ..o_§.=3c8

8.2. 93 8m 8w 8.» on.» 8.» So 8d 8.... 8m

 

 

 

 

 

 

— b p p h _ _ h p p o

. ....

.N...
a... czarmsnll u
”8......" w. 2 u
a... Na ... as... I 88... . .38... u .. m
E: 03:. o m.
.2. o
e: E... . a

. no

98..."... ......

88... + .88... u ..
-3

 

.2532. 2.9.2.: .8 «ca 33 «.38

52

82m 0. Race m_ EoBEooo
:ozoczxm .mEEEfiE 3:93-: .8 BE .sm_ m.:omm ”we 9:9:

33 5:22.88

8.9 and oo.m 8d 8d 8.: 8s 8.0 8m 8d 8m

— h _ _ b P _ h p b o

- Nod

23
O

8
o
aoueqlosqe

   
  

 

58.0"». 4:...
:8... . fig... n ..

.2...

-2...

3:522: 3:23.: .3 .2: 26.. {com

53

coefficient for N-pyrenylmaleimide was found to be 65300 L/mol-cm at 242 nm
and 32400 Umol-cm at 340 nm.

What we intend to do now is to sample the reaction mixture and to analyze
the samples by absorbance spectroscopy. By finding the absorbance at specific
wavelengths and using Beer’s Law the concentrations of each monomer can be
found in solution. These concentrations will be examined if each monomer is
being reacted at the same rate. The monomer reactivity information can be used
to find if the polymer forming is blocky or statistical.

After this aspect has been examined the polymers will be adsorbed to a
silica surface using the chemistry described previously. The excimer to
monomer ratio for bound polymer on the surface will be examined and this
information will allow evaluation of whether or not the morphology of the polymer

on the silica surface is related to that in solution.

54

, ..... / T . 04 II"

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

llllflllllllllllllllllllljllllllllllllllljjlll