Citation
Acyclic diene metathesis (ADMET) polymerization

Material Information

Title:
Acyclic diene metathesis (ADMET) polymerization the synthesis of well-defined model polymers for polyolefin materials
Creator:
Valenti, Dominick J., 1969-
Publication Date:
Language:
English
Physical Description:
ix, 173 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Alcohols ( jstor )
Alkenes ( jstor )
Catalysts ( jstor )
Dienes ( jstor )
Metathesis ( jstor )
Molecular weight ( jstor )
Monomers ( jstor )
Polyethylenes ( jstor )
Polymerization ( jstor )
Polymers ( jstor )
Alcohols ( lcsh )
Chemistry thesis, Ph. D ( lcsh )
Dissertations, Academic -- Chemistry -- UF ( lcsh )
Metathesis ( lcsh )
Polyethylene ( lcsh )
Polymerization ( lcsh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 161-172).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Dominick J. Valenti.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
028074777 ( ALEPH )
37863393 ( OCLC )

Downloads

This item has the following downloads:


Full Text










ACYCLIC DIENE METATHESIS (ADMET) POLYMERIZATION:
THE SYNTHESIS OF WELL-DEFINED MODEL POLYMERS FOR
POLYOLEFIN MATERIALS











By
DOMINICK J. VALENTI


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA



























This dissertation is dedicated to my family: Anthony, Patricia, and Julia
Valenti, all of whom have provided me with unlimited support, confidence,
love and strength to achieve any goal or challenge in the known universe.













ACKNOWLEDGMENTS


I would like to first acknowledge Dr. Guy Mattson for introducing me

to the world of polymer chemistry and for his faith in my abilities from the

beginning.

Special thanks are extended to Dr. Chris Marmo and Dr. Jason Portmess

who instigated my interest in ADMET chemistry. Sincere thanks are

extended to all the members of the Wagener group, past and present: Dr. K.
Brzezsinka, Dr. J. Anderson, Dr. J. O'Gara, Dr. D. Tao, Dr. H. Tamura, Dr. J.

Konzelman, J. Reichwein, F. Gomez, S. Cummings, T. Davidson, D. Tindall,

M. Watson, S. Wolfe and L. Williams.

I would also like to recognize the cooperation of Drs. R. Duran, J.

Reynolds, K. Wagener and G. Butler that has created an institute of vast

resources, knowledge and funding for this and other scientific pursuits.

Sincere thanks are also extended to the past and present members of
the Duran and Reynolds groups for their contributions and support of my

research goals. Specific recognition is extended to J. Batten, Dr. A. Kumar, D.

Cameron, Dr. P. Balanda, and Dr. M. Diverti for significant contributions of

scientific knowledge and philosophical ideas.

Grateful acknowledgment for funding is given to the National Science
Foundation and The Dow Chemical Company. Special recognition is

extended to Steve Hahn from Dow for his contributions, ideas, and faith in
this project from start to finish.








Respects go to D. Panosian, T. Nguyen, J. Miller, L. Wolert, and N.
Morales for boundless friendship, support and thoughtfulness.
Finally I would like to thank Dr. Wagener for believing in my abilities
and providing nearly unlimited resources, freedom and knowledge that have
made this research and my graduate education possible.














TABLE OF CONTENTS




ACKNOWLEDGMENTS.............................................. iii

A BST R A C T ........................................................................ ..................................viii

CHAPTERS

1 INTRODUCTION......................................................... 1

History of Polyethylene and Related Polymers............................... ......3
Discovery and Synthesis of "Polymethylene".................... .............. 3
Synthesis of Polyethylene via Radical Chemistry..................................... 6
Synthesis of "Polyethylene" (Heterogeneous Ziegler Catalysts).............9
Synthesis of "Polyethylene" (Homogeneous Ziegler Catalysts)........... 12
Copolymerization of a-olefins............................ .............. ...........15

Structure-Property Relationships of Polyolefins...................... ............ 18
Crystallinity, Morphology, and Melting of Flexible Chain
M acrom olecules................................... ..... .......................... 20
Thermal Behavior and Chain Structure.............................................. 23

M etathesis Polym erization......................................................................... 27
Ring Opening Metathesis Polymerization (ROMP)
and Well Defined Alkylidenes...........................................................27
Acyclic Diene Metathesis Polymerization (ADMET)..............................30
ADMET: Step Condensation Method to Producing a-Olefin
Precursor Polym ers................................... ....................................34


2 EXPERIM EN TA L.............................. ........................................................ 36

Instrumentation and Analysis......................... .............................36

M materials and Techniques..................................... ..................................... 38

Synthesis and Characterization.............................. ......... ........................ 39
Synthesis of Symmetrical Alkyl-Substituted Terminal Dienes............39








Alternative Enolate Displacement Reactions for the
Synthesis of Compounds la, and lb................................................45
Alternative Methods for the Synthesis of Compounds (5a)-(5c)..........52
W ittig M ethod............................................. .... .........................53
Tertiary alcohol synthesis and Reduction............................................54

ADMET Polymerizations of Monomers (5a) and (5b)................................ 57
General Metathesis Conditions............................. ........................ 57

Hydrogenation of Polymers (P5a) and (P5b)...................................................60

Alcohol Monomer Synthesis.................................................61

ADMET Polymerizations of Alcohol Monomers (Al), (A2), (A3).............64
General M etathesis Conditions................................................................64
Hydrogenation of Poly [1,10-Undecadiene-6-ol] (HPA2).......................66

Preliminary Alcohol Dimerization Experiments..................................68


3 DESIGN AND SYNTHESIS OF SYMMETRICAL ALKYL-
SUBSTITUTED TERMINAL DIENES...................................................71

Designing the Target Monomers........................................ .........................72

Synthesizing the Target Monomer.......... ............. ...........................75
3-Keto Ester Substitution Method....................... ............ ............... 75
Enolate Alkylation of Ethyl Acetoacetate and the
Retro-Claisen Condensation.................................................... 80
Reduction, Tosylation, and Hydride Displacement.............................. 87
Dealkoxycarbonylation of the Keto Ester.................................. .....88
W ittig M ethod............................................................. .........................91
Malonic Ester, Di-Grignard and Lactone Methods.................................93
Organometallic/Tert-Alcohol Method................. ......... ........... .. 95

C onclusions........................................... ................................................... 100


4 ADMET MODELING OF BRANCHING IN POLYETHYLENE:
THE SYNTHESIS OF MACROMOLECULES WITH
PERFECTLY SPACED METHYL BRANCHES...................................... 101

Modeling Polyethylene............................................ .........................101
ADMET Modeling of Branching in Polyethylene............................... 104
Polymer synthesis via ADMET......................... ..... ........... 106
Hydrogenation of the unsaturated polymer...................................113
Thermal Analysis of Methyl Substituted Polyethylene..................... 120








Thermal Analysis of ADMET Methyl Substituted Polyethylene.......122

C onclusions........................................ ...................................................... 128


5 THE DIRECT SYNTHESIS OF WELL-DEFINED ALCOHOL
FUNCTIONALIZED POLYMERS VIA ACYCLIC DIENE
METATHESIS (ADMET) POLYMERIZATION................................... 130

Alcohol Functionalized Polymers via Metathesis.................................... 132
The Direct Synthesis of Alcohol Functionalized Polymers
via A D M ET................................................ ............. 135
Monomer Design and Synthesis......................................... .......... 136

ADMET Polymerization of Hydroxy Functionalized Dienes..................140
General ADM ET Polym erizations.......................................................... 140
Polymerization of 6-(4-Pentene)-l-Heptene-7-ol (Al)...........................141
Polymerization of 6-Methyl-l,10-Undecadiene-6-ol (A3).....................145
Polymerization of 1,10-Undecadiene-6-ol (A2).......................................146
Hydrogenation of Poly(6-Hydroxynonenylene) PA2...........................150
Thermal Analysis of the Alcohol Containing Polymers................... 152
Preliminary Dimerization Experiments.....................................156

Conclusions................. .................................................................. 160

R EFE R E N C ES........................................................................ ............ ............ 161

BIOGRAPHICAL SKETCH......................... ...... .........................173













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ACYCLIC DIENE METATHESIS (ADMET) POLYMERIZATION:
THE SYNTHESIS OF WELL-DEFINED MODEL POLYMERS FOR
POLYOLEFIN MATERIALS

By

Dominick J. Valenti

August 1997


Chairman: Professor Kenneth B. Wagener
Major Department: Chemistry

A synthetic method has been designed to facilitate the synthesis of
target monomers required for the preparation of perfectly branched

polyethylene model polymers. These polymers can be used to better
understand the relationships between branching in polyethylene and its
ultimate behavior. Symmetrical alkyl-substituted alpha omega dienes were
synthesized using a six-step procedure involving two enolate displacement

reactions, a retro-Claisen condensation, an ester reduction followed by a
tosylation, and a hydride displacement reaction. Purity of the resulting
monomers was greater than 99%.

The alkyl substituted monomers were condensed using Schrock's
molybdenum alkylidene (CF3)2CH3CO]2(N-2,6-C6H3-i-Pr2)Mo=CHC(CH3)2Ph
forming well-defined alkyl-substituted polyethylene prepolymers.
Quantitative hydrogenation (by NMR analysis) of the resulting unsaturated
polymers was achieved using a modification of typical diimide reduction


viii








procedures. The resulting polyethylene polymers' contain branch points
occurring at every 9 or 15 carbons, respectively. These samples are ideal with
regard to well-defined chain branching and precise branch distribution.
Presently, this high order of regularity in polyethylene copolymers cannot be

obtained by any other method. Thermal analysis of these polymers
demonstrates the unique behavior of these materials. The presence of

perfectly spaced methyl branches results in a significant drop in the polymers'

melting point within a very narrow range.

The synthesis of well-defined primary, secondary, and tertiary alcohol
functionalized ADMET polymers was also accomplished using a similar

methodology. Unsaturated polymers of this type can be hydrogenated in a
similar manner producing a new series of ethylene-vinyl alcohol copolymers
exhibiting precise chemical regularity.














CHAPTER 1

INTRODUCTION


Research in polymer chemistry during the past five decades has

resulted in the significant advancement of macromolecules as materials in

technical and nontechnical applications. We all have become everyday

consumers of the product of this research, from the clothes we wear to the

materials used to print this document. Advancements such as these do not

occur rapidly, but rather are derived from patient work leading to the

fundamental understanding needed to create polymers and tailor their

physical properties.

The dissertation represents a contribution to this fundamental
research. The work presented herein describes the creation of new techniques

that offer unique control over a polymer's microstructure, which in turn can

have an important effect on ensuing physical properties. This control of

microstructure is demonstrated by the synthesis of model polymers with well

defined structures that can be used to elucidate the effect of molecular changes

on the behavior of the polymer. Also demonstrated herein is the ability to

use Lewis basic/protic groups toward the synthesis of unsaturated,
functionalized polymers via acyclic diene metathesis (ADMET)
polymerization. These polymers then can be further hydrogenated, resulting
in the direct synthesis of saturated polyalcohols.






2



The model polymers are structurally similar to the industrially
significant a-olefin polymers; polyethylene and polyethylene-polyvinyl

alcohol copolymers. This work models the chemical/structural imperfections

or branch points that result from uncontrollable events which are present in

typical polymerization reactions. Branch points often are purposely
incorporated into polymer chains so that desired physical properties are

achieved and are typically incorporated by random copolymerization with

higher order a-olefins.

a-Olefin copolymerization (as well as uncontrollable chain transfer
events) causes random branch placement, thus producing materials with

variable physical properties. The resulting randomness makes it difficult to

study structure-property relationships of industrially important materials.

This dissertation presents a methodology that controls polymer structure

therefore producing analogous materials that can be used to investigate the

effect of structure on the physical properties of polyethylene and related

materials.

It is appropriate to describe polyethylene (PE) and related a-olefin
polymers as the main focus of this introduction. This will be accomplished
with a brief historical overview and a discussion of the important

polymerization mechanisms of PE. The concept of branching, related PE

studies and the crystallinity of flexible chain polymers are also presented.
Since ADMET is the chosen polymerization route, a discussion of metathesis

chemistry is also warranted for a complete understanding of the chemistry
within.








History of Polyethylene and Related Polymers


Discovery and Synthesis of "Polymethylene"


The repetitive joining of molecules of ethylene forming linear
macromolecules with the general formula -(CH2-CH2)n is known today as the
polymerization of polyethylene. The first samples of polyethylene were made

in 1898 by von Pechmann.1 He observed that small amounts of a white,
flocculent substance separated from an etheral solution of diazomethane on

standing, and while he was able to crystallize the substance from chloroform
there was not enough product for further study. Two years later, Bamberger
and Tschirner2 were able to precipitate a large quantity of this solid, which
was described as a white powder with a melting point of 128 'C and had a
corresponding chemical analysis consistent with the structure (CH2). This

compound was termed polymethylene and was the product of the reaction of
diazomethane with itself (Figure 1.1).3


n(CH2N2) -- (CH)n + n(N2)


Figure 1.1. The polymerization of diazomethane: the synthesis of
polymethylene.


Further investigations of this polymerization showed that the addition
of boron4 or copper5 catalysts resulted in the production of straight-chain,
high molecular weight, highly crystalline polymethylene. Further, the first
investigation of branching and polymethylene was done via the
homopolymerization of diazoethane or 1-diazopropane to produce
polyethylidene (methyl branched) and polypropylidene (ethyl branched). The










fH -H H2 H H H,

CH3 CH3 CH3 CH2CH3 CH2CH CH2CH3
n n
A B

Figure 1.2. A: Polyethylidene B: Polypropylidene



materials were described as rigid, amorphous, brittle glasses similar to

polystyrene (Figure 1.2).3 At this time, the structure-property relationship of
branching was investigated by the copolymerization (decomposition) of

diazomethane with varying amounts of diazoethane and
1-diazopropane.3 These samples demonstrated the influence of chain

branching on the crystallinity and therefore the physical properties of the
polymers. These copolymers would later be used as models to clarify the
structures of polyethylenes made by the more involved processes of high-
pressure, high-temperature polymerizations.3

In 1953 Kantor and Osthoff6 reported the synthesis of polymethylene
with a molecular weight greater than 3 million, chemistry which was

achieved by treating diazomethane in diethyl ether with a diethyl ether-boron
trifluoride complex at 0 "C.7 The polymer had a crystal melting point of 132
"C, which is close to what is expected for linear PE, while X-ray examination

showed a high degree of crystallinity. These results indicated a high
molecular weight and a low degree of branching. This assumption was
confirmed by comparison of these samples with those produced by high
pressure, high temperature radical polymerization.








Similar to the decomposition of diazomethane, this dissertation

presents the synthesis of polyethylene and its branching by an alternate

mechanism, and so it is appropriate to briefly discuss some of the details of

the mechanism to form polymethylene from diazo-decomposition. The first

such discussion was provided by Kantor and Osthoff6 who believed that the

polymerization took place through an ionic mechanism, a conclusion

derived from the fact that the catalyst and solvent were ionic reaction

promoters. Bawn and Rhodes8 also produced high molecular weight

polymethylene in a toluene solution with copper wire, copper stearate, and

boron trifluoride as catalysts. They derived the second order rate equation by

following the reaction via titration and measuring the evolving nitrogen,

and their results also indicated an ionic mechanism. Further investigations9

proposed two mechanisms for this catalytic conversion using boron

compounds, and in both cases the first step was the formation of an
intermediate by the nucleophilic attack of diazomethane on boron. The

difference in these mechanisms was that one was cationic in nature and the

other anionic. The accepted mechanism of today is anionic in nature.3

Alternate, yet historically less significant, routes to polyethylene have
also been described. In order to synthesize straight chain hydrocarbons,

Carothers et al.10 used Wurtz coupling to polymerize decamethylene
bromide. This method proved to be messy and did not successfully produce

polymers with more than 100 carbon atoms. The Fisher-Tropsch reduction
was also successful at producing polymethylene type paraffins. These

products had average molecular weights of 7,000-9,000 g/mole, and melting

points ranging from 117 to 132 'C.1,12 Other synthetic methods were derived
from the desire to study similar polymer structures. Hahn and Miiller,13 in
order to better understand the structure of poly(vinyl chloride), converted








PVC to PE by total reduction using excess lithium aluminum hydride under

high pressure and temperature. It was later determined that the total

reduction of poly(vinyl bromide) was much easier to accomplish.


Synthesis of Polyethylene via Radical Chemistry


The direct polymerization of ethylene to polyethylene was first
achieved in March of 1933 by Imperial Chemical Industries, Ltd. (I.C.I.),14

where a trace of white powder was discovered during the systematic study of
high-pressure chemistry of organic compounds. Soon after this discovery,
I.C.I. developed a commercial process to produce this polymer, and a fully

operational plant was erected in 1942 with the first applications being that of a

wire insulator for radar systems. The process I.C.I. developed was termed the

high-pressure free-radical process, for it involved the use of pressures greater
than 124 MPa, temperatures between 100-300 "C and a free radical catalyst.15

This method of polyethylene production is still used today and
produces a wide range of polymers called low density polyethylene (LDPE).
Further, this process has been modified in order to produce medium and
high density polyethylene. The first synthesis of "linear" polyethylene using
this method was accomplished by Dupont using temperatures between 50-80
"C and ethylene pressures of 707 MPa.16 The resulting polymer had a density
of 0.955 g/cm3, with less than 0.80 alkyl substituents per 1000 carbon atoms.17
However, the polymers that are industrially produced using this method are
low density (0.912-0.935 g/cm3) and usually contain 15-30 variable length alkyl
substituents per 1000 carbon atoms.








The radical polymerization of ethylene consists of a chain addition

mechanism and follows the classical description for free radical

polymerization (Figure 1.3).



I* + CH,=CH- k I-CH2-CH2'


I-CH2-CH2* + n CH2=CH2 k RCH2CH2'

Figure 1.3. Radical polymerization of ethylene. Initiation and Propagation.


As with other typical radical reactions, termination consists of radical

coupling and disproportionation (Figure 1.4 a ). Chain transfer events are

also evident with transfer to ethylene to form a vinyl end group, or the

transfer to solvent to form a terminated chain (Figure 1.4 b, c ).



2 RCH2CH2- k- RCH2CH2CH2CH2R or RCH=CH2 + RC2CH3 (a)



RCH2CH2- + CH2=CH, kt --, RCH=CH2 + CH3CH2* (b)



RCH2CH2* + SH kt -- RCH2CH3 + S* (c)

Figure 1.4. (1) Termination of radical chain growth.
(2) Chain transfer to ethylene. (3) Chain transfer to solvent.




In radical systems branching can occur as a result of chain transfer
events: intermolecular and intramolecular. Intermolecular transfer causes
the formation of long chain branches (> 6 carbons) and is the result of normal








chain transfer to polymer18,19 via transfer of a radical to a completed polymer
chain causing the growth of a grafted chain. The formation of long chains
causes changes in the melt flow viscosity of the polymer.
Intramolecular chain transfer, which is about 40 times more prevalent
than intermolecular chain transfer, causes the formation of short chain
branches. The number, size, and distribution of these branches have a
significant effect on the polymer's crystallinity. A methodology to study this
effect is the primary focus of this dissertation.


.
SCH,2CH2CH2CH2 CH2
CH2 /CH2
CH2
b


^'CH2CH2CH2 H CH3
CH2 /CH2
CH2
n-Butyl branch


-.CH2CHCH2ZH2 CH3
CH2 CH2
CH2
n-Hexyl branch




"CH2CH2CHCIH2 CH3
CH2 CH2
CHz

n-Amyl branch


maximum crystallinity 60-70%

Figure 1.5. Intramolecular chain transfer during the radical
polymerization of polyethylene.








The intramolecular chain transfer event of interest is caused by the
transfer of the propagating radical on the end of the chain to a more stable
secondary radical on the same chain which occurs by the extraction of a

hydrogen atom via a cyclization or "backbiting" mechanism (Figure 1.5). The
resulting species can continue to undergo radical propagation with ethylene,
which leads to the formation of an alkyl branch. This uncontrollable

backbiting results in the formation of ethyl, n-butyl, n-amyl, and n-hexyl

branches, where the relative abundance of these different branches depends

on the synthetic conditions.20 The formation of the n-butyl branch is the
most prevalent and is attributed to the favorable formation of a six-
membered transition state shown in Figure 1.5.

There is evidence that the polymerization of vinyl acetate follows the
same radical reactions and side reactions as those for ethylene.21,22 The
resulting polymer (polyvinyl acetate) is a precurser to alcohol functionalized
a-olefin polymers which will be discussed further in chapter 5 of this
dissertation.


Synthesis of "Polyethylene" (Heterogeneous Ziegler Catalysts)


In the fall of 1953, Ziegler and co-workers observed the polymerization
of ethylene at atmospheric pressure using titanium halides and alkyl
aluminum compounds,23,24,25 a landmark discovery which has had a
dramatic effect on the world of chemistry and is still the basis of extensive
organometallic research. These catalysts are formed by the reaction between a
metal alkyl or a metal hydride of the main group metals and a reducible
compound of the transition elements of groups 4 to 7.3 The most preferred
combination is an aluminum alkyl or hydride with a titanium-IV compound.








Natta and co-workers termed these new catalysts, Ziegler catalysts.26,27 These

systems are commonly referred to as classical or heterogeneous Ziegler
catalysts.

Ziegler's discovery made it possible to produce polyethylene,
previously only obtainable by extreme conditions of pressure and
temperature, in a simple flask or test tube.28 This was first demonstrated by
Ziegler in a Weck-glass vessel under atmospheric pressure and room
temperature using gasoline as the solvent.25 The polymer produced was more
rigid, stronger, and more heat resistant than those made by the radical
process. This was attributed to the fact that the polymers were made up of
nearly linear molecules which resulted in a material that was more dense
(0.95 g/cm3) and more crystalline. The branches that did occur were identified
as being mostly ethyl groups. The contributions that Ziegler made in this area
of organometallic/polymer chemistry resulted in his receipt of the Nobel
Prize for chemistry in 1963, along with Natta.

About the same time as Ziegler's discovery, two medium-pressure
polymerization methods using transition metal oxide catalysts were
discovered by the Phillips Petroleum Co. and the Standard Oil Co. of Indiana.3
The polyethylenes produced by these methods were more linear in nature
and therefore had a higher density of 0.96 g/cm3.

Ziegler polymers, as well as the medium pressure polymers, are
generally accepted to occur through an insertion mechanism. A
representation of this chain addition coordination polymerization is outlined
in Figure 1.6.29 This reaction initiates with the transfer of an alkyl group
from the alkyl aluminum to the titanium chloride, followed by the t-
complexation of the monomer to the transition metal's vacant d-orbital
(Figure 1.6 A-B). A four-centered anionic coordination insertion process that








inserts the monomer into the metal-carbon bond (C) then occurs. The result
of this process produces a vacant orbital with the opposite configuration of
the original complex (D). This mechanism is referred to as the monometallic
mechanism for the stereo specific polymerization of polypropylene and was
proposed by Cossee and Arlman.30,31



FPropagation


C ,,, I Cl

Cl


CH=CH2


C i/,,.. ,,\c i

Cl T
B


c i,,,. I Cl c Ci/,, ..,,\ -I .
Cl i O Cl L
Cl Cl '
D C

Figure 1.6. The insertion of ethylene as the propagation step of the
coordination polymerization of Polyethylene. The monometallic
mechanism.


The Ziegler and Phillips (chromium oxide-based) catalysts are used to
manufacture virtually all of the linear polyethylene made today. These








systems, as mentioned earlier, produce nearly linear high molecular weight

paraffins. These linear flexible chains are able to arrange in such a fashion

that they can acquire long range three dimensional order (crystalline regions)

which allows the polymer to pack better. This resulting order produces a

material that is termed high density polyethylene (HDPE). A density of 0.95

g/cm3 is the result of having 1 to 3 alkyl groups (branches) per 1000 backbone
carbon atoms (Figure 1.7), while a density of 0.97 g/cm3 results from, on

average of 1.5 alkyl groups per 1000 atoms.32



TiC14 + RA1CI2 RTiCI3 + AIC13

RTiC3 --- TiCl3 + R-


*^C^H^-C CF- I H2- CH H-- CHr- CH2-Vn
R v498R

0.5 3 R's per 1000 Carbons R = CH3
-CH2CH3
-CH2CH2CH3
etc.
Figure 1.7. High density polyethylene (HDPE) via Ziegler catalysts.



Synthesis of "Polyethylene" (Homogeneous Ziegler Catalysts)


The latest step in polyethylene research has been the introduction of a

family of transition metal complexes called metallocenes. These catalysts

have been extensively used to investigate the long debated mechanism of

Ziegler catalysts. Elucidating the elementary steps of a polymerization is

simplified by studying soluble well characterized systems;33 therefore,

metallocenes have contributed the most definitive results toward this








cause.32 Much of the drive to understand the details of the mechanism

originated from the special demands on understanding the structure-property

relationship of c-olefin systems.


x = C2H4, Me2Si


M = Zr, Hf
x = C2H4, Me2Si


x = C2H4, Me2Si M = Ti, Zr, Hf
RI= Me, Ph, Naph RI = H, 5*Me, neomenthyl
R2= H, Me R2 = Cl, Me

Figure 1.8. Examples of metallocene catalysts.33



The first structure of a metallocene was described in 1952 by Wilkinson

et al.34 and Fischer35 with the introduction of ferrocene. These molecules

(metallocenes) were commonly referred to as sandwich compounds due to

their unique spatial arrangement (Figure 1.8).33 This new class of compounds

showed considerable promise in the way of advancing organometallic








chemistry, but at the time failed to have an industrial impact. They were

frequently used as a transition metal catalyst in combination with either

triethyl or diethylaluminum chloride for the polymerization of olefins.27,36

The poor catalytic activity that they exhibited limited there use to mechanistic

studies.


This was the case until 1975 when water was accidentally added to a
NMR tube which contained biscyclopentadienyltitaniumdemethyl,

trimethylaluminum and ethylene.33 Upon this addition, the fast

polymerization of ethylene was observed in a system that was thought to be

inactive toward this process.37 It was soon determined that the best cocatalyst
for these systems was methylaluminoxane (MAO) (Figure 1.9). Today,

zirconocenes (zirconium transition metal metallocenes) that are combined

with MAO result in olefin catalysts that are 10-100 times more active than

classical Ziegler catalysts.38



H3K 3CH3 CH3
Al-O-41-O-Al-O-A,
H3C CH3 CH3

Figure 1.9. A proposed oligimer of methylaluminoxane (MAO). Cocatalyst
for homogeneous Ziegler-Natta polymerizations.33




Kaminsky states that the polymerization of ethylene with
bis(cyclopentadienyl)zirconium dichloride (Cp2ZrC12) and MAO can result in
the production of up to 40,000,000 g PE/g Zr h.33 The resulting polymers
have typical molecular weight distributions of M,/M. = 2 (a much higher








distribution is obtained with heterogeneous systems)39 with 0.9 to 1.2 pendent

methyl groups per 1000 backbone carbons atoms.

Arguably, the most important contribution that metallocenes have
brought to polymer chemistry is the opportunity to develop plastics with

variable control over their structure/property relationships. Metallocenes

grant structural control by defining a single site at which the building blocks

of the polymer monomerss) are joined together linearly. Single site

polymerization is in contrast to the nonuniform catalytic action that takes

place in heterogeneous systems which follow a similar mechanism. Due to
their solubility, the active sites of homogeneous systems are more accessible

for analytical examination, therefore provide the ability to make rational

catalyst modifications.33 The impact that this type of control can provide is

demonstrated by the design and synthesis of chiral zirconocenes, which are

used for the synthesis of stereoregular polypropylenes.40


Copolymerization of (c-olefins


For the past fifty years the majority of low density polyethylene has
been produced by the high pressure, free radical polymerization of ethylene.

This process inherently produces side chains (Figure 1.5), both long and short,

hence producing a large distribution of chains that pack poorly (low %

crystallinity) or have no semblance of order (amorphous).

The homopolymerization of ethylene using Ziegler catalysts results in
a very marketable plastic termed high density polyethylene (HDPE). These

polymers do contain some degree of short chain branching but are essentially
considered linear. In contrast, when ethylene and higher order olefins are
copolymerized in the presence of Ziegler catalysts, a new class of material is





16

obtained that is termed linear low density polyethylene (LLDPE) (Figure 1.10).

The copolymerization of ethylene with a-olefins like 1-butene, 1-hexene or 1-

octene has opened up a new chapter in polyethylene based materials and

currently shows a higher growth rate than the homopolymer. These new

products offer an advantage over those made via radical chemistry in that

there is no significant uncontrollable branching. This inherent linearity

produces molecules in which the only significant branch points are from the

incorporation of the a-olefins (Figure 1.11).



R R
CH,- CH, + = Ca\ C CH
R



R CH3










Figure 1.10. The copolymerization of higher order a-olefins with ethylene.
The polymerization of linear low density polyethylene (LLDPE).



LLDPE was introduced in 1977 with Union Carbide's implementation
of a low pressure, gas-phase, fluidized-bed process.7 This process was later
termed UnipolTM and was soon followed by Dow Chemical with a new line of

LLDPE polymers termed DowlexTM. Today these and other companies have





17

shifted gears toward the development and production of large amounts of

LLDPE.

The use of heterogeneous catalyst systems for these copolymerizations,

results in LLDPE products with large polydispersities and little to no control



HDPE Polymerized by Zeigler Natta or Metallocene chemistry


CH2 CH ca- CH2 CH2 CH- HH

n>>>m R
t

R= infrequent short
chain branches
LDPE --- Polymerized by radical addition mechanism


ICH2CH, + n CH, CH2 --- 4CH,--CH2 -CHf--jH- -

n>m

R= frequent short and
long chain branches
LLDPE Copolymerized by Zeigler type catalysts


nCH2== CH2 + m CH =CH t CH-- CH2 CH2--
R R


R= the identity of R
from monomer feed
Figure 1.11. The induced or non-induced branching in polyethylene.


over comonomer sequence distributions.41,42,43,44 Contrary to this, the

development and use of homogeneous catalysts have produced LLDPE
samples with narrow molecular-weight and more uniform comonomer








possible to design new materials that are driven by special demands on

properties. The resulting short branches on these polymers produce materials
with lower melting points, lower crystallinities, and lower densities as
compared to HDPE.33


Structure-Property Relationships of Polyolefins


The ability to design systems from a molecular standpoint
(comonomer identity and ratio) in order to tailor physical properties has
benefited tremendously from homogeneous systems. By adjusting
comonomer composition or catalytic activity/affinity, a macromolecule can
be designed and produced with known but irregular molecular composition.
The use of homogeneous systems provides a method to assemble these
mixtures of comonomers while eliminating nearly all the undesirable side
reactions and producing narrower molecular weight distributions.

This lack of "regular" order is due to the variable reactivities of
different monomers towards the polymerizing catalyst. The affinity for any

given olefin to insert depends on the last inserted monomer and is defined by
a parameter termed the monomer reactivity ratio. For a two monomer

system these ratios are represented by rl and r2 and defined in equation (1).
The term ri is expressed as the ratio of the rate constant


r,= r2 = (1)



of inserting an ethylene unit into the growing chain versus the rate constant
of inserting an a-olefin, when the last monomer inserted was an ethylene.
Parameter r2, in this case, would be the analogous ratio for the a-olefin.








Equation (2) defines the copolymer composition and is related to the
concentrations of the two monomers in the feed [ml] and [m2] and their

reactivity ratios.


d[m,] [m,](r,[m,]+[m,])
d[mz] -[mr ]([, + r2[lm 2])


The tendency for a comonomer to homopolymerize, randomly
copolymerize, or alternatively copolymerize can be inferred by reactivity
ratios. The product of the two reactivity ratios rl r2 represents the
distribution of the comonomers. When this product equals one it represents
a randomly distributed copolymer.33 A product value considerably less than



Table 1. Examples of copolymerization parameters of metallocene/MAO
catalysts in combination with ethylene and a-olefins.


Metallocene Temp. a-Olefin rl r2 ri r2
in C

Cp2ZrMe2 20 propene 31 0.005 0.25
[En(Ind)2]ZrCl2 50 propene 7 0.06 0.40

Cp2ZrCl2 40 butene 55 0.017 0.93

Cp2ZrCl2 80 butene 85 0.010 0.85
[En(Ind)2]ZrC12 30 butene 8 0.07 0.59

Cp2ZrMe2 60 hexene 69 0.02 1.38
Data taken from Kaminsky, W. Macromol. Chem. Phys. 1996, 197, 3907.


one represents a somewhat alternating structure.48 The use and modification
of metallocene catalysts have provided limited control over these reactivity








ratios, by modifying the symmetry and steric crowding of the active center as

well as the temperature of the polymerization. Some examples of this are

shown in Table 1.33


The modification of catalyst and conditions does afford some control of

comonomer distribution but the placement of monomers is still irregular.

These systems do, though, provide samples with better control over the

intermolecular distribution of branches than previously possible. This

inherent variability limits the usefulness of these systems toward the

fundamental understanding of structure, morphology and property

relationships.


Crystallinity, Morphology, and Melting of Flexible Chain Macromolecules


The ability of flexible chain polymers to orient in such a fashion as to
display long range three dimensional order is referred to as their ability to

crystallize. This is similar to the crystallization process in low molecular

weight molecules but differs in that the crystallizing molecule is larger than

the unit cell. The increased size of polymers enables a single chain to

crystallize in multiple unit cells. Multi-participation and its corresponding

thermodynamics are due to the connectivity of a vast number of chain atoms

which result in the unique yet complex crystallization properties of polymer

systems.

Polyethylene and its copolymers are considered semicrystalline
because they contain both crystalline regions and viscous glassy regions.49,50
The viscous glassy regions are termed amorphous and are composed of
unordered molecules or repeat units.







The ir.rrl',h..l of the crystalline regions of polymers has long been
debated. The fr'..'., d-'...ii.' theory, developed in the 1930s, proposed that a
polymer's crystallinity came from the occasional alignment of chains in an
extended fashion creating order. This theory was supported by the belief that
the most thermodynamically stable crystal was one involving extended
chains.51 Twenty years later, upon the isolation of a polymer single crystal, a
new theory on chain morphology arose called folded-chain lamella.52,53,54,55
This conformation consists of the chains folding back on themselves
lamellaee) creating ordered crystalline regions. The lamellae nucleate from a
central point making up a crystal region within the polymer termed a
spherulite. (Figure 1.12).


*


onPsL


****** *i *** 1

1*** ***

PE Spherulite Lamellae
Figure 1.12. Representation of PE Spherulite and Lamellae.


Mandelkern,56 in 1986, acknowledged that the crystallite morphology is
clearly lamella but that the molecular morphology and chain structure cannot
be determined by the analytical techniques currently available. The theory of
folded chain lamella is widely accepted today, but is still the subject of
extensive investigation.57,58,59,60,61


II


'""' ~"""""-'~""' "~^








The molecular requirement for crystallinity in macromolecules is that

the chain atoms or repeat units must be capable of adopting an ordered
configuration in which the adjacent units can lie parallel to one another

creating well defined lattice sites.62 This ability to conform is directly related

to the repeat unit's degree of regularity. The three forms of regularity are 1)

chemical, 2) geometrical, and 3) spatial.

Chemical regularity is defined by the frequency of the repeat units. For

example, in linear polyethylene the repeat unity is -(CH2-CH2)- and its degree

of chemical regularity is defined by how often this repeat unit exists without

the appearance of branches. When branching occurs, the degree of

crystallinity is noticeably changed. Linear polyethylene contains 2 to 3 side

branches per 1000 carbon atoms and is -90% crystalline,63 at 30 branches per

1000 carbons the crystallinity decreases to 50%.64 The decrease in crystallinity
due to an increase in branching is thought to be a result of the exclusion of

branched units from the crystalline regions.65

Geometrical regularity is defined as the regular placement of an
unsymmetrical repeat unit. If an unsymmetrical repeat unit is incorporated

into a chain in both directions, it results in what is termed head to tail (HT),

tail to tail (TT), head to head (HH), and tail to head (TH) placement. These

different connectivities result in a geometrical irregularity which disrupts a

chain's conformational abilities.

Spatial regularity results from the systematic placement of substituents

that can attain different spatial arrangements (stereo isomerism). This occurs
during the polymerization of monosubstituted alkenes such as higher order

a-olefins. This can be seen by examining samples of spatially regular (tactic)
and non-regular (atactic) oriented polypropylenes. The ordered (tactic)








polypropylene results in a highly crystalline polymer while the atactic version
is a completely amorphous material.

These three molecular -cguaij. r. requirements--chemical, geometrical,
and spatial--are determining factors for the ability of a material to crystallize

based solely on the makeup of the chain itself. However, there are many

other variables that affect the crystallinity of the material as a whole. These

include molecular weight, molecular distribution, end groups, diluents

(impurities), and thermal history.


Thermal Behavior and Chain Structure


Much of the research on polymer crystallinity concerns the resulting
thermal behavior. Thermal behavior relates to order because the chain

structure influences the melting point of crystalline regions (Tm) through

conformational properties and thus the entropy of fusion.66 The melting-

crystallization process of long chain polyethylenes is a first-order phase

transition and can be related to the same transition in small molecules.57

Small molecules, though, lack the conformational constraints of a large

polymer therefore resulting in sharp transitions when its equilibrium
temperature is attained. The longer chains introduce structural or

morphological "impurities" causing the broadening of this fusion process

(Figure 1.13).62 This broadening effect can also be observed by comparing the

melts of fractionated and polydisperse samples67 thus demonstrating that the
fusion process is interrupted by varying molecular constitution while the

peak of the fusion curve remains relatively the same.

Examination of linear, low molecular weight PE samples shows that
comparable fusion curves can be obtained that are similar to those of








hydrocarbons62 having a limited critical length of only 150 carbons.68

Therefore, relating folding and/or fusion effects from low molecular weight

samples to those of larger PE samples of similar molecular makeup is a viable

methodology. This is possible because the molecular weight (length) at which


















400 380 360 340 320
Sr(KI

Figure 1.13. Differential Scanning Calorimetric endotherms:62 (a) low MW
PE Mn= 725 g/mole; (b) n-hydrocarbon C44H90 MW= 618 g/mole



chain folding can occur is the same for n-alkanes and PE polymer

fractions.68'69,70,71 A detailed study comparing the melting temperatures of n-

alkanes with those of linear polyethylenes was done by Mandelkern et al.,72

demonstrating that identical melting points could be obtained for most

samples but a lack of correlation for infinitely large samples. Mandelkern et

al. explain this as being a result of a large increase in the interfacial free energy

with a large increase in size.72

The research in this dissertation is concerned with the effect of a

regularly distributed structural irregularity (branches). Therefore, it is








appropriate to briefly discuss how copolymerization affects the crystallization
and melting points of these types of materials.

It can be generally stated for random copolymers that the concentration
of crystallizable units decreases via the addition of a copolymer impurity, the
melting range becomes progressively broader and the level of crystallinity
decreases. The amount of theoretical temperature depression for random
copolymers was predicted by the Flory equation which is shown as equation
373 here where Tm and Tm are the melting temperatures of the unbranched




(/T.)-(TO )= -(R/AH,)lnX (3)


and branched polymers, R is the gas constant, Hu is the heat of fusion per
repeating unit, and XA is the mole fraction of crystallizable units. Results,
using this equation, have shown that the identity of the alkyl pendent groups
does not affect the amount of depression as related to the temperature
composition relationship.74,75 Alamo and Mandelkern point out that this
result is an example of the fundamental principle that the melting
temperature of a copolymer does not depend on the composition but on the
details of their sequence distribution.75

There have been numerous studies describing the physical effects of
these randomly copolymerized units.75,76,77,78 As described earlier, chain
copolymerization results in an inhomogeneous distribution of alkyl
branching. Compared to heterogeneous Ziegler catalysts, the metallocenes
grant some control over the intermolecular distribution of side chains.
Therefore, they produce polymers in which each individual molecule
possesses the same distribution of branches as another, but the distribution of








the branches is not uniform along the individual polymer backbones. The

lack of order in both cases has recently been termed intra and intermolecular

heterogeneity, respectively.79






CH3

Hydrogenated
butadiene/propene
co-polymer



CH3 CH3

Hydrogenated
butadiene/ethene/propene
terpolymer

Figure 1.14. Methyl branched model polymers for PE.80,81


Both types of distributions influence the superstructure and the

crystallization behavior of these materials; therefore, model systems that can

control both of these variables would be suitable for detailed investigations.

Gerum et al. 80,81 modeled this control by studying the short branching

achieved by strictly alternating hydrogenated poly[butadiene-alt-(1-olefin)]

copolymers and butadiene/ethene/1-olefin terpolymers (Figure 1.14). For the
copolymers, a limit of 167 short chain branches per 1000 back bone carbons

could be achieved with a intrahomogeneity of a branch point appearing every

6 backbone carbons, and a strictly alternating content greater than 97%. The

copolymers related to this dissertation are produced when propene (methyl

branch) is used as the copolymer. The copolymers with propene from Gerum








et al. show by X-ray scattering and thermal analysis that a completely

amorphous material is produced. It is important to note that a glass

transition was detected for these samples at -66 C.

Terpolymers subsequently were produced in order to increase the
number of methylene units between branch points by the incorporation of

ethene as a spacer. This type of ter-unit addition is able to dilute the number

of branch points per chain resulting in spacers between branches being on

average 5, 11, 17, 23 etc. The resulting terpolymers, using propene as the 1-

olefin, had as many as three thermal transitions with the first two being very

broad. Even though these models proved to be interesting to study, the

samples still contain a significant degree of intra- and intermolecular

heterogeneity. Until complete control of the copolymerization can be

achieved, the use of Ziegler chain addition reactions will not provide the

necessary control over the microstructure in these polymers. Therefore, the

design of well defined models for the fundamental study of structure property

relationships will not be possible using these means.


Metathesis Polymerization



Ring Opening Metathesis Polymerization (ROMP) and Well Defined
Alkylidenes


Olefin metathesis, over the last 25 years, has become a well established
means of cleanly producing linear unsaturated polymers. The majority of
this research lies in the area of a chain growth mechanism termed ring

opening metathesis polymerization (ROMP).82 ROMP is represented by the
polymerization of a strained cyclic olefin to its corresponding linear








unsaturated polymer (Figure 1.15). The resulting polymers (Figure 1.15) are
similar in structure to those produced by the Ziegler polymerization of
dienes (butadiene, isoprene) in that both types of products contain a site of
unsaturation and are linear carbon backbone chains.




O ROMP initiator _CH CH= CHf.



Figure 1.15. The ROMP of Cyclooctene.


The metathesis polymerization is made possible by a transition metal catalyst
(initiator) in the form of a metal carbene. This reaction, as are all metathesis
reactions, is governed by a competing ring chain equilibrium.82c The
feasibility of shifting the equilibrium toward chain production is usually
driven by enthalpy; therefore, strained cyclics, such as 3, 4, and 8 membered
rings and norbornenes provide the necessary energetic for polymerization.
Similar to the use of classical Ziegler-Natta catalysts, the synthesis of
ROMP polymers was first achieved by polymerization using heterogeneous
catalyst systems. These systems were typically composed of various transition
metal halides and oxohalides accompanied by a wide range of metal alkyl and
organohalide Lewis acid cocatalysts. Again, due to the unknown nature of
the true catalytic species, the mechanism of polymerization could not be
directly studied and therefore no rational changes in the system could be
made. The heterogeneous catalysts required the addition of a Lewis acid
cocatalyst giving these systems poor functional group tolerance. Some
examples of short term living systems were observed but upon longer








reaction times would undergo secondary metathesis reactions, therefore

broadening their molecular weight distribution.82c,83

In 1986, Shaverien, Dewan, and Schrock developed a well defined

alkylidene complex with a highly active four coordinate dO tungsten center.84

This complex did not require a Lewis acid cocatalyst and was capable of olefin

metathesis. Due to the homogeneous nature, these systems provided the

arena for structure elucidation and mechanistic pathway discovery.85s86

Following this discovery by Schrock and co-workers, similar alkylidene

systems were able to perform ROMP in a living manner as well as resist
"poisoning" functionalities. This enabled the design and synthesis of new

block copolymers and new functionalized polymers via a metathesis
mechanisms.




X Ru]




X = CO2H, CH2OH, CHO

Figure 1.16. Functionalized ring closing metathesis (RCM) using Grubbs'
ruthenium alkylidene.


Taking the design of homogeneous alkylidines one step further
Nguyen et al., in 1992, presented the synthesis of a well defined ruthenium
alkylidene which initiated ROMP in both organic and protic/aqueous
solvents.87 This design also showed living characteristics as well as multiple

functional group tolerances, therefore opening up new doors for olefin

metathesis chemistry. Unique to this system was its ability to undergo








metathesis in the presence of aldehydes, alcohols, carboxyclic acids and

quarternary amines.88 This was demonstrated by Fu, Nguyen, and Grubbs

using ring closing metathesis for the formation of six membered

functionalized rings (Figure 1.16).88 Further investigations of the ruthenium

systems have developed new structural reactivity relationships and isolatable

intermediates which have provided a tremendous impact from a mechanistic

standpoint.89 Chapter 5 in this dissertation examines a ruthenium alkylidene

as a catalyst for the direct synthesis of alcohol functionalized polymers.


Acyclic Diene Metathesis Polymerization (ADMET)


The inherent control over polymer structure that Schrock and Grubbs'

well-defined alkylidene catalysts provide has significantly broadened the

scope and applications of ROMP chemistry, but it has had an explosive effect
on the evolution of a rather new polymerization method called Acyclic Diene

Metathesis (ADMET) polymerization.90 ADMET has been under

investigation for 25 years, but upon the advent of Schrock's alkylidenes the

viability and understanding of these systems have become a reality. Prior to

their introduction the metathesis reaction was plagued with problematic side

reactions and an undefined mechanism. Based on theoretical and

experimental observations of Schrock's well defined tungsten alkylidene,

Wagener, Boncella, and Nel proposed the accepted mechanism of ADMET.90

The ADMET and ROMP mechanisms both proceed through the
equilibrium polymerization of olefins via metathesis. However, ADMET
differs by involving the polymerization of acyclic dienes, with its equilibrium

shifted to polymer by the removal of a small alkene (entropically) (Figure
1.17). ROMP is typically shifted to high polymers via a favorable enthalpy








change via ring strain release. The major mechanistic difference is that

ROMP follows a chain growth process where the active alkylidene acts as a

catalytic initiator. In contrast, ADMET follows a step condensation

mechanism and the alkylidene behaves as a true catalyst.



Catalyst _
%' _R + H2C:=CH2

Figure 1.17. ADMET polymerization: General reaction scheme.


The accepted mechanism for ADMET is outlined below in Figure 1.18.
Upon addition of the catalyst to a neat solution of dienes, the olefin will first

coordinate in a it complex with the transition metal center followed by an

insertion reaction forming either a productive (d) and nonproductive (c)

metallacyclobutane intermediate. The nonproductive cyclic (d) can collapse

to form the original it complex, whereas (c) undergoes a productive

rearrangement eliminating the precatalyst fragment (e) while forming the

monomer as part of the new alkylidene (f). Another monomer can then

coordinate and insert in the same fashion as before, producing the

metallacyclobutane (g). On the first trip around this cycle, collapse of this

metallacyclobutane forms the dimer, with ensuing trips forming powers
thereof. The collapse of (g) leaves the catalyst as a methylidene (h) which is

considered to be the true active catalysts species.89 This methylidene then
coordinates with another monomer, or appropriate chain end, forming the

metallocycle (i), which upon its collapse produces the small molecule









a

LnM=CR2


4-"?s


coordination LnM=CR, /

of olefin (CH, non-productive
insertion

productive (CH2)6
insention C

R R c


Ln. d

R R
formation of
new alkylidene


R,C=CH2

HC=CH2 /CH 2
HLnM=C second monomer
release or polymer olefin insertion
release of
ethylene
(driving force)

( /,2,, i Ln


Ln ,'/-- (CH2)6



generation of /
active methylidene
insertion onef eand free polymer
or polymer
LnM=CH,

(CH CH
^^(ru^\ ^^ru 'a


Figure 1.18. ADMET polymerization cycle.








ethylene while returning back to the top of the cycle with the alkylidene

having the identity of the stepwise growing chain.

The clockwise movement of this cycle is caused by the removal of the

small molecule, in this case ethylene, producing (depending on the number

of cycles) a high molecular weight polymer. The movement of this cycle in a

counter clockwise manner has also been demonstrated by the

depolymerization of high molecular weight polymer in solution via the

addition of excess ethylene91,92 or various functionalized monoenes.93,94,95

Unlike ROMP polymerizations, ADMET condensations are performed under

neat conditions so that competing equilibria are avoided and the release of

ethylene is favored.

The step condensation process of ADMET requires stringent conditions
to achieve high molecular weight polymer. Under step conditions, high

polymer can only be achieved if the system obtains an overall conversion

greater than 98%. This kinetic behavior is in contrast to the chain methods

used to make PE and the corresponding copolymers were a conversion of 90%

is considered excellent.51 In these chain growth systems, monomers add to a

highly energized chain end, which results in high molecular weight polymer

almost instantly. Step condensations proceed by monomers first coming

together to form dimers, which then condense to form the tetramer and so

on. Carothers defines the stepwise growth in an equation which calculates

the degree of polymerization (DP) as it relates to the extent of reaction (p)
(equation 4).51 The number of these repetitive condensations



DP- = (4)
(1-p)








(DP) is also related to the stoichiometric balance of the system. However, the

ADMET system typically involves condensing two identical functional

groups in an A-B type manner, which eliminats the stoichiometry concerns.
ADMET polymerizations also follow other notable stepwise constraints,

including the formation of molecules during equilibrium conditions that
have a most probable distribution (polydispersity index (PDI)) of 2. This
distribution is commonly referred to as a Flory distribution, and is
represented in Equation 5 as the weight-average molecular weight (M,)
divided by the number-average molecular weight (M,).51 Fortuitously, PE

samples synthesized by various metallocenes have a similar most probable

distribution. This will allow for better comparisons between the step samples
of ADMET polymerization and those produced by metallocenes.


M" =2.0 (5)


ADMET: Step Condensation Method to Producing a-Olefin Precursor
Polymers


The establishment of ADMET occurred via the metathesis
polymerization of 1,9-decadiene using the tungsten96 and molybdenum
versions of Schrock's alkylidene. The investigation of this and other
hydrocarbon monomers has demonstrated the clean conversion of a, (o-
hydrocarbon dienes into linear unsaturated polymers.97,98 Using Grubbs'
ruthenium alkylidenes, similar results showing the clean conversion of
hydrocarbons can also be obtained.99 The resulting polymers again, like the
ROMP polymers, produce linear unsaturated carbon backbone polymers
similar to their ct-olefin counterparts (Figure 1.16). For example, the
quantitative hydrogenation of polyoctenelene produced via ADMET results






35

in a saturated hydrocarbon chain that is analogous to completely linear PE.100

This dissertation is an extension of this work, using the same well established

clean conversion of a, co-hydrocarbodienes in order to produce PE samples

with inter and intramolecular homogeneity of branch inclusion. The

durability of Grubbs' ruthenium benzylidene toward alcohol functionalities is

also investigated as a means to directly synthesize well defined models of

alcohol functionalized a-olefin polymers similar to those of

polyvinylalcohol-polyethylene copolymers.















CHAPTER 2

EXPERIMENTAL
Instrumentation and Analysis


1H NMR 300 MHz and 13C NMR 75 MHz spectra were recorded on a
General Electric QE-Series NMR Superconductiong Spectrometer system or a
Varian Associates Gemini 300 Spectrometer. All NMR spectra were recorded
in CDC13 with v/v 0.03% TMS as an internal reference. Chemical shifts

reported were internally referenced to residual chloroform. Infrared data was

recorded on a BioRad FTS/40A infrared spectrometer. Analyses were

performed between NaC1 plates neat or with chloroform as a solvent. Purity

of compounds and reaction conversions were determined on either a

Hewlett-Packard HP5880A gas chromatograph using a capillary column with a

flame ionization detector or on silica coated tic plates with mixtures of

pentane and ethyl acetate as the mobile phase. Micro-extractions were used in

order to monitor reactions by GC. This was done in order to remove all water

soluble salts before injection into the GC. All pertinent GC peaks were
confirmed by mass spectrometry or NMR on the isolated compound. Low

and high Resolution Mass Spectrometry (LRMS), (HRMS) was recorded on a
Finnigan 4500 Gas Chromatography/Mass Spectrometer using either electron
ionization or chemical ionization conditions. Elemental analyses were

performed by Atlantic Microlab, Inc., Norcross, GA.








Gel permeation chromatography (GPC) was performed using a Waters
Associates liquid chromotograph U6K equipped with a tandem ABI

Spectroflow 757 UV absorbance detector and a Perkin-Elmer LC-25 RI detector.
All molecular weights are relative to polybutadiene or polystyrene standards.
Polymer samples were prepared in HPLC grade THF or CHC13 (-1% w/v) and
filtered before injection (a volume of 20-40 uL). The GPC was equipped with a
Ultrastyragel linear mixed bed column (CHC13) or two successive 5 x 103 A
and 5 x 104 A (THF) Phenogel columns (crosslinked polystyrene gel). HPLC

grade chloroform or THF were used as the eluent at a constant flow rate of 1.0
ml/min. Retention times were calibrated against polystyrene standards

(Scientific Polymer Products, Inc.) or polybutadiene standards (Polysciences,

Inc.) All standards Mp or Mw were selected to be well beyond the expected
polymers range. A minimum of 5 data points were achieved for a calibration
curve. On noted samples an internal standard of acetone was used.

Differential scanning calorimetry (DSC) analysis were performed using
a Perkin-Elmer DSC 7 at a heating rate varying between 20- 2'C/min. All
samples were first cooled to -120 "C (using liquid nitrogen as the coolant with
a helium flow at a rate of 30 ml/min.) and underwent isothermal cooling for

2 5 minutes before scanning up to 200 C followed by isothermal heating for
2 5 minutes. Multiple cycles were performed with data collection on the

second heating cycle or later. When transitions were identified the samples
were then slowly scanned over the pertinent temperature range. Reported
values are Tm peak (first order transition peak position), Tm onset and Tg
(glass transition). Thermal calibrations were done using indium and
cyclohexane as standards for both peak temperature transitions as well as for
heats of fusion. All samples were run using an empty pan as a reference and
empty cells as a subtracted baseline. Thermogravimetric analysis was








performed on a Perkin-Elmer TGA 7. All samples were heated from room
temperature to 700 C in nitrogen at a scan rate of 10 'C/min. The onset of

weight loss was taken as the reported value.




Materials and Techniques


Schrock's molybdenum metathesis catalyst [(CF3)2CH3CO]2(N-2,6-C6H3-
i-Pr2)Mo=CHC(CH3)2Ph, was synthesized according to published methods.101

Grubbs' ruthenium catalyst, RuCl2(=CHR)(PCy3)2 were Cy = cyclohexyl, was
provided by the Wagener group members, specifically Mark Watson, Shane
Wolfe, and Dr. John D. Anderson via literature procedure.89 All catalyst
systems employed in this study will be graphically depicted during their
pertinent discussions. p-Toluenesulfonohydrazide (TSH) was purchased
from Aldrich and was recrystallized from methanol prior to use.

Tripropylamine (TPA) and 0-xylene were purchased from Aldrich and
distilled from CaH2 prior to use. Two molar potassium tert-butoxide was

prepared in a dry schlenk tube by combining the salt (Aldrich) with THF
distilled from NaK alloy. 5-bromo-l-pentene, 8-bromo-l-octene (Aldrich) and

10-bromo-l-decene (Alfa Aesar) were used without further purification.
Tetrahydrofuran THF and dimethoxyethane DME were first distilled from

NaK alloy using benzaldehyde as an indicator. "Super dry" ethanol was
prepared as described in the literature.102 All other reagents mentioned in
the experimental were used as received.
Micro-extractions were done by using approximately 1/2 ml of the
reaction mixture with an equal amount of water or acid, followed by
vibration. Frequently 1/2 ml of diethyl ether would also be added to this








solution. After vibrating for 1 min. the mixture was allowed to separate into

layers with the upper (organic) layer used for GC analysis.


Synthesis and Characterization


Synthesis of Symmetrical Alkyl-Substituted Terminal Dienes


Step A (one pot two step synthesis): Synthesis of ethyl-2-acetyl-2-(4-
pentene)-6-heptenoate (la). 10.9g (84 mmoles) of ethyl acetoacetate (Aldrich)

and 200 ml of dry DME (Aldrich) were placed in an argon purged 500 ml three
neck flask equipped with a magnetic stir bar and condenser. To this stirring
solution, 42 ml of a 2 molar solution of potassium tert-butoxide in THF was

added. Upon this addition the solution turned lime-green in color. This

solution was allowed to stir for 0.5 hours at room temperature followed by
slowly adding 12.5g (84 mmoles) of 5-bromo-l-pentene by syringe and raising
the temperature to reflux. White salts were formed upon reflux. The first

addition was complete in 18 hrs as shown by GC. The reaction was then
cooled to room temperature and the second addition of 42 ml of the 2 molar
solution of potassium tert-butoxide in THF was added followed by the alkenyl
bromide as above. The reaction was again followed by GC with the majority
of product formation done in 24 hrs. The reaction was then quenched with
3M HCI and extracted with ether. The ether was dried over MgSO4 and then

evaporated yielding -80% of la. The following spectral properties were
observed: 1H NMR 1.11 ppm (m, br, 4H), 1.19 ppm (t, 3H), 1.79 ppm (m, br,
4H), 1.98 ppm (q, 4H), 2.04 ppm (s, 3H), 4.12 ppm (q, 2H), 4.91 ppm (m, 4H), 5.70

ppm (m, 2H); 13C NMR 14.06 ppm, 23.21, 26.62, 30.79, 33.84, 61.09, 63.36, 115.1,








137.94, 172.5, 205.1. The Low Resolution Mass Spectrum (LRMS, EI) also
confirms structure with a parent ion at 266, calcd for C16H2603 266.

Ethyl-2-acetyl-2-(7-octene)-9-decenoate (Ib). Synthesized as above.

Product not isolated for NMR spectral analysis. LRMS (EI) 350, calcd. for

C22H3803 350.
Ethyl-2-acetyl-2(9-decene)-11-dodecenoate (Ic). Synthesized as above.

Product not isolated before subsequent step. The product of this reaction was

a thick oil. GC retention times were consistent with what was expected for

this compound and the resulting side products.


Step B: Retro-Claisen condensation: Synthesis of ethyl-2-(4-pentene)-6-
heptenoate (2a). In an argon purged dry 250 ml 3 neck round bottom flask,

equipped with a condenser and a magnetic stir bar, 22.34g (84 mmoles) of 1
and 100 ml of dry ethanol were added. To this solution 35ml of a 21%

solution of sodium ethoxide in ethanol (Aldrich) was added. (This solution

was also prepared by the addition of 21 weight percent of sodium metal into

"dry" ethanol. Caution: This addition should be done under an inert

atmosphere using "super dry" ethanol. Any contact with water in the

presence of oxygen can likely result in a fire. Super dry ethanol was prepared

as described in the purification of organic compounds.102) The solution was

allowed to reflux for 3.5 hours and turned a dark yellow in color. After

cooling to room temperature the solution was quenched with water and 3M
HC1 followed by extraction with pentane or ether. The organic layer was then
dried over MgSO4 and evaporated under reduced pressure yielding -90% of

ester 3. The product was vacuum distilled through a short path vacuum
distillation apparatus. The product was collected between 130-150 "C at
4mmHg. The following spectral properties were observed: 1H NMR (CDC13):








1.25 ppm (t, 3H), 1.40 ppm (m, br, 6H), 1.62 ppm (m, br, 2H), 2.05 ppm (q, 4H),
2.31 ppm (m, 1H), 4.14 ppm (q, 2H), 4.95 ppm (m, 4H), 5.79 ppm (m, 2H); 13C

NMR: 14.5 ppm, 26.9, 32.0, 33.8, 45.5, 60.0, 114.9, 138.6, 176.0. The Low

Resolution Mass Spectrum (LRMS, EI) also confirms structure with a parent
ion at 244, calcd for C14H2502 244.

Ethyl-2-(7-octene)-9-decenoate (2b). Synthesized as above with high
conversion indicated by GC. 1H NMR (CDC13): 1.24- 1.55 ppm (m, br, 24H),

2.01 ppm (q, 4.6H), 2.28 ppm (m, 1H) 4.10 ppm (m, 2H) 4.92 ppm (m, 4H) 5.75

ppm (m, 2H); 13C NMR: 14.31, 25.80, 27.31, 28.09, 28.79, 28.89, 29.33, 32.44,
32.61, 33.70, 45.69, 46.50, 59.88, 64.05, 114.12, 139.04, 175.93, 176.54. Spectra

consistent with structure. LRMS 308, calcd. for C20H3602 308. HRMS 308.2739,
calcd. for C20H3602 308.2715.

Ethyl-2-(9-decene)-11-dodecenoate (2c). Synthesized as above with high
conversion indicated by GC. Product not isolated before subsequent step.

Product was a thick oil with consistent GC retention times with what was
expected.


Step C: Reduction of ester: Synthesis of 2-(4-pentene)-6-heptene-l-ol
(3a). In a flamed dry 250 ml three neck round bottom flask equipped with a
stir bar and condenser, 9.301g of 2 and 125 ml of dry THF were placed. This
solution was kept under an inert atmosphere and cooled to 0 'C. To this
stirring solution, 25 ml (2 eq) of a 1M solution of LiAIH4 (Aldrich) in THF

was added over a period of 5 min. Some bubbling was observed during this
addition. The reaction was allowed to warm to room temperature and stirred
for 2 hours. The reaction was then slowly quenched with water followed by
3M HC1. The solution was extracted with ether dried over MgSO4 and rotary

evaporated. A clear oil was recovered and vacuum distilled at 69-72 "C at








1mm Hg. The reduction resulted in 91% yield. The following spectral
properties were observed: 1H NMR (CDC13): 1.18 ppm (m, br, 9H), 1.79 ppm

(s, 1H), 1.98 ppm (q, 4H), 3.49 ppm (d, 2H), 4.88 ppm (m, 4H) 5.78 ppm (m 2H);
13C NMR: 26.08 ppm, 30.29, 33.98, 34.01, 40.22, 65.36, 114.28, 138.74. IR (CDCl3,

cm-1): 3383.8 (br), 3078.1, 3013.4, 2931.0, 2860.9, 1640.4, 1460.3, 1217.1, 1030.5,
996.1, 913.1, 759.5. Elemental analysis C12H230 C(calc=79.06, found=78.99),

H(calc=12.16, found=12.16).

2-(7-octene)-9-decene-l-ol (3b). Synthesized as above with high
conversion by GC. Short path vacuum distillation was done for purification
with the main fraction collected at -100-110 'C at lmmHg. Gas cromatograph

showed a purity of 98%. Isolated yield was 61%. Spectral analysis: 1H NMR

1.27 ppm (m, br, 22H), 2.03 ppm (br, 4H), 3.50 ppm (m, 2H) 4.93 ppm (m, 4H),
5.78 ppm (m, 2H); 13C NMR: 26.77 ppm, 28.87, 29.05, 29.84, 30.86, 33.72, 40.45,

65.50, 114.06, 139.06.
2-(9-decene)-11-dodecene-1-ol (3c). Synthesized as above with high
conversion by GC. Due to the size, and therefore the boiling point, of this
product a high vacuum short path distillation was done using a short path
high vacuum distillation apparatus equipped with a dry ice isopropanol

condenser. The product, while stirring with a Teflon stir bar, was brought to
full vacuum (>10-5 mmHg) at room temperature. The solution was then

slowly heated until condensing was observed. Due to the design of this
apparatus the temperature of the condensing gas could not be measured.
During the distillation three fractions were taken. The temperature of the oil
bath was approximately 180 'C. A purity of 96% was determined by GC.
Spectral analysis: 1H NMR (CDC13): 1.29 ppm (s, br, 32H), 1.71 ppm (s, 1H),

2.02 ppm (q, 4H), 3.49 ppm (d, 2H), 4.95 ppm (m, 4H), 5.38 ppm (br, 0.2H,
elimination product), 5.78 ppm (m, 2H); 13C NMR: 25.71, 26.83, 28.88, 29.09,








29.45, 29.54, 30.02, 30.88, 32.72, 33.75, 40.46, 65.51, 114.03, 139.08. LRMS (CI)

M+1=323, calcd. for C22H420 322.


Step D: Tosylation: 6-p-toluenesolfonyl methyl-l,10-undecadiene (4a).
In a flame dried and argon purged 100 ml three neck flask, 8.290g of the

alcohol 3a (46 mmoles) and 60 ml of CHC13 were added with a stir bar. This

solution was then cooled to 0 "C followed by the addition of 7.28g (92

mmoles) of pyridine. After stirring for 15 min., 13.03g (69 mmoles) of p-
toluenesolfonyl chloride (Aldrich) dissolved in 35 ml of CDCL3 was slowly
added (15 min.) by syringe or addition funnel. A brown yellow color was

observed following this addition. This solution was then allowed to warm to
room temperature and stirred for 8 hours. The reaction was followed by GC
until high conversions were observed. The reaction was then stopped and

washed with 3 M HCI to remove the pyridine. The organic layer was then
washed with water followed by a wash with K2C03 sat. in order to remove
unreacted tosyl-chloride. The water layers were all extracted with chloroform

which was then combined with the original organic layer. The organic layer
was then dried over MgSO4, filtered through a glass fitted funnel followed by

rotary evaporation. The resulting product was a viscous oil with a yellow

tint. The product was not further purified due to fear of elimination. A yield
of -30% was determined by GC with solvent subtraction. The following
spectral properties were observed: 1H NMR (CDC13): 1.29 ppm (m, 9H) 1.79

ppm (s, br, 1H), 1.97 ppm (s, br, 4H), 2.45 ppm (s, 3H), 3.92 ppm (m, 2H), 4.94
ppm (m, 4H), 5.72 ppm (m, 2H), 7.35 ppm (d, 2H), 7.80 ppm (d, 2H); 13C NMR:
21.62 ppm, 25.73, 30.07, 33.79, 37.47, 72.63, 114.62, 127.85, 127.91, 129.81, 138.46,
144.67. LRMS 336, calcd for C19H28SO3 336.








9-p-toluenesolfonyl methyl-1,16-heptadecadiene (4b): Synthesized as
above, with similar yields. The crude product had the appearance of a viscous
brown cloudy solution. Product not isolated before subsequent reaction.
11-p-toluenesolfony methyl-1,18-uneicosadiene (4c) Synthesized as
above. Yield was not determined by GC due to its high boiling point. The
crude product had the appearance of a viscous brown cloudy solution.
Spectral analysis: 1H NMR (CDC13): 1.23 ppm (m, br, 46H), 2.05 ppm (q, 5H),
2.42 ppm (s, 5.4H) 3.50 ppm (d, unreacted alcohol) 3.90 ppm (d, 0.9H), 4.95 ppm
(m, 4H), 5.40 ppm (s, br, blip, elimination product), 5.80 ppm (m, 1.6H), 7.33

ppm (d, 2.9H), 7.79 ppm (d, 2.66H), some unreacted tosyl chloride or acid was
also detectable in this region of the proton NMR.


Step 6: 6-methyl-l,10-undecadiene (5a). Approximately 9.66g (27
mmoles) of 4a was placed in a flame dried argon purged 250 ml three neck
flask equipped with a condenser and stir bar. 100 ml of dry THF was then
added and the solution was cooled to 0 "C. To this stirring solution 20 ml
(16.2 mmoles) of a 1 molar solution of lithium aluminum hydride (LAH) was
slowly syringed in (15 min.). Bubbling occurred upon first addition. After the
complete addition of the LAH the solution was brought to reflux for 5 hours
and monitored by GC. A white precipitate was formed during the reflux. The
reaction was then cooled and quenched slowly, first with water, followed by 3
N HCI (Caution: Addition of water and HC1 should be done slowly due to the
highly exothermic reaction and the copious production of hydrogen gas).
This mixture was then extracted three times with ether, followed by the
washing of the ether layer with sat K2C03. (In some cases the washing with
K2CO3 was omitted without change.) The organic layer was then dried over
MgSO4 and then filtered through a glass fritted funnel and rotary evaporated.








The product was a clear oil and isolated by short path vacuum distillation (28-

30'C at 2 mmHg) in approximately 60% yield. The following Spectral

properties were observed: 1H NMR (CDC13): 0.77 ppm (d, 3H), 1.09 ppm (m,

1H), 1.32 ppm (m, br, 10H), 1.99 ppm (q, 4H), 4.91 ppm (m, 4H), 5.78 ppm (m,

2H); 13C NMR 19.50 ppm, 26.29, 32.46, 33.98, 36.39, 113.99, 139.06, 139.11. Low

resolution mass spectrometry (EI) = 166, calcd for C12H22 166.

9-methyl-1,16-heptadecadiene (5b). Synthesized as above. Spectral

properties: 1H NMR 0.79 ppm (d, 3H), 1.29 ppm (m, br, 22H), 2.01 ppm (q, 4H),

4.94 ppm (m, 4H), 5.79 ppm (m, 2H); 13C NMR 19.74 ppm, 27.07, 29.01, 29.26,

29.91, 32.78, 33.87, 37.12, 114.12, 139.28. Elemental analysis C18H34 C(calc=86.32,

found=86.31) H(calc=13.68, found=13.60).

11-methyl-1,20-uneicosadiene (5c). Synthesized as above with only

three hours of reflux. Spectral properties: 1H NMR 0.82 ppm (d, 3H), 1.23

ppm (br, m, 40H), 2.05 ppm (q, 4.5H), 4.98 ppm (m, 4H), 5.49 ppm (m, 0.09H),
5.81 ppm (m, 2H); 13C NMR 19.55 ppm, 26.91, 28.83, 28.98, 29.33, 29.43, 29.46,

29.83, 32.65, 33.60, 36.98, 65.39, 113.84, 139.01. Low resolution mass

spectrometry (EI) = 306, calcd for C22H42 306. High resolution mass

spectrometry (CI M+1) = 307.3396, calcd for C22H43 = 307.3365.



Alternative enolate displacement reactions for the synthesis of compounds
la, and lb


Ethyl-2-acetyl-6-heptenoate (ml) was prepared by an enolate
displacement of a alkenyl bromide. Approximately 100 ml of "super dry"
ethanol was distilled into a 200 ml three neck flask with a condenser. Na

metal (Aldrich 1.035 g, 0.04501 moles) was then added to the argon filled flask
at room temperature. Caution: This addition should be done under an inert








atmosphere using "super dry" ethanol. Any contact with water in the

presence of oxygen can likely result in a fire. This was then refluxed until all

of the Na dissolved (1 hour). The solution was allowed to cool then ethyl

acetoacetate (Aldrich 9.0 g, 0.069 moles) was added slowly via an addition

funnel or syringe. This solution was then allowed to reflux for 1 hour. A

color change from colorless to lime green was observed. The solution was

allowed to cool to room temperature and then 5-bromo-l-pentene (Aldrich

6.7 g, 0.045 moles) was added over a period of 30 min. A color change to

milky yellow was observed (incipient precipitation of NaBr salts). This
solution was allowed to reflux for 24 hours. The solution was transferred to a

200 ml RB flask and the ethanol evaporated off. The remaining oil was then
extracted with water and ether. The ether layer was then dried with MgSO4

followed by rotary evaporation. A crude yield of 94% was attained. The

product was then purified by short path vacuum distillation. The products

fraction came over at the range 70-74C at a pressure of -2mmHg. The
following spectral properties were observed: 1H NMR(CDC13) 1.28 ppm (t,

3H), 1.39 ppm (m, 2H), 1.85 ppm (m, 2H), 2.08 ppm (q, 2H), 2.23 ppm (s, 3H),

3.41 ppm (t, 1H), 4.20 ppm (q, 2H), 5.00 ppm (m, 2H), 5.78 ppm (m, 1H); 13C

NMR: 13.96 ppm, 14.14, 20.96, 30.04, 50.02, 59.74, 61.18, 89.65, 166.96, 200.22,
200.26.

Ethyl-2-(4-pentene)-6-heptenoate (2a) and ethyl-2-acetyl-2-(4-pentene)-6-

heptenoate (la) were prepared and worked up in the same manner as 6.
Compound 6 was used as a reactant in place of ethyl acetoacetate. Upon

addition of 6 the solution turned from clear to a dark red. After the addition
of 5-bromo-l-pentene the solution again turned a milky yellow due to

precipitating salts. The reaction was refluxed for 112 hours, followed by

evaporation, extraction, and distillation. A GC trace showed that there were 2








major products, 34% (2a) and 47% (la). The following spectral properties were
observed: (2a) 1H NMR (CDC13): 1.25 ppm (t, 3H), 1.40 ppm (m, br, 6H), 1.62

ppm (m, br, 2H), 2.05 ppm (q, 4H), 2.31 ppm (m, 1H), 4.14 ppm (q, 2H), 4.95
ppm (m, 4H), 5.79 ppm (m, 2H); 13C NMR: 14.5 ppm, 26.9, 32.0, 33.8, 45.5, 60.0,
114.9, 138.6, 176.0. LRMS: 224 calcd. for C14H2502 224. (la) 1H NMR 1.11

ppm (m, br, 4H), 1.19 ppm (t, 3H), 1.79 ppm (m, br, 4H), 1.98 ppm (q, 4H), 2.04
ppm (s, 3H), 4.12 ppm (q, 2H), 4.91 ppm (m, 4H), 5.70 ppm (m, 2H); 13C NMR
14.06 ppm, 23.21, 26.62, 30.79, 33.84, 61.09, 63.36, 115.1, 137.94, 172.5, 205.1.
LRMS 266, calcd for C16H2603 266.

The above reactions using ethanol and Na metal, was also done in
order to produce 2a in a one pot procedure. After the first addition of Na and
alkene bromide (as stated above) the reaction was allowed to cool to room
temperature followed by an addition of a second equivalent of Na metal. The
solution was allowed to stir for 1/2 hour forming a dark orange color. The
solution was then brought to reflux, dissolving the remaining sodium,
followed by the slow addition of one equivalent of 5-bromo-l-pentene (1/2
hour). The reaction was followed by GC producing both products 2a and la.
All of the product could then be converted to 2a by an addition (1/2
equivalent) of sodium metal. The reaction was then cooled to room
temperature followed by quenching with water and 3N HCL. The reaction
was extracted as listed above followed by isolation by vacuum distillation as
listed before. An isolated yield of 36% was obtained using this method.
Spectral analysis was consistent with the structure and as listed before.


Ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la) was also pursued by an
enolate displacement of a bromine in an aprotic solvent. Sodium ethoxide
was made by the addition of sodium (Aldrich, 0.116g 0.005 moles) into a 100








ml three neck flask containing 50 ml of "super dry" ethanol under argon.

Caution: This addition should be done under an inert atmosphere using
"super dry" ethanol. Any contact with water in the presence of oxygen can

result in a fire. This mixture was allowed to mix until all of the sodium

dissolved. The ethanol was then distilled off under reduced pressure. 6 (1.00g

0.005 moles) and an excess of diethyl carbonate was added to the cooled solid.

This was then mixed at room temperature until most of the salt dissolved.
This solution was then fractionated under reduced pressure (150 mmHg at 60-

70C) to remove any formed ethanol. 5-Bromo-l-pentene (0.0745g 0.005
moles) was then added to the solution. A condenser was then placed in the
center neck and the temperature was then raised to 1150C at standard
pressure. The solution was then stirred at this temperature for 104 hours.
The diethyl carbonate was then evaporated off and water was added, la was
extracted with ether and dried over MgSO4. A GC trace was done, which

showed two products. The products were distilled under reduced pressure
(10-2 mmHg 90-100C). The spectral properties were the same as 2a, and la

above. The yield on this reaction was 13% for la and 4% for 2a. The yields
were never optimized. Spectral analysis was the same as reported before.

Compound la was pursued again by use of a hydride base and an
aprotic solvent. A solution of 6 (3.0 g 0.015 moles) in dry 1,2-
dimethoxyethane (DME) (20 ml) was added dropwise under nitrogen to a
stirred mixture of potassium hydride (1.72 g of 35% suspension 0.015 moles
Aldrich) in DME (100 ml). When evolution of hydrogen stopped a solution
of 5-bromo-l-pentene (2.31 g 0.015 moles Aldrich) in DME (20 ml) was added
dropwise. The solution was heated under reflux for 24 hours. The DME was
then evaporated and la was again worked up as before. Compound la was








recovered in a 50% yield before optimizing. The spectral properties were the

same as reported before.


Ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la) and ethyl-2-(4-pentene)-6-
heptenoate (2a) using KH (one pot two step method): In flame dried argon
purged 100 ml 3 neck round bottom flask with condenser, 3.5 g (0.03M) of a
35% dispersion of KH in mineral oil was washed three times with 20 ml of

dry pentane. The remaining white salt was pumped to dryness under
vacuum. Approximately 50 ml of dry dimethoxy ethane (DME) was then

cannula transferred to this flask followed by stirring. No evolution of
hydrogen was observed. To this stirring solution 3.9g (0.03M) of

ethylacetoacetate was slowly added (20 min.) resulting in the evolution of a

gas (H2) and a change from colorless to a clear lime green color. The solution
was then heated to 70 *C followed by the slow addition (15 min.) of 4.47g

(0.03M) 5-bromo-l-pentene. Upon this addition the mixture turned yellow in
color with a suspension of white salts. The solution was refluxed for 20 hours
followed by cooling to 50 'C. A second equivalent of washed KH was added
under argon. Vigorous bubbling was observed with a darkening of the yellow
color to an orange. The second addition of 5-bromo-l-pentene was then
added followed by refluxing for 72 hrs. The reaction was then cooled followed

by evaporation of the solvent. The remaining orange liquid was then
extracted 3 time with water and ether. The organic layer was dried over

MgSO4 and rotary evaporated. A 60% yield was obtained (41% la, 18% 2a by
GC). A high vacuum distillation and a low temperature recrystallization
were attempted with poor results. A column was run on 200 mg of the
product using 80% pentane and 20% ethyl acetate through silica gel. A purity
of 94.4 was attained for la as determined by GC. 1H and 13C NMR are








consistent with structure la. This procedure was also done in two steps with

complete isolation of 6 and la after each step. Yields were lower due to loss

during isolation.


Ethyl-2-(4-pentene)-6-heptenoate (2a) using NaH (one pot two step
method): In a 500 ml 3 neck flask equipped with a condenser and stir bar,

10.9g (0.085 M) of ethylacetoacetate and 250 ml of DMF were stirred. To this

solution 8.16g (4 equivalents) of powdered sodium hydride was added.

Bubbling (H2 gas) occurred immediately upon this addition. The solution

was stirred at room temperature for 0.5 hours followed by the addition of 25g

(0.17 M) of 5-bromo-l-pentene. The temperature of the solution was then

raised to 120 "C. The reaction was followed by GC, using micro-extractions,

showing the production of both 2a and la. The reaction was continued for 20

hrs. At this time GC showed almost full conversion to the ester 2a. Full

conversion was often achieved by the addition of dry ethanol to the basic

solution. The solution was then cooled to room temperature followed by

slowly quenching with water and 3N HCL (Caution: Addition of water and

HCI should be done slowly due to the highly exothermic reaction and the

copious production of hydrogen gas.) The solution was then extracted 3 times

with 250 ml portions of ether. The organic phase was then dried over MgSO4

followed by filtering and rotary evaporation. GC showed that all the product

had been converted to 4. The product was isolated by a short path vacuum
distillation (74-85 "C at lmmHg) with approximately a 51% yield. Spectral
analysis was consistent with the structure and as listed before.


Ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la) and ethyl-2-(4-pentene)-6-
heptenoate (2a) using KOH, NaOH or 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)








(one pot two step method); general procedure. In a 500 ml three neck flask

10.9g (0.085 moles) of ethyl acetoacetate and 250 ml of DMF (anhydrous) were

stirred. Approximately 23.8g (5 eq) of potassium hydroxide pellets were added

followed by the addition of 25g (0.17 moles) of 5-bromo-l-pentene. The

solution was then stirred and heated to 120 'C upon which all of the KOH

pellets dissolved. A yellow brown color was observed. After three hours GC

showed the productions of both products with product (la) in majority. After
24 hours the pH of the solution was taken resulting in a pH of -10. No

change in the product ratio was detected by GC. More KOH was then added

until the pH was over 12. The reaction was then allowed to stir for 12 more

hours. At this time GC showed a mixture of products with the majority

product being 2a when the bases KOH of NaOH were used. When DBU was

used alone, only product la was formed. The reaction was then cooled to

room temperature and quenched with 3M HCL and water. This solution was

then extracted three times with 100 ml portions of ether. The organic layer

was then dried with MgS04 followed by rotary evaporation. The neat product
was dark brown orange in color. A short path vacuum distillation was then
done with the ester product distilling between 90-110 "C at 6 mmHg. An

isolated yield of 43% was obtained. The remaining product was not

identified. Spectral analysis was consistent with the structure and as listed
before.


Ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la) using NaNH2. In a dry
100 ml 3 neck flask with stir bar and condenser, 2g (0.015 moles) of ethyl
acetoacetate was mixed with 50 ml of dry dimethoxy ethane (DME). To this

stirring argon purged solution, 1.2 g (0.015 moles) of NaNH2 in a 50% weight

solution in toluene, was pipetted in. A lime green color appeared








immediately upon this addition. The solution was then heated to 60 'C

followed by the addition of 0.015 moles of the alkene-bromide. The

production of white salts upon this addition was immediately observed. The

solution was then allowed to reflux for 8 hrs. and followed by GC. The

reaction mixture was then cooled to room temperature followed by a second

addition of base and alkene-bromide in the same manner as before. A yellow

color and a significant production of salts were observed. The product

mixture was again followed by GC and the reaction was stopped after 12 hrs.

of reflux. The reaction was then quenched under an inert atmosphere using 3

M HC1. Diethyl ether was then added to this solution and 3 extractions were

done. The organic layer was then dried over anhydrous magnesium sulfate,
followed by filtration and evaporation. A EI-GC/MS was done showing that

the crude solution was a mixture of previously isolated products and a tri-

alkylated product. The product mixtures are illustrated in Figure 3.14. No

further isolation was done on the mixture.


Alternative Methods for the Synthesis of Compounds 5a-5c


Attempted decarbonylation of ethyl-2-acetyl-2-(4-pentene)-6-heptenoate
to form 6-acetyl-l,10-undecadiene (7). The decarbonylation was attempted by

the addition of an acid, base or a salt with the appropriate solvent (DMF,

DMSO, H20). These reactions were brought to reflux and monitored by

micro-extractions followed by GC. After the reactions were cooled to room
temperature, they were extracted with pentane and water. The pentane was
then rotary evaporated followed by spectral analysis if appropriate. No
reaction or decomposition was observed for these reactions. When base

(KOH) was used the expected increase in (2a) was observed by GC.










6-acetyl-l,10-undecadiene (7). In a 25 ml round bottom flask with 10 ml
of NMP and 0.5g of water, 2g of LiCI and Ig of the keto-ester (la) was added.

The solution was heated to reflux noting a color change from colorless to

yellow by 150 "C. All salts were dissolved by 160 C. Reflux was maintained

for 6 h forming a brown solution. The reaction was monitored by GC using

micro-extractions and showed the disappearance of the starting material and

the appearance of a new peak with a retention time of 10.78 minutes. The

starting material has a retention time of 12.77 minutes. The reaction was

allowed to reflux for a total of 9 hours. The reaction was cooled to room

temperature and extracted with 200 ml of water and 50 ml of pentane. The

organic layer was then dried over anhydrous magnesium sulfate followed by

filtering and rotary evaporation. A crude yield of 90% was obtained. Low

resolution mass specrometry calculated M + 1 = 195 actual M + 1 = 195. 1H
NMR:1.19 ppm (m, br, 8H), 1.51 ppm (m, 2H), 1.96 ppm (q, 4H), 2.04 ppm (s,

3H), 4.89 ppm (m, 4H), 5.69 ppm (m, 2H).


Wittig Method


(1-Pentene)triphenylphosphonium bromide (8) was synthesized as a
Wittig salt. 5-Bromo-l-pentene (Aldrich, 15 ml, 0.1 moles) was added to a 250

ml 3-neck flask containing 145 ml DMF and triphenylphosphine (Aldrich,
28.85 g, .11 moles). A TeflonTM coated stir bar and a condenser were then
added. The solution was refluxed at 1700C for 20.5 hours. The salt was then

precipitated out with diethyl ether. The salt was vacuum filtered and the
filtrate was treated again with ether for further precipitation. A yellow white
solid was recovered. The salts were then washed with hot THF then mixed in








this solvent until they dissolved partially. The solution was then allowed to

cool to room temperature and then cooled in a freezer. The salts were then

collected cold by vacuum filtration to give 1 as white crystals in a 93% yield.
Compound 1 was dried in a schlenk tube at room temperature at 10-2 mmHg
for 12 hours. 1H NMR (CDC13): 1.68 ppm (m, 2H), 2.23 ppm (q, 2H), 3.80 ppm

(m, 2H), 5.02 ppm (t, 2H), 5.65 ppm (m, 1H), 7.80 ppm (m, 15H). Elemental
calcd: C, 67.00; H, 6.11; P, 7.51; Br, 19.38. Obsd: C, 67.07; H, 6.05.


Tertiary Alcohol Synthesis and Reduction


6-methyl-l,10-undecadiene-6-ol (9), was synthesized by the Grignard
reaction with the ethyl acetate. 5-Bromo-l-pentene was dried over freshly
ground CaH2 for 3 hrs, then vacuum transferred to a schlenk flask with 4 A

molecular sieves. In a flame dried 100 ml three neck round bottom flask 1.63g
(67 mmoles) of freshly ground magnesium turnings were placed followed by
the addition of 50 ml of dry ether. A crystal of iodine, and a small amount of
the bromide was added in order to initiate the reaction. The balance of the
bromide was then added at such a rate as to maintain reflux. After complete
addition, the mixture was stirred and refluxed for 0.5 hours. After cooling to

room temperature 33.5 mmoles of dry ethyl acetate was added to produce (9).
Refluxing occurred upon addition of the electrophiles and was maintained by

heating for 2 hours. The reaction was then cooled and quenched with 3
molar HC1. The reaction mixture was extracted with ether and dried over

anhydrous magnesium sulfate, followed by filtering and evaporation yielding
5.18g (84%) of a clear liquid 9. The monomer was distilled under reduced
pressure (70-80 'C at lmmHg) using a short path distillation apparatus. The
following spectral properties were observed: 1H NMR (9) 1.14 ppm (s, 3H) 1.43








ppm (m, 9H), 2.05 ppm (m, 4H), 4.98 ppm (m, 4H), 5.80 ppm (m, 2H); 13C
NMR: 23.12 ppm, 26.80, 34.07, 41.21, 72.52, 114.48, 138.61; Elemental calcd: C,
79.06; H, 12.16; 0, 8.78. Obsd: C, 79.16; H, 12.15.


5-methyl-1,9-decadiene-5-ol (10) was prepared by using a Grignard
reagent with a ketone. 1-pentene-5-magnesium bromide (0.1 moles) was

prepared in the usual manner as a Grignard reagent. The reagent was
prepared in a 200 ml three neck flask and condenser with 20 ml of dry ether.

5-hexene-2-one (Aldrich 9.8 g, 0.1 moles) and 20 ml of ether were then slowly
added via an addition funnel. A 15 min. reflux and a color change from

brown to green was observed. The solution was allowed to stir for 7 hours
and then transferred to a 250 ml Erlenmeyer. Approximately 10 ml of 1N
H2S04 was added and 10 was extracted using water and ether. The ether layer
was dried over MgSO4, evaporated, and then vacuum distilled using a short

path vacuum apparatus. The alcohol fraction came over at 900C at 10-1
mmHg. 10 was recovered in a 78% yield and a 99.92% purity determined by
GC. The following spectral properties were observed: 1H NMR (CDC13): 1.17

ppm (s, 3H), 1.45 ppm (d, br, 4H), 1.55 ppm (p, 2H) 2.10 ppm (m, 4H), 5.00 ppm
(m, 4H), 5.82 ppm (m, 2H); 13C NMR: ppm 23.20, 26.75, 26.81, 28.29, 34.13,
40.80, 41.37, 72.55, 114.30, 114.6, 138.6, 138.9. Elemental calcd: C, 79.51; H, 11.98;

0, 9.51. Obsd: C, 76.16; H, 11.36.


Attempted synthesis of 5-methyl-1,9-decadiene (11) or 6-methyl-l,10-
undecadiene (5a) (Carey method). In a dry 250 ml 3 neck flask 0.786 (4.6
mmoles) of 2-(3-butene)-6-heptene-2-ol (10) and 1.34g (5.1 mmoles) of
triphenyl silane (weighed out in nitrogen purged glove bag) were placed with
15 ml of methylene chloride. To this stirring solution 0.80g (7.0 mmoles) of








trifloroacetic acid was dripped in. No physical change was observed upon

addition. The reaction was followed by GC using micro-extractions. After 24

hours no significant reduction in the starting material was observed so an

additional 1.5 equivalents of trifloroacetic acid was added. This solution was

then allowed to mix for an additional 12 hours. The reaction was then

quenched with an excess of saturated potassium carbonate followed by

extraction with 3 equivalent amounts of diethyl ether. The organic phase was

then dried over anhydrous magnesium sulfate followed by filtration then

rotary evaporation. The crude reaction mixture was then run through an

alumna column using pentane as the mobile phase. A GC/MS-CI/methane

was done showing that one of the minor products (5%) had a base peak of 165.

This indicates that there is some eliminated product present. 1H NMR was

relatively messy, but the olefin region did show some internal olefin

resonance's at 5.6 ppm. The target product could not be identified by GC/MS

or NMR. The reaction was repeated again using the same scale with shorter

reacting times, but resulting in the same results.


Attempted synthesis of 5-methyl-1,9-decadiene or 6-methyl-l,10-
undecadiene (5a) (Ireland method). Approximately Ig (5.5 mmoles) of

alcohol 9 was placed in a 100 ml three neck flask containing 40 ml of a 4 : 1

mixture of dry THF and tetramethylethylene diamine (TMEDA) respectively.

To this stirring solution at room temperature 2.5 ml (6 mmoles) of n-butyl
lithium in hexanes was slowly added by syringe. After mixing for 20 min. 0.9

ml (5.5 mmoles) of Diethyl chlorophosphate was dripped in via syringe.
After the first 0.5 hour white salts were noticed. The solution was allowed to
mix at room temperature for 3 hours followed by quenching with excess 1 M

HCI in anhydrous ether. This was done in order to remove the TMEDA and








LiCI salts. After the addition of the acid, large amounts of salt precipitated out

which were then filtered by using a fine frit funnel. The pH was taken

followed by additional acid washes until the solution was acidic. The ether

was then evaporated giving 850 mg of the crude reaction mixture. A GC/MS

was done on this crude mixture. The crude mixture was then dissolved in 20

ml of THF and 800 mg of t-butanol. This solution was then dripped into a

mixing solution of 20 ml EtNH2 with 200 mg of Li metal cooled to 0 'C.

Some bubbling was noted upon this addition. The reaction was stirred for 3

hours without dissolving the lithium metal. The reaction was quenched

with water and the excess lithium metal was filtered out. The reaction was

extracted with ether, dried, filtered and rotary evaporated. A GC and a

GC/MS was done showing no conclusive results of product formation.


ADMET Polymerizations of Monomers 5a and 5b


General Metathesis Conditions


All glassware was thoroughly cleaned and flame dried under vacuum
before use. The monomers were vacuum fractionally distilled or column

chromatographed to a purity of > 99% as determined by GC and elemental

analysis. The monomers were then dried over calcium hydride for a

minimum of 8 hours and then vacuum transferred onto a sodium-potassium
alloy to insure dryness. They were stirred on this alloy for a minimum of 4
hours. During this period the clear solutions would turn blue in color.
Monomers were degassed by subjecting them to several freeze pump thaw
cycles under high vacuum (<10-4 torr). The dry degassed monomers were








then vacuum transferred into a clean dry flask fitted with a TeflonTM vacuum

valve.

All metathesis reactions were initiated in the bulk, under argon glove
box conditions. Monomers were introduced into a 25 or 50 ml round-

bottomed flask equipped with a high vacuum TeflonTM valve and magnetic

stir bar. On occasion, treatment with catalyst for purification was necessary.

All monomers (0.4-2.0 g) were opened in the dry box followed by the addition

of catalyst (2-20 mg). After the addition of catalyst, the reactions were exposed

to intermittent vacuum at room temperature, until a noticeable increase in

viscosity was apparent. When the stirring of the reaction became hindered it

was then placed under high vacuum in order to remove the continuous

generation of ethylene. All the reactions were started at room temperature

then ramped to 60 "C over a period of 2 to 3 days. When no further

evolution of ethylene was apparent (bubbling stopped) the solutions were

then cooled to room temperature and quenched by exposure to air. Polymers

were isolated in high yield (-90%), followed by NMR, GPC, and elemental

analysis on the crude reaction mixture. The polymers were then precipitated

from toluene into cold methanol producing a clear, colorless to white

semisolid. If the polymer retained any color (due to catalyst) it could be

removed by dissolving the sample into toluene followed by passing it

through a short alumina column.


Polymerization of 6-methyl-l,10-undecadiene (P5a). The addition of
approximately 5-10 mg of Schrock's Mo catalysts resulted in the immediate

evolution of ethylene. The reaction was then placed onto a high vacuum
line and exposed to intermittent vacuum until there was a noticeable increase

in viscosity (30 min.). The reaction was then exposed to high vacuum (<10-5








mmHg) and stirred. As the viscosity of the system increased the solution was
slowly heated in order to facilitate stirring, in the manner stated above. After
precipitation the polymer was a clear white nonflowable tacky solid. The

polymer was dried by exposure to schlenk line vacuum at approximately
60'C. The following spectral analysis were observed: 1H NMR 0.78 ppm (d,
3H), 1.03 ppm (m, 1H), 1.28 ppm (m, 8H), 1.89 ppm (m, br, 4H), 5.29 ppm (m,

2H); 13C NMR 19.68 ppm, 27.13, 27.54, 32.58, 32.95, 36.54, 36.70, 129.92, 130.38.
Elemental analysis for repeat C10H18 C(calc=86.88 found=85.59) H(calc=13.12

found=13.01). Molecular weight determination for all the batches is discussed

in the results and discussion (Table 4.1).


Polymerization of 9-methyl-1,16-heptadecadiene (P5b). The same
general procedure was followed as for P5a. In this batch of monomer, some
impurities were present so the reaction was restarted by further addition of
Schrock's Mo catalyst in order to purify. The polymer was precipitated and
dried as stated above. A clear to white tacky solid was obtained with the
following spectral analysis: 1H NMR 0.81 ppm (d, 3H), 1.06 ppm (s, br, 1H),
1.29 ppm (s, br, 10H), 1.98 ppm (m, br, 4H), 4.94 ppm (m, end group) 5.37 ppm

(m, br 2H), 5.80 ppm (m, br, =CH2); 13C NMR 19.53 ppm 26.91, 27.10, 29.04,
29.19, 29.54, 29.64, 29.72, 32.39, 32.67, 37.00, 129.73, 130.21. Elemental analysis

for repeat C16H30 C(calc=86.4 found=84.9) H(calc=13.6 found=13.4). Molecular
weight determination for all the batches is discussed in the results and
discussion (Table 4.2).








Hydrogenation of Polymers P5a and P5b


Hydrogenation of poly [6-methyl-l,10-undecadiene] (HP5a). The
hydrogenation was performed in a flame dried, 20 ml round-bottom three-
neck flask equipped with a reflux condenser, TeflonTM stir bar, and supplied
with a positive Argon pressure. To this flask 50 mg of polymer P5a, 5 ml of
dry reagent grade o-xylene, 135 mg (2.5 eq) of TSH, and 103 mg (2.5 eq) of TPA
were added. The mixing solution was then heated to reflux. All of the TSH
did not dissolve until the solution was close to reflux. Upon heating some
gas evolution could be observed (bubbling of N2) before reflux was achieved.
The solution was allowed to reflux for 6 hours followed by cooling to room
temperature. Some white precipitation was noted at this time. A second
addition of 68 mg (1.25 eq) of TSH, and 51 mg (1.25) of TPA was administered,
followed by refluxing for 3 hours. The polymer was then precipitated in cold
methanol. The methanol was then decanted off followed by drying the
polymer in a schlenk flask under reduced pressure at 60 'C for 18 hours.
Approximately 70% of the polymer was recovered. A semi-clear tacky solid
was recovered with the following spectral analysis: 1H NMR 0.83 ppm (3H,
d), 1.10 ppm (2H, br), 1.27 ppm (14H s); 13C NMR 18.8 ppm, 26.2, 28.8, 29.5, 30.0,
36.3. Molecular weight determination is displayed in Figure 4.14.


Hydrogenation of poly [9-methyl-1,16-heptadecadiene] (HP5b). The
hydrogenation was performed in a flame dried, 100 ml round-bottom three-
neck flask equipped with a reflux condenser, TeflonTM stir bar, and supplied
with a positive Argon pressure. To this flask 500 mg of polymer P5b, 30 ml of
dry reagent grade o-xylene, 1.05g (2.5 eq) of TSH, and 0.80g (2.5 eq) of TPA
were added. The mixing solution was then heated to reflux. All of the TSH








did not dissolve until the solution was close to reflux. Upon heating some

gas evolution could be observed (bubbling of N2) before reflux was achieved

-110 C. The solution was allowed to reflux for 3 hours followed by cooling

to room temperature. Some white precipitation was noted at this time. A
second addition of 1.05g (2.5 eq) of TSH, and 0.8g (2.5 eq) of TPA was

administered, followed by refluxing for 3 more hours. The polymer was then
precipitated twice from cold methanol. The methanol was then decanted off

followed by drying the polymer in a schlenk flask under reduced pressure at

80 "C for 24 hours. Approximately 70% of the polymer was recovered. A

opaque yellow brown hard waxy solid was recovered with the following

spectral analysis: 1H NMR 0.83 ppm (3H, d), 1.09 ppm shoulder, 1.27 ppm

(28H s); 3C NMR 18.8 ppm, 26.1, 28.7, 29.1, 31.9, 36.2. Molecular weight

determination is discussed in Figure 4.15.




Alcohol Monomer Synthesis


6-methyl-l,10-undecadiene-6-ol (A3), and 1,10-undecadiene-6-ol (A2)
were synthesized by the Grignard reaction with the appropriate ester. 5-
Bromo-1-pentene was dried over freshly ground CaH2 for 3 hrs, then vacuum

transferred to a schlenk flask with 4 A molecular sieves. In a flame dried 100

ml three neck round bottom flask 1.63g (67 mmoles) of freshly ground
magnesium turnings were placed followed by the addition of 50 ml of dry

ether. A crystal of iodine, and a small amount of the bromide was added in

order to initiate the reaction. The balance of the bromide was then added at
such a rate as to maintain reflux. After complete addition, the mixture was
stirred and refluxed for 0.5 hours. After cooling to room temperature 33.5








mmoles of either dry ethyl format was slowly added to produce (A2), or

alternatively dry ethyl acetate was added to produce (A3). Refluxing occurred

upon addition of the electrophiles and was maintained by heating for 2 hours.
The reaction was then cooled and quenched with 3 molar HC1. The reaction

mixture was extracted with ether and dried over anhydrous magnesium
sulfate, followed by filtering and evaporation yielding 5.18g (84%) of a clear

liquid A2. Similar yields were obtained for A3. Both monomers were

distilled under reduced pressure (70-80 "C at ImmHg) using a short path

distillation apparatus. The secondary alcohol diene (A2) was further purified
by column chromatography using silica gel with a solvent gradient elution
using 1-5% ethyl acetate and pentane. The column was monitored by TLC
using a 5% ethyl acetate 95% pentane mobile phase on silica plates. The
following spectral properties were observed: 1H NMR (A2) 1.35 ppm (m, 9H),
1.99 ppm (m, 4H), 3.54 ppm (s, 1H), 4.89 ppm (m, 4H), 5.73 ppm (m, 2H); 13C
NMR: 24.79, 33.56, 36.79, 71.53, 114.41, 138.54; Elemental Calcd: C, 78.51; H,
11.98; 0, 9.51. Obsd: C, 77.61; H, 11.80. 1H NMR (A3) 1.14 ppm (s, 3H) 1.43

ppm (m, 9H), 2.05 ppm (m, 4H), 4.98 ppm (m, 4H), 5.80 ppm (m, 2H); 13C
NMR: 23.12 ppm, 26.80, 34.07, 41.21, 72.52, 114.48, 138.61; Elemental calcd: C,

79.06; H, 12.16; 0, 8.78. Obsd: C, 79.16; H, 12.15.


Synthesis of 6-(4-pentene)-l-heptene-7-ol (Al) was synthesized in a
three step reaction sequence as shown below.


Step 1: Synthesis of ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la).
10.9g (84 mmoles) of ethyl acetoacetate (Aldrich) and 200 ml of dry DME were
placed in an argon purged 500 ml three neck flask equipped with a stir bar and
condenser. To the stirring solution 42 ml of a 2 molar solution of potassium








tert-butoxide in DME was added. Upon addition the solution turned a lime

green color and was allowed to stir for 0.5 hours at room temperature.

Approximately 12.5g (84 mmoles) of 5-bromo-l-pentene was then slowly

added and the temperature raised to reflux upon which white salts began to
precipitate. The first addition was complete in 18 h as shown by GC. The

reaction was then cooled to room temperature and the second addition of the
alkenyl bromide was repeated as above and completed in 24 h. The reaction
was then quenched with 3M HCI and extracted with ether. The ether was
dried over MgSO4 and then evaporated yielding -80% of la. The following

spectral properties were observed: 1H NMR 1.11 ppm (m, br, 4H), 1.19 ppm (t,

3H), 1.79 ppm (m, br, 4H), 1.98 ppm (q, 4H), 2.04 ppm (s, 3H), 4.12 ppm (q, 2H),
4.91 ppm (m, 4H), 5.70 ppm (m, 2H); 13C NMR 14.06 ppm, 23.21, 26.62, 30.79,

33.84, 61.09, 63.36, 115.1, 137.94, 172.5, 205.1. The Low Resolution Mass

Spectrum (LRMS) also confirms structure with a parent ion at 266, calcd for

C16H2603 266.


Step 2: Retroclaisen condensation: Synthesis of ethyl-2-(4-pentene)-6-
heptenoate (2a). In an argon purged dry 250 ml 3 neck round bottom flask

equipped with a condenser and a stir bar, 22.34g (84 mmoles) of la and 100 ml
of dry ethanol were added. To this solution 35ml of a 21% solution of sodium
ethoxide in ethanol was added. The solution was allowed to reflux for 3.5

hours and turned a dark yellow in color. After cooling to room temperature
the solution was quenched with water and 3M HCI then extracted with
pentane. The organic layer was then dried over MgSO4 and evaporated

yielding 90% of ester 2a. The following spectral properties were observed: 1H
NMR: 1.25 ppm (t, 3H), 1.40 ppm (m, br, 6H), 1.62 ppm (m, br, 2H), 2.05 ppm

(q, 4H), 2.31 ppm (m, 1H), 4.14 ppm (q, 2H), 4.95 ppm (m, 4H), 5.79 ppm (m,








2H); 13C NMR: 14.5 ppm, 26.9, 32.0, 33.8, 45.5, 60.0, 114.9, 138.6, 176.0. The Low

Resolution Mass Spectrum (LRMS) also confirms structure with a parent ion
at 224: calcd. for C14H2502 = 224.


Step 3: Reduction of ester: Synthesis of 6-(4-pentene)-l-heptene -7-ol

(3a) (Al). Approximately 9.301g of 2a and 125 ml of dry THF were placed in a
flamed dried 250 ml three neck round bottom flask with stir bar and

condenser. This solution was kept under argon and cooled to 0 "C. To this
stirring solution, 25 ml of a 1M solution of LiAlH4 in THF was added over a

period of 5 min. Some bubbling was observed during this addition. The

reaction was allowed to warm to room temperature and stirred for 2 hours.

The reaction was then slowly quenched with water followed by 3M HC1. The
solution was extracted with ether, dried over MgSO4 and evaporated. A clear

oil was recovered and vacuum distilled at 69-72 "C at 1mm Hg. The
following spectral properties were observed: 1H NMR (CDC13): 1.18 ppm (m,

br, 9H), 1.79 ppm (s, 1H), 1.98 ppm (q, 4H), 3.49 ppm (d, 2H), 4.88 ppm (m, 4H)
5.78 ppm (m 2H); 13C NMR: 26.08 ppm, 30.29, 33.98, 34.01, 40.22, 65.36, 114.28,
138.74. IR (CHC13, cm-1): 3383.8 (br), 3078.1, 3013.4, 2931.0, 2860.9, 1640.4,

1460.3, 1217.1, 1030.5, 996.1, 913.1, 759.5. Elemental analysis C12H230

C(calc=79.06, found=78.99), H(calc=12.16, found=12.16).


ADMET Polymerizations of Alcohol Monomers Al, A2, A3


General Metathesis Conditions


All glassware was thoroughly cleaned and flame dried under vacuum
before use. The monomers were vacuum fractionally distilled (from CaH2 if








needed) prior to polymerization. The monomers, if stored, were placed over

4A molecular sieves in order to preserve dryness. The purity of the

monomers was >98% as determined by GC. Monomers were degassed by

subjecting them to several freeze pump thaw cycles under high vacuum (<10-
4 torr). The dry, degassed monomers were then vacuum transferred into a

clean dry reaction flask fitted with a TeflonTM vacuum valve and containing a

magnetic stir bar.

All metathesis reactions were initiated in the bulk, in an argon glove
box conditions. The addition of a few drops of dry CDC13 was occasionally
done in order to help initiate the reaction. Monomers (0.5-1.5 g), while in the

glove box, were introduced into a 25 or 50 ml round-bottomed flask equipped

with a high vacuum TeflonTM valve and magnetic stir bar. After the addition

of catalyst (catalyst to monomer ratios discussed below), the reactions were

first exposed to intermittent vacuum until the viscosity increased, followed

by exposure to high vacuum in order to remove the continuous generation of

ethylene. All the reactions were started at room temperature and maintained

there until the increase in viscosity prevented stirring. At this time the

reaction temperature was slowly ramped to 70 "C over a period of 2 to 3 days.

The solutions were then cooled to room temperature and quenched by the
addition of excess ethyl vinyl ether or by exposure to air. Reactions were run

on a 0.5 1.5g scale, with a monomer to catalyst ratio of 500:1, 300:1 or 200:1, as
noted.


Polymerization of [6-(4-pentene)-l-heptene-7-ol] (PAl). The monomer
Al was synthesized and dried as previously described. For a monomer to

catalyst ratio of 200:1, 23 milligrams of catalyst, RuC12(=CHPh)(PCy3)2 (C3), was

added to 1 g of the monomer. The reaction was performed under typical








metathesis conditions, until the contents could no longer be stirred or

ethylene evolution had stopped. The reaction was quenched by exposure to
air. The polymers were not precipitated before undergoing characterization.

The following spectral properties were observed: 1H NMR (CDC13): 1.09 ppm
(br, 8H); 1.35 (br, 1H); 1.71 (br, 4.4H); 3.19 (br, 1.9H); 4.67 (br, 0.04H end group);

5.15 (br, 2H); 5.60 (br, 0.003H end group). 13C NMR: Anal. Calcd for CloH10s:
C, 77.87; H, 11.76. Found: C, 75.55; H, 11.41. For GPC analysis see table 5.1. IR

(neat, cm-1) 3341, 3005, 2926, 2856, 1726, 1457, 1440, 1036, 967, 511.


Polymerization of [1,10-undecadiene-6-ol] (PA2). The monomer A2
was synthesized and dried as previously described. For a monomer to catalyst

ratio of 200:1, 24 mg of catalyst, RuCl2(=CHPh)(PCy3)2, was added to Ig of the
monomer. The procedure was followed as in (PA1). The following spectral

properties were observed: 1H NMR: 1.45 ppm (br, d, 9H); 2.00 (br, 4H); 3.55
(br, .7H); 4.98 (br, 0.29H end group); 5.40 (br, 2H); 5.80 (br, m, 0.04H end group).
13C NMR: 24.8-33.4 (multiple signals), 36.4, 71.0, 129.5-130.5 (multiple signals).

Anal. Calcd for C9H160: C, 77.09; H, 11.50. Found: C, 76.48; H, 11.33. For GPC
analysis see table 5.1.


Polymerization of [6-methyl-l,10-undecadiene-6-ol] (PA3) The
monomer A3 was synthesized and dried as previously described. For a
monomer to catalyst ratio of 500: 15 mg of catalyst C3 was added to 500 mg of
the monomer. The procedure was followed as in PA1. The following spectral
properties were observed: 1H NMR: 1.12 ppm (s, 3H); 1.40 (br, 8H); 1.99 (br,
4H); 4.94 (br, m, 0.25H end group); 5.39 (br, 2H); 5.79 (br, m, 0.09H end group).
13C NMR: 23.90, 26.94, 27.67, 33.01, 41.37, 72.72, 129.92, 130.42. Anal. Calcd for








C10H180: C, 77.87; H, 11.76. Found: C, 76.61; H, 11.50. For GPC analysis see
table 5.1.


Hydrogenation of Poly [1,10-Undecadiene-6-ol] (HPA2)


The hydrogenation was performed in a flame dried, 50 ml round-

bottom three-neck flask equipped with a reflux condenser, TeflonTM stir bar,

and supplied with a positive Argon pressure. To this flask 266 mg of

polymer PA2, 25 ml of dry reagent grade toluene, 907 mg of TSH, and 0.932 ml
of TPA were added. The mixing solution was then heated to reflux. All of

the TSH did not dissolve until the solution was close to reflux. Upon heating
some gas evolution could be observed (bubbling of N2) before reflux was

achieved. The solution was allowed to reflux for 6 hours followed by cooling

to room temperature. Some white precipitation was noted at this time

(insoluble hydrogenated polymer). A second addition of 907 mg of TSH, and

0.932 ml of TPA was administered, followed be refluxing for 3 hours. Once

the solution cooled to room temperature the product could be observed as a

white precipitant. The polymer was then precipitated in cold methanol.

Approximately 60% of the polymer was recovered. The following spectral

properties were observed: 1H NMR (Toluene d8 at 100 'C): 1.55 ppm and 1.62

ppm (br, s, 18H); 3.72 (s, 1H); 5.18 (br, m, residual vinilic CH2); 5.62 (br, m,
residual internal olefin). 13C NMR: 26.60, 30.50, 30.66, 38.67, 72.41. GPC
analysis was not performed due to the insoluble nature of this polymer.








Preliminary Alcohol Dimerization Experiments


Dimerization of 4-pentene-l-ol (12). Monoene 12 was purchased from

Aldrich in Ig bottles and used on that scale with no further purification. The

monoene was placed in a 50 ml reaction flask followed by degassing under

high vacuum using three freeze pump thaw cycles. The monoene was then

placed into the dry box along with the addition of a magnetic stir bar. To this

stirring liquid, 23 mg of Grubbs' ruthenium catalyst C3 (a monoene to catalyst

ratio of 400 : 1) was added. Upon this addition the reaction turned purple

brown in appearance follow by a small amount of bubbling. The reaction was

then removed from the dry box, placed on the high vacuum line, followed by

intermittent exposure to low vacuum (<10-2 torr) while stirring. At this time

noticeable degassing (bubbling) occurred and continued. After two hours the

solution turned an orange in color followed by a yellow color with a white

precipitate after 18 hours. After 18 hours a sample was taken for a proton

NMR. Approximately 80% conversion was determined by proton NMR.

This was determined by the ratio (proton integration) of internal to external

olefin peaks. 1H NMR peaks of the mixture are as follows: 1.56 ppm (qn), 2.05

ppm (m), 2.17 ppm (qrt.), 2.25 ppm (qrt.), 2.75 ppm (br OH), 3.54 ppm (d), 4.93

ppm (m, =CH2), 5.38 ppm (m, trans internal olefin), 5.47 ppm (m, cis internal
olefin), 5.72 ppm (m, CH= terminal olefin).


Attempted dimerization of 2-propene-l-ol (allylic alcohol) (13). The

attempted dimerization was performed in a similar manner as that described
for monoene 12. The addition of 35 mg of catalyst C3 (M : C = 400 : 1) resulted
in an immediate color change to yellow with what appeared to be a white
precipitate. The evolution of ethylene (bubbling) was not observed upon this








addition. Some bubbling was noticed while exposure to the intermittent
vacuum. The reaction was allowed to stir for 18 hours, followed by NMR
analysis. The 1H NMR represents a mixture of isomers: 0.94 ppm (m), 1.10

ppm (m), 1.61 ppm (br, m), 2.32 ppm (qrt.), 2.47 ppm (qrt.), 3.72 ppm (br, OH),
4.12 ppm (shp, m), 5.20 ppm (m), 5.82 ppm (s), 5.97 ppm (m), 9.78 ppm (shp, s).


Dimerization of 5-hexene-2-ol (14). The monoene 14 was provided by
Shane Wolfe fully characterized and at a purity of >99% by GC. This sample
was degassed and placed in a dry box atmosphere as before. To this stirring
liquid 21 mg of catalyst C3 (M : C = 200 : 1) was added followed by immediate
bubbling and an orange purple color. The reaction was then exposed to
intermittent vacuum as before with a noticeable increase in the evolution of
gas. Samples were taken for NMR after 1 hour and after 24 hours showing a
52 and 97% conversion, respectively. 1H NMR peaks of the mixture after 24
hours are as follows: 1.18 ppm (d), 1.49 ppm (m), 1.61 ppm (d of d), 2.10 ppm
(br, m), 3.78 ppm (q), 4.98 ppm (m) (unreacted =CH2), 5.40 ppm (m) (cis and
trans), 5.8 ppm (m) (terminal =CH).


Attempted dimerization or polymerization of 1, 5-hexadiene-2, 3-diol
(15). Monoene 15 (Aldrich) was purchased in Ig bottles and used on that scale
with no further purification. One gram of 15 was placed into a reaction flask
with a stir bar and degassed as before. Under dry box conditions 42 mg of
catalyst C3 (M : C = 200 : 1) was added to this stirring solution. Upon this
addition the catalyst and solution maintained there original color (purple).
No bubbling was observed upon this addition. The reaction was transferred
to the vacuum line as before, and allowed to stir for 12 hours. No outward
sign of condensation was observed (increase in viscosity and bubbling). After








12 hours an aliquot was removed for NMR analysis. No reaction was

observed by NMR at this time. The reaction was returned to dry box

conditions upon which a few drops (-0.3 ml) of CDC13 was added. This was

done in order to solvate the catalyst. Upon this addition the solution

immediately turned brown in color. This mixture was allowed to stir for an

additional 12 hours. 1H NMR analysis again showed no detectable reaction

(no detection of internal olefin). The NMR was consistent with the unreacted
monoene.


Attempted dimerization of 1, 5-hexadiene-3-ol (16). As stated before Ig

of 16 (Aldrich) was degassed and placed in dry box conditions. To this stirring
solution 42 mg (M: C = 200: 1) of catalyst C3 was added. Upon this addition

the reaction immediately turned brown in color with no detectable evolution

of gas (bubbling of ethylene). Intermittent vacuum was applied with no

increase in viscosity or continued bubbling. After 12 hours an aliquot was

removed for 1H NMR analysis. A small amount of internal olefin was

detected at 5.5 ppm. The remaining proton shifts remained consistent with
the starting material.














CHAPTER 3


DESIGN AND SYNTHESIS OF SYMMETRICAL ALKYL-SUBSTITUTED
TERMINAL DIENES


This chapter is concerned with the design and synthesis of the
symmetrical alkyl substituted monomers that are required for the productive

conversion towards a linear regular branched acyclic diene metathesis

(ADMET) polymer (Figure 3.1).




,,C, Catalyst c ~ + CH-CH2C


Figure 3.1. ADMET polymerization of terminal dienes



The goal of this research has been to model branching in polyethylene
in order to better understand how branching affects the physical properties of

a polymer. The branching of linear polyethylene that is induced by the
random copolymerization of a-olefins or chain transfer events can be

mimicked by the ADMET condensation of alkyl substituted dienes followed
by hydrogenation. The resulting copolymer is the product of a
homopolymerization; therefore, the amount and location of the branch
points is controlled by the monomer used. The synthetic methodology
presented, via the homopolymerization of ADMET monomers, provides
both inter and intramolecular homogeneity with regard to branch








distributions (imperfections) along the polymer backbone. Polymer samples

containing these qualities are excellent models for the study of structure-

property relationships with regard to branching in polyethylene. This type of
system with a perfectly-spaced alkyl branch is illustrated in Figure 3.2.



H





Control over # of methylenes

Control over length and identity of "R"

Figure 3.2. Target polymer for the synthesis of perfectly branched
polyethylene.


Designing the Target Monomers


The clean nature of polymers produced by the homopolymerization of

a, co-dienes, enables the design of the appropriate monomer to be deduced

from the target unsaturated polymer (prepolymer). This is represented by the

retrosynthesis of a target prepolymer in Figure 3.3.



H H



Figure 3.3. The retrosynthesis of the target prepolymer.



To achieve control over the distribution of branch points in the

resulting polymer, it is necessary to design a synthetic methodology that will









produce symmetrical hydrocarbon a, co-dienes. This type of synthesis is

challenging due to the lack of functional groups present and the symmetrical

nature of the final product (Figure 3.4). Further, it is necessary to be able to

synthesize a homologous series of the models in which both the frequency

and identity of this branch point can be varied (Figure 3.4).

A symmetrically substituted diene is necessary to prevent the
scrambling of branches (imperfections) in the polymer chain. The metathetic



H


R
Control over the frequency R= Variable length alkyl groups;
of the branch -\Control over the branch identity


Figure 3.4. Target monomer; symmetrically substituted diene


reactivity of the terminal olefins in alkyl substituted dienes is essentially

equivalent. Therefore, the polymerization of an unsymmetrical diene lacks

the necessary head to tail preference required for ordered assembly. The

resulting placement isomerism can be demonstrated by the

homopolymerization of 4-alkyl-1,8-nonadiene (Figure 3.5). This isomerism is

illustrated by the 1,6 and 1,7 placement of the alkyl branch in a trimer that

contains one head to head (HH) and one tail to head (TH) connection (Figure

3.5). Due to functional group equivalence, these types of connections will

occur in a random fashion providing only intermolecular homogeneity. This

type of distribution would be considerably more ordered than that obtained in

a-olefin copolymerizations, but precise control cannot be achieved.









H
I
R

Mo

H H



HH THR

1. 6 placement
1,7 placement

Figure 3.5. Trimer of unsymmetrical 4-alkyl-1,8-nonadiene; 1,6 and 1,7
placement of branch due to head to head (HH) and tail to head (TH) type
placements.


Purification is also an important consideration during the design of the

monomers due to the need for 99%+ purity in order for polycondensation to

proceed to high molecular weights. This is a result of the statistics involved

in step polymerizations as defined by the Carothers equation (equation 4

chapter 1). Further, the catalyst used for ADMET polymerizations (Figure 3.5.)

is vulnerable to attack by Lewis bases.







CF3 N
CCF ,CH3
CHO 0 NH3

CH3 F3 Ph
CF3
C1
Figure 3.5. ADMET catalyst; Schrock's Molybdenum Alkylidene.








Synthesizing the Target Monomer


Various synthetic methods were investigated toward the production of

a homologous series of target monomers. The priorities in this consideration

were to obtain a relatively high yielding procedure with the variability to

synthesize a reasonable number of the desired monomer derivatives (Figure

3.4). The synthetic methods that were attempted (Figure 3.6) have displayed

interesting mechanistic implications that were considered in the

development and optimization of monomer construction.


P-Keto Ester Substitution Method


The carbon-carbon bond-forming reaction of enolate substitutions has
been shown to be a synthetically useful method toward the production of the

target monomer. Carbanions or enolates can be formed by the deprotonation

of an a-carbon by a strong Br6nsted base. These protons exhibit significantly

higher acidities compared to hydrocarbons, where their increased acidity

results from a combination of the inductive effect from the alpha carbonyl

and the resonance stabilization of the anion formed by removal of a proton
(Figure 3.7). The formation of a significant concentration of the enolate

requires that the base used has a weaker conjugate acid than the active

methylene compound.103 The solvent must also be a weaker acid than the
conjugate acid of the base to avoid solvent deprotonation.'03

Enolate anions are useful in a variety of base catalyzed condensation
reactions of carbonyl compounds, specifically aldol/Claisen condensations

and alkylations. This type of enolate substitution is common for the anions









O2Et

H + =CH, Br
c= n


CH


3



Dicarbonyl substitution t



/ Wittig chemistry


R

H2 H2
H


Malonic ster


0 R 0

RO- k(CHY -CH2 "OR
1 n v n


DiGrignard coupling


Lactone ubstitution





R


R
BrMg V L MgBr + BCH2')B


Figure 3.6. Pursued routes of monomer synthesis.




of P-ketoesters with the majority of the examples involving the

monosubstitution of stabilized enolates with alkyl halides.103 The di-


0

LCHig + RC-OEt

M=Li.MgX


Organometallic/ter


t-alcohol


/


-^,








alkylation of ethyl acetoacetate, via enolate anions, is the basis for a multistep

synthetic procedure for the generation of the target monomer.






/C-OC2H /C-OC2HS C-OCH5
H2C\ CH -CH3
CCH3 CCH3 \ CH3



Figure 3.7. The equilibrium and resonance structures of enolate formation.



A general procedure for the synthesis of the target monomer was
devised which involved the enolate displacement of terminal alkene

bromides. This method of synthesizing the target monomer is illustrated in

Figure 3.8 and is termed the p-keto-ester method. The first reaction, (A)

Figure 3.8, involves the generation of the anion from a diprotic active

methylene group followed by subsequent alkylation of the 0-keto-ester with

an alkene bromide forming the mono-substituted p-keto-ester. This can then

be repeated for the other proton forming the dialkylated product (1). This

method provides a single center were both alkylations can occur creating a

symmetrical intermediate. The frequency of the repeat unit in the target

polymer can be controlled by the alkylation of various length alkenyl

bromides. The intermediate (1) could then follow two routes (B or C) to

generate the alkyl group. Route B is the process of eliminating the acetyl

group by the use of a retro-Claisen condensation reaction. The ester

produced, (2), can then be reduced to form the primary alcohol (3), which can

be removed by conversion to a leaving group (tosylate) followed by hydride














'O2Et

H2 + ^;Hk Br
C= O
CH3


COEt


H 2
Reduction D
OH
CH,

H
3


R

H
target monomer
0
COEt
Base H H
A O Dealkoxycarbonylation
D CH3 c
Deacylation CHC 1
B H


c=o 6
CH3
E Reduction
Sto Methylene

1. Tosylation- H J.z -~jH-
2. Reduction 2yn 11

5 CH
7


Figure 3.8. P-Keto-ester method: synthesis of target monomer.



displacement to form the methyl substituted target monomer (5). The

alternative route, C, removes the ester group via dealkoxycarbonylation. This

can be done directly or by decarboxylation of the saponified ester. The

resulting ketone (6) can then be reduced to the methylene by a Wolf-Kishner

reaction to form the ethyl substituted target monomer (7). Both

intermediates (3) and (6) can be used to extend the symmetrical alkyl

substituent via further alkylations.

Route B (Figure 3.8) provided the means to construct a series of the

target monomers and was used to synthesize monomers 5a, 5b, and 5c (where









C(O)OCHCH
A CH3CHOO)C-CH,-C(O)CH, 2 eq (CH)CO-K/DME
2 eq CH--CH(CHI,,Br +(CH,)n (CH2)n''
C(O)CH,
1





C(O)OCH2CH3 C(O)OCHCH3
B (CH)n (CH2n E Na / OH (CH)n C

C(O)CH3
1 2





OH
C(O)OCH,CH3 IH
S (CH) (CH) LiAIH4/THF 2
(CH). ( (CH2) (CH,)./,-


2 3





OH OTs
CH2 CH,
D ..(CH) 2(CH2) TosC I/Et3N (CH 2)n
D --(CH2) l^(CH,),-`- --- .)---- t(CH2) j,) (CH,).----


OTs

E '(CH. CH,


CH3
LiAIH4/THF .
---------- '^CH-^tC^-


Figure 3.9. The synthesis of the methyl substituted target monomer.








n=3, 6, and 8 respectively). The reactions used to accomplish this 6 step

conversion are shown in Figure 3.9.


Enolate Alkylation of Ethyl Acetoacetate and the Retro-Claisen Condensation


The enolate anion of ethyl acetoacetate was investigated as a means to
add alkene halides via subsequent nucleophilic displacements. The anion

generated by deprotenation of a p-keto-ester can exist in two resonance

structures (Figure 3.7); the anion on the oxygen, or the anion on the carbon.

The type of base and solvent used in these reactions can favor either O-

alkylation or C-alkylation, and has a direct effect on competing side reactions.

If the base is too nucleophilic (Lewis basic), the direct displacement of the

alkene bromide or the deacylation of the P-keto-ester by a reverse Claisen

condensation can occur. If the base exhibits a strong Bronsted basicity,

elimination of the alkene bromide, alkylation of the ketone, or dianion

formation of the P-keto ester can occur.

In order to optimize the conditions for this substitution reaction, both
the type of base and solvent were investigated using 5-bromo-l-pentene as the
halo-alkene (Figure 3.10). The major products in these reactions were the di-

alkylated P-keto ester (A Figure 3.10) and the di-alkylated ester (B Figure 3.10).

These products correspond to compounds la, b, c and 2a, b, c in the

experimental section.

The syntheses of la, Ib, and Ic were designed to be a one pot, two step
procedure (A Figure 3.9), which involved the deprotonation of ethyl
acetoacetate by the bulky Br6nsted base potassium tert-butoxide in dimethoxy

ethane (DME) (entry 12 Figure 3.10). This was accomplished by the room









O 0
H2Et COEt COEt
2 H Br5--BS^ ICH.-(CH2 3.CH,)--(CH
0 +2 CHBr +Bae H 1 +iH
c=o 3 C-=0 H
CH, CH3

A B

Entry Base Solvent A% B%
1 2 eq EtO-Na+ EtOH 17 22
2 4 eq EtO-Na+ EtOH 0 48
3 2 eq EtO-Na+ EtO(CO)OEt 13 4
4 2eqKH THF 69 4
5 4 eq NaH DMF 16 51
6 3 eq NaH DMF 60 0
7 2 eq NaH EtO(CO)OEt 42 0
8 4 eq NaOH DMF 0 52
9 2 eq KOH DMF 39 21
10 2 eq DBU DMF 42 0
11 2 eq NaNH2 THF 57 7
12 1.9 eq (CH3)3CO- K+ DME 86 0

Figure 3.10. The enolate displacement reaction of 5-bromo-l-pentene.
Displayed percentages were attained by GC and show % of A or B within each
sample.



temperature addition of one equivalent of base to a stirring solution of the P-

keto-ester. Formation of the enolate could be detected by the lime green color

of the solution, which is the apparent result of the highly delocalized anion of

a p-keto ester. After acid-base equilibrium was obtained, 5-bromo-l-pentene

was added (or higher homologue) to the basic solution. The reaction was

monitored by GC using a micro-extraction method and was allowed to

continue until a high conversion was achieved. Mono-alkylation of the keto-

ester was a relatively fast reaction as shown by detection by GC within the first

10 minutes of refluxing. A small amount of the di-alkylated product was

detected before the addition of a second equivalent of base or alkene-bromide.








The second alkylation was facilitated by the addition of another

equivalent of base at room temperature, followed by a second equivalent of

the alkene-bromide. Again, the enolate was detected by a dark green color but

was usually only evident at increased temperatures, due to the decreased

acidity of the second proton. The second alkylation was also monitored by GC

and required much longer reaction times to achieve good conversions. The

retardation of the second alkylation probably is due to a more sterically

hindered enolate anion and a shifted acid-base equilibrium. A large amount

of potassium bromide salt was apparent by the end of the second reaction.
Product, la, was isolated using high vacuum distillation via a short path

distillation apparatus and resulted in a 65% yield, while products lb and Ic

were not isolated due to their low vapor pressures. GC retention times of the

crude products were consistent with what was expected, and mass spectra

were consistent with their formula weight. The crude mixtures were used in

subsequent reactions without further purification.

The addition of excess base (potassium tert-butoxide), or increased
reaction times also resulted in the conversion of la to the deacylated product

2a (Figure 3.10). Similar conversions occurred for lb and Ic consequently

producing compounds 2a, b, c, as the major products. Cooler temperatures
and longer reaction times may facilitate the production of the keto-ester only.

This deacylation of a P-keto esters was identified as a retro-Claisen

condensation.104,105,106 The cleavage of P-keto esters at the ketone has been
observed during alkylation reactions and was most pronounced when the
alkylated product had two alpha substituents,107 as do products la, b, and c.
Literature reports that this cleavage can be minimized by three methods; low

temperature reactions using sodium ethoxide, the use of sterically hindered

bases such as potassium tert-butoxide,10s or the use of sodium hydride109 in








polar aprotic solvents. The extended exposure of nonenolizable ketones to

tert-butoxide in DME has also shown a similar ketone cleavage.110111 The

proliferation of this cleavage has become the basis for the second step in the

synthetic sequence (Figure 3.11).



0 o
OEt Et o
SH BOCC

EtO CH3 H'
la 2a
Figure 3.11. The conversion of la to 2a by the reverse Claisen condensation.



The retro-Claisen condensation of the products la, Ib, and Ic, was

induced by the addition of an excess of sodium ethoxide (Figure 3.11 and

Figure 3.9 B). This deacylation occurred with near quantitative conversion,

i.e. > 95% and was monitored by GC (Figure 3.12). Chromatogram A is the

product mixture after the enolate alkylation reaction and chromatograms B-D

are the product mixtures after 1, 2, and 3 hours of exposure to sodium

ethoxide, respectively. Compound percentages were determined by

subtracting out the known solvent peaks. The GC peaks were assigned by the

isolation and characterization of the compounds by NMR and GC/MS. It was

also demonstrated the conversion could be obtained by direct addition of

sodium ethoxide to the reaction pot after the completion of the di-alkylation.

Higher yields were achieved, though, by first isolating the crude products. It

was also observed that the use of more nucleophilic bases for the enolate

alkylation resulted in higher percentages of the deacylated product before the

addition of sodium ethoxide.









2a la 2a la 2a la 2a
78% 16.8% 82% 13% 94% 5.3% 99%







i i I I la
1%





A B C D

Figure 3.12. Gas chromatograms demonstrating the conversion of la to 2a by
a reverse Claisen condensation using sodium ethoxide.


Other less successful methods of di-alkylating P-keto esters were

examined with their respective results listed in the table of Figure 3.10.

Entries 1 and 2 show the results when sodium ethoxide was used as the base

in the two step reaction where both the mono and di-alkylated products were

isolated between steps. The monoalkylated product was isolated with

relatively high yields (-80%) while the dialkylated product was isolated with a

yields of 30% or less. Previous investigators have shown that the failure to

efficiently di-alkylate was due to the incomplete formation of the sodio

derivative of the mono-alkylated product,112 which leaves large amounts of

sodium ethoxide in solution which could then directly displace the alkyl

bromide. A modification of this reaction presented by Wallingford et a112 is

shown in entry 3. In this case the reaction is performed in a high boiling

aprotic solvent where ethanol is distilled off as it is formed therefore driving

the reaction toward the production of the sodio derivative. However, poor








yields were also obtained using this technique. Both of these methods using

sodium ethoxide, resulted in deacylation of the product in a uncontrollable

fashion, which is consistent with the use of a nucleophilic base.

The use of dimethylformamide (DMF) and other aprotic solvents
results in the acceleration of enolate alkylations.113 Aprotic solvents show a

clear advantage over the use of protic solvents for the alkylation of enolates

of monosubstituted acetoacetic esters,114 but occasionally have the

detrimental effect of favoring alkylation at the oxygen rather than at the

carbon. Zaugg et al offers an excellent review of specific solvent effects in the

alkylation of enolate anions.115

Zaugg115 also demonstrates the use of hydride as the base in DMF.
Hydride has the advantage of shifting the acid base equilibrium in favor of

the enolate anion due to the release of hydrogen upon deprotenation. Entries

4-7 (Figure 3.10) are the results of using a hydride base with ethyl acetoacetate

in an aprotic solvent. The yields were moderate to low and varied with

reaction times. These reactions were repeated multiple times and were

performed as either a two step process with isolation of each step or as a one

pot synthesis. If an excess of hydride was used and or reaction times were

dramatically increased, the deacylated product 2a was the major product.

Further, the production of 2a was improved by the addition of ethanol after

alkylation was complete. Higher boiling residues were detected for each of
these runs.

The di-alkylation of activated methylenes is reported to react in the
presence of 1,8-diazabicyclo[5.4.0]undec-7-en (DBU) and DMF.116 Using this
method (entry 10, Figure 10), longer reaction times were required for the di-
alkylation, while producing only monomer la. This retention of the P-keto








ester functionality is consistent with the use of the non-nucleophilic base

DBU.

The use of potassium and sodium hydroxide in DMF was investigated

as an alkylating medium (entries 8 and 9, Figure 10). Usually, these types of

bases are avoided due to their tendency to cause saponification and

decarboxylation of the ester group.103 The nucleophilic nature of the

hydroxide can also induce cleavage of the ketone functionality.103 The di-

alkylated product la was obtained using this method but resulted in low

yields. Quantitative conversion, as monitored by GC, of la to 2a was obtained

and was promoted by increasing the pH of the solution. The production of

the keto acid and the decarboxylated product was not investigated.

Sodium amide, due to its high Brbnsted basicity, was investigated as a

means to increase the rate of the second alkylation. Its basicity also creates the

possibility of forming the dianion of the 1, 3,-keto ester. It has been shown

that the addition of two equivalents of sodium amide to ethyl acetoacetate can

be used to generate the dianion which results in the activation of the acetyl

methylene for displacement reactions (Figure 3.13).117 The acetyl anion will

be the "hotter" anion and will usually undergo a high yielding alkylation.



0 O0 O 0 0
0 NaNH,
COACH5 NH OCH5 ----- OC2H

Figure 3.13. The generation of the enolate dianion by sodium hydride.


Results obtained from using sodium amide as the base in a two step
method (entry 11, Figure 10), resulted in acetyl substitution. The products

masses (GC/MS) were consistent with a mixture of the mono, di, di-
deacylated, and tri-alkylated products (Figure 3.14 A, B, C, D respectively). The









percentages of products were determined by GC with subtraction of the

known solvent peaks.




coEt COEt COEt COEt

C--o C- 3 H 3 C-O
CH3 CH3 CH

A B C
20% 50% 6.6% II


D
16%
Figure 3.14. Product percentages of the enolate displacement using NaNH2.


Reduction, Tosylation, and Hydride Displacement



The reduction of 2a, b, and c was accomplished by the addition of excess

lithium aluminum hydride (LAH) (> 2 eq) to a solution of the ester in THF118

(Figure 3.9 C). This reaction proceeded as expected with near quantitative

conversion at room temperature producing the alcohols 3a, b, and c. The

reduction did not require the extensive purification of the ester starting

material before reacting. Due to the high boiling points of the other alcohols,

only the intermediate 3a was purified by vacuum distillation. Alcohols 3b

and 3c were characterized and used in their crude form in the subsequent

reaction.

Now that the primary alcohol could be obtained, it was necessary to

remove the alcohol functionality so that the methyl substituted target

monomer could be synthesized. This was accomplished by first converting

the alcohol into a tosylate and then displacing it by a hydride.








Tosylation of the alcohol (Figure 3.9 D) was achieved by using a
technique similar to that developed by Kabalka et al,119 in which the tosylates

of long chain aliphatic alcohols were obtained by reacting the alcohols with a

1 : 1.5 : 2 ratio of alcohol/tosyl chloride/pyridine in chloroform. These

reactions were cooled to 0 "C for the addition of reagents followed by

warming to room temperature. This was done to avoid the undesirable

elimination product. The reaction was followed by GC until no further

conversion was observed (approx. 12 hrs.). The products recovered (4a, b, c)

were thick oils and were used without further purification. The tosylate (4a)

was purified by column chromatography for complete characterization.

Tosylation methods using different equivalent amounts of reactant
and reagents were also attempted. Using THF as the solvent and NaH or

triethyl amine as bases were somewhat successful but resulted in relatively

low yields.

The final reaction was the nucleophilic displacement of the tosyl group
by a hydride (Figure 3.9 E). Two equivalents of lithium aluminum hydride in

THF were added to a mixing solution of compound 4 followed by refluxing.

The conversion was complete in less than 5 hours and gave moderate yields

of the target monomers 5a, b, and c. The monomers 5a and 5b were purified

by short path vacuum distillation to a purity of 99% by GC, followed by full

characterization. The 1H NMR spectra of the target methyl substituted

monomers 5a, 5b and 5c are shown in Figure 3.15.


Dealkoxycarbonylation of the Keto Ester

Route C (Figure 3.8) was investigated as a means to produce the ethyl
substituted monomer 7. Alkylated products of P-keto esters may be
hydrolyzed and decarboxylated to form the corresponding acids and ketones,















CH,

-(CH2,)3 ,-(CH2)f-

5a









ppm




CH3

"'-(CH2) (CH2)6 I

5b











CH3

',-(CH2) 3(CH2)g

5c





ppm
Figure 3.15. 1H NMR of the methyl substituted target monomers 5a and 5b.








respectively.103 The decarboxylation of saponified esters (acids) is known to

proceed through a six-center transition state which initially forms the enol

(Figure 3.16).107


R- C-- CHC0,H A M H2
SHH
R--C-CcoH RC O 2C- R--=CH, R-C--CH,


Figure 3.16. Decarboxylation of 1, 3-keto acid.


Saponification of di-substituted p-keto esters is often complicated by the
competing attack of the hydroxide anion at the ketone functionality, which
leads to the reverse-Claisen cleavage instead of the saponification. This is due

to the presence of a nucleophilic group with a dialkylated 1, 3-keto ester.
Hydrolysis and decarboxylation are typically accomplished with aqueous acid
and heat to avoid this side reaction.107

Decarbonylation of la to 6 (C Figure 3.8) was attempted by heating la in
the presence of either an acid, base, or salt (NaC1) in DMF, DMSO, or H20. No
detectable reaction or significant decomposition was observed using these

methods. The expected conversion of la to 2a (deacylation) was observed
with the addition of KOH and heating.

Krapcho reports a method to affect dealkoxycarbonylation of
disubstituted p-keto esters by the use of a salt such as lithium chloride or
sodium cyanide in a dipolar aprotic media.120 This method has the
advantage of proceeding with neutral conditions, therefore preventing side
reactions such as acid addition to olefins and ketone cleavage. Using this
method, la was reacted with LiC1 salts in NMP and water under reflux
conditions. The GC results showed the appearance of a new major peak.
Further, low resolution GC/MS resulted in a M+1 ion that was consistent
with the production of 7. The 1H NMR showed the retention of a 3 proton








singlet at 2.04 ppm from the retention of the methyl ketone as well as the

disappearance of the ethoxy protons. Work on the dealkoxycarbonylation was

stopped due to the difficulty in isolating large amounts of la.

The P-keto ester substitution method has provided the most success in

the generation of the target monomers. Target symmetrical methyl

substituted monomers 5a, 5b, and 5c have been generated using the general

methodology outlined in Figure 3.9. The following discussions are based on

other synthetic methods that have been investigated but are currently not

being pursued.


Wittig Method


The Wittig reaction between a phosphorus ylid and an aldehyde or
ketone results in carbon-carbon bond-formation and is extensively employed

in synthesis. The use of this reaction was investigated for the synthesis of the

target monomers.

The synthetic design was to first make a phosphonium salt from an
alkene bromide followed by the formation of the ylid and the coupling with

an alkene-one (Figure 3.17). The carbon-carbon bond-formation results in the
formation of an olefin at the joining site. The product, therefore, contains a

tri-substituted olefin in place of the carbonyl (Figure 3.17 C).

This method provides the ability to make monomers of multiple sizes
and variable length branches. The size (length) and symmetry of the diene

product can be modified by choosing longer chain functionalized alkenes.
The branch point identity can be varied by modification of the alkane side of
the ketone. Figure 3.17 demonstrates this utility with the synthesis of a
methyl branched monomer by reacting the ylid with a methyl ketone.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E1CWPW1F2_R4NHYT INGEST_TIME 2013-01-23T15:45:10Z PACKAGE AA00012941_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

ACYCLIC DIENE METATHESIS ( ADMET ) POLYMERIZATION : THE SYNTHESIS OF WELL-DEFINED MODEL POLYMERS FOR POLYOLEFIN MATERIALS By DOMINICK J. VALENTI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNNERSITY OF FLORIDA 1997

PAGE 2

This dissertation is dedicated to my family: Anthony Patricia and Julia Valenti all of whom have provided me i vith unlimited s upport c onfidence love and strength to achieve any goal o r challenge in the known universe.

PAGE 3

ACKNOWLEDGMENTS I would like to first acknowledge Dr. Guy Mattson for introducing me to the world of polymer chemistry and for his faith in my abilities from the beginning. Special thanks are extended to Dr. Chris Marmo and Dr. Jason Portmess who instigated my interest in ADMET chemistry. Sincere thanks are extended to all the members of the Wagener group, past and present: Dr. K. Brzezsinka, Dr. J. Anderson, Dr. J O'Gara, Dr. D. Tao, Dr. H. Tamura, Dr. J. Konzelman, J. Reichwein, F. Gomez, S. Cummings, T. Davidson, D. Tindall, M. Watson, S. Wolfe and L. Williams. I would also like to recognize the cooperation of Drs. R. Duran, J. Reynolds, K. Wagener and G. Butler that has created an institute of vast resources, knowledge and funding for this and other scientific pursuits. Sincere thanks are also extended to the past and present members of the Duran and Reynolds groups for their contributions and support of my research goals. Specific recognition is extended to J. Batten, Dr A. Kumar, D. Cameron, Dr. P. Balancia, and Dr. M. Diverti for significant contributions of scientific knowledge and philosophical ideas. Grateful acknowledgment for funding is given to the National Science Foundation and The Dow Chemical Company. Special recognition is extended to Steve Hahn from Dow for his contributions, ideas, and faith in this project from start to finish. 111

PAGE 4

Respects go to D. Panosian, T. Nguy en J. Miller, L. Wolert, and N. Morales for boundless friendship, support and thoughtfuli1ess. Finally I would like to thank Dr. Wagener for believing in my abilities and providing nearly unlimited resources, freedom and knowledge that have made this research and my graduate education possible IV

PAGE 5

TABLE OF CONTENTS page A CKN O WLEDG MENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii ABSTRACT ................................................. .............................................................. viii CHAPTERS 1 WTRODUCTION ............................................................. .............. ...................... 1 History of Polyethylene and Related Polymers ................................................ 3 Discovery and Synthesis of "Polymethylene" ............................................ 3 Synthesis of Polyethylene via Radical Chemistry .......... .......................... 6 Synthesis of "Polyethylene" (Heterogeneous Ziegler Catalysts) ............ 9 Synthesis of "Polyethylene" (Homogeneous Ziegler Catalysts) ........... 12 Copolymerization of cx-olefins ......... ........................................................... 15 Structure Property Relationships of Polyolefins .......................................... 18 Crystallinity, Morphology, and Melting of Flexible Chain Macromolecules ........................................................................................ 20 Thermal Behavior and Chain Structure ................................................. .. 23 Metathesis Polymerization ............................................................................. ... 27 Ring Opening Metathesis Polymerization (ROMP) and Well Defined Alkylidenes .............................................................. 27 Acyclic Diene Metathesis Polymerization (ADMET) ................. ............ 30 ADMET: Step Condensation Method to Producing a-Olefin Precursor Po 1 ymers ........................ .......................................................... 34 2 EXPERIMENT AL ...... ... ............................... ....................................................... 36 Instrumentation and Analysis ........................................................................... 36 Materials and Techniques ......................................... ..................................... ... 38 Synthesis and Characterization ............ .......... .......................... ...................... 39 Synthesis of Symmetrical Alkyl-Substituted Terminal Dienes ............ 39 V

PAGE 6

Alternative Enolate Displacement Reactions for the Synthesis of C o mpounds la and 1 b ... ..... ... .... ... .. ....................... ........ 45 Alternative Method s for the Synthesis of Compounds (Sa) (Sc) .. .. ..... 52 Wittig Method . ... ...... .... . .... ......... ......... .... .. . ... ..... .. .. ................................... 53 Tertiary a lcohol synthesis and Reduction ....... ..... ... .. ... .......................... 54 ADMET Polymerizations of Monomers (Sa) and (Sb) ........ .... ..... .. .... ..... ..... 57 General Metathesis Conditions ............................ .... ..... ....... ..... ..... ..... .. ... 57 Hydrogenation of Polymers (PSa) and (PSb ) ..... ............. .. . .... .......................... 60 Alcohol Monomer Synthesis .. ...... .... ........... . .... ........... ... ..... ....... ....... ... . ......... 61 ADMET Polymerizations of Alcohol Monomers (Al), ( A2), (A3) ............. 64 General Metathesis Conditions .............................................................. .. ... 64 Hydrogenation of Poly [1 10-Undecadiene -6 -ol] (HP A2 ) ......................... 66 Preliminary Alcohol Dimerization Experiments ............................... .... .. .. ... 68 3 DESIGN AND SYNTHESIS OF SYMMETRICAL ALKYLSUBS'I'I I 'U TED TERMINAL DIENES ........................................................ 71 Designing the Target Monomers ................................. ........ .............................. 72 Synthesizing the Target Monomer ... .... ............................ .... ......... .. ................ 75 ~-Keto Ester Substitution Method ........................................ .. .................... 75 Enolate Alkylation of Ethyl Acetoacetate and the Retro-Claisen Condensation .................. . ..... ............. ............ ..... ....... .. 80 Reduction, Tosylation and Hydride Displacement ..... ..... ......... .. . ....... . 87 Dealko xyca rbonylation of the Keto Ester ..... ............................................. 88 Wittig Method .......... ........ ..... ........ ............................................................... 91 Malonic Ester, Di-Grignard and Lactone Methods ........ .. ............ ... ...... . 93 Organometallic / Tert-Alcohol Method ....... ... . . .. ............... .. ......... .. .. ..... ... 95 Con cl us ions .... ...................................... ................................. ............................ 100 4 ADMET MODELING OF BRANCIDNG IN POLYETHYLENE: THE SYNTHESIS OF MACROMOLECULES WITH PERFECTLY SP ACED METHYL BRANCHES .......... ....... .......... .. .......... 101 Modeling Polyethylene ............................................ ........ .................................. 101 ADMET Modeling o f Branching in Polyethylene ... .. ... .................... .... 104 Polymer synthesis via ADMET ........ ...... ........ .......... .. ............... .... .. .. 106 Hydrogenation of the unsaturated polymer . ... .. ...... ................... .. ... 113 Thermal Analysis of Methyl Substituted Polyethylene ....................... 120 V I

PAGE 7

Thermal Analysis of ADMET Methyl Substituted Polyethylene ....... 122 Con cl us ions ... ........ . .. ............... ... .. ........... ..... ................................ ....... ............ 128 5 THE DIRECT SYNTHESIS OF WELL-DEFINED ALCOHOL FUNCTIONALIZED POLYMERS VIA ACYCLIC DIENE MET A THESIS (AD MET) POLYMERIZATION ...................................... 130 Alcohol Functionalized Polymers via Metathesis ...................................... 132 The Direct Synthesis of Alcohol Functionalized Polymers via ADMET .............................................................................................. 135 Monomer Design and Synthesis ..... ............. ................................ .. ........... 136 ADMET Polymerization of Hydroxy Functionalized Dienes .................... 140 General ADMET Polymerizations ............................................................ 140 Polymerization of 6-(4-Pentene)-1-Heptene-7-ol (Al) ........................... 141 Polymerization of 6-Methyl-1,10-Undecadiene-6-ol (A3) ...... .... ........... 145 Polymerization of 1,10-Undecadiene-6-ol (A2) ....................................... 146 Hydrogenation of Poly(6-Hydroxynonenylene) P A2 ............................. 150 Thennal Analysis of the Alcohol Containing Polymers ..................... 152 Preliminary Dimerization Experiments .................................................. 156 Conclusions ............................................... ...... ......... .. .......... ... ......... ........ ............ 160 REFERENCES .................................... .... .................................... ............ ........ .... ........ 161 BIOGRAPHICAL SKETCH .... ........ .. ........................ .... .. ....... ......... ............. .... ..... .. 173 Vll

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ACYCLIC DIENE METATHESIS (ADMET) POLYMERIZATION: THE SYNTHESIS OF WELL-DEFINED MODEL POLYMERS FOR POL YOLEFIN MATERIALS By Dominick J Valenti August 1997 Chairman: Professor Kenneth B. Wagener Major Department: Chemistry A synthetic method has been designed to facilitate the synthesis of target monomers required for the preparation of perfectly branched polyethylene model polymers. These polymers can be used to better understand the relationships between branching in polyethylene and its ultimate behavior. Symmetrical alkyl-substituted alpha omega dienes were synthesized using a six-step procedure involving two enolate displacement reactions, a retro-Claisen condensation, an ester reduction followed by a tosylation, and a hydride displacement reaction. Purity of the resulting monomers was greater than 99/o. The alkyl substituted monomers were condensed using Schrock's molybdenum alkylidene (CF3)2CH3C0]2(N-2,6-C6H3-i-Pr2)Mo=CHC(CH3)2Ph forming well-defined alkyl-substituted polyethylene prepolymers. Quantitative hydrogenation (by NMR analysis) of the resulting unsaturated polymers was achieved using a modification of typical diimide reduction Vlll

PAGE 9

procedures. The resulting polyethylene polymers co ntain branch poi nts occurring at every 9 or 15 ca rbons, respectively. These samples are ideal with regard to well-defined chain branching and precise branch distribution. Presently this high order of regularity in polyethylene copolymers cannot be obtained by any other method Thermal analysis of these polymers demonstrates the unique behavior of these materials The presence of perfectly spaced methyl branches results in a significant drop in the polymers melting point within a very narrow range. The synthesis of well-defined primary, secondary, and tertiary alcohol functionalized ADMET polymers was also accomplished using a similar methodology. Unsaturated polymers of this type can be hydrogenated in a similar manner producing a new series of ethylene-vinyl alcohol copolymers exhibiting precise chemical regularity lX

PAGE 10

CHAPTER 1 INTRODUCTION Research in polymer chemistry during the past five decades has resulted in the signifigant advancement of macromolecules as materials in technical and nontechnical applications. We all have become everyday consumers of the product of this research, from the clothes we wear to the materials used to print this document. Advancements such as these do not occur rapidly, but rather are derived from patient work leading to the fundamental understanding needed to create polymers and tailor their physical properties. The dissertation represents a contribution to this fundamental research. The work presented herein describes the creation of new techniques that offer unique control over a polymer's microstructure, which in turn can have an important effect on ensuing physical properties. This control of microstructure is demonstrated by the synthesis of model polymers with well defined structures that can be used to elucidate the effect of molecular changes on the behavior of the polymer. Also demonstrated herein is the ability to use Lewis basic/protic groups toward the synthesis of unsaturated, functionalized polymers via acyclic diene metathesis (ADMET) polymerization. These polymers then can be further hydrogenated, resulting in the direct synthesis of saturated polyalcohols. 1

PAGE 11

2 The model polymers are st ructurally similar to the industrially significant a-olefin polymers; polyethylene and polyethylene-polyvinyl alcohol copolymers. This work models the chemical/ structural imperfections or branch points that result from uncontrollable events which are present in typical polymerization reactions. Branch points often are purposely incorporated into polymer chains so that desired physical properties are achieved and are typically incorporated by random copolymerization with higher order a-olefins. a-Olefin copolymerization (as well as uncontrollable chain transfer events) causes random branch placement, thus producing materials with variable physical properties. The resulting randomness makes it difficult to study structure-property relationships of industrially important materials This dissertation presents a methodology that controls polymer structure therefore producing analogous materials that can be used to investigate the effect of structure on the physical properties of polyethylene and related materials. It is appropriate to describe polyethylene (PE) and related a-olefin polymers as the main focus of this introduction. This will be accomplished with a brief historical overview and a discussion of the important polymerization mechanisms of PE. The concept of branching, related PE studies and the crystallinity of flexible chain polymers are also presented. Since ADMET is the chosen polymerization route, a discussion of metathesis chemistry is also warranted for a complete understanding of the chemistry within.

PAGE 12

3 History of Polyethylene and Related Polymers Discovery and Synthesis of ''Polymethylene'' The repetitive joining of molecules of ethylene forming linear macromolecules with the general formula -(CH2-CH2)nis known today as the polymerization of polyethylene. The first samples of polyethylene were made in 1898 by von Pechmann. 1 He observed that small amounts of a white, flocculent substance separated from an etheral solution of diazomethane on standing, and while he was able to crystallize the substance from chloroform there was not enough product for further study. Two years later, Bamberger and Tschirner 2 were able to precipitate a large quantity of this solid, which was described as a white powder with a melting point of 128 C and had a corresponding chemical analysis consistent with the structure (CH2). This compound was termed polymethylene and was the product of the reaction of diazomethane with itself (Figure 1.1). 3 Figure 1.1. The polymerization of diazomethane: the synthesis of pol ymethy lene Further investigations of this polymerization showed that the addition of boron 4 or copper 5 catalysts resulted in the production of straight-chain, high molecular weight, highly crystalline polymethylene Further, the first investigation of branching and polymethylene was done v ia the homopolymerization of diazoethane or 1-diazopropane to produce polyethylidene ( methyl branched) and polypropylidene (ethyl branched). The

PAGE 13

4 H----r H-~ CH 2 CH 3 CH 2 CH CH 2 CH 3 n A B Figure 1.2. A: Polyethylidene B: Polypropylidene materials were described as rigid amorphous, brittle glasses similar to polystyrene ( Figure 1 2). 3 At this time, the structure-property relationship of branching was investigated by the copolymerization ( decomposition) of diazomethane with varying amounts of diazoethane and 1-diazopropane. 3 These samples demonstrated the influence of chain branching on the crystallinity and therefore the physical properties of the polymers. These copolymers would later be used as models to clarify the structures of polyethylenes made by the more involved processes of high pressure high-temperature polymerizations. 3 In 1953 Kantor and Osthoff6 reported the synthesis of polymethylene with a molecular weight greater than 3 million, chemistry which was achieved by treating diazomethane in diethyl ether with a diethyl ether-boron trifluoride complex at O C. 7 The polymer had a crystal melting point of 132 C, which is close to what is expected for linear PE, while X-ray examination showed a high degree o f crystallinity These results indicated a high molecular weight and a low degree of branching. This assumption was confirmed by comparison of these samples with those produced by high pressure, high temperature radical polymerization.

PAGE 14

5 Similar t o the decomposition of diazomethane, this dissertation presents the s ynthesis of polyethylene and its branching by an a lternate mechanism and s o it is appropriate to briefly discuss s ome of the details of the mechanism to form polymethylene from diazo-decomposition The first such discussion was provided by Kantor and Osthoff 6 who believed that the polymerization t ook place through an ionic mechanism, a c onclusion derived from the fact that the c atalyst and solvent were ionic reaction promoters. Bawn and Rhodes 8 also produced high molecular weight polymethylene in a toluene solution with copper wire, c opper stearate and boron trifluoride as catalysts. They derived the second order rate equation by following the reaction via titration and measuring the evolving nitrogen and their results also indicated an ionic mechanism. Further investigation s 9 proposed two mechanisms for this catalytic conversion using boron compounds, and in both cases the first step was the formation of an intermediate by the nucleophilic attack of diazomethane on boron. The difference in these mechanisms was that one was cationic in nature and the other anionic. The accepted mechanism of today is anionic in nature. 3 Alternate y et historically less s ignificant, routes to polyethylene have also been described. In order to synthesize s traight c hain hydrocarbons Carothers et al. 1 0 used Wurtz coupling to polymerize decamethylene bromide This method proved to be messy and did not s uccessfully produce polymers with more than 100 carbon atoms. The Fisher-Tropsch reduction was also successful at producing polymethylene type paraffins. These products had a verage molecular weights of 7 000-9 000 g / mole and melting points ranging from 117 to 132 C. 1 1 12 Other synthetic methods were derived from the desir e t o s tudy similar polymer s tructures. Hahn and Miiller ,1 3 in order to better understand the s tructure of poly ( vinyl c hloride), c onverted

PAGE 15

6 PVC to PE by total reduction using excess lithium aluminum hydride under high pressure and temperature. It was later determined that the total reduction of poly(vinyl bromide) was much easier to accomplish. Synthesis of Polyethylene via Radical Chemistry The direct polymerization of ethylene to polyethylene was first achieved in March of 1933 by Imperial Chemical Industries, Ltd. (I.C.I.),14 where a trace of white powder was discovered during the systematic study of high-pressure chemistry of organic compounds. Soon after this discovery I.C.I. developed a commercial process to produce this polymer, and a fully operational plant was erected in 1942 with the first applications being that of a wire insulator for radar systems. The process I.C.I. developed was termed the high-pressure free-radical process, for it involved the use of pressures greater than 124 MPa, temperatures between 100-300 C and a free radical catalyst.15 This method of polyethylene production is still used today and produces a wide range of polymers called low density polyethylene (LDPE). Further, this process has been modified in order to produce medium and high density polyethylene. The first synthesis of "linear polyethylene using this method was accomplished by Dupont using temperatures between 50-80 C and ethylene pressures of 707 MPa. 16 The resulting polymer had a density of 0.955 g / cm 3 with less than 0.80 alkyl substituents per 1000 carbon atoms. 17 However, the polymers that are industrially produced using this method are low density (0.912-0.935 g/ cm3) and usually contain 15-30 variable length alkyl substituents per 1000 carbon atoms.

PAGE 16

7 The radical polymerization of ethylene consists o f a chain addition mechanism and follows the c lassical description for free radical polymerization ( Figure 1.3 ). k 1 I-CH2-CH 2 Figure 1 3. Radical polymerization of ethylene. Initiation and Propagation. As with other typical radical reactions termination consists of radical coupling and disproportionation (Figure 1.4 a ). Chain transfer events are also evident with transfer to ethylene to form a vinyl end group, or the transfer to solvent to form a terminated chain (Figure 1.4 b, c ). ( b) (c) Figure 1.4. (1) Termination of radical chain growth. (2) Chain transfer to ethylene. (3) Chain transfer to solvent. In radical systems branching can occur as a result of chain transfer events: intermolecular and intramolecular. Intermolecular transfer causes the formation of long chain branches (> 6 carbons) and is the result of normal

PAGE 17

8 chain transfer to polymer 1 8 ,1 9 via transfer of a radical to a completed polymer chain causing the g rowth of a grafted chain. The formation of long ch ains c auses changes in the melt flow viscosity of the polymer. Intramolecular chain transfer, which is about 40 times more prevalent than intermolecular c hain transfer, ca uses the formation of short chain branches. The number size, and distribution of these branches have a significant effect on the polymer 's crystallinity. A methodology to study this effect is the primary focus of this dissertation. a -A/'CH 2 CH 2C H2)H 2 ~H2 CH 2 / CH2 "-cH 2 C ~H 2 CH 2 CH2f H S:H 3 CH 2 /C H 2 "cH 2 n-Butyl branch ~H2CHCH2~H2 S:H 3 a C1 /CH 2 CH 2 n-Hexyl branch JV'CH2CH2CH~H2 S:H 3 C1 / CH2 CH 2 n-Amyl branch maximum crystallinity 60-70/o Figure 1.5. Intramolecular chain transfer during the radical polymerization of polyethylene.

PAGE 18

9 The intramolecular chain transfer event of interest is caused by the transfer of the propagating radical on the end of the chain to a more stable secondary radical on the same chain which occurs by the extraction of a hydrogen atom via a cyclization or backbiting mechanism (Figure 1.5). The resulting species can continue to undergo radical propagation with ethylene, which leads to the formation of an alkyl branch. This uncontrollable backbiting results in the formation of ethyl, n-butyl, n-amyl, and n-hexyl branches, where the relative abundance of these different branches depends on the synthetic conditions. 20 The for1nation of the n-butyl branch is the most prevalent and is attributed to the favorable formation of a six membered transition state shown in Figure 1.5. There is evidence that the polymerization of vinyl acetate follows the same radical reactions and side reactions as those for ethylene.21,22 The resulting polymer (polyvinyl acetate) is a precurser to alcohol functionalized a-olefin polymers which will be discussed further in chapter 5 of this dissertation. Synthesis of ''Polyethylene'' (Heterogeneous Ziegler Catalysts) In the fall of 1953, Ziegler and co-workers observed the polymerization of ethylene at atmospheric pressure using titanium halides and alkyl aluminum compounds, 23 24 2 5 a landmark discovery which has had a dramatic effect on the world of chemistry and is still the basis of extensive organometallic research. These catalysts are formed by the reaction between a metal alkyl or a metal hydride of the main group metals and a reducible compound of the transition elements of groups 4 to 7. 3 The most preferred combination is an aluminum alkyl or hydride with a titanium-IV compound.

PAGE 19

10 Natta and co-workers termed these new catalysts Ziegler catalysts .26, 2 7 These s ystems a re c ommonly referred to as classical or heterogeneous Ziegler catalysts. Ziegler 's discovery made it possible to produce polyethylene previously only obtainable by extreme c onditions of pressure and temperature, in a simple flask or test tube 2 8 This was first demonstrated by Ziegler in a Weck-glass vessel under atmospheric pressure and room temperature using gasoline as the solvent. 25 The polymer produced was more rigid, stronger and more heat resistant than those made by the radical process This was attributed to the fact that the polymers were made up of nearly linear molecules which resulted in a material that was more dense (0.95 g / cm 3 ) and more crystalline. The branches that did occur were identified as being mostly ethyl groups The contributions that Ziegler made in this area of organometallic / polymer chemistry resulted in his receipt of the Nobel Prize for chemistry in 1963 along with Natta. About the same time as Ziegler s discovery, two medium-pressure polymerization methods using transition metal oxide c atal y sts were discovered by the Phillips Petroleum Co. and the Standard Oil Co. o f Indiana. 3 The polyethylenes produced by these methods were more linear in nature and therefore had a higher density of 0.96 g / cm 3 Ziegler polymers as well a s the medium pressure polymers are generally a ccepted to o c c ur through an insertion mechanism. A representation o f this chain a ddition c oordination polymerization i s outlined in Figure 1.6 2 9 This reaction initiates with the transfer of an a lkyl group from the alkyl aluminum to the titanium c hloride, followed by the 7t complexation o f the monomer to the transition metal' s vacant d -orbital (Figure 1.6 A-B) A four-centered a nionic c oordination insertion process that

PAGE 20

11 inserts the monomer into the metal-carbon bond (C) then occurs. The result of this process produces a vacant orbital with the opposite configuration of the original complex (D). This mechanism is referred to as the monometallic mechanism for the stereo specific polymerization of polypropylene and was proposed by Cossee and Arlman. 30,31 Propagation Cl,,,,. ,,,,Cl Ti'' Cl ,. I Cl A CH 2 ,,.._ -CH 2 B C Figure 1.6. The insertion of ethylene as the propagation step of the coordination polymerization of Polyethylene The monometallic mechanism The Ziegler and Phillips (c hromium oxide-based) catalysts are used to manufacture virtually all of the linear polyethylene made today. These

PAGE 21

12 systems, as mentioned earlier produce nearly linear high molecular weight paraffins. These linear flexible chai n s are able to arrange in such a fashion that they can acquire long range three dimensional order (crystal line regions) which allows the polymer to pack better This resulting order produces a material that is termed high density polyethylene ( HDPE). A density of 0.95 g/ cm 3 is the result of having 1 to 3 alkyl groups (branches) per 1000 backbone carbon atoms ( Figure 1 .7), whi l e a density of 0.97 g/ cm 3 results from, on average of 1.5 alkyl groups per 1000 atoms. 3 2 RTiCI 3 -TiCI 3 + R 0.5 3 R's per 1000 Carbons R=CH 3 -CH 2 CH 3 -CH 2 CH 2 CH 3 etc. Figure 1.7. High density polyethylene (HDPE) via Ziegler catalysts. Synthesis of ''Polyethylene'' (Homogeneous Ziegler Catalysts) The latest step in polyethylene research has been the introduction of a family of transition metal complexes called metallocenes. These catalysts have been extensively used to investigate the long debated mechanism of Ziegler catalysts. El u cidating the elementary steps of a polymerization is simplified by studying so lubl e well characterized systems; 33 therefore, metallocenes have contributed the most definitive results toward this

PAGE 22

13 cause. 32 Much of the drive to understand the details of the mechanism originated from the special demands on understanding the structure-property relationship of a-olefin systems. CI -\M-CI = I Cl x x = C2Iit, Me 2 Si R1= Me, Ph, Naph R2=H,Me --Cl M= Zr, Hf x = C2Iit, Me2Si M = Ti, Zr Hf R1 = H, 5Me, neomenthyl R2 = Cl, Me Figure 1.8. Examples of metallocene catalysts.3 3 The first structure of a metallocene was described in 1952 by Wilkinson et al. 3 4 and Fischer 35 with the introduction of ferrocene. These molecules (metallocenes) were commonly referred to as sandwich compounds due to their unique spatial arrangement (Figure 1.8). 33 This new class of compounds showed considerable promise in the way of advancing organometallic

PAGE 23

14 c hemistry but at the time failed to have an industrial impact. They were frequently used as a transition metal catalyst in combination with e ither trieth y l or diethylaluminum chloride for the polymerization of o lefins. 27 3 6 The p o or c atalytic activit y that they exhibited limited there use to mechanistic s tudies This was the case until 1975 when water was accidentally a dded to a NMR tube which c ontained biscyclopentadienyltitaniumdemethyl, trimethylaluminum and e thylene. 33 Upon this addition the fast polymerization of ethylene was observed in a system that was thought to be inactive toward this process. 37 It was soon determined that the best cocatalyst for these s ystems was methylaluminoxane ( MAO) (Figure 1 9). Today, zirconocenes (z irconium transition metal metallocenes) that are combined with MAO result in olefin catalysts that are 10-100 times more active than classical Ziegler catalysts. 38 Figure 1.9. A proposed oligimer of methylaluminoxane (MAO). Cocatalyst for homogeneous Ziegler-Natta polymerizations. 3 3 Kaminsk y states that the polymerization of e thylene with bis(cyclopentadienyl)zirconium dichloride ( Cp2ZrCl2) and MAO c an result in the production of up to 40 000,000 g PE / g Zr h 33 The resulting polymers have typical molecular weight distributions of M w I M n = 2 (a much higher

PAGE 24

15 distribution is obtained with heterogeneous systems) 39 with 0.9 to 1.2 pendent methyl groups per 1000 backbone carbons atoms. Arguabl y the most important contribution that metallocenes have brought to polymer chemistry is the opportunity to develop plastics with variable control over their s tructure / property relationships. Metallocenes grant structural control by defining a single site at which the building blocks of the polymer (monomers) a re joined together linearly. Single site polymerization is in contrast to the nonuniform catalytic action that takes place in heterogeneous s ystems which follow a similar mechanism. Due to their solubility the active sites of homogeneous systems are more accessible for analytical examination therefore provide the ability to make rational catalyst modifications. 3 3 The impact that this type of control can provide is demonstrated by the design and synthesis of chiral zirconocenes, which are used for the synthesis of stereoregular polypropylenes 40 Copolymerization of a-olefins For the past fifty years the majority of low density polyethylene has been produced by the high pressure free radical polymerization of ethylene. This process inherently produces side chains (Figure 1.5), both long and short, hence producing a large distribution of chains that pack poorly (low 0 /o crystallinity) or have no semblance of order (amorphous). The homopolymerization of ethylene using Ziegler catalysts results in a very marketable plastic termed high density polyethylene (HDPE). These polymers do contain some degree of short chain branching but are essentially considered linear. In contrast when ethylene and higher order olefins are copolymerized in the presence of Ziegler c atalysts a new class of material is

PAGE 25

16 obtained that is termed linear low density polyethylene ( LLDPE) (Figure 1.10). The copolymerization of ethylene with a-olefins like 1-butene, 1-hexene or 1octene has opened up a new chapter in polyethylene based materials and currently shows a higher growth rate than the homopolymer. These new products offer an advantage over those made via radical chemistry in that there is no significant uncontrollable branching. This inherent linearity produces molecules in which the only significant branch points are from the incorporation of the a-olefins (Figure 1.11). R R + \ Cat CH R \ R Figure 1.10. The copolymerization of higher order a-olefins with ethylene. The polymerization of linear low density polyethylene (LLDPE). LLDPE was introduced in 1977 with Union Carbide's implementation of a low pressure, gas-phase, fluidized-bed process.1 7 This process was later termed Unipol and was soon followed by Dow Chemical with a new line of LLDPE polymers termed Dowlex Tl\ -1. Today these and other companies have

PAGE 26

17 s hifted gears toward the development and production of large amo unts of LLDPE. The use of heterogeneous catalyst syste ms for these copolymerizations, results in LLDPE products with large polydispersities and little to no control HDPE Polymeri ze d by Zeigler Natta or Metallocene c h e mi stry ca t n CH 2 -1 H n>>>m A t LDPE ----.. Polymerized by radical addition mechanism R= infrequent s hort c hain branches -+-+-CH_ -CH 2 -n-CH2 -,H m A t n~m R= frequent s hort and long chain branches LLDPE Copolymerized by Zeig ler t ype catalysts nCH2,== CH2 + m CH,;::.2 = CH ca t ., A CH-2 -CH 2 -0 -CH --,H A t R= the identity of R from mon o mer feed Figure 1.11. The induced or non-induced branching in polyethylene. over comonomer s equence distributions. 4 1, 42 ,43, 4 4 Contrary to this the development and use of homogeneous catalysts have produced LLDPE samples with narrow molecular-weight and more uniform comonomer

PAGE 27

18 possible to design new materials that are driven by special demands on properties. The resulting s hort branches on these polymers produce materials with lower melting points, lower crystallinities, a nd lower densities as compared to HDPE. 33 Structure-Property Relationships of Polyolefins The ability to design systems from a molecular standpoint (c omonomer identity and ratio) in order to tailor physical properties has benefited tremendously from homogeneous systems. By adjusting comonomer composition or catalytic activity/ affinity, a macromolecule can be designed and produced with known but irregular molecular composition. The use of homogeneous systems provides a method to assemble these mixtures of comonomers while eliminating nearly all the undesirable side reactions and producing narrower molecular weight distributions. This lack of "regular" order is due to the variable reactivities of different monomers towards the polymerizing catalyst. The affinity for any given ole fin to insert depends on the last inserted monomer and is defined by a parameter termed the monomer reactivity ratio. For a two monomer system these ratios are represented by r 1 and r2 and defined in equation (1). The term r1 is expressed as the ratio of the rate constant k r 11 1 k1 2 k r 22 ') "21 (1) of inserting an ethylene unit into the growing chain versus the rate constant of inserting an a-olefin, when the last monomer inserted was an ethylene. Parameter r2 in this case, would be the analogous ratio for the a-olefin.

PAGE 28

19 Equation ( 2) defines the c opolymer composition and is related to the concentrations o f the two monomers in the feed [m1] and [m2] and their reactivity ratios. d[ m 1 ] [ !1li ]( r [ fni] + [ m 2 ]) d[ m 2 ] [ ]([ fni] + r 2 [ m 2 ]) ( 2) The tendency for a comonomer to homopolymerize, randomly copolymerize, or alternatively copolymerize can be inferred by reactivity ratios The product of the two reactivity ratios r 1 r2 represents the distribution of the comonomers. When this product equals one it represents a randomly distributed copolymer. 3 3 A product value considerably less than Table 1. Examples of copolymerization parameters of metallocene/MAO catalysts in combination with ethylene and a-olefins. Metallocene Cp2ZrMe2 [En(lnd)2]ZrCl2 Cp2ZrCl2 Cp2ZrCl2 [En(Ind )2] ZrCl2 Temp. in C 20 50 40 80 30 a-Olefin propene propene butene butene butene 3 1 7 55 85 8 0.005 0.06 0.017 0.010 0.07 0.25 0.40 0.93 0.85 0.59 Cp2ZrMe2 60 hexene 69 0.02 1.38 Data taken from Kamin s ky, W. Macr o m o l. Cl1em Phy s 1996 197 3907. one represents a somewhat alternating structure. 48 The use and modification of metallocene catalysts have provided limited control over these reactivity

PAGE 29

20 ratios, by modifying the symmetry and steric crowding of the active center as well as the temperature of the polymerization. Some examples of this are shown in Table 1. 33 The modification of catalyst and conditions does afford some control of comonomer distribution but the placement of monomers is still irregular. These systems do, though, provide samples with better control over the intermolecular distribution of branches than previously possible. This inherent variability limits the usefulness of these systems toward the fundamental understanding of structure, morphology and property relationships. Crystallinity, Morphology, and Melting of Flexible Chain Macromolecules The ability of flexible chain polymers to orient in such a fashion as to display long range three dimensional order is referred to as their ability to crystallize. This is similar to the crystallization process in low molecular weight molecules but differs in that the crystallizing molecule is larger than the unit cell. The increased size of polymers enables a single chain to crystallize in multible unit cells. Multi-participation and its corresponding thermodynamics are due to the connectivity of a vast number of chain atoms which result in the unique yet complex crystallization properties of polymer systems. Polyethylene and its copolymers are considered semicrystalline because they contain both crystalline regions and viscous glassy regions.49,SO The viscous glassy regions are termed amorphous and are composed of unordered molecules or repeat units.

PAGE 30

21 The morphology of the crystalline regions of polymers has long been debated. The f ri11 g ed-micelle theory developed in the 1930s, proposed that a polymer s crystallinity came from the occasional alignment of chains in an extended fashion creating order. This theory was supported by the belief that the most thermodynamically s table c rystal was one involving extended chains. 5 1 Twenty years later, upon the isolation of a polymer single crystal, a new theory on chain morphology arose called folded-chain lamella. 5 2,53,54,55 This conformation consists of the chains folding back on themselves (lamellae) creating ordered crystalline regions. The lamellae nucleate from a central point making up a crystal region within the polymer termed a spherulite. (Figure 1.12). PE Spherulite Larnellae Figure 1.12. Representation of PE Spherulite and Lamellae. Mandelkern,5 6 in 1986, acknowledged that the crystallite morphology is clearly lamella but that the molecular morphology and chain structure cannot be determined by the analytical techniques currently available. The theory of folded chain lamella is widely accepted today, but is still the subject of extensive investigation. 5 7 58 5 9,6 0 61

PAGE 31

22 The molecular requirement for crystallinity in macromolecules is that the c hain ato ms or repeat units must be ca pable of adopting an ordered co nfigurati o n in which the a djacent units ca n lie parallel to one another creating well defined lattice s ites. 6 2 This ab ility to conform i s directly related to the repeat unit 's degree of regularity. The three forms of regularity a re 1 ) chemical, 2 ) geometrical, and 3) spatial. Chemical regularity is defined by the frequency of the repeat units. For example, in linear polyethylene the repeat unity is -(CH2-CH2)and its degree of chemical regularity is defined by how o ften this repeat unit exists without the appearance of branches. When branching occurs, the degree of crystallinity is noticeably changed. Linear polyethylene contains 2 to 3 side branches per 1000 carbon atoms and is ~90/o crystalline ,63 at 30 branches per 1000 carbons the crystallinity decreases to 50/o.64 The decrease in crystallinity due to an increase in branching is thought to be a result of the exclusion of branched units from the crystalline regions .65 Geometrical regularity is defined as the regular placement of an unsymmetrical repeat unit If an unsymmetrical repeat unit is incorporated into a chain in both directions it results in what i s termed head to tail ( HT) tail to tail (TT), head to head ( HH ), and tail to head (TH) placement. These different connectivities result in a geometrical irregularity which disrupts a chain's c onformational abilities. Spatial regitlarity results from the systematic placement of substituents that can attain different spatial arrangements (s tereo isomerism) This occurs during the polymerization of monosubstituted alkenes s uch as higher order a-olefins. This can be seen by ex amining s amples of spatially regular (tactic) and non-regular ( atactic ) oriented polypropylenes. The ordered ( tactic)

PAGE 32

23 polypropylene results in a highly crystalline polymer while the atactic ve rsion is a completely a morphous material These three mole c ular regularity requirements--chemical, geometrical, and spa tial-a re determining factors for the ability of a material to c r ys tallize based so lely on the makeup of the c hain itself However there a re many other va riables that affect the crystallinity of the material as a whole. These include molecular weight, mole c ular distribution end groups, diluents ( impurities) and thermal history. Thermal Behavior and Chain Structure Much of the research on polymer crystallinity concerns the resulting thermal behavior. Thermal behavior relates to order because the chain structure influences the melting point of crystalline regions (Tm) through conformational properties and thus the entropy of fusion. 66 The melting crystallization process of long c hain polyethylenes is a first-order phase transition and c an be related to the sa me transition in s mall mole c ules. 57 Small molecules, though, lack the co nformational co nstraints o f a la rge polymer therefore resulting in sharp transitions when its eq uilibrium temperature is attained. The longer c hains introduce structural or morphological "i mpurities" ca using the broadening of this fusion process ( Figure 1.13). 62 This broadening effect can also be observed by comparing the melt s of fractionated a nd polydisperse sa mples 6 7 thus demonstrating that the fusion process is interrupted by va rying molecular constitution while the peak of the fusion c urve remains relatively the sa me. Examination of linear low molecular weight PE samples s how s that c omparable f u sio n c urve s can be ob tained that are s imiliar to those of

PAGE 33

24 hydrocarbons 6 2 having a limited critical length of only 150 ca rbons. 68 Therefore, relating folding and/ or fusion effects from low molecular weight samples to those of larger PE samples of sim ilar molecular makeup is a viable methodology This is possible because the molecular weight ( length) at which ( 0) ( b) 400 380 360 340 320 T (Kl Figure 1.13. Differential Scanning Calorimetric endotherms: 62 (a) low MW PE Mn= 725 g/mole; (b) n-hydrocarbon C44H90 MW= 618 g/mole chain folding can occur is the same for n-alkanes and PE polymer fractions. 6B,69, 7 o, 7 1 A detailed study comparing the melting temperatures of a lkan es with those of linear polyethylenes was done by Mandelkern et al.,7 2 demonstrating that identical melting points could be obtained for most samples but a lack of correlation for infinitely large samples Mandelkern et al. explain this as being a result of a large increase in the interfacial free energy with a large increase in size.72 The research in this dissertation is concerned with the effect of a regi,larly distributed s tructural irregularity (b ranches). Therefore, it is

PAGE 34

25 appropriate to briefly discuss how copolymerization affects the crystallization and melting points of these types of materials. It can be generally stated for random copolymers that the concentration of crystallizable units decreases v ia the addition of a copolymer impurity, the melting range becomes progressively broader and the level of crystallinity decreases. The amount of theoretical temperature depression for random copolymers was predicted by the Flory equation which is shown as equation 3 73 here where Tm 0 and Tm are the melting temperatures of the unbranched and branched polymers, R is the gas constant, Hi, is the heat of fusion per repeating unit, and XA is the mole fraction of crystallizable units. Results, using this equation, have shown that the identity of the alkyl pendent groups does not affect the amount of depression as related to the temperature composition relationship. 74 7 5 Alamo and Mandelkern point out that this result is an example of the fundamental principle that the melting temperature of a copolymer does not depend on the composition but on the details of their sequence distribution.75 There have been numerous studies describing the physical effects of these randomly copolymerized units. 7 5, 7 6, 7 7, 7 8 As described earlier, chain copolymerization results in an inhomogeneous distribution of alkyl branching. Compared to heterogeneous Ziegler catalysts, the metallocenes grant some control over the intermolecular distribution of side chains. Therefore, they produce polymers in which each individual molecule possesses the same distribution of branches as another, but the distribution of

PAGE 35

26 the branches is not uniform along the individual polymer backbones. The lack of order in both cases has recently been termed intra and intermolecular heterogeneity respectively.7 9 X Hydrogenated butadiene / propene co-polymer Hydrogenated butadiene/ ethene / propene terpolymer Figure 1.14. Methyl branched model polymers for PE.80,81 Both types of distributions influence the superstructure and the crystallization behavior of these materials; therefore, model systems that can control both of these variables would be suitable for detailed investigations. Gerum et al. 8 0, 81 modeled this control by studying the short branching achieved by strictly alternating hydrogenated poly[butadiene-alt-(1-olefin)] copolymers and butadiene/ethene/1-olefin terpolymers (Figure 1.14). For the copolymers, a limit of 167 short chain branches per 1000 back bone carbons could be achieved with a intrahomogeneity of a branch point appearing every 6 backbone carbons, and a strictly alternating content greater than 97 /o The copolymers related to this dissertation are produced when propene (methyl branch) is used as the copolymer. The copolymers with propene from Gerum

PAGE 36

27 et al. s how by X-ray scattering and thermal analysis that a completely amorphous material is produced. It is important to note that a glass transition was detected for these samples at -66 C. Terpolymers subsequently were produced in order to increase the number of methylene units between branch points by the incorporation of ethene as a spacer. This type of ter-unit addition is able to dilute the number of branch points per chain resulting in spacers between branches being on average 5, 11, 17, 23 etc. The resulting terpolymers, using propene as the 1olefin, had as many as three thermal transitions with the first two being very broad. Even though these models proved to be interesting to study, the samples still contain a significant degree of intraand intermolecular heterogeneity. Until complete control of the copolymerization can be achieved, the use of Ziegler chain addition reactions will not provide the necessary control over the microstructure in these polymers. Therefore, the design of well defined models for the fundamental study of structure property relationships will not be possible using these means. Metathesis Polymerization Ring Opening Metathesis Polymerization (ROMP) and Well Defined Alkylidenes Olefin metathesis, over the last 25 years, has become a well established means of cleanly producing linear unsaturated polymers. The majority of this research lies in the area of a chain growth mechanism termed ring opening metathesis polymerization (ROMP).8 2 ROMP is represented by the polymerization of a strained cyclic olefin to its corresponding linear

PAGE 37

28 unsatur ated polymer (F igure 1.15). The r es ulting polymers (F igure 1.15) are similiar in st ructure to those p r od uced by the Zieg ler polymerization of dienes ( butadiene, isoprene) in tha t both types o f products contain a site of unsaturation and a re linear ca rbon backbone chai ns ROMP initiator Figure 1. 15. Th e ROMP of Cyclooctene. The metathesis polymerization is made possible by a transition metal catalyst (initiator) in the form of a metal carbene. This reaction, as are all metathesis reactions, is go verned by a c ompeting ring chain equilibrium. 8 2 c The feasibility of shifting the equilibrium toward chain production is usually driven by enthalpy; therefore, s trained cyclics, such as 3, 4, and 8 membered rings and norbornenes provide the necessary energetics for polymerization. Si milar to the use of c lassical Ziegler-Natta cat alysts the synthesis of ROMP po lymer s was first achieved by polymerization using heterogeneous catalyst systems. These systems were typically composed of various transition metal halide s and oxohalides acco mpanied by a wide range of metal alkyl and organohalide Lewis acid co catalysts. Again, due to the unknown nature of the true ca talyti c species, the mechanism of polymerization could not be directly studied and therefore no rational changes in the system could be made. The heterogeneou s ca talysts required the a ddition of a Lewis acid cocatalyst giving these sys tem s poor fun c tional g roup tolerance. So me examples of shor t term living sys tems were o bserved but upon longer

PAGE 38

29 reaction times would undergo secondary metathesis reactions, theref o re broadening t heir molecular weight distribution. 8 2 c,83 In 1986 S haverien Dewan a nd Schrock developed a well defined alkylidene complex with a highly active four coo rdinate do tungsten ce nter. 8 4 This complex did not require a Lewis acid cocatalyst and was capable of olefin metathesis. Due to the homogeneous nature, these systems provided the arena for s tructure elucidation and mechanistic pathway discovery. 85 ,86 Following this discovery by Schrock and co-workers, similar alkylidene systems were able to perform ROMP in a living manner as well as resist "poisoning functionalities. This enabled the design and s ynthesis of new block copolymers and new functionalized polymers via a metathesis mechanisms X X [Ru] Figure 1.16 Functionalized ring closing metathesis (RCM) using Grubbs ruthenium alkylidene. Taking the design of homogeneous alkylidines o ne step further Nguyen et al., in 1992, presented the sy nthesis of a well defined ruthenium alkylidene which initiated ROMP in both organic and protic/ aqueous solvents. 87 This design also showed living characteristics as well as multiple functional group tolerances therefore opening up new doors for olefin metathesis chemistry. Unique to this sys tem was its ability to undergo

PAGE 39

30 met at he s i s in the presen ce of aldehydes, alco hols ca rbo xycl ic ac ids a nd quarternary a mines. 88 T his was demonstrated by Fu, Nguyen, a nd Grubbs usin g ring clos ing metathesis fo r the fo rmation of s ix membered functionalized rings (F igure 1 16) 88 Further in ves tigations of the ruthenium sys t e ms have developed new str uctural reacti vity relationships and isolatable inter1nediates which have provided a tremendou s impact from a mechanistic standpoint 89 C hapter 5 in this dissertation examines a ruthenium alkylidene as a c atalyst fo r the direct sy nthesis of alcohol functionalized polymers. Acyclic Diene Metathesis Polymerization (ADMET) The inherent control over polymer structure that Schrock and Grubbs well-defined alkylidene catalysts provide has significantly broadened the scope and applications of ROMP chemistry, but it has had an explosive effect on the evolution of a rather new polymerization method called Acyclic Diene Metathesis ( ADMET) polymerization .90 ADMET has been under investigation for 25 years, but upon the advent of Schrock 's alkylidenes the vi ability and understandin g of these sys tems have become a reality. Prior to their introduction the metathesi s reaction was plagued with problematic s ide reactions and an undefined mechanism Based o n theoretical and experimental o bservations of Schrock's well defined tungsten alkylidene, Wagener Bonc e lla, and Nel proposed the accepted mechanism of ADMET. 9 0 The ADMET a nd ROMP mechanisms both proceed through the equilibrium polymerization of olefins via metathesis However ADMET differs by involving the polymerization of acyclic dienes, with its equilibrium s hift ed to pol yme r by the removal of a s mall alkene (entropically) (Fi gure 1.17). ROMP is typically s hifted to high polymers via a favorable e nthalpy

PAGE 40

3 1 change v ia ring strain release. The major mechanistic difference is that ROMP follows a chain growth process where the active alkylidene acts as a catalytic initiator. In contrast, ADMET follows a step condensation mechanism and the alkylidene behaves as a true catalyst. Catalyst R R n Figure 1.17. ADMET polymerization: General reaction scheme. The accepted mechanism for ADMET is outlined below in Figure 1.18. Upon addition of the catalyst to a neat solution of dienes, the olefin will first coordinate in a 1t complex with the transition metal center followed by an insertion reaction forming either a productive (d) and nonproductive (c) metallacyclobutane intermediate. The nonproductive cyclic (d) can collapse to form the original 1t complex, whereas (c) undergoes a productive rearrangement eliminating the precatalyst fragment (e) while forming the monomer as part of the new alkylidene (). Another monomer can then coordinate and insert in the same fashion as before, producing the metallacyclobutane (g). On the first trip around this cycle, collapse of this metallacyclobutane forms the dimer, with ensuing trips forming powers thereof. The collapse of (g) leaves the catalyst as a methylidene (h) which is considered to be the true active catalysts species. 89 This methylidene then coordinates with another monomer, or appropriate chain end, forming the metallocycle (i), which upon its collapse produces the small molecule

PAGE 41

a + b release of ethylene ( driving force) , , , , , I coo rdinati on of olefin I I t I productive insertion A R f onnation of new alkylidene insertion o f mon o mer or polymer LnM=CH '.) h Figure 1.18. ADMET polymerization cycle. d e non-productive insertion R R c -econd monomer 32 o r polymer olefin insertion ' ' ',, / ' g generation of active methylidene and free polymer n

PAGE 42

33 ethy lene while returning back to the top of the cycle with the alky lidene having the identity of the stepwise growing chain. The clockwise movement of this cyc le is ca used by the removal of the small molecule, in this case ethylene, producing (de pending on the number of cycles) a high molecular weight polymer. The movement of this cycle in a co unter c lockwise manner has also been demonstrated by the depolymerization of high molecular weight polymer in so lution via the addition of excess ethylene 91 92 or various functionalized monoenes.93,94,95 Unlike ROMP polymerizations, ADMET condensations are performed under neat conditions so that co mpeting equilibria are avoided and the release of ethylene is favored. The step condensation process of ADMET requires stringent conditions to achieve high molecular weight polymer. Under step co nditions, high polymer can only be achieved if the system obtains an overall conversion greater than 98/o. This kinetic behavior is in contrast to the chain methods used to make PE and the corresponding copolymers were a conversion of 90/o is considered excellent. 51 In these chain growth systems, monomers add to a highly energized chain end, which results in high molecular weight polymer almost instantly. Step condensations proceed by monomers first coming together to form dimers, which then condense to form the tetramer and so on. Carothers defines the stepwise growth in an equation which calculates the degree of polymerization (DP) as it relates to the extent of reaction (p) (eq uation 4). 51 The number of these repetitive co ndensations DP= l ( 1-p ) (4)

PAGE 43

3 4 ( DP) is also related t o the stoichiometric balance of the system. However the ADMET s ystem typic a ll y involve s c onden s ing two identical functional groups in an A-B type manner which e liminats the stoichiometry concerns. ADMET polymerizations also follow o ther notable stepwise constraints, including the formation o f molecules during equilibrium conditions that have a most probable distribution ( polydispersity index (PDI)) of 2. This distribution is commonly referred to as a Flory distribution and is represented in Equation 5 as the weight-average molecular weight ( M w ) divided by the number-average molecular weight ( Mn). 5 1 Fortuitously PE samples synthesized by various metallocenes have a similar most probable distribution. This will allow for better comparisons between the step samples of ADMET polymerization and those produced by metallocenes. ADMET: Step Condensation Method to Producing a-Olefin Precursor Polymers The e stablishment o f ADMET occurred V la the meta thesis polymerization of 1,9-decadiene using the tungsten9 6 and molybdenum versions of Schrock's alkylidene. The investigation of this and other hydrocarbon monomers has demonstrated the clean conversion of a ro hydrocarbon dienes into linear unsaturated polymers. 9 7 ,9 8 Using Grubbs' ruthenium alkylidenes, similar results showing the clean conversion of hydrocarbons c an also be obtained. 9 9 The resulting polymers again, like the ROMP polymers produce linear unsaturated carbon backbone polymers similar to their a-olefin counterparts ( Figure 1.16). For example, the quantitative hydrogenation of polyoctenelene produced via ADMET results

PAGE 44

3 5 in a saturated hydrocarbon chain that is analogous to completely linear PE 100 This dissertation is an extension of this work, using the same well established clean conversion of a, m-hydrocarbodienes in order to produce PE samples with inter and intramolecular homogeneity of branch inclusion. The durability of Grubbs' ruthenium benzylidene toward alcohol functionalities is also investigated as a means to directly synthesize well defined models of alcohol functionalized a -olefin polymers similar to those of polyvinylalcohol-polyethylene copolymers. t

PAGE 45

CHAPTER2 EXPERIMENTAL I n strumentation and Analysis ltt NMR 300 MHz and 13c NMR 75 MHz spectra were recorded on a General Electric QE Series NMR Superconductiong Spectrometer system or a Varian Associates Gemini 300 Spectrometer. All NMR spectra were recorded in CDCl3 with v /v 0 03 /o TMS as an internal reference. Chemical shifts reported were internally referenced to residual chloroform. Infrared data was recorded on a BioRad FTS/ 40A infrared spectrometer. Analyses were performed between NaCl plates neat or with chloroform as a solvent. Purity of compounds and reaction conversions were determined on either a Hewlett-Packard HP5880A gas chromatograph using a capillary column with a flame ionization detector or on s ilica coated tlc plates with mixtures of pentane and ethyl acetate as the mobile phase. Micro-extractions were used in order to monitor reactions by GC. This was done in order to remove all water soluble salts before injection into the GC. All pertinent GC peaks were confirmed by mass spectrometry or NMR on the isolated compound. Low and high Resolution Mass Spectrometry (LRMS), ( HRMS) was recorded on a Finnigan 4500 Gas Chromatography / Mass Spectrometer using either electron ionization or chemical ionization co nditions. Elemental analyses were performed by Atlantic Microlab, Inc., Norcross GA 36

PAGE 46

3 7 Gel permeation chromatography (G PC ) was performed using a Waters Associates l iquid chromotograph U6K eq uipped with a tandem ABI Spectroflow 757 UV absorbance detector and a Perkin-Elmer LC-25 RI detector. All molecular weights are relative to polybutadiene or polystyrene standards. Polymer samples were prepared in HPLC grade THF or CHCl3 (~ 1 /o w / v) and filtered before injection (a volume of 20-40 uL) The GPC was equipped with a Ultrastyragel linear mixed bed column (CHCl 3) or two successive 5 x 103 A and 5 x 10 4 A ( THF) Phenogel columns (crosslinked polystyrene gel). HPLC grade chloroform or THF were used as the eluent at a constant flow rate of 1.0 ml/min. Retention times were calibrated against polystyrene standards ( Scientific Polymer Products, Inc.) or polybutadiene standards (Polysciences, Inc.) All standards Mp or Mw were selected to be well beyond the expected polymers range. A minimum of 5 data points were achieved for a calibration curve On noted samples an internal standard of acetone was used. Differential scanning calorimetry (DSC) analysis were performed using a Perkin-Elmer DSC 7 at a heating rate varying between 202 C / min All samples were first cooled to -120 C ( using liquid nitrogen as the coolant with a helium flow at a rate of 30 ml / min. ) and underwent isothermal cooling for 2 5 minutes before scanning up to 200 C followed by isothermal heating for 2 5 minutes. Multiple cycles were performed with data collection on the second heating cycle or later. When transitions were identified the samples were then slowly scanned over the pertinent temperature range. Reported values are Tm peak (first order transition peak position), Tm onset and T g (glass transition). Thermal c alibrations were done using indium and cyclohexane as standards for both peak temperature transitions as well as for heats of fusion All samples were run using a n empty pan as a reference and empty cells as a subtracted baseline. Thermogravimetric analysis was

PAGE 47

3 8 performed on a Perkin-Elmer TGA 7. All samples were heated from room temperature to 700 C in nitrogen at a scan rate of 10 C / min. The onset of weight loss was taken as the reported value. Materials and Techniques Schrock's molybdenum metathesis catalyst [(CF3)2CH 3C 0]2(N-2,6-C6H3i-Pr2)Mo=CHC(CH3)2Ph, was synthesized according to published methods. 1 01 Grubbs' ruthenium catalyst, RuCl2( =CHR)(PCy3)2 were Cy = cyclohexyl, was provided by the Wagener group members, specifically Mark Watson, Shane Wolfe, and Dr. John D Anderson via literature procedure. 89 All catalyst systems employed in this study will be graphically depicted during their pertinent discussions. p-Toluenesulfonohydrazide (TSH) was purchased from Aldrich and was recrystallized from methanol prior to use. Tripropylamine (TP A) and 0-xylene were purchased from Aldrich and distilled from CaH2 prior to use. Two molar potassium tert-butoxide was prepared in a dry sch lenk tube by combi ning the salt (Aldrich) with THF distilled from NaK alloy. 5-bromo-1-pentene, 8-bromo-1-octene (Aldrich) a nd 10-bromo-1-decene (A lfa Aesar) were used without further purification. Tetrahydrofuran THF and dimethoxyethane DME were first distilled from NaK alloy using benzaldehyde as an indicator. "S uper dry" ethanol was prepared as described in the literature. 1 0 2 All other reagents mentioned in the experimental were used as received. Micro-extractions were done by using a pproximately 1 /2 ml of the reaction mixture with a n eq ual amount of water or acid, followed by vibration Frequently 1 / 2 ml of diethyl ether would also be added to this

PAGE 48

39 solution. After vibrating for 1 min. the mixture was al lowed to separate into layers with the upper (o rganic ) l aye r used for GC a naly sis Synthesis and Characterization Synthesis of Symmetrical Alkyl-Substituted Ter1ninal Dienes Step A (one pot two step synthesis): Synthesis of ethyl-2-acetyl-2-(4pentene)-6-heptenoate (la). 10.9g (84 mmoles) of ethyl acetoacetate ( Aldrich ) and 200 ml of dry DME (Aldrich) were placed in an argon purged 500 ml three neck flask equipped with a magnetic stir bar and condenser To this stirring solution, 42 ml of a 2 molar solution of potassium tert-butoxide in THF was added. Upon this addition the solution turned lime-green in color. This solution was allowed to stir for 0 .5 hours at room temperature followed by slowly adding 12.5g (84 mmoles) of 5-bromo-1-pentene by syringe and raising the temperature to reflux. White salts were formed upon reflux. The first addition was co mplete in 18 hr s as shown by GC. The reaction was then c ooled to room temperature and the seco nd addition of 42 ml of the 2 molar solution of potassium tert -but oxide in THF was added followed by the alkenyl bromide as above. The reaction was again followed by GC with the majority of product formation done in 24 hrs. The reaction was then quenched with 3M HCl and extracted with ether. The ether was dried ove r MgS04 and then evaporated y ielding ~80/o of la. The following spectral properties were observed: 1 H NMR 1.11 ppm (m, br, 4H), 1.19 ppm ( t 3 H), 1.79 ppm (m, br 4H), 1 .98 ppm (q, 4H), 2.04 ppm (s, 3 H), 4 12 ppm (q, 2H) 4.91 ppm (m, 4H) 5.70 ppm (m, 2H); 13c NMR 14.06 ppm 23.21, 26.62, 30.79, 33.8 4, 61.09, 63.36, 115.1,

PAGE 49

40 137.94 172.5 205.1. The Low Resolution Mass S pectrum ( LRMS EI) a l so confirms structure with a parent ion at 266, c alcd for C16H2603 266. Ethyl-2-acetyl-2-(7-octene)-9-decenoate (lb). Synthesized as above. Product not isolated for NMR spectral analysis LRMS ( EI) 3 50 calcd. for C22H3g0 3 3 50 Ethyl-2-acetyl-2(9-decene)-11-dodecenoate (le). Synthesized as above. Product not isolated before subsequent step. The product of this reaction was a thick oil. GC retention times were c onsistent with what was expected for this compound and the resulting side products. Step B: Retro-Claisen condensation: Synthesis of ethyl-2-(4-pentene)-6heptenoate (2a). In an argon purged dry 250 ml 3 neck round bottom flask, equipped with a condenser and a magnetic stir bar, 22.34g (84 mmoles) of 1 and 100 ml of dry ethanol were added. To this solution 35ml of a 21 /o solution of sodium ethoxide in ethanol (Aldrich) was added. (This solution was also prepared by the addition of 21 weight percent of sodium metal into "dry ethanol. Caution : This addition should be done under an inert atmosphere using "s uper dry" e thanol. Any c ontact with water in the presence of oxygen can likely result in a fire. Super dry ethanol was prepared as described in the purification of organic compounds. 1 02) The solution was allowed to reflux for 3.5 hours and turned a dark yellow in color. After cooling to room temperature the solution was quenched with water and 3M HCl followed by extraction with pentane or ether. The organic layer was then dried over MgS04 and evaporated under reduced pressure yielding ~90 /o of ester 3. The product was vacuum distilled through a short path vacuum distillation apparatus. The product was collected between 130-150 C at 4mmHg. The following spectral properties were observed: 1 H NMR (CDCl3):

PAGE 50

41 1.25 ppm ( t 3H), 1.40 ppm ( m br, 6H), 1.62 ppm ( m br, 2H), 2.05 ppm (q, 4H), 2.31 ppm ( m lH), 4.14 ppm (q, 2H), 4.95 ppm ( m, 4H), 5.79 ppm (m, 2H); 13c NMR: 1 4.5 ppm, 26.9, 32 .0 33.8, 45.5, 60.0, 114.9, 138.6, 176.0. The Low Resolution Mass Spectrum ( LRMS EI) also co nfirms structure with a parent ion at 244 calcd for C14H2502 244. Ethyl-2-(7-octene)-9-decenoate (2b). Synthesized as above with high conversion indicated by GC. 1 H NMR (C DCl 3) : 1.241.55 ppm (m, br, 24H), 2.01 ppm (q, 4.6H), 2.28 ppm (m, lH) 4.10 ppm ( m, 2H) 4.92 ppm (m, 4H) 5.75 ppm (rn, 2H); 13c NMR: 14.31, 25.80 27.31, 28.09, 28.79 28.89, 29.33, 32.44, 32.61, 33.7 0, 45 69 46.50, 59.88, 64.05, 114.12 139.04, 175.93, 176.54. Spectra consistent with structure. LRMS 308, calcd. for C20H3602 308. HRMS 308.2739, calcd. for C20H3602 308.2715. Ethyl-2-(9-decene)-11-dodecenoate (2c). Synthesized as above with high conversion indicated by GC. Product not isolated before subsequent step. Product was a thick oil with consistent GC retention times with what was expected. Step C: Reduction of ester: Synthesis of 2-(4-pentene)-6-heptene-1-ol (3a). In a flamed dry 250 ml three neck round bottom flask equipped with a stir bar and condenser, 9.301g of 2 and 125 ml of dry THF were placed. This solution was kept under an inert atmosphere and cooled to O C. To this s tirring so lution, 25 ml (2 eq) of a lM solution of LiAlH4 (Aldrich) in THF was added over a period of 5 min. Some bubbling was observed during this addition. The reaction was allowed to warm to room temperature and stirred for 2 hours. The reaction was then slowly quenched with water followed by 3M HCl. The solution was extracted with ether dried over MgS04 and rotary evaporated. A clear oil was recovered and vacuum distilled at 69-72 C at

PAGE 51

42 1mm Hg. T he reduction resulted in 9 1 /o yie ld. The fo llowing spectral properties were o bserved: 1 H NMR (CDCl3): 1.18 ppm (m, br 9H), 1.79 ppm (s l H ), 1.98 ppm ( q, 4H) 3. 49 ppm (d, 2H), 4.88 ppm (m, 4H) 5.78 ppm (m 2H) ; 13c NMR : 26.08 ppm, 30.29, 33.98, 3 4.01 40 .22 65.36, 114.28 138.74 IR (CDCl 3 1 cm1 ): 33 83.8 (b r), 3078.1 30 13 .4, 2931.0 2860.9 1640.4, 1460.3, 1217.1 1030 5 996.1, 913.1, 759.5. Elemental a nalysis C12H230 C(calc=79.06, found=78.99), H(calc=12.16, found=12.16 ) 2-(7-octene)-9-decene-1-ol (3b). Synthesized as above with high conversion by GC Short path vacuum distillation was done for purification with the main fraction collected at ~ 100-110 C at lmmHg. Gas cromatograph showed a purity of 98/o. Isolated yield was 61 o/o Spectral analysis: lH NMR 1.27 ppm (m, br 22H), 2.03 ppm (br 4H), 3.50 ppm (m, 2H) 4.93 ppm (m, 4H), 5.78 ppm (m, 2H); 13C NMR: 26.77 ppm, 28.87, 29.05, 29.84 30.86, 33.72, 40.45, 65.50, 114.06, 139.06. 2-(9-decene)-11-dodecene-1-ol (3c). Synthesized as above with high conversion by GC. Due to the size, and therefore the boiling point, of this product a high vac uum s hort path distillation was done using a short path high vac uum distillation ap paratus e quipped with a dry ice isopropanol condenser. The product, while s tirring with a Teflon stir bar, was brought to full vacuum (> 105 mmHg) at room temperature. The so lution was then slowly heated until condensing was observed. Due to the design of this apparatus the temperature of the co ndensing gas could not be measured. During the distillation three fractions were taken. The temperature of the oil bath was approximately 180 C. A purity of 96 /o was determined by GC. Spectral analysis : 1 H NMR (C DCI 3): 1.29 ppm (s, br, 32H), 1.71 ppm (s, lH), 2.02 ppm ( q 4H) 3. 49 ppm ( d 2H), 4.95 ppm ( m 4H), 5.38 ppm (br, 0.2H, elimination produ c t ), 5.78 ppm ( m 2H); 13C N MR : 25 71, 26.83, 28.88, 29.09,

PAGE 52

43 29.45, 29.54, 30. 02 30.88, 32.72, 33 .75, 40.46, 65.51, 114.03, 139.08. LRMS (CI) M+ 1=323, calcd. for C22Hi20 322 Step D: Tosylation: 6-p-toluenesolfonyl methyl-1,10-undecadiene (4a). In a flame dried and argon purged 100 ml three neck flask, 8.290g of the alcohol 3a (46 mmoles) and 60 ml of CHCl3 were added with a stir bar. This solution was then cooled to O C followed by the addition of 7.28g (92 mmoles) of pyridine. After stirring for 15 min., 13.03g (69 mmoles) of toluenesolfonyl chloride (Aldrich) dissolved in 35 ml of CDCL3 was slowly added (15 min.) by syringe or addition funnel. A brown yellow color was observed following this addition. This solution was then allowed to warm to room temperature and stirred for 8 hours. The reaction was followed by GC until high conversions were observed. The reaction was then stopped and washed with 3 M HCl to remove the pyridine. The organic layer was then washed with water followed by a wash with K2C03 sat. in order to remove unreacted tosyl-chloride. The water layers were all extracted with chloroform which was then combined with the original organic layer. The organic layer was then dried over MgS04, filtered through a glass fritted funnel followed by rotary evaporation. The resulting product was a viscous oil with a yellow tint. The product was not further purified due to fear of elimination. A yield of ~30/o was determined by GC with solvent subtraction. The following spectral properties were observed: 1 H NMR (CDCl3): 1.29 ppm (m, 9H) 1.79 ppm (s, br, lH), 1.97 ppm (s, br 1 4H), 2.45 ppm (s, 3H), 3.92 ppm (m, 2H), 4.94 ppm (m, 4H), 5.72 ppm (m, 2H), 7 35 ppm (d, 2H), 7.80 ppm (d, 2H); 13C NMR: 21.62 ppm, 25.73, 30.07, 33.79, 37.47, 72.63, 114.62, 127.85 127.91, 129.81, 138.46, 144.67. LRMS 336, calcd for C19H2sS03 336.

PAGE 53

44 9-p-toluenesolfonyl methyl-1,16-heptadecadiene (4b): Syn thesized as above, with simi lar yields. The crude product had the appearance of a viscous brown cloudy solution. Product not isolated before s ubsequent reaction. 11-p-toluenesolfony methyl-1,18-uneicosadiene (4c) Syn thesized as above. Yield was not determined by GC due to its high boiling point. The cr ude product had the appearance of a visco us brown cloudy so lution S pectral analysis: 1 H NMR (CDC l 3): 1.23 ppm (m, br, 46H), 2.05 ppm (q, SH), 2.42 ppm (s, 5.4H) 3.50 ppm (d, unreacted alcohol) 3.90 ppm (d, 0.9H), 4.95 ppm ( m, 4H), 5.40 ppm (s, br blip, el imination product), 5.80 ppm ( m, 1.6 H), 7.33 ppm ( d, 2.9H), 7.79 ppm (d, 2.66H), so me unreacted tosyl chloride or acid was also detectable in this region of the proton NMR. Step 6: 6-methyl-1,10-undecadiene (Sa). Approximately 9.6 6g ( 27 mmoles) of 4a was placed in a flame dried argon purged 250 ml three neck flask equipped with a condenser and s tir bar. 100 ml of dry THF was then added and the solution was cooled to O C. To this stirring so lution 20 ml (16.2 mmoles) of a 1 molar so lution of lithium aluminum hydride (LA H) was slowly sy ringed in (15 min.). Bubbling occurred upon first addition. Af ter the co mplete addition of the LAH the solution was brought to reflux for 5 h o urs and monitored by GC. A white precipitate was formed during the reflux. The reaction was then cooled and quenched slowly first with water followed by 3 N HCl (Caution: Addition of water and HCl sho uld be done slowly due to the highly exo thermi c reaction and the co pious production of hydrogen gas). This mixture was then extracted three times with ether, followed by the washing of the ether layer with sa t K2C0 3. ( In some cases the washing with K2C03 was omitted without cha nge.) The organic layer was then dried over MgS04 and then fi ltered through a g lass fritted funnel and rotary evaporated.

PAGE 54

45 The pr od u c t was a clear oil and isolated by short path vacu um distillation ( 28300 C at 2 mmHg ) in a pproximately 6 /o y ield. The fo llowin g Spect ra l properties were observed : 1 H NMR (CDCl 3) : 0.7 7 ppm (d, 3H), 1.09 ppm ( m lH), 1.32 ppm ( m br, l O H ), 1.99 ppm ( q, 4H), 4.91 ppm (m, 4H), 5.78 ppm ( m 2H); 13 C NMR 19.50 ppm 26.29 32.46, 33.98, 36.39, 113.99, 139.06, 139.11. Low resolution mass spectrometry (EI) = 166, ca lcd for C 12H22 166. 9-methyl-1,16-heptadecadiene (Sb). Synthesized as above Spectral properties: 1 H NMR 0.79 ppm (d, 3H), 1.29 ppm (m, br, 22H), 2.01 ppm (q, 4H), 4.94 ppm ( m 4H), 5 .7 9 ppm (m, 2H); 13C NMR 19.74 ppm 27.07, 29.01, 29.26 29.91 3 2.78 33.87, 37.12, 114.12, 139.28. Elemental analysis C1gH34 C(calc=86.32, found=86.31) H(calc=l3.68, found=l3.60 ). 11-methyl-1,20-uneicosadiene (Sc). Synt h esized as above with only three hours of reflux. Spectral properties: 1 H NMR 0.82 ppm (d, 3H), 1.23 ppm (br, m, 40H) 2.05 ppm (q, 4.5H), 4.98 ppm ( m, 4H), 5.49 ppm (m, 0.09H), 5.81 ppm (m, 2H); 13C NMR 19 55 ppm, 26.91, 28.83, 28.98, 29.33, 29.43, 29.46, 29.83 3 2.65, 33.60, 36.98, 65.39, 11 3.8 4 139.01. Low resolution mass spectrometry (EI) = 306, calcd for C22H 42 306. High resolution mass s pectr ome try (C I M+ 1 ) = 30 7 .3396, calcd for C22f43 = 3 07.3365. Alternative enolate displacement reactions for the synthesis of compounds la, and lb Ethyl-2-acetyl-6-heptenoate (ml) was prepared by an enolate displacement of a alkenyl bromide Approximate ly 100 ml of "super dry ethanol wa s distilled into a 200 ml three neck flask with a condenser. Na metal (Al drich 1 .035 g, 0.04501 moles ) was then added to the argon filled flask at room temperature. Caution: This add ition s hould be done under an inert

PAGE 55

46 atmosphere using "s uper dry" ethanol. Any contact with water in the presence of oxygen ca n likely result in a fire. This was then refluxed until all of the Na dissolved (1 hour). The solution was allowed to cool then ethyl acetoacetate (Aldrich 9.0 g, 0 .069 moles) was added slowly via an addition funnel or syringe. This solution was then allowed to reflux for 1 hour. A color change from colorless to lime green was observed. The solution was allowed to cool to room temperature and then 5-bromo-1-pentene (Aldrich 6.7 g, 0.045 moles) was added over a period of 30 min. A color change to milky yellow was observed (incipient precipitation of NaBr salts). This solution was allowed to reflux for 24 hours. The solution was transferred to a 200 ml RB flask and the ethanol evaporated off. The remaining oil was then extracted with water and ether. The ether layer was then dried with MgS04 followed by rotary evaporation. A crude yield of 94/o was attained. The product was then purified by short path vacuum distillation. The products fraction came over at the range 70-74C at a pressure of ~2 mmHg. The following spectral properties were observed: lH NMR(CDCl3) 1.28 ppm (t, 3H), 1.39 ppm (m 2H), 1.85 ppm (m, 2H), 2.08 ppm (q, 2H), 2.23 ppm (s, 3H), 3.4 1 ppm (t, lH) 4.20 ppm (q, 2H), 5.00 ppm (m, 2H), 5.78 ppm (m, lH); 13c NMR: 13.96 ppm, 14.14, 20.96, 30.04, 50.02, 59.74, 61.18, 89.65, 166.96, 200.22, 200.26. Ethyl-2-(4-pentene)-6-heptenoate (2a) and ethyl-2-acetyl-2-(4-pentene)-6heptenoa te (la) were prepared and worked up in the same manner as 6. Compound 6 was used as a reactant in place of ethyl acetoacetate. Upon addition of 6 the solution turned from clear to a dark red. After the addition of 5-bromo-1-pentene the solution again turned a milky yellow due to precipitating salts. The reaction was refluxed for 112 hours, followed by evaporation, extraction, and distillation. A GC trace showed that there were 2

PAGE 56

47 major products, 34/o (2a) and 47 /o (la). The following spectral properties were observed: (2a) ltt NMR (CDCl3): 1.25 ppm ( t 3H), 1 40 ppm (m, br 6H) 1.62 ppm (m, br 2H), 2.05 ppm ( q, 4H ), 2.31 ppm ( m, lH), 4.14 ppm (q, 2H) 4.95 ppm (m, 4H) 5.79 ppm ( m, 2H); 13c NMR: 14.5 ppm, 26 .9, 32.0, 33.8, 45.5 60.0, 114.9, 138.6, 176.0. LRMS: 224 calcd. for C14H2502 224. (la) lH NMR 1.11 ppm (m, br, 4H), 1.19 ppm (t, 3H), 1.79 ppm (m, br, 4H) 1.98 ppm (q, 4H), 2.04 ppm (s, 3H), 4.12 ppm (q, 2H) 4.91 ppm ( m, 4H), 5.70 ppm (m, 2H); 13c NMR 14.06 ppm, 23.21, 26.62, 30.79, 33.84, 61.09, 63.36, 115.1, 137.94, 172.5 205.1. LRMS 266, calcd for C16H2603 266. The above reactions using ethanol and Na metal, was also done in order to produce 2a in a one pot procedure. After the first addition of Na and alkene bromide (as stated above) the reaction was allowed to cool to room temperature followed by an addition of a second equivalent of Na metal. The solution was allowed to stir for 1/2 hour forming a dark orange color. The solution was then brought to reflux, dissolving the remaining sodium, followed by the slow addition of one equivalent of 5-bromo-1-pentene (1/2 hour). The reaction was followed by GC producing both products 2a and la. All of the product could then be converted to 2a by an addition (1 / 2 equivalent) of sodium metal. The reaction was then cooled to room temperature followed by quenching with water and 3N HCL. The reaction was extracted as listed above followed by isolation by vacuum distillation as listed before. An isolated yield of 36/o was obtained using this method. Spectral analysis was consistent with the structure and as listed before. Ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la) was also pursued by an enolate displacement of a bromine in an aprotic solvent. Sodium ethoxide was made by the addition of sodium (Aldrich, 0.116g 0.005 moles) into a 100

PAGE 57

48 ml three neck flask containing 50 ml of 's uper dry" ethanol under argon. Caution: This addition s hould be done under an inert atmosphere using super dry ethanol. Any contact with water in the presence of oxygen can result in a fire. This mixture was allowed to mix until all of the sodium dissolved. The ethanol was then distilled off under reduced pressure 6 (l.OOg 0.005 moles) and an excess of diethyl carbonate was added to the cooled solid. This was then mixed at room temperature until most of the salt dissolved. This solution was then fractionated under reduced pressure (150 mmHg at 60700C) to remove any formed ethanol. 5-Bromo-1-pentene (0.0745g 0.005 moles) was then added to the solution. A condenser was then placed in the center neck and the temperature was then raised to 115 C at s tandard pressure. The solution was then stirred at this temperature for 104 hours. The diethyl carbonate was then evaporated off and water was added la was extracted with ether and dried over MgS04. A GC trace was done, which showed two products. The products were distilled under reduced pressure ( 1 o-2 mmHg 90-100C). The spectral properties were the same as 2a, and 1 a above. The yield on this reaction was 13 /o for la and 4 /o for 2a. The yields were never optimized. Spec tral a nalysi s was the sa me as reported before. Compound la was pursued again by use of a hydride base and an aprotic solvent. A solution of 6 (3.0 g 0.015 moles) in dry 1,2dimethoxyethane ( DME ) ( 20 ml ) was added dropwise under nitrogen to a s tirred mixture of potassium hydride ( 1 .7 2 g of 35o/o suspension 0.015 moles Aldrich) in DME ( 100 ml ) When evolution of hydrogen stopped a solution of 5-bromo-1-pentene ( 2.31 g 0.015 moles Aldrich) in DME (20 ml) was added dropwise. The solution was heated under reflux for 24 hours. The DME was then evaporated and la was again worked up as before. Compound la was

PAGE 58

49 recovered in a 50 /o yield before optimizing. The spectral properties were the same as reported before. Ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la) and ethyl-2-(4-pentene)-6heptenoate (2a) using KH (one pot two step method): In flame dried argon purged 100 ml 3 neck round bottom flask with condenser 3.5 g (0.03M) of a 35 o dispersion of KH in mineral oil was washed three times with 20 ml of dry pentane. The remaining white salt was pumped to dryness under vacuum. Approximately 50 ml of dry dimethoxy ethane (DME) was then cannula transferred to this flask followed by stirring. No evolution of hydrogen was observed. To this stirring solution 3.9g (0.03M) of ethylacetoacetate was slowly added (20 min.) resulting in the evolution of a gas (H2) and a change from colorless to a clear lime green color. The solution was then heated to 70 C followed by the slow addition (15 min.) of 4.47g (0.03M) 5-bromo-1-pentene. Upon this addition the mixture turned yellow in color with a suspension of white salts. The solution was refluxed for 20 hours followed by cooling to 50 C. A second equivalent of washed KH was added under argon. Vigorous bubbling was observed with a darkening of the yellow color to an orange The second addition of 5-bromo-1-pentene was then added followed by refluxing for 72 hrs. The reaction was then cooled followed by evaporation of the solvent. The remaining orange liquid was then extracted 3 time with water and ether. The organic layer was dried over MgS04 and rotary evaporated. A 60 o yield was obtained (41 /o la, 18/o 2a by GC). A high vacuum distillation and a low temperature recrystallization were attempted with poor results. A column was run on 200 mg of the product using 80 /o pentane and 20 /o ethyl acetate through silica gel. A purity of 94.4 was attained for la as determined by GC. lH and 1 3 C NMR are

PAGE 59

50 consistent with structure la. This procedure was also done in two steps with complete isolation of 6 and la after each step. Yields were lower due to loss during isolation. Ethyl-2-(4-pentene)-6-heptenoate (2a) using NaH (one pot two step method): In a 500 ml 3 neck flask equipped with a condenser and stir bar, 10.9g (0.085 M) of ethylacetoacetate and 250 ml of DMF were stirred. To this solution 8.16g (4 equivalents) of powdered sodium hydride was added. Bubbling (H2 gas) occurred immediately upon this addition. The solution was stirred at room temperature for 0.5 hours followed by the addition of 25g (0.17 M) of 5-bromo-1-pentene. The temperature of the solution was then raised to 120 C. The reaction was followed by GC, using micro-extractions, showing the production of both 2a and la. The reaction was continued for 20 hrs. At this time GC showed almost full conversion to the ester 2a. Full conversion was often achieved by the addition of dry ethanol to the basic solution. The solution was then cooled to room temperature followed by slowly quenching with water and 3N HCL (Caution: Addition of water and HCl should be done slowly due to the highly exothermic reaction and the copious production of hydrogen gas.) The solution was then extracted 3 times with 250 ml portions of ether. The organic phase was then dried over MgS04 followed by filtering and rotary evaporation. GC showed that all the product had been converted to 4. The product was isolated by a short path vacuum distillation (74-85 C at lmmHg) with approximately a 51 o/o yield. Spectral analysis was consistent with the structure and as listed before. Ethy 1-2-acetyl-2-( 4-pen tene)-6-heptenoate (la) and ethyl-2-( 4-pentene )-6heptenoate (2a) using KOH, NaOH or 1,8-diazabicyclo[S.4.0]undec-7-ene (DBU)

PAGE 60

51 (one pot two step method); general procedure. In a 500 ml three neck flask 10.9g (0.085 moles) of ethyl acetoacetate and 250 ml of DMF (anhydrous) were stirred. Approximately 23.8g (5 eq) of potassium hydroxide pellets were added followed by the addition of 25g (0.17 moles) of 5-bromo-1-pentene. The solution was then stirred and heated to 120 C upon which all of the KOH pellets dissolved. A yellow brown color was observed. After three hours GC showed the productions of both products with product (la) in majority. After 24 hours the pH of the solution was taken resulting in a pH of ~ 10. No change in the product ratio was detected by GC. More KOH was then added until the pH was over 12. The reaction was then allowed to stir for 12 more hours. At this time GC showed a mixture of products with the majority product being 2a when the bases KOH of NaOH were used. When DBU was used alone, only product la was formed. The reaction was then cooled to room temperature and quenched with 3M HCL and water. This solution was then extracted three times with 100 ml portions of ether. The organic layer was then dried with MgS04 followed by rotary evaporation. The neat product was dark brown orange in color. A short path vacuum distillation was then done with the ester product distilling between 90-110 C at 6 mmHg. An isolated yield of 43o/o was obtained. The remaining product was not identified. Spectral analysis was consistent with the structure and as listed before. Ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la) using NaNH2. In a dry 100 ml 3 neck flask with stir bar and condenser, 2g (0.015 moles) of ethyl acetoacetate was mixed with 50 ml of dry dimethoxy ethane (DME). To this stirring argon purged solution, 1.2 g (0.015 moles) of NaNH2 in a 50/o weight solution in toluene, was pipetted in. A lime green color appeared

PAGE 61

52 immediately upon this addition. The solution was then heated to 60 C followed by the addition of 0.015 moles of the alkene-bromide. The production of white salts upon this addition was immediately observed. The solution was then allowed to reflux for 8 hrs. and followed by GC. The reaction mixture was then cooled to room temperature followed by a second addition of base and alkene-brornide in the same manner as before. A yellow color and a significant production of salts were observed. The product mixture was again followed by GC and the reaction was stopped after 12 hrs. of reflux. The reaction was then quenched under an inert atmosphere using 3 M HCl. Diethyl ether was then added to this solution and 3 extractions were done. The organic layer was then dried over anhydrous magnesium sulfate, followed by filtration and evaporation. A EI-GC / MS was done showing that the crude solution was a mixture of previously isolated products and a tri alkylated product. The product mixtures are illustrated in Figure 3.14. No further isolation was done on the mixture. Alternative Methods for the Synthesis of Compounds Sa-Sc Attempted decarbonylation of ethyl-2-acetyl-2-(4-pentene)-6-heptenoate to form 6-acetyl-1,10-undecadiene (7). The decarbonylation was attempted by the addition of an acid, base or a salt with the appropriate solvent (DMF, DMSO, H20). These reactions were brought to reflux and monitored by micro-extractions followed by GC. After the reactions were cooled to room temperature, they were extracted with pentane and water. The pentane was then rotary evaporated followed by spectral analysis if appropriate. No reaction or decomposition was observed for these reactions. When base (KOH) was used the expected increase in (2a) was observed by GC.

PAGE 62

53 6-acetyl-1,10-undecadiene (7) In a 25 ml round bottom flask with 10 ml of NMP and 0.5g of water, 2g of LiCl and lg of the keto-ester (la) was added. The solution was heated to reflux noting a co lor change from colorless to yellow by 150 C. All salts were dissolved by 160 C. Reflux was maintained for 6 h forming a brown solution. The reaction was monitored by GC using micro-extractions and showed the disappearance of the starting material and the appearance of a new peak with a retention time of 10.78 minutes. The starting material has a retention time of 12.77 minutes. The reaction was allowed to reflux for a total of 9 hours The reaction was cooled to room temperature and extracted with 200 ml of water and 50 ml of pentane. The organic layer was then dried over anhydrous magnesium sulfate followed by filtering and rotary evaporation. A crude yield of 90/o was obtained. Low resolution mass specrometry c alculated M + 1 = 195 actual M + 1 = 195 lH NMR:1.19 ppm (m, br, 8H), 1.51 ppm (m, 2H), 1.96 ppm (q, 4H), 2 .0 4 ppm (s, 3H), 4.89 ppm ( m, 4H), 5.69 ppm (m, 2H). Wittig Method (1-Pentene)triphenylphosphonium bromide (8) was synthesized as a Wittig sal t . 5-Bromo-1-pentene ( Aldrich, 15 ml 0.1 moles) was added to a 250 ml 3-neck flask containing 145 ml DMF a nd triphenylphosphine (Aldrich, 28.85 g, 11 moles ) A Teflon coated s tir bar and a condenser were then added. The sol ution was refluxed at 170 C fo r 20.5 hours. The salt was then precipitated out with diethyl ether The sa lt was vacuum filtered and the filtrate was treated again with ether for further precipitation A yellow white solid was recovered The sa lt s were then washed with hot THF then mixed in

PAGE 63

54 this solvent until they dissolved partially The solution was then allowed to cool to room temperature and then c ooled in a freezer The salts were then colle c ted cold by vacuum filtration to give 1 as white c r y stals in a 9 3 /o yield. Compound 1 was dried in a schlenk tube at room temperature at 10-2 mmHg for 12 hours. ltt NMR (CDCl3): 1.68 ppm (m, 2H) 2.23 ppm (q, 2H), 3.80 ppm (m, 2H), 5.02 ppm (t 2H), 5.65 ppm (m, lH), 7.80 ppm ( m 15H). Elemental calcd: C, 67 00; H 6 11; P, 7 51 ; Br 19.38 Obsd: C, 67.07; H 6.05. Tertiary Alcohol Synthesis and Reduction 6-methyl-1,10-undecadiene-6-ol (9), was synthesized by the Grignard reaction with the ethyl acetate. 5-Bromo-1-pentene was dried over freshly ground CaH2 for 3 hrs, then vacuum transferred to a schlenk flask with 4 A molecular sieves In a flame dried 100 ml three neck round bottom flask 1.63g (67 mmoles) of freshly ground magnesium turnings were placed followed by the addition of 50 ml of dry ether. A crystal of iodine and a small amount of the bromide was added in order to initiate the reaction The balance of the bromide was then added at su c h a rate as to maintain r e flux. After complete addition, the mixture was stirred and refluxed for 0.5 hours. After cooling to room temperature 33.5 mmoles of dry ethyl acetate was added to produce (9). Refluxing occurred upon addition of the electrophiles and was maintained by heating for 2 hours. The reaction was then cooled and quenched with 3 molar HCl. The reaction mixture was extracted with ether and dried over anhydrous magnesium sulfate followed by filtering and evaporation yielding 5.18g ( 84 /o) of a clear liquid 9. The monomer was distilled under reduced pressure (70-80 C a t lmmHg) u s ing a short path distillation apparatus. The following spectral properties w e re o bserved: 1 H NMR ( 9 ) 1.14 ppm ( s, 3H) 1.43

PAGE 64

55 ppm ( m, 9H) 2.05 ppm (m, 4H) 4.98 ppm ( m 4H ), 5.80 ppm ( m, 2H) ; 1 3 C NMR: 23.12 ppm 26.80 34.07 41.21 72.52 114 48, 138.61; Elemental calcd: C, 79.06; H, 12.16; 0 8.78. Obsd: C, 79 16; H 12.15. 5-methyl-1,9-decadiene-5-ol (10) was prepared by using a Grignard reagent with a ketone. 1 pentene-5-magnesium bromide (0 1 moles) was prepared in the usual manner as a Grignard reagent. The reagent was prepared in a 200 ml three neck flask and condenser with 20 ml of dry ether. 5-hexene-2-one ( Aldrich 9.8 g, 0 .1 moles) and 20 ml of ether were then slowly added via an addition funnel. A 15 min. reflux and a color change f rom brown to green was observed. The solution was allowed to stir for 7 hours and then transferred to a 250 ml Erlenmeyer. Approximately 10 ml of lN H2S04 was added and 10 was extracted using water and ether. The ether layer was dried over MgS04, evaporated and then vacuum distilled using a short path vacuum apparatus. The alcohol fraction came over at 90 C at 10l mmHg. 10 was recovered in a 78 /o yield and a 99 .92 /o purity determined by GC. The following spectral properties were observed: lH NMR (CDCl3): 1.17 ppm (s, 3H) 1.45 ppm ( d br 4H ), 1.55 ppm ( p 2H) 2.10 ppm (m 4H ), 5.00 ppm (m, 4H) 5.82 ppm (m, 2H) ; 1 3 c NMR: ppm 23.20, 26.75, 26 81, 28.29 3 4.13 40.80, 41.37, 72.55 114.30, 114 6 138.6, 138 9. Elemental calcd: C, 79.51; H, 11.98; 0, 9.51. Obsd: C, 76.16; H 11.36. Attempted synthesis of 5-methyl-1,9-decadiene (11) or 6-methyl-1,10undecadiene (Sa) (Carey method). In a dry 250 ml 3 neck flask 0.786 ( 4.6 mmoles ) of 2 -( 3-butene)-6 heptene-2-ol ( 10 ) and 1.34g ( 5.1 mmoles ) of triphenyl silane ( weighed out in nitrogen purged glove bag) were placed with 15 ml of meth y lene chloride To this stirring s olution 0.80g ( 7.0 mmoles ) of

PAGE 65

56 trifloroacetic acid was dripped in. No physical change was observed upon addition. The reaction was followed by GC using micro-extractions. After 24 hours no significant reduction in the starting material was observed so an additional 1.5 equivalents of trifloroacetic acid was added. This solution was then allowed to mix for an additional 12 hours. The reaction was then quenched with an excess of saturated potassium carbonate followed by extraction with 3 equivalent amounts of diethyl ether. The organic phase was then dried over anhydrous magnesium sulfate followed by filtration then rotary eva poration. The c rude reaction mixture was then run through an alumna column using pentane as the mobile phase. A GC/MS-CI/methane was done showing that one of the minor products (5/o) had a base peak of 165. This indicates that there is some eliminated product present. lH NMR was relatively messy, but the olefin region did show some internal olefin resonance's at 5.6 ppm. The target product could not be identified by GC/MS or NMR. The reaction was repeated again using the same scale with shorter reacting times, but resulting in the same results. Attempted synthesis of 5-methyl-1,9-decadiene or 6-methyl-1,10undecadiene (Sa) (Ireland method). Approximately lg (5.5 mmoles) of alcohol 9 was placed in a 100 ml three neck flask containing 40 ml of a 4 : 1 mixture of dry THF and tetramethylethylene diamine (TMEDA) respectively. To this stirring so lution at room temperature 2 5 ml (6 mmoles) of n-butyl lithium in hexanes was s lowly added by sy ringe After mixing for 20 min 0.9 ml (5 5 mmoles) of Diethyl chlorophosphate was dripped in via syringe. After the first 0.5 hour white sa lts were noticed. The solution was allowed to mix at room temperature for 3 hours followed by quenching with excess 1 M HCl in anhydrous ether. This was done in order to remove the TMEDA and

PAGE 66

57 LiCl salts. After the addition of the acid, large amounts of salt precipitated out which were then filtered by using a fine frit funnel. The pH was taken followed by additional acid washes until the solution was acidic. The ether was then evaporated giving 850 mg of the crude reaction mixture. A GC/MS was done on this crude mixture. The crude mixture was then dissolved in 20 ml of THF and 800 mg of t-butanol. This solution was then dripped into a mixing solution of 20 ml EtNH2 with 200 mg of Li metal cooled to O C. Some bubbling was noted upon this addition. The reaction was stirred for 3 hours without dissolving the lithium metal. The reaction was quenched with water and the excess lithium metal was filtered out. The reaction was extracted with ether, dried, filtered and rotary evaporated. A GC and a GC/MS was done showing no conclusive results of product formation. ADMET Polymerizations of Monomers Sa and Sb General Metathesis Conditions All glassware was thoroughly cleaned and flame dried under vacuum before use. The monomers were vacuum fractionally distilled or column chromatographed to a purity of > 99/o as determined by GC and elemental analysis. The monomers were then dried over calcium hydride for a minimum of 8 hours and then vacuum transferred onto a sodium-potassium alloy to insure dryness. They were stirred on this alloy for a minimum of 4 hours. During this period the clear solutions would turn blue in color. Monomers were degassed by subjecting them to several freeze pump thaw cycles under high vacuum (
PAGE 67

58 then vacuum transferred into a clean dry flask fitted with a Teflon vacuum valve. All metathesis reactions were initiated in the bulk, under argon glove box conditions. Monomers were introduced into a 25 or 50 ml round bottomed flask equipped with a high vacuum Teflon TM valve and magnetic stir bar. On occasion, treatment with catalyst for purification was necessary. All monomers (0.4-2.0 g) were opened in the dry box followed by the addition of catalyst (2-20 mg). After the addition of catalyst, the reactions were exposed to intermittent vacuum at room temperature, until a noticeable increase in viscosity was apparent. When the stirring of the reaction became hindered it was then placed under high vacuum in order to remove the continuous generation of ethylene. All the reactions were started at room temperature then ramped to 60 C over a period of 2 to 3 days. When no further evolution of ethylene was apparent (bubbling stopped) the solutions were then cooled to room temperature and quenched by exposure to air. Polymers were isolated in high yield (~90/o), followed by NMR, GPC, and elemental analysis on the crude reaction mixture. The polymers were then precipitated from toluene into cold methanol producing a clear, colorless to white semisolid If the polymer retained any color (due to catalyst) it could be removed by dissolving the sample into toluene followed by passing it through a short alumina column. Polymerization of 6-methyl-1,10-undecadiene (PSa). The addition of approximately 5-10 mg of Schrock's Mo catalysts resulted in the immediate evolution of ethylene. The reaction was then placed onto a high vacuum line and exposed to intermittent vacuum until there was a noticeable increase in viscosity (30 min.). The reaction was then exposed to high vacuum (
PAGE 68

59 mmHg) and stirred. As the viscosity of the system increased the solution was slowly heated in order to facilitate stirring, in the manner stated above. After precipitation the polymer was a clear white nonflowable tacky solid. The polymer was dried by exposure to schlenk line vacuum at approximately 60C. The following spectral analysis were observed: 1 H NMR 0.78 ppm ( d, 3H), 1.03 ppm (m, lH), 1.28 ppm (m, 8H), 1.89 ppm (m, br, 4H), 5.29 ppm (m, 2H); 13C NMR 19.68 ppm, 27.13, 27.54, 32.58, 32.95, 36.54, 36.70, 129.92, 130.38. Elemental analysis for repeat C10H1s C(calc=86.88 found=85.59) H(calc=13.12 found=13.01). Molecular weight deter1nination for all the batches is discussed in the results and discussion (Table 4.1). Polymerization of 9-methyl-1,16-heptadecadiene (PSb). The same general procedure was followed as for PSa. In this batch of monomer, some impurities were present so the reaction was restarted by further addition of Schrock's Mo catalyst in order to purify. The polymer was precipitated and dried as stated above. A clear to white tacky solid was obtained with the following spectral analysis: lH NMR 0.81 ppm (d, 3H), 1.06 ppm (s, br, lH), 1.29 ppm (s, br, lOH), 1.98 ppm (m, br, 4H), 4.94 ppm (m, end group) 5.37 ppm (m, br 2H), 5.80 ppm (m, br, =CH2); 1 3C NMR 19.53 ppm 26.91, 27.10, 29.04, 29.19, 29.54, 29.64, 29.72, 32.39, 32.67, 37.00, 129.73, 130.21. Elemental analysis for repeat C16H30 C(calc=86.4 found=84.9) H(calc=13.6 found=13.4). Molecular weight determination for all the batches is discussed in the results and discussion (Table 4.2).

PAGE 69

6 0 Hydrogenation of Polymers P5a and P5b Hydrogenation of poly [6-methyl-1,10-undecadiene] (HP5a). The hydrogenation was performed in a flame dried, 20 ml round-bottom three neck flask equipped with a reflux condenser Teflon TM stir bar, and supplied with a positive Argon pressure. To this flask 50 mg of polymer PSa, 5 ml of dry reagent grade a-xylene, 135 mg (2.5 eq) of TSH, and 103 mg (2.5 eq) of TP A were added. The mixing solution was then heated to reflux. All of the TSH did not dissolve until the solution was close to reflux. Upon heating some gas evolution could be observed (bubbling of N2) before reflux was achieved. The solution was allowed to reflux for 6 hours followed by cooling to room temperature. Some white precipitation was noted at this time. A second addition of 68 mg (1.25 eq) of TSH, and 51 mg (1.25) of TP A was administered, followed by refluxing for 3 hours. The polymer was then precipitated in col d methanol. The methanol was then decanted off followed by drying the polymer in a schlenk flask under reduced pressure at 60 C for 18 hours. Approximately 70/o of the polymer was recovered. A semi-clear tacky so lid was recovered with the following spectral analysis: lH NMR 0.83 ppm (3 H d), 1.10 ppm (2H, br), 1.27 ppm ( 14H s); 1 3C NMR 18.8 ppm, 26.2, 28.8, 29.5, 30 .0 36.3. Molecular weight determination is displayed in Figure 4.14. Hydrogenation of poly [9-methyl-1,16-heptadecadiene] (HPSb). The hydrogenation was performed in a flame dried, 100 ml round-bottom three neck flask equipped with a reflux condenser, Teflon stir bar, a nd s upplied with a positive Argon pressure. To this flask 500 mg of polymer PSb 30 ml of dry reagent g rade a-x ylene 1.05g (2.5 eq) of TSH, and 0.80g (2.5 eq) of TP A were added. The mixing solution was then heated to reflux. All of the TSH

PAGE 70

61 did not dissolve until the solution was close to reflux. Upon heating s ome gas evolution co uld be observed (bubbling of N2) before reflux was achieved ~ 110 C. The sol ution was allowed to reflux fo r 3 hours followed by cooling to room temperature. Some white precipitation was noted at this time. A second addition of 1.05g (2.5 eq) of TSH, an d 0.8g (2.5 eq) of TPA was administered, followed by refluxing for 3 more hours. The polymer was then precipitated twice from cold methanol The methanol was then decanted off followed by drying the polymer in a schlenk flask under reduced pressure at 80 C for 24 hours. Approximately 70 /o of the polymer was recovered. A opaque yellow brown hard waxy solid was recovered with the following spectral analysis: lH NMR 0.83 ppm (3H, d), 1.09 ppm shoulder, 1.27 ppm (28H s); 13 C NMR 18.8 ppm, 26.1, 28.7, 29.1, 31.9, 36.2. Molecular weight determination is discussed in Figure 4.15. Alcohol Monomer Synthesis 6-methyl-1,10-undecadiene-6-ol (A3), and 1,10-undecadiene-6-ol (A2) were synthesized by the Grignard reaction with the appropriate ester. 5Bromo-1-pentene was dried over freshly ground CaH2 for 3 hrs, then vacuum transferred to a schlenk flask with 4 A molecular sieves. In a flame dried 100 ml three neck round bottom flask 1.63g (67 mmoles) of freshly ground magnesium turnings were placed followed by the addition of 50 ml of dry ether. A crystal of iodine, and a small amount of the bromide was added in order to initiate the reaction. The balance of the bromide was then added at such a rate as to maintain reflux. After co mplete addition, the mixture was stirred and refluxed for 0.5 hours. After cooling to room temperature 33 .5

PAGE 71

62 mmoles of either dry ethyl formate was slowly ad ded to produce (A2), or alternatively dry ethyl acetate was added to produce ( A3). Refluxing occurred upon addition of the electrophiles and was maintained by heating for 2 hours. The reaction was then cooled and quenched with 3 molar HCl. The reaction mixture was extracted with ether and dried over anhydrous magnesium sulfate, followed by filtering and evaporation yielding 5.18g (84/o) of a clear liquid A2. Similar yields were obtained for A3. Both monomers were distilled under reduced pressure (70-80 C at lmmHg) using a short path distillation apparatus. The secondary alcohol diene ( A2) was further purified by column chromatography using silica gel with a solvent gradient elution using 1-5/o ethyl acetate and pentane. The column was monitored by TLC using a 5o/o ethyl acetate 95/o pentane mobile phase on silica plates. The following spectral properties were observed: lH NMR (A2) 1.35 ppm (m, 9H), 1.99 ppm (m, 4H), 3 .54 ppm (s, lH), 4.89 ppm (m, 4H), 5.73 ppm (m, 2H); 13C NMR: 24.79, 33.56, 36.79, 71.53, 114.41, 138.54; Elemental Calcd: C, 78.51; H 11.98; 0, 9.51. Obsd: C, 77.61; H, 11.80. 1 H NMR (A3) 1.14 ppm (s, 3H) 1.43 ppm (m, 9H), 2.05 ppm (m, 4H), 4.98 ppm (m, 4H), 5.80 ppm (m, 2H); 13C NMR: 23.12 ppm, 26.80, 34.07, 41.21, 72.52, 114.48, 138.61; Elemental calcd: C, 79.06; H, 12.16; 0, 8.78. Obsd: C, 79.16; H, 12.15. Synthesis of 6-(4-pentene)-1-heptene-7-ol (Al) was synthesized in a three step reaction sequence as shown below. Step 1: Synthesis of ethyl-2-acetyl-2-(4-pentene)-6-heptenoate (la). 10.9g (84 mmoles) of ethyl acetoacetate (Aldrich) and 200 ml of dry DME were placed in an argon purged 500 ml three neck flask equipped with a stir bar and condenser. To the stirring solution 42 ml of a 2 molar solution of potassium

PAGE 72

6 3 te rt-butoxide in DME was added. Upon addition the s olution turned a lime green color and was allowed to stir f or 0 5 hours a t room temperature. Approximately 12.Sg (84 mmoles) of 5-bromo-1-pentene was then slowly added and the temperature raised to reflux upon which white salts began to precipitate. The first addition was complete in 18 h as shown by GC. The reaction was then cooled to room temperature and the second addition of the alkenyl bromide was repeated as above and completed in 24 h. The reaction was then quenched with 3M HCl and extracted with ether. The ether was dried over MgS04 and then evaporated yielding ~80 /o of la. The following spectral properties were observed: ltt NMR 1.11 ppm ( m, br, 4H), 1.19 ppm (t, 3H), 1.79 ppm (m, br, 4H), 1.98 ppm (q, 4H), 2.04 ppm ( s, 3H), 4.12 ppm (q, 2H) 4.91 ppm (m, 4H), 5.70 ppm (m, 2H); 13c NMR 14.06 ppm, 23.21, 26.62, 30.79, 33.84, 61.09, 63.36, 115.1, 137.94, 172.5, 205 1. The Low Resolution Mass Spectrum (LRMS) also confirms structure with a parent ion at 266, calcd for C16H2603 266. Step 2: Retroclaisen condensation: Synthesis of ethyl-2-(4-pentene)-6heptenoate (2a). In an argon purged dry 250 ml 3 neck round bottom flask equipped with a condenser and a stir bar, 22.34g (84 mmoles) of la and 100 ml of dry ethanol were added. To this solution 35ml of a 21 /o solution of sodium ethoxide in ethanol was added. The solution was allowed to reflux for 3.5 hours and turned a dark yellow in color. After cooling to room temperature the solution was quenched with water and 3M HCl then extracted with pentane. The organic layer was then dried over MgS04 and evaporated yielding 90 /o of ester 2a. The following s pectral properties were observed: 1 H NMR: 1.25 ppm (t, 3H), 1.40 ppm (m, br, 6H), 1.62 ppm (m, br, 2H), 2.05 ppm ( q, 4H) 2.31 ppm (m, lH), 4.14 ppm (q, 2H), 4 95 ppm (m, 4H), 5.79 ppm (rn,

PAGE 73

6 4 2H); 13c NMR: 14.5 ppm 26. 9 3 2.0 3 3 8 45. 5, 60 .0, 114.9 138.6 176 0. The Low Resolution Ma ss Spectrum ( LRMS ) also c onfirms structure w ith a parent ion at 224: calcd. fo r C14H2502 = 2 24. Step 3: Reduction of ester: Synthesis of 6-(4-pentene)-l-heptene -7-ol (3a) (Al). Approximately 9. 3 01g of 2a and 125 ml of dry THF were placed in a flamed dried 250 ml three neck round bottom f lask with s tir bar and condenser. This solution was kept under argon and cooled to O C To this stirring soluti o n 25 ml of a lM solution of LiAlH4 in THF was added over a period of 5 min. Some bubbling was observed during this addition. The reaction was allowed to warm to room temperature and stirred for 2 hours. The reaction wa s then slowly quenched with water followed by 3 M HCl The solution was e x tracted with ether, dried over MgS04 and evaporated. A clear oil was recovered and vacuum distilled at 69-72 C at 1mm Hg. The following spectral properties were observed: lH NMR (CDCl3): 1.18 ppm (m br 9H) 1.79 ppm (s, lH) 1.98 ppm ( q 4H), 3.49 ppm (d, 2H), 4.88 ppm (m 4H ) 5.78 ppm (m 2H ); 13c NMR: 26.08 ppm, 3 0.29 33.98, 34 01 40.22 65 .3 6 114 28 138.74 IR ( CHCl 3 1 c m1 ) : 3 3 8 3.8 ( br ) 3 078.1, 3 013.4 29 3 1.0 2860. 9, 1 6 40.4 1460.3, 1217 1 1 030 5 9 96.1 9 13.1 759.5. Elemental a naly s is C 12H 23 0 C(calc=79.06, f ound=78.99), H( c alc=12.16, found=l2.16). ADMET Polymerizations of Alcohol Monomers Al, A2, A3 General Metathesis Conditions All gla ss ware was thoroughly cleaned and flame dried under vacuum before use. The monomer s wer e vacuum fractionally distilled ( from CaH 2 if

PAGE 74

65 neede d ) prior to polymerization. The monomer s, if stored, were placed ov er 4A m o lec ul a r sieves in orde r t o preserve dryness. The purity of the mon ome rs was >98/o as de termined by GC. Monomers were degassed by s ubje cting them to several freeze pump thaw cyc les under high vacuum ( < l04 torr ) The dry, degassed monomer s were then vac uum transferred in to a clean dry reaction flask fitted with a Teflon vacuum valve and co ntaining a magn etic stir bar. A ll metathesis reactions were initiated in the bulk in an argon gl ove box cond itions. The addition of a few drops of dry CDCl3 was occ asionally done in o rder to help initiate the reaction Monomers (0. 5-1.5 g), while in the glove box, were introduced into a 25 or 50 ml round-bottomed flask equipped with a high vac uum Teflon TM valve and magnetic stir bar. After the addition of catalyst (ca talyst to monomer ratios discussed below), the reactions were first expose d to intermittent vacuum until the viscosity increased, followed by exp os ure to high vacuum in order to remove the continuous generation of ethylen e All the reactions were started at room temperature and maintained there u ntil the increase in viscosity prevented stirring. At this time the reaction temperature was slowly ramped to 70 C over a period of 2 to 3 days. The solu tion s were then co oled to room temperature and quenched by the addition of excess ethyl vinyl ether o r b y ex posure to air. Reactions were run on a 0.5 1.5g scale, with a monomer to ca taly s t ratio of 500:1, 300: 1 or 200:1 as noted Polymerization of [6-(4-pentene)-l-heptene-7-ol] (PAl). The mon o mer Al was sy nthesi zed and dried as previously described For a monom e r to catalyst ratio of 200:1, 23 milligram s of catalyst, RuCl2(=CHPh)(PCy 3)2 (C3) was added to 1 g of the monomer The reaction was performed under t ypica l

PAGE 75

66 metathesis conditions, until the contents could no longer be stirred or ethylene evolution had stopped. The reaction was quenched by exposure to air. The polymers were not precipitated before undergoing characterization. The following spectral properties were observed: 1 H NMR (CDCl3): 1 09 ppm (br, 8H); 1.35 (br, lH); 1.71 (br, 4.4H); 3.19 (br, 1.9H); 4.67 (br, 0.04H end group); 5.15 (br, 2H); 5.60 (br, 0.003H end group). 13 C NMR: Anal. Calcd for C10H1sO: C, 77.87; H, 11.76. Found: C, 75.55; H, 11.41. For GPC analysis see table 5.1. IR (neat, cm1 ) 3341, 3005, 2926, 2856, 1726, 1457, 1440, 1036, 967, 511. Polymerization of [1,10-undecadiene-6-ol] (PA2). The monomer A2 was synthesized and dried as previously described. For a monomer to catalyst ratio of 200:1, 24 mg of catalyst, RuCl2( =CHPh)(PCy3)2, was added to lg of the monomer. The procedure was followed as in (PAl). The following spectral properties were observed: lH NMR: 1.45 ppm (br, d, 9H); 2.00 (br, 4H); 3.55 (br, .7H); 4.98 (br, 0.29H end group); 5.40 (br, 2H); 5.80 (br, m, 0.04H end group). 13 C NMR: 24.8-33.4 (multiple signals), 36.4, 71.0, 129.5-130.5 (multiple signals). Anal. Calcd for C9H160: C, 77.09; H, 11.50. Found: C, 76.48; H, 11.33. For GPC analysis see table 5.1. Polymerization of [6-methyl-1,10-undecadiene-6-ol] (PA3) The monomer A3 was synthesized and dried as previously described. For a monomer to catalyst ratio of 500: 15 mg of catalyst C3 was added to 500 mg of the monomer. The procedure was followed as in PAl. The following spectral properties were observed: lH NMR: 1.12 ppm (s, 3H); 1.40 (br, 8H); 1.99 (br, 4H); 4.94 (br, m, 0.25H end group); 5.39 (br, 2H); 5.79 (br, m, 0.09H end group). 13 C NMR 23.90, 26.94, 27.67, 33.01, 41.37, 72.72, 129.92, 130.42. Anal. Calcd for

PAGE 76

67 C10H1sO: C, 77.87; H 11 76. Found: C, 76.61; H, 11.50. For GPC analysis see table 5.1. Hydrogenation of Poly [1,10-Undecadiene-6-ol] (HP A2) The hydrogenation was performed in a flame dried 50 ml round bottom three-neck flask equipped with a reflux condenser, Teflon stir bar and supplied with a positive Argon pressure. To this flask 266 mg of polymer P A2, 25 ml of dry reagent grade toluene 907 mg of TSH and 0.932 ml of TP A were added. The mixing solution was then heated to reflux. All of the TSH did not dissolve until the solution was close to reflux. Upon heating some gas evolution could be observed (bubbling of N2) before reflux was achieved. The solution was allowed to reflux for 6 hours followed by cooling to room temperature. Some white precipitation was noted at this time (insoluble hydrogenated polymer). A second addition of 907 mg of TSH, and 0.932 ml of TP A was administered, followed be refluxing for 3 hours. Once the solution cooled to room temperature the product could be observed as a white precipitant. The polymer was then precipitated in co ld methanol. Approximately 60/o of the polymer was recovered The following spectral properties were o bserved : 1 H NMR (Toluene d8 at 100 : 1.55 ppm and 1.62 ppm ( br, s, 18H); 3. 72 (s, lH); 5.18 (br, m residual vinilic CH2); 5.62 (br, m, residual internal olefin). 13 C NMR: 26.60, 30.50, 30.66, 38 67, 72.41. G PC analysis was not performed due to the insoluble nature of this polymer.

PAGE 77

68 Preliminary Alcohol Dimerization Experiments Dimerization of 4-pentene-1-ol (12). Mo n oe ne 12 was purchased from Aldrich in lg bottles and used on that scale with no further purification. The monoene was placed in a 50 ml reaction flask followed by degassing under high vac uum using three freeze pump thaw cycles. The monoene was then placed into the dry box along with the addition of a magnetic stir bar. To this stirring liquid, 23 mg of Grubbs ruthenium catalyst C3 (a monoene to catalyst ratio of 400 : 1) was added. Upon this addition the reaction turned purple brown in appearance follow by a small amount of bubbling. The reaction was then removed from the dry box placed on the high vacuum line, followed by intermittent exposure to low vacuum (
PAGE 78

6 9 addition. Some bubbling was noticed while exposure to the intermittent vacuum. The reaction was allowed to stir for 18 hours, followed by NMR analysis. The 1 H NMR represents a mixture of isomers: 0.94 ppm (m), 1.10 ppm (m ) 1.61 ppm (br m ), 2.32 ppm (qrt. ), 2.47 ppm ( qrt.), 3.72 ppm (br, OH) 4.12 ppm (shp, m), 5.20 ppm ( m) 5.82 ppm (s), 5.97 ppm (m), 9.78 ppm ( shp, s ) Dimerization of 5-hexene-2-ol (14). The monoene 14 was provided by Shane Wolfe fully characterized and at a purity of >99 / o by GC. This sample was degassed and placed in a dry box atmosphere as before. To this stirring liquid 21 mg of catalyst C3 ( M : C = 200 : 1 ) was added followed by immediate bubbling and an orange purple color The reaction was then exposed to intermittent vacuum as before with a noticeable increase in the evolution of gas. Samples were taken for NMR after 1 hour and after 24 hours showing a 52 and 97/o conversion, respectively. 1 H NMR peaks of the mixture after 24 hours are as follows: 1 18 ppm (d), 1.49 ppm (m) 1.61 ppm (d of d), 2.10 ppm (br, m), 3.78 ppm (q), 4.98 ppm (m) (unreacted =CH2), 5.40 ppm (m) (cis and trans), 5.8 ppm ( m) (terminal =CH). Attempted dimerization or polymerization of 1, 5-hexadiene-2, 3-diol (15). Monoene 15 (Aldrich) was purchased in lg bottles and used on that scale with no further purification. One gram of 15 was placed into a reaction flask with a stir bar and degassed as before. Under dry box conditions 42 mg of catalyst C3 (M : C = 200 : 1) was added to this stirring solution. Upon this addition the catalyst and solution maintained there original color (purple). No bubbling was observed upon this addition. The reaction was tran sferred to the vacuum line as before and allowed to stir for 12 hours No outward sign of condensation was o bserved (increase in viscosity and bubbling). After

PAGE 79

70 12 hours a n aliqu ot was removed for NMR a nalysis. No reaction was observed by NMR a t this time The reaction was returned to dry box c onditions upon whi c h a few drops ( ~0 .3 ml) of CDCl 3 was added. This was done in order to so lvate the c atalyst. Upon this addition the s olution i mmediately turned brown in color. This mixture was allowed to stir f o r an additional 12 hours. 1 H NMR analysis again showed no detectable reaction ( no detection of internal olefin ) The NMR was consistent with the unreacted monoene. Attempted dimerization of 1, S-hexadiene-3 -ol (16). As stated before lg of 16 (Aldrich) was degassed and placed in dry box conditions To this stirring solution 42 mg (M : C = 200 : 1) of catalyst C3 was added. Upon this addition the reaction immediately turned brown in color with no detectable evolution of gas (bubbling of ethylene). Intermittent vacuum was applied with no increase in viscosity or continued bubbling. After 12 hours an aliquot was removed for 1 H NMR analysis. A small amount of internal olefin was detected at 5 5 ppm. The remaining proton shifts remained consistant with the s tarting material

PAGE 80

CHAPTER 3 DESIGN AND SYNTHESIS OF SYMMETRICAL ALKYL-SUBSTITUTED TERMINAL DIENES This chapter is concerned with the design and synthesis of the symmetrical alkyl substituted monomers that are required for the productive conversion towards a linear regular branched acyclic diene metathesis (ADMET) polymer (Figure 3.1) R I Cataly s t .. /:'-.... c I R R I n Figure 3.1. ADMET polymerization of terminal dienes The goal of this research has been to model branching in polyethylene in order to better understand how branching affects the physical properties of a polymer. The branching of linear polyethylene that is induced by the random copolymerization of a-olefins or chain transfer events can be mimicked by the ADMET condensation of alkyl substituted dienes followed by hydrogenation. The resulting copolymer is the product of a homopolymerization; therefore the amount and location of the branch points is controlled by the monomer used. The synthetic methodology presented, via the homopolymerization of ADMET monomers provides both inter a nd intramolecular homogeneity with regard to branch 71

PAGE 81

72 distributions (imperfections) along the polymer backbone. Polymer samples containing these qualities are excellent models for the study of structure property relationships with regard to branching in polyethylene. This type of system with a perfectly-spaced alkyl branch is illustrated in Figure 3.2. H I L---f"\Ht-f-fcHa--L_ R i I I I Control over# of methylenes : X I I I Control over length and identity of "R" Figure 3.2. Target polymer for the synthesis of perfectly branched polyethylene. Designing the Target Monomers The clean nature of polymers produced by the homopolymerization of a, ro-dienes, enables the design of the appropriate monomer to be deduced from the target unsaturated polymer (prepolymer). This is represented by the retrosynthesis of a target prepolymer in Figure 3.3. H H ____ ;> ~~~H~~-fcH ~_-.::;, n n R R X Figure 3.3. The retrosynthesis of the target prepolymer. To achieve control over the distribution of branch points in the resulting polymer, it is necessary to design a synthetic methodology that will

PAGE 82

73 produce sy mmetri cal hydrocarbon a, ro-dienes. This type of synthesis is c hallenging due to the lack of functional groups present and the symmetrical nature of the final product (Fig ure 3.4). Further, it is necessary to be able to sy nthesize a homologous se rie s of the models in which both the frequency and identity of this branch point ca n be varied (Figure 3.4). A symmetrically su bstituted diene is necessary to prevent the sc rambling of branches (imperfections) in the polymer chain. The metathetic H ~r H'Tt"-""'--l -{cH2.r-n-.:;,,, Control ove r the frequency t--" of the branch R = Variable length alkyl groups; Control over the branch identitY, Figure 3.4. Target monomer; symmetrically substituted diene reactivity of the terminal olefins in alkyl substituted dienes is essentially equivalent. Therefore, the polymerization of an unsymmetrical diene lacks the necessary head to tail preference required for ordered assembly. The resulting placement isomerism ca n be demonstrated by t he homopolymerization of 4-alkyl-1,8-nonadiene (Figure 3.5). This isomerism is illustrated by the 1,6 and 1,7 placement of the alky l branch in a trimer that contains one head to head ( HH ) and one tail to head (T H) connection (Figure 3.5). Due to functional gro up eq uivalence these types of connections will occur in a random fashion providing only intermolecular homogeneity This type of distribution would be considerably more ordered than that obtained in a-olefin co polymerization s, but precise control cannot be achieved.

PAGE 83

H I -------c ----I R H :::::::-------d-----------..:::--;:::::::--.......... ----HH I 6 pla ce ment 1 R H I ----c I M o TH 1 7 placement H I C I R Figure 3 .5. Trimer of unsymmetrical 4-alkyl-1 ,8 -nonadiene ; 1,6 and 1 7 placement of branch due to head to head (HH) and tail to head (TH) type placements 7 4 ------Purification is also an important consideration during the design of the monomers due to the need for 99/o+ purity in order for polycondensation to proceed to high molecular weights. This is a result of the statistics involved in step polymerizations as defined by the Carothers equation (equation 4 chapter 1). Further the catalyst used for ADMET polymerizations (Figure 3 .5.) is vulnerable to attack by Lewis bases. Ph Cl Figure 3.5. ADMET c atalyst; Schrock's Molybdenum Alkylidene.

PAGE 84

75 Synthesizing the Target Monomer Various synthetic methods were investigated toward the production of a homologous series of target monomers. The priorities in this consideration were to obtain a relatively high yielding procedure with the variability to synthesize a reasonable number of the desired monomer derivatives (Figure 3.4 ) The synthetic methods that were attempted (Figure 3 .6) have displayed interesting mechanistic implications that were considered in the de, 1 elopment and optimization of monomer construction. ~-Keto Ester Substitution Method The carbon-carbon bond-forming reaction of enolate substitutions has been shown to be a synthetically useful method toward the production of the target monomer. Carbanions or enolates can be formed by the deprotonation of an a-carbon by a strong Bronsted base These protons exhibit significantly higher acidities co mpared to hydrocarbons, where their increased acidity results from a combination of the inductive effect from the a lpha carbonyl and the resonanc e stabilization of the a nion formed by removal of a proton (Figure 3.7). The formation of a significant concentration of the enolate requires that the base used has a weaker conjugate acid than the active methylene compo und 103 The solve nt must also be a weaker acid than the conjugate acid of the base to avoid solvent deprotonation.1 03 Enolate anions are useful in a variety of base catalyzed condensation reactions of carbo nyl co mpounds specifica lly aldol/ Claisen co ndensations and alkylations. This type of enolate s ubstitution is common for the anions

PAGE 85

76 jH2 + ~Br CH 2 C -o n 10 C H 3 I B II CH~.:M ~;_,PPh 3 + RC OEt CH + RC--.(CH2 n Dicarbonyl ubstitution n M = Li, MgX Organometallic/tert-alcohol Wittig chemistry R Lactone ub s titution o R o DiGrignard co uplin g R R BrMg MgBr Br Figure 3.6. Pursued routes of monomer synthesis. of -ketoesters with the majority of the examples involving the monosubstitution of stabilized enolates with alkyl halides. 1 03 The di

PAGE 86

77 alkylation of ethyl acetoacetate, via enolate anions i s the basis for a multistep synthetic procedure for the generation of the target monomer. 0 0 0 11 11 I /C-OC 2 H 5 Base /C-OC 2 H 5 CH1/ H 2 C~CCH .. 'ccH I 3 11 3 11 3 0 0 0 Figure 3.7. The equilibrium and resonance structures of enolate formation. A general procedure for the synthesis of the target monomer was devised which involved the enolate displacement of terminal alkene bromides. This method of synthesizing the target monomer is illustrated in Figure 3.8 and is termed the P-keto-ester method. The first reaction, (A) Figure 3.8, involves the generation of the anion from a diprotic active methylene group followed by subsequent alkylation of the P-keto-ester with an alkene bromide forming the mono-substituted P -keto-ester. This can then be repeated for the other proton forming the dialkylated product (1). This method provides a single center were both alkylations can occur creating a symmetrical intermediate. The frequency of the repeat unit in the target polymer can be c ontrolled by the alkylation o f various length alkenyl bromides. The intermediate (1) could then follow two routes (B or C) to generate the alkyl group. Route B is the process of eliminating the acetyl group by the use of a retro-Claisen condensation reaction. The ester produced, (2), can then be reduced to form the primary alcohol (3), which can be removed by conversion to a leaving group (tosylate) followed by hydride

PAGE 87

H 2 + ~ HyBr C :o t I C H 3 f? COEt ~H 2 ~-{cH 2 ~ n H n 2 Reduction i D OH I C H 2 ~H 2 0 -{cH 2 ~ H 3 R ~CH ~ I H ~,o H target mon o m er 0 11 Ba s e A C OEt ~HH-{cH~ n I n G=O I Deacylation C H 3 I B F C H 3 1 T osy lati o n 1 ~ K-{c .,,......,,,~---..."'--1'-C? n 1 I H 2 /,/ 2. Redu ct ion n n H 5 78 D ea lk ox y c arbonylation C H E Redu ctio n to Methylene H ~H '" H~ n I n C H 2 I C H -i 7 Figure 3. 8. B-Keto-ester method: s ynthesis of target monomer. displacement to form the methyl s ubstituted target monomer (5). The alternative route, C, removes the ester group v ia dealkoxycarbonylation This ca n be done directly or by decarboxylation of the saponified ester. The resulting ketone (6) c an then be reduced to the methylene by a Wolf-Kishner reaction to form the ethyl s ubstituted target monomer ( 7). Both intermediates ( 3) and (6) can be used to extend the symmetrical alkyl substituent v ia further alkylations. Route B ( Figure 3.8) provided the means to construct a se ries of the target monomers and was used to synthesize monomers Sa, Sb, a nd Sc (w here

PAGE 88

A B C D E C(O)OCH2CH 3 ~ (C H 2)n ( CH 2) 0 C(O)CH3 1 C(O)OCH CH 3 ,.,,._ ( CH:,) 0 ;:_ ( CH :,) :,,........, 2 OH I CH 2 ~ ( C H z) 0 A ( CH 2) n 3 OTs I CH, ~ ( CH 2) 0 A ~ CH2)n~ 4 LiAIH 4 /THF LiAI H 4 /T HF (C H 1)n ~ C(O)CH3 1 C(O)OCH 2 CH 3 ,.,,._ ( CH 2ln ;:_ (C H :,), ,,........, 2 3 OTs I C H 2 ~ ( CH2)0 A ( CH 2)0 4 CH 3 ( CH2),A ( CH 2) ,,,........, 5 Figure 3.9. The synthesis of the methyl s ubstituted target monomer. 79

PAGE 89

80 n=3, 6, and 8 respectively ) The reactions used to accomplish this 6 step conversion are shown in Figure 3.9. Enolate Alkylation of Ethyl Acetoacetate and the Retro-Claisen Condensation The enolate anion of ethyl acetoacetate was investigated as a means to add alkene halides via subsequent nucleophilic displacements. The anion generated by deprotenation of a P-keto-ester can exist in two resonance structures (Figure 3.7); the anion on the oxygen, or the anion on the carbon. The type of base and solvent used in these reactions can favor either 0alkylation or C-alkylation, and has a direct effect on competing side reactions. If the base is too nucleophilic (Lewis basic), the direct displacement of the alkene bromide or the deacylation of the P-keto-ester by a reverse Claisen condensation can occur. If the base exhibits a strong Bronsted basicity, elimination of the alkene bromide, alkylation of the ketone, or dianion formation of the P-keto ester can occur. In order to optimize the conditions for this s ubstitution reaction, both the type of base and solvent were investigated using 5-brorno-1-pentene as the halo-alkene (Figure 3.10). The major products in these reactions were the di alkylated P-keto ester (A Figure 3.10) and the di-alkylated ester (B Figure 3.10). These products correspond to compounds la, b, c and 2a, b, c in the experimental section. The syntheses of la, lb, and le were designed to be a one pot, two step procedure (A Figure 3.9), which involved the deprotonation of e thyl acetoacetate by the bulky Bronsted base potassium tert-butoxide in dirnethoxy ethane (DME) (e ntry 12 Figure 3.10). This was accomplished by the room

PAGE 90

8 1 A B E n try Base Solvent A 0 /o B 0 /o 1 2 eq EtO-Na+ EtOH 17 22 2 4 eq EtO-Na+ EtOH 0 48 3 2 eq EtO-Na+ EtO(CO)OEt 13 4 4 2eqKH THF 69 4 5 4eqNaH DMF 16 51 6 3 eqNaH DMF 60 0 7 2eqNaH EtO(CO)OEt 42 0 8 4eqNaOH DMF 0 52 9 2eqKOH DMF 39 21 10 2 eq DBU DMF 42 0 11 2 eq NaNH2 THF 57 7 12 1.9 eq (CH3)3COK+ DME 86 0 Figure 3.10. The enolate displacement reaction of 5-bromo-1-pentene. Displayed percentages were attained by GC and show 0 / o of A or B within each sample. temperature addition of one equivalent of base to a stirring solution of the keto-ester. Formation of the enolate could be detected by the lime green color of the solution, which is the apparent result of the highly delocalized anion of a P-keto ester. After acid-base equilibrium was obtained, 5-bromo-1-pentene was added ( or higher homologue) to the basic solution. The reaction was monitored by GC using a micro-extraction method and was allowed to continue until a high conversion was achieved. Mono-alkylation of the keto ester was a relatively fast reaction as shown by detection by GC within the first 10 minutes of refluxing. A small amount of the di-alkylated product was detected before the addition of a second equivalent of base or alkene-bromide.

PAGE 91

8 2 The seco nd alkylation wa s facilitated by the addition of another equivalent of base at room temperature fo llowed by a second equivalent of the alkene-bromide. Again, the enolate was detected by a dark green color but was usually only evident at increased temperatures due t o the decreased acidity of the second proton. The second alkylation was also monitored by GC and required much longer reaction times to achieve good conversions. The retardation of the second alkylation probably is due to a more sterically hindered enolate anion and a shifted acid-base equilibrium. A large amount of potassium bromide salt was apparent by the end of the second reaction. Product, la, was isolated using high vacuum distillation via a short path distillation apparatus and resulted in a 65 /o yield, while products lb and le were not isolated due to their low vapor pressures. GC retention times of the crude products were consistent with what was expected, and mass spectra were consistent with their formula weight. The crude mixtures were used in subsequent reactions without further purification. The addition of excess base (potassium tert-butoxide), or increased reaction times also resulted in the conversion of la to the deacylated product 2a (Figure 3. 10). Similar co nversions occurred for lb and le co nsequently producing compounds 2a, b c as the major products. Cooler temperatures and longer reaction times may facilitate the production of the keto-ester only. This deacylation of a ~-keto esters was identified as a retro-Claisen condensation. 104 105 10 6 The cleavage of ~-keto esters at the ketone has been observed during alkylation reactions and was most pronounced when the alkylated produ c t had two alpha substituents, 1 0 7 as do products la, b, and c. Literature reports that this cleavage can be minimized by three methods; low temperature reactions using sodium ethoxide, the use of sterically hindered bases such as potassium tert-butoxide, 1 08 or the use of sodium hydride1 09 in

PAGE 92

83 polar aprotic solvents. The extended exposure of nonenolizable ketones to tert-butoxide in DME has also shown a similar ketone cleavage. 1 10, 11 1 The proliferation of this cleavage has become the basis for the second step in the syn thetic sequence (F igure 3 .11 ). 0 II jOEt ~H 2 t-b-(cH~ ......... C=O 3 I EtOCH 3 la 0 T 11 EtOCCH 3 2a Figure 3 11. The conversion of la to 2a by the reverse Claisen condensation. The retro-Claisen condensation of the products la, lb, and le, was induced by the addition of an excess of sodium ethoxide (Figure 3.11 and Figure 3.9 B). This deacylation occurred with near quantitative conversion, i.e. > 95/o and was monitored by GC (Figure 3.12). Chromatogram A is the product mixture after the enolate alkylation reaction and chromatograms B-D are the product mixtures after 1, 2, and 3 hours of exposure to sodium ethoxide, respectively Compound percentages were determined by subtracting out the known solvent peaks The GC peaks were assigned by the isolation and characterization of the compounds by NMR and GC/MS. It was also demonstrated the co nversion could be obtained by direct addition of sodium ethoxide to the reaction pot after the completion of the di-alkylation Higher yields were achieved, though, by first isolating the crude products It was also observed that the use of more nucleophilic bases for the enolate alkylation resulted in higher percentages of the deacylated product before the addition of sodium ethoxide.

PAGE 93

84 2a la 2a la 2a la 2a 78% 16.8% 82% 13% 94% 5.3 % 99% \.. I n 11 j i It I J I I I i 'J i 1 I, I I I la 11 I 1 % I : I 11 __ ,->. f .. r,..... ""-.J' ,.j_ I. .. J\.. A B C D Figure 3.12. Gas chromatograms demonstrating the conversion of la to 2a by a reverse Claisen condensation using sodium ethoxide. Other less successful methods of di-alkylating ~-keto esters were examined with their respective results listed in the table of Figure 3.10. Entries 1 and 2 show the results when sodium ethoxide was used as the base in the two step reaction where both the mono and di-alkylated products were isolated between steps. The monoalkylated product was isolated with relatively high yields (~80/o) while the dialkylated product was isolated with a yields of 30o/o or less. Previous investigators have shown that the failure to efficiently di-alkylate was due to the incomplete formation of the sodio derivative of the mono-alkylated product, 112 which leaves large amounts of sodium ethoxide in solution which could then directly displace the alkyl bromide. A modification of this reaction presented by Wallingford et al 1 1 2 is shown in entry 3. In this case the reaction is performed in a high boiling aprotic solvent where ethanol is distilled off as it is formed therefore driving the reaction toward the production of the sodio derivative. However, poor

PAGE 94

8 5 yields were also obtained using this technique Both of these methods using sodium ethoxide resulted in deacylati o n o f the product in a uncontrollable fashion, which i s consistent with the use of a nucleophilic base. The use of dimethylformamide ( DMF ) and other aprotic solvents results in the acceleration of enolate alkylations. 1 1 3 Aprotic solvents show a clear advantage over the use of protic s olvents for the alkylation of enolates of monosubstituted a cetoacetic esters 114 but occasionally have the detrimental effect of favoring alkylation at the oxygen rather than at the carbon. Zaugg et al offers an excellent review of specific solvent effects in the alkylation of enolate anions .115 Zaugg 1 1 5 also demonstrates the use of hydride as the base in DMF. Hydride has the advantage of shifting the acid base equilibrium in favor of the enolate anion due to the release of hydrogen upon deprotenation. Entries 4-7 (Figure 3.10) are the results of using a hydride base with ethyl acetoacetate in an aprotic solvent. The yields were moderate to low and varied with reaction times. These reactions were repeated multiple times and were performed as either a two step process with isolation of each step or as a one pot synthesis. If an excess of hydride was used and or reaction times were dramatically increased, the deacylated product 2a was the major product. Further, the production of 2a was improved by the addition of ethanol after alkylation was complete. Higher boiling residues were detected for each of these runs The di-alkylation of activated methylenes is reported to react in the presence of 1 8-diazabicyclo[S 4.0]undec-7-en ( DBU) and DMF 1 1 6 Using this method (entry 10 Figure 10) longer reaction times were required for the di alkylation, while producing only monomer la. This retention of the P-keto

PAGE 95

86 ester functionality is consistent with the tise of the non-nucleophilic base DBU. The use of potassium and sodium hydroxide in DMF was investigated as an alkylating medium (entries 8 and 9, Figure 10). Usually, these types of bases are avoided due to their tendency to cause saponification and decarboxylation of the ester group. 103 The nucleophilic nature of the hydroxide can also induce cleavage of the ketone functionality.103 The di alkylated product la was obtained using this method but resulted in low yields. Quantitative conversion, as monitored by GC, of la to 2a was obtained and was promoted by increasing the pH of the solution. The production of the keto acid and the decarboxylated product was not investigated. Sodium amide, due to its high Bronsted basicity, was investigated as a means to increase the rate of the second alkylation. Its basicity also creates the possibility of forming the dianion of the 1, 3,-keto ester. It has been shown that the addition of two equivalents of sodium amide to ethyl acetoacetate can be used to generate the dianion which results in the activation of the acetyl methylene for displacement reactions (Figure 3.13). 117 The acetyl anion will be the hotter anion and will usually undergo a high yielding alkylation. 0 NaNH2 o o 0 .. 0 OC2H s Figure 3.13. The generation of the enolate dianion by sodium hydride. Results obtained from using sodium amide as the base in a two step method (entry 11, Figure 10), resulted in acetyl substitution. The products masses (GC / MS) were consistent with a mixture of the mono, di, di deacylated, a11d tri-alkylated products (Figure 3 14 A, B, C, D respectively) The

PAGE 96

87 percent ag es of products were determined by GC with subtraction of the known so lvent peaks. 0 II A 20% 0 I I B 50% 0 I I COEt ~H H-tH ~ 3 H 3 C 6.6% D 16 % Figure 3.14. Product percentages of the enolate displacement using NaNH2. Reduction, Tosylation, and Hydride Displacement The reduction of 2a, b, and c was accomplished by the addition of excess lithium aluminum hydride (LAH) (> 2 eq) to a solution of the ester in TifFl 18 (Figure 3.9 C). This reaction proceeded as expected with near quantitative conversion at room temperature producing the alcohols 3a, b, and c. The reduction did not require the extens ive purification of the ester starting material before reacting. Due to the high boiling points of the other alcohols, only the intermediate 3a was purified by vacuum distillation. Alcohols 3 b and 3c were characterized and used in their crude form in the subsequent reaction. Now that the primary alcohol could be obtained, it was necessary to remove the alcohol functionality so that the methyl substituted target monomer could be synthesized. This was accomplished by first converting the alcohol into a tosylate and then displacing it by a hydride.

PAGE 97

8 8 Tosylation o f the alcohol ( Figure 3 .9 D) w as achieved by using a technique similar to that developed by Kabalka et a l 119 in which the tosylates of long chain aliphatic alcohols were obtained by r e acting the alcohols with a 1 : 1.5 : 2 ratio o f alcohol / tosyl c hloride / pyridine in chloroform. These reactions were c ooled to O C for the addition o f reagents followed by warming to room temperature. This was done to avoid the undesirable elimination product The reaction was followed by GC until no further conversion was observed (approx. 12 hrs.). The products recovered (4a, b, c ) were thick oils and were used without further purification. The tosylate ( 4a ) was purified by c olumn chromatography for complete characterization Tosylation methods using different equivalent amounts of reactant and reagents were also attempted. Using THF as the solvent and NaH or triethyl amine as bases were somewhat successful but resulted in relatively low yields. The final reaction was the nucleophilic displacement of the tosyl group by a hydride (Figure 3.9 E). Two equivalents of lithium aluminum hydride in THF were added to a mixing solution of compound 4 followed by refluxing The conversion wa s complete in less than 5 hour s and gave moderate yields of the target monomers Sa b, and c. The monomers Sa and Sb were purified by short path vacuum distillation to a purity of 99 /o by GC, followed by full characterization. The 1 H NMR spectra of the target methyl substituted monomers Sa Sb and Sc are shown in Figure 3 .15. Dealkoxycarbonylation of the Keto Ester Route C ( Figure 3.8) was investigated as a means to produce the ethyl substituted monomer 7. Alkylated products o f ~ keto esters may be hydrolyzed and de c arboxylated to f orm the corre s ponding a cids and ketones

PAGE 98

C H 3 ~(C H 2lJA ( C H 2)0 Sa ,+ ,. ,. -,, ''I ''I' '' i ll 1 1, 6.0 II II 5.0 4 ~ 4 .0 3.5 S .O 2.5 2.0 1~ t O 0.5 0 0 ppm Sb t I I I I I l 11 I IJ~ ,;..,.__ J \ ____ : \) I.. tL -J -' ' l ... Sc l I I A ~ Jw ' J 2 1 -0 ppm Figure 3.15 1 H NMR of the methyl subst itut ed target monomers Sa and Sb. 89

PAGE 99

90 respecti, ely. 103 The decarboxylation of s aponified esters (acids) is known to pro c eed through a sixc enter transition state which initially forms the e nol ( Figure 3 16). 107 R C C H --CO ., H II 0 R f = C H 2 -~--- R O H C C H II 3 0 Figure 3. 16. Decarboxylation of 1, 3-keto acid. Saponification of di-substituted P-keto esters is often complicated by the competing attack of the hydroxide anion at the ketone functionality, which leads to the reverse-Claisen cleavage instead of the saponification. This is due to the presence of a nucleophilic group with a dialkylated 1, 3-keto ester. Hydrolysis and decarboxylation are typically accomplished with aqueous acid and heat to avoid this side reaction.107 Decarbonylation of la to 6 (C Figure 3.8) was attempted by heating la in the presence of either an acid, base, or salt (NaCl) in DMF, DMSO, or H20. No detectable reaction or significant decomposition was observed using these methods The expected c onversion of la to 2a (deacylation) was observed with the a ddition of KOH and heating. Krapcho reports a method to affect dealkoxycarbonylation of disubstituted P-keto esters by the use of a salt such as lithium chloride or sodium c yanide in a dipolar aprotic media. 120 This method has the advantage of proceeding with neutral c onditions, therefore preventing s ide reactions such as acid addition to olefins and ketone cleavage. Using this method la was reacted with LiCI salts in NMP and water under reflux conditions. The GC results showed the appearance of a new major peak. Further low resolution GC / MS resulted in a M+l ion that was consistent with the production of 7. The lH NMR s howed the retention of a 3 proton

PAGE 100

91 singlet at 2.04 ppm from the retention of the methyl ketone as well as the disappearance of the ethoxy protons. Work on the dealkoxycarbonylation was stopped due to the difficulty in isolating large amounts of la. The P-keto ester substitution method has provided the most success in the generation of the target monomers Target symmetrical methyl substituted monomers Sa, Sb, and Sc have been generated using the general methodology outlined in Figure 3.9. The following discussions are based on other synthetic methods that have been investigated but are currently not being pursued. Wittig Method The Wittig reaction between a phosphorus ylid and an aldehyde or ketone results in carbon-carbon bond-formation and is extensively employed in synthesis. The use of this reaction was investigated for the synthesis of the target monomers. The synthetic design was to first make a phosphonium salt from an alkene bromide followed by the formation of the ylid and the coupling with an alkene-one (Figure 3.17). The carbon-carbon bond-formation results in the formation of an olefin at the joining site. The product, therefore, contains a tri-substituted olefin in place of the carbonyl (Figure 3.17 C). This method provides the ability to make monomers of multiple sizes and variable length branches. The size (length) and symmetry of the diene product can be modified by choosing longer chain functionalized alkenes. The branch point identity can be varied by modification of the alkane side of the ketone. Figure 3.17 demonstrates this utility with the synthesis of a methyl branched monomer by reacting the ylid with a methyl ketone.

PAGE 101

A 0 11 C H 3 C --(c H 2 3 B nBuL i THF H C H 2yc 2 C Figure 3 .17 Wittig reaction forming methyl s ubstituted triene. 9 2 While undergoing ADMET, there exists the possibility that the tri substituted o lefin will undergo metathesis Sita has s hown that tri substituted olefins will undergo ring closing metathesis when exposed to Schrock's molybdenum a lkylidene. 12 1 Konzelman and Wagener using ADMET with the same catalyst system have formed tri-substituted olefins via dimerization reactions 12 2 If the tri-substituted olefin is metathetically active, scrambling of the branch point will occur If the substituted olefin does prove to be metathetically inactive the resulting polymer will contain tri-substituted olefins in the backbone. In order to produce the polyethylene-like target polymer, the unsaturated prepolymer must undergo quantitative hydrogenation. In general the ease of hydrogenating a lkenes decreases with increasing substitution. 1 2 3 Hahn has recently demon s trated the incomplet e hydrogenation of polyisoprene v ia the diimide hydro g enation t e chnique. 124 Therefore, some unsaturation may remain in polymer samples made with this monomer (Figure 3.17 C). These types of system s would be poor samples for modeling PE. This route was initially pursued by the s y nthesis of the Wittig salt by reacting 5-bromo-1-pentene with triphenylphosphine using typical synthetic conditions. An i s olated yield of 9 3 /o was obtained. Due to the possibility of the above consequences, this route did not prove useful after the synthe s is of (1-pentene)triphenyl phosphonium bromide (Figure 3.17 A)

PAGE 102

93 Malonic Ester, Di-Grignard and Lactone Methods The synthesis of substituted dienes through the use of diesters was considered. This method appeared promising due to the co mmercial availability of methyl substituted diesters ( R= Me) were m and n range from 0 to 2 ( Figure 3.18 1) Other diesters derivatives could also be prepared through enolate substitution of malonic esters or lactones. H I R R H~ ( C H 2) ~ ( C H 2) 0 ,.............. 0 H II ( CH 2) mA ( CH 2) :---= K R T s O,.............. ( CH2 )m A ( CH v n~ OTs R X~ ( CH 2)m A ( CH2)n~ X V IV m M R ( C H 2) ~ ( CH 2) 0 ~ VI H. reduction of the ester I tosylation of the diol J. bromination or chlorination of the diol K. s ubstitution with acetylene nu c le o phile L. s elective r educ ti o n to form the alkene from the alkyne M s ubstitution with alkene nucleophile (vinylic or allylic) N. production of di-Grignard and then s ub s titution with alkenyl halid es. Figure 3.18. Diesters in the synthesis of substituted dienes The synthetic scheme outlined in Figure 3 18 was attempted without complete isolation of the steps. It was soon realized that each s tep produced

PAGE 103

94 relatively low yields and resulted in mixtures of products that were difficult to isolate, which was also the case for the di-Grignard and lactone methods (Figure 3.19 (1) and (2) respectively). These methods were also limited in the variability of diene size and the symmetrical branch point identity. Due to the poor yields, these synthetic methods were not pursued further. Most of the yields and product mixtures of these reactions were determined by GC and or crude 1 H NMR's. Isolation of the product mixtures was not done due to time constraints involved with the isolation of complex mixtures of high boiling oils. The ultimate goal of this project was to synthesize and characterize the target polymers. The use of preparative high performance liquid chromatography (HPLC) may facilitate further investigation of these methods. R Br ~,.,___ Br VII 0 I Base 2.RI IX 1. Mg 0 R X R ~(CH2)2A (CH2)2~ VID R HO/"-,.,.(CH2)mA (CH2)n,OH XI where R' = Me ; m and n = 1 Figure 3.19 (1) Di-Grignard method (2) Lactone method for diol intermediate. (1) (2)

PAGE 104

9 5 Organometallic/Tert-Alcohol Method Another route to the target monomers is the use of the carbon-carbon bond forming reaction of Grignards or lithiates on carbonyl compounds. In order to incorporate the branch point, these reactions would result in tertiary alcohols that theoretically could be reduced to the corresponding methine (Figure 3.20). Synthesis of the tertiary alcohol 9 (Figure 3.20) was achieved by forming the Grignard reagent from 5-bromo-1-pentene and reacting it with ethyl acetate. Using ethyl acetate as the ester reagent produces the methyl branched alcohol. The advantage of this simple monomer synthesis is the ability to design a whole se ries of tertiary alcohol pre-monomers with precise control over the identity and frequency of the alkyl group. The alkyl substituent can be easily changed by changing the identity of the ester (higher order), and similarly, the frequency of the substituents can be controlled by changing the identity of the alkene bromide (Figure 3.20). The monomer was isolated by distilling under reduced pressure to achieve a purity of >99/o. Use of the Grignard reagent gave a satisfactory purity with a 95 /o yield, while the lithium reagent resulted in a low yield due to the poor conversion of the bromide to the lithiate and was no longer pursued The unsymmetrical diene 5-methyl-1,9-decadiene-5-ol (10) was also synthesized via this Grignard method. The Grignard reagent of 5-bromo-1pentene was reacted with the ketone, 5-hexene-2-one producing 10 in a 78/o yield. The unsymmetrical diene-ol was made because of its ease of synthesis and was used to model reduction reactions of tertiary alcohols.

PAGE 105

Br + 2 Li Li + LiBr + 0 I I Mg 1.) 0.5 EtOCCH 3 4% 0 H I I MgBr 1.) 0.5 EtOCCH 3 H 2 --{<:H 2 2.) H+ 3 3 95% CH3 9 ''R'' can be varied by changing the ester-~ Figure 3.20. Organolithium and Grignard synthesis of symmetrical tertiary alcohol diene. 96 The alcohol reduction was pursued by many different methods. A method derived from Carey and Tremper 12 5 was attempted due to its high reported yields and ease of synthesis. This reaction was designed to reduce tertiary alcohols to the corresponding alkane through a carbonium ion. 12 5 The carbonium intermediate is generated by protonation of the alcohol with a strong acid, followed by a hydride transfer reaction via an organosilicon hydride. This reaction was attempted using the tertiary alcohols 9 or 10 as the starting materials (Figure 3.21). The GC and GC/MS results obtained were not consistent with the target product ( Sa) The GC chromatogram

PAGE 106

O H H I HSi ( Ph )3 ){I H ,, H ? CH 2 I C F 3 COOH THF I 3 3 3 C H 3 C H 3 9 Sa Figure 3 .21. Attempted reduction o f tertiar y alcohol via carbonium intermediate. 9 7 H 2 3 showed multiple products which indicated the possibility of the protonation of the olefins by trifloroacetic acid An internal olefin peak at 5.6 ppm was detectable in the 1 H NMR suggesting s ome degree of elimination This eliminated product was also detectable in CI-LR GC / MS by a base peak of 165 which is consistent with the m+ 1 of the eliminated product. Another method put forth by Ireland et al was investigated for the reduction of tertiary alcohols to alkanes. 126 The general reaction is performed by first converting the tertiary alcohol to the diethyl phosphate, followed by reduction to the corresponding alkane using a lithium-ethyl amine solution The tertiary alcohol (9) was first exposed to diethyl chlorophosphate in the presence of the s trong base n-butyl lithium to undergo phosphorylation ( Figure 3 22 ( 1 )) This product was not isolated before proceeding to the reduction The diethyl phosphate derivative from (1) was then exposed to a lithium metal, ethyl amine solution for the reduction ((2) Figure 3 .22). The GC results of the worked up reaction showed a complex mixture o f products with the major constituent having the s ame retention time as s tarting material. A CI-GC / MS confirmed this result by the base peak of the major product having a mass c onsistent with s tarting material. Similar GC / MS results were o btained when the r e a c tion mixture from (1) (Figure 3 .22) was analyzed. This provided eviden c e that the phosphate derivative was never produ c ed. The reaction wa s nev e r repeated but it was assumed that the

PAGE 107

9 8 absence of the phosphate was due to the se v ere steric hindrance of the tertiary alcohol Further investigation o f this effect would require longer reaction times as well as reacting the c yclic analog of the tertiary alcohol. 2 .5M N BuLi 4 : 1 THF : EDA H I ~"i' H 2 t-1-f:H 2 r3 ........;., CH 3 Sa X Li EtNH 2 t-butan o l ( 2) Figure 3.22 Attempted diethyl phosphate reduction of (9). 0 11 ClP ( OEt )2 ( 1 ) The deoxygenation of the alcohol was also attempted via a mesylated intermediate. The purpose of this reaction was to make the alcohol a g ood leaving group to facilitate its replacement by excess hydride. This method i s similar to the Carey method but does not require a strong acid therefore avoiding interaction with the olefins. The mesylation of alcohol 9 was performed by e xposure to methanesulfonyl chloride in the presence of triethyl amine at O C ( Figure 3.23). Due to the stability of the tertiary carbocation formed, this reaction resulted in the eliminated product. This was indicated in the 1 H NMR spectrum by internal olefin resonances As with the Wittig this triene cannot be used a s a target monomer.

PAGE 108

9 2( Et 3N ), CH 2 Cl2 0 C fH3 O =S=O 6 I "--'1' H 2 t---c-f:H 2 r---..:,,,, 3 3 CH3 uncontrollable elimination ~"T" H 2 r-1~cH7CH 2 -,-._.::.,,, CH3 internal ole fin dete c ted by proton NMR Figure 3.23. Mesylation and elimination of alcohol 9. 99 Other methods of deoxygenation were attempted including first converting the alcohol to a halogen, followed by reduction to the alkane. Two methods developed by Olah et al. were attempted without success (Figure 3.24). 127 The first method involved exposing alcohol 9 or 10 to hexamethyldisilane in the presence of pyridinium bromide perbromide. The reaction resulted in a mixture of brominated products due to the presence of molecular bromine in the reagent The brornination of the olefins was detected by GC / MS The second method involve reacting the alcohol with chlorotrimethylsilane followed by exposure to LiBr. 127 This reaction resulted in the recovery of starting material.

PAGE 109

OH I H -, C H 2 3 3 CH3 TMSCI/LiBr CH3CN No rea c ti o n Br HMDS py ridinium bromide pe rbr o mide Br Br I H 2 H 2 I 3 3 CH3 Mixture of brominated product s Figure 3.24. Attempted conversion of alcohol 9 to halogen homologue. Conclusions 100 A synthetic method was designed to facilitate the synthesis of the target monomers needed in the synthesis of perfectly branched polyethylene (Figure 3.4). This method is shown in Figure 3.8 and has demonstrated the ability to successfully produce the symmetrical target molecules were n=3, 6, 8 and R = methyl group (Sa, Sb, Sc respectively). The reactions used for this conversion are displayed in Figure 3.9. This method exemplifies the potential to synthesize an entire series of the desired monomers where n and R can be varied to the desired integer and a lkyl identity respectively. Using these techniques monomers can be assembled to grant full control over the frequency and identity of the repeat unit in the ensuing polymer. Other methods, including the use of the Grignards, Wittigs, and lactone / malonic ester alkylations, were pursued with limited success ( Figure 3.6).

PAGE 110

CHAPTER4 ADMET MODELING OF BRANCHING IN POLYETHYLENE: THE SYNTHESIS OF MACROMOLECULES WITH PERFECTLY SP ACED METHYL BRANCHES. Modeling Polyethylene In order to better understand the crystallization behavior of chain molecules, it is important to study systems in which the microstructure is controlled and understood. The fundamental study of chain folding and melting in linear macromolecules has been made possible by examining samples of high molecular weight normal paraffins. 1 28,129,69 These systems are void of side chains as well as being monodisperse in nature. Therefore their size, linearity, and end group identity can be precisely controlled. These model compounds have been used to determine the critical chain length with regard to lamellae type folding 6 8 and comparing their melt behavior with those of similarly sized PE samples.7 2 The correlation between monodisperse models (n-alkanes) and fractionated PE samples is apparent by the molecular weight at which chain folding occurs is the same for both systerns.68,69,7 2 Further, it has been found that the melting behavior-molecular weight relationship between these samples is also the same. Therefore n-alkanes, having a molecular weight limitation of 5,462 g/mole, make excellent models to study the melting behavior of low molecular weight linear PE samples. 101

PAGE 111

102 Metathesis has recently been investigated as a new method of producing linear samples of PE. The resulting polymers via ROMP or ADMET mechanisms, are analogous to the syn thesis of polybutadiene by a Ziegler system, in that the resulting polymer is linear in nature and still co ntains a site of unsaturation. Therefore, these polymerization techniques produce a prepolymer that can be converted into the saturated target polymer by quantitative hydrogenation. Wu and Grubbs 1 30 have recently investigated the use of ROMP to synthesize linear monodisperse PE. They achieved a narrow polydispersity (Mw /Mn < 1.1) by the use of a living polymerization system based on the ROMP of cyclobutene (Figure 4.1). 1 3 1 The resulting polymer has the same repeat unit as polybutadiene, with the added advantage of being more linear in nature. The absence of complete linearity in anionically polymerized 1,3butadiene is the result of a small degree of 1,2 addition. This side reaction leads to a small degree of C2 branching on the polymer backbone. 51 Similar to the ROMP polymers, the polydispersity of the anionically polymerized polymers also approach 1. Me co NAr 3 '11,,, Ii CHCMe3 MeaCOr n Toluene/ PMe3 --+-==--{cH~-+---=::lN 2 n xylene TsNHNH 2 120 C Figure 4.1. The synt he sis of narrow dispersed linear polyethylene via ROMP.

PAGE 112

10 3 The polymer produced in this s tudy had a number average molecular weight o f ~85,000 g/ mole and a melting temperature of 126 C, which is lower than expected for such a linear high molecular weight sample. Typical melt transitions for linear PE are u s ually in the range of 131-136 C. 3 The low e ring of the melting t e mperature may be due to a number of factors including the possibility of incomplete hydrogenation as well as the incorporation of bulky e nd groups Prior to the study by Wu and Grubbs, O'Gara et al 100 synthesized linear PE via AD MET chemistry. This method not only used metathesis as the polymerization medium, but generated the model polymers through the use of a step condensation process. This process involved the clean conversion of 1,9-de c adiene to polyoctenylene via the olefin co ndensation reaction (Figure 4.2). lOO The prepolymer was then exhaustively reduced by the diimide hydrogenation reaction perfected by Hahn. 12 4 The resulting linear PE samples display an absence of residual olefinic sites by 1 H NMR spectroscopy and a high degree of crystallinity as determined by thermal analysis.1 00 This s ample also displayed a melting temperature of 134 C which is c lo se t o that o f the theoretical values for perfectly crystal line polyethylene. 57,132 Poly(octenylene ) 6 10 4 torr C H .., 6 X Diirnide Hydro2enation T s NHNH -, 6 X nPr 3 N a-xy l e n e n Figur e 4.2 The sy nthe sis of perfectly linear polyethylene, with Tm = 134 C

PAGE 113

104 To date i t has been much more difficult to model branching with regard to its effects on cyrstallinity in chain molecules. 66 The synthesis of PE models using c onventional ca talysts shows a very inhomogeneous distribution of alkyl branching. 7 9 Branched products polymerized using metallocene catalys ts provide samples with a high degree of intremolecular homogeneity but the distribution of branches a long the backbone of an individual molecule is uncontrollable 79 To investigate the influence of branching on material properties (specifically melting), it is required that the models have well defined chain branching. Optimally, these models should grant full control over the b ranch identity as well their inter and intramolecular distribution.80 ,S l This is necessary because it has been found that the melting temperatures of copolymers are c rucially dependent on the sequence distribution of the branch points.1 33 Most models designed for the study of branching still require the use of chain polymerization techniques. These systems include the copolymerization of diazomethane with higher order diazo compounds (diazoethane), 134 the copolymerization of butadiene monomers with olefins followed by hydrogenation,B0, 81 and the metallocene polymerization of ethylene with higher order olefins. 79 All of these systems still exhibit so me degree of randomness regardless of the reactivity ratios of the comonomers. Fractionation of these systems has been extensively used in order to isolate samples that are more homogeneous in nature. ADMET Modeling of Branching in Polyethylene The research presented in this chapter is essentially an extension of the ADMET method in that the sa me type of chemistry as in Figure 4.2 is used to

PAGE 114

105 produce hydrocarbon polymers with perfectly spaced methyl branches. As stated in chapter 3, polymer samples containing this highly ordered quality are excellent models for the study of the structure-property relationships with regard to branching in polyethylene. This is accomplished herein by the homopolymerization of the symmetrically substituted terminal dienes Sa and Sb (Figure 4.3) v ia ADMET polymerization. The resulting mechanism achieves precise control over the branch points identity and distribution along the polymer backbone due to the nature of using a condensation mechanism with a single monomer feed (Figure 4.3). H H I ~H 2 ti--f:H 2 ~ ---"-H 2 \ 1 -f:H 2 -.-...._ + C H 2 ::CH 2 n R Sa (n= 3), Sb (n=6 ) R= Variable l e ngth alkyl gr o up s Control over# of methylen~ Control over length and identity of ".n.--" -R X PSa,PSb Diimide Hydr o genation HPSa, HPSb Figure 4 3. The synthetic scheme for the synthesis of polyethylene with perfectly-spaced alkyl groups

PAGE 115

106 Polymer synthesis via ADMET As with the polymerization of perfectly linear polyethylene ( Figure 4.2), Schrock s molybdenum alkylidene (Fig ure 3 .5 Cl) was chose n as the metathesis ca talyst for the c lean ste pwise co nversion of the branched hydrocarbons (Fig ure 4.4 Sa Sb and Sc ) to the co rresponding linear polymer. Therefore the polymers produced are linear in nature and devoid o f s ide reactions that cause uncontrollable branching. CH 3 -::r'--(C H 2h,,,l_(C H 2}3/, Sa C H 3 ~(C H 2) (C H 2)6~ Sb C H 3 -::r'--(C H2 )s,,,l_ ( CH2 )s/, Sc Figure 4.4. Symmetrically methyl s ubstituted ADMET monomers. The ADMET polymerization of the branched monomers (Figure 4.4) produces a prepolyrner (Fig ure 4.5) were (n) can be varied by se lection of the appropriate monomer. The a bility of these nonfunctionalized alkyl substituted dienes to undergo ADMET polymerization can be equated to the well documented hydrocarbon s tudies performed by Konzelman 13 5 and O Gara 100 n Figure 4.5. Target unsaturated prepolymer where n and R ca n be varied.

PAGE 116

107 ADMET polymerizations are performed in the absence of solvent, (bulk co nditions) and because this reaction is a step condensation equilibrium, the production and removal of ethylene (for terminal olefins ) is essential for attaining l1igh molecular weights To facilitate these requirements, polymerizations are performed under high vacuum ( > 104 torr) with typical monomer to catalyst ratios of 500-1000:1. In the presence of nonfunctionalized hydrocarbon dienes, the catalyst-monomer mixture results in the immediate evolution of ethylene at room temperature along with an obvious increase in viscosity within half an hour. Reactions are typically continued for a period of 2 to 3 days or until evolution of ethylene (bubbling) ceases. When the viscosity increases to a point were the mobility of the catalyst is reduced, the reaction is heated to 60 C to facilitate further stirring. This increase in temperature does not introduce any detectable side reactions as shown by NMR. Monomers Sa and Sb (Figure 4.4) were purified in order to remove residual monoene and any protic or Lewis basic impurities which could limit the ultimate molecular weight or terminate the polymerization. Most of the impurities were removed by a flash filtration using alumina in a fine frit funnel with pentane as the solvent. The filtrate was then further purified by a vacuum fractional distillation to ob t ain greater than 99/ o purity as determined by gas chromatography on a neat sample. The purification of monomer Sc could not be achieved by distillation due to its higher boiling point. This sample contains a small percentage of eliminated product which was indicated by the presence of a disubstituted olefinic proton s ignal at 5.42 ppm (Figure 3.15 s pectrum Sc). The existence of the eliminated product is the result of the competing elimination reaction

PAGE 117

108 with that of nucleophilic substitution. Polymeriz atio n of this monomer has currently not been performed The pure monomers Sa and Sb were first degassed by several freeze pump thaw cycles followed by exposure to an alloy of sodium and potassium vi a a vacuum transfer. This was necessary for the removal of oxygen and to ensure dryness respectively. The monomer s were stirred over the alloy for a minimum of 1 hour and then transferred to a clean dry reaction flask with a Teflon TM vacuum valve. A catalytic amount of alkylidene Cl (Figure 3.5) was added in an inert atmosphere and upon ca talyst addition an immediate evolution of pure ethylene was observed. The reaction of monomer Sb had to be restarted several times by the addition of more catalyst. This is likely the result of a catalytic amount of a poisoning impurity, therefore the monomer underwent further purification by excess catalyst addition. Upon successful initiation, the reactions were exposed to intermittent vacuum until the viscosity had noticeably increased. The reactions were considered to be complete when the high viscosity of the material prevented stirring at ~60 C and the release of ethylene ceased. The polymerizations were cooled to room temperature and terminated by exposure to air. The clean co nversion of monomer to polymer was demonstrated by the examination of the 1 H and 1 3C NMR's before, during and after the polymerization. The conversion of monomer Sa to polymer PSa is illustrated in Figure 4.6. A high degree of polymerization for the reaction is indicated by the detection of only internal overlapping cis and trans olefinic signals in the 1 H NMR (~5 .3 ppm s pectrum B Figure 4.6). Following the accepted mechanism for the ADMET cyc le ( Figure 1.17), the co ndensation of monomers in a stepwise fashion results in an increase in the number of

PAGE 118

109 Polyethylene Prepolymer With 8 carbons Between Branch Points [ Mo] <104 rnrnHg P5a + ncH ;;=:: c H ., A 'I ', ,11 11 111111 11 ,,11 1 111r11111 r 1 11111111 1 11111r11111111111,1 8 0 5 5 5 0 4 5 4 0 3.5 3.0 2.5 2 0 1.5 1.0 0 5 0 0 ppm I I I B I l l J i I I I ,,,, fl 1111 111 1 1 11 + ''l '' ''l''''l'liijillll''''l''t'I' e .o s.s s o 4 .5 4 o 3.S 3.o 2.s 2.0 1 .s 1 0 o.s o o ppm Figure 4.6. Polymerization of monomer Sa (n=3).

PAGE 119

110 internal olefins, while leaving the terminal olefins as end groups. Therefore the degree of polymerization can be determined by the relative ratio of internal to terminal olefins. In the polymerization of Sa, the absence of signals in the 1 H NMR (4.9 ppm CH2= and 5.8 ppm =CH A Figure 4.6) indicates a high number average molecular weight (Mn)The maintenance of methyl branch can be observed by the expected doublet at ~0.8 ppm for both the monomer and polymer. Further comparison of the up-field monomer to polymer proton signals shows no change in their expected values and hence indicates the absence of detectable side reactions. 13C NMR and elemental analysis show similar results. The excellent correlation between the expected and the actual data provides proof of the prepolymer's exact structure, which possesses 8 carbons between branch points. The polymer P5a is a clear tacky solid; molecular weight determination was done by gel permeation chromatography (GPC) relative to polystyrene standards. Table 4.1 shows the number average (Mn) and weight average (Mw) molecular weights of a typical batch of P5a. As with this sample, the molecular weight distribution (PDI) of ADMET polymers is governed by the same statistical relationship as for other step condensations, and typically approaches 2. Thermal analysis of this polymer by DSC was also performed and will be discussed in a later section. Table 4.1 GPC results of methyl substituted polymer P5a (n=3) Sample g of sample Mn Mw PDI PSa 0.9 40,000 73,000 1.8 Reported value s are relative t o poly s t y r e ne s tandard s.

PAGE 120

Polyethylene Prepolymer With 14 Carbons Between Branch Points > ppm [Mo] < 1 04 mrnHg H C H 3 I H 2 b-r-(cH 2 H 6 o X H PSb J \ + XCH 2 CH 2 11 s 4 ppm C l D 3 Figure 4.7. The polymerization of monomer Sb (n=6) 111 l l I I I I I 2 1

PAGE 121

112 The polymerization of monomer Sb proceeded in the same manner as for Sa, except a larger amount of monomer was used (3g) The purification of monomer Sb via vacuum distillation was complicated due to its increased molecular weight, which resulted in the presence of some residual impurities. This could be visually detected by the lack of ethylene evolution (bubbling) early in the reaction. The 1 H NMR spectrum and reaction scheme are presented in Figure 4.7. The 1 H NMR of the polymer PSb (Figure 4.7 D) shows residual end groups (~4.9 ppm, =CH2; ~5.8 ppm, CH=) which can be identified by comparison with that of the monomer's (Figure 4.7 C). The quantification of the end group signals could not be accomplished due to their very low intensity, hence end group analysis to obtain number average molecular weight was not possible. GP<;:: results show that the polymer has a Mn of approximately 15,000 (Table 4.2) Previous examples of end group analysis with ADMET polymers by 1 H NMR have shown good correlation with GPC data when Mn was under 20,00Q.90 134 136,137 Table 4.2 GPC results of methyl substituted polymer PSb (n=6) Sample g of sample POI P5b 3 15,000 26,500 1.8 Reported value s are relative to polystyrene standard s The crystallization of the polymer PSb is not visually apparent and was a viscous but stirrable fluid at 60 C. At room temperature, PSb was similar to PSa in that both were clear tacky solids. Additional purification of this

PAGE 122

113 monomer should result in polymers with mole c ular weights sim ilar to that o f P5a ( Mn > 20 000). Spectrum D (Figure 4 7) is of the polymer's final reaction mixture without any purification o r precipitation. This NMR shows the absence of byproducts in the reaction mixture, and demonstrates the exceptionally clean mode of propagation that is co mmon for ADMET polymerizations. All the proton and carbon signals can be easily assigned as a result of the perfectly regular structure. The regularity of this structure is paramount for the generation of a series of well-defined polyethylene models. Hydrogenation of the unsaturated polymer The viability of ADMET chemistry to synthesize a prepolymer with a perfectly regular structure has been demonstrated. Olefin metathesis results in a prepolymer that has regularly spaced sites of unsaturation within the pol y mer backbone which can be hydrogenated (Figure 4.3). For laboratory sc al e reactions the use of the inorganic reagent diimide ( N2H2) is typically used to accomplish this co nver s ion (F igure 4.8).124 HN=N H + n Figure 4.8. The diimide hydrogenation of backbone unsaturation The thermolysis of p-toluenesulfonyl hydrazide ( TSH) is commonly used for the in-situ generation of diimide. 1 38, 1 3 9 ,140 The generation of side products (p-toluensulfinic ac id ) that can cause the addition of the s ulfone and

PAGE 123

1 14 the degradation of the polymer backbone is a draw back to this reaction. 141 1 42 Hahn 124 has developed a modification of this method that uses an equivalent amount of tri-n-propyl amine ( TP A) with the TSH. This rather new method has shown to eliminate these side reactions in butadiene containing polymers and copolymers and results in greater than 99/o hydrogenation. 124 Hahn s diimide hydrogenation technique was successfully used to generate perfectly linear polyethylene. 100 This was demonstrated by the >9 /o hydrogenation by 1 H NMR of the ADMET polymer polyoctenylene (Figure 4.2). 100 Therefore, this method was chosen to hydrogenate the structurally similar alkyl substituted polyene's PSa and PSb. The fully characterized unsaturated polymer PSa was subjected to exhaustive hydrogenation via diimide chemistry. Quantitative conversion by 1 H NMR was achieved by the addition of 2.5 mole equivalents of TSH and TP A per mole of olefin. Subsequently an additional 1.5 equivalents of TSH and TP A were added after 3 hours of reflux. The need for the excess reagent was due to the self hydrogenation reaction of diimide.124 Upon spectral analysis, the higher molecular weight sa mples typically showed olefinic signals which resulted from residual unsaturation. These samples were precipitated followed by a subsequent treatment of the diimide. These repetitive treatments produce polymer HPSa, a completely saturated polymer chain which is clearly demonstrated by analysis of the 1 H NMR before and after hydrogenation (Figure 4.9). Examination of the hydrogenated spectrum (Figure 4.9 spectrum F) shows the quantitative disappearance of both the internal olefinic protons at 5.3 ppm and the allylic methylene protons at 1.9 ppm. The branch point is also easily identified as the doublet proton signal at 0.8 ppm and the integration of all the peaks are consistent with the repeat unit identity

PAGE 124

H C H 3 I H 2H ~H 2 H PSa n-Pr 3 N o-xylene HPSa 115 ~ X E \_ . 1 ,1 ,,,1 1 'I , 11 ' ' 1 ' '11 ,, '' ''1''''1'''11''''1' 'I' 8 0 5.5 5.0 4.5 o 3.5 3.0 2.5 2.0 1.5 1 o 0.5 0 0 ppm F --------------------~~ _._,_ 6 5 4 ppm I 3 2 1 -0 Figure 4.9. The 1 H NMR spectra of the diimide hydrogenation of PSa (E) to HPSa (F).

PAGE 125

116 Spectral analysis of the sample using 1 3 C NMR a lso indicated no detectable unsaturated groups. The carbon s pectrum is particularly revealing with the presence of only six expected resonances (36.3, 30.0 29.5, 28.8 26.2 and 18.8 ppm) with resulting values that are similar to those predicted using a model designed for branched alkanes by Carman, Tarpley and Goldstein. 143 The branched alkane values also help to identify the correct order of labeling for the resonances (Figure 4.10). These data suggest the lack of side reactions from hydrogenation as well as a regular repeat structure. This type of spectral consistency is impossible to achieve via the typical copolymerization of olefins. 1 44 2 1 43 t 2 3 CH3 f 6 t t t 5 6 4 L 5 . . ' .. . 1 1 Ir t 1, 1 11111 ( I 11 1 1 I I I I I' 111,1 1 11 I 1t11rr,, ) 1111, 111I'111 1 1111 I', 11 tl 11, t 1, 111111 80 70 60 SO 40 30 20 10 -0 Figure 4.10. 13 C NMR of hydrogenated polymer HPSa with assigned resonances.

PAGE 126

117 The hydrogenation of P5b (n=6) proceeded in much the same fashion with the elimination of the olefinic and allylic proton signals. Quantitative hydrogenation of this sample was not achieved as evident upon expansion of the olefinic region in the 1 H NMR (Figure 4.11 inset). Concentration of the residual olefinic groups was too low to obtain an accurate integration indicating a small abundance. Further treatments of the diimide reaction should remove these residual signals. HPSb's appearance was very different from HP5a in that it was an opaque, yellow brown colored, hard waxy solid. I 7 5 4 fH 3 --y-(CH2) 14 H I 3 n 2 1 Figure 4.11. lH NMR of HPSb; inset shows residual olefin signals. The resulting l3C NMR of this sample again shows the perfect regularity of this repeat structure with only six resonances present (36.2, 31.9, 29.1, 28.7, 26.1 and 18.8 ppm) (Figure 4.12), which are corroborated by the results obtained from calculations and the experimentally determined signals from HPSa. The similarities between the 1 3 C NMR signals of HP5a and HPSb

PAGE 127

118 ( Figures 4.10 and 4.12 respectively ) illustrates the ability of this technique to produce a series of perfectly intra and inter-homogeneously d istributed branched copolymers of polyethylene. These models provide the perfect control of the polymers se quence distribution that is needed to s tudy the effects of linear c hain branching or imperfections 80 70 IH 3 -+--T ( CH 2) 14 H HPSb eo n 60 Figure 4.12. 13C of polymer HPSb. 40 30 :zo ,o -0 ppm Gel permeation chromatography (GPC) of these samples ( HPSa and HPSb) compared to their unsaturated prepolymers (PSa and PSb) indicated no chain degradation as evident by the absence of new peaks or broadening of the molecular weight distribution. This is demonstrated by the comparison of the chromatagraphs for PSa and HPSa (Figure 4 13). The molecular weight values of the unsaturated polymer PSa and the hydrogenated polymer HPSa are in good agreement with respect to the peak molecular weight (Mp) and the higher order z-average ( Mz) (Figure 4.13) which indicates that the hydrodynamic vo lumes of the sa turated and

PAGE 128

119 unsaturated polymers are similar in chloroform. The value fo r the number average ( Mn) molecular weight ( HPSa) i s ca. three times greater as a result of J0. 00 . z e oo r Ace tone int s td HPSa RI det )II I ~f\ I r' V 2, 0 0 . 24.00PSa RI det )II . H 2 0 22. 00\ Z0.00 1 1 .0 0 I \ Ir \ . \ y I/ r PSa UV det ... r HP5a UV der ; ~ l J 01,.0 0.00 2.00 1 00 6,00 . . ' '~ .. I .D O 10. 00 1 2.00 1 1 00 1 6.00 11.0 0 20,00 H1.aut e1 Polymer Mn Mp Mw Mz PDI P5a 13,000 30,100 33,500 59,200 2.6 HP5a 30,600 34,100 41 500 61,400 1.4 o btained value s are relative t o polystyrene s tandards. Figure 4.13. GPC c hromatograms and curve results of sa mples P5a and HP5a. the fractionation of the hydrogenated sample from reprecipitations of this particular sample. Each precipitation resulted in the loss of the lower molecular weight molecules c reating a higher mole fraction of the larger constituents which results in an increase in the Mn and a decrease in the polydispersity (Figure 4.13). Examination of the refractive index data shows similar retention times but a narrower peak width. The absence of the sulfonate chromaphore incorporation is also suggested by the lack of new peaks via the ultraviolet detector.

PAGE 129

120 The GPC res ults f o r HPSb are also co nsi ste nt by displaying no detectable decomposition. As illustrated by the peak ove rlay in Figure 4.1 4 the unsaturated and sa turated samples are also essen tially eq uivalent with respect the their hydrodynamic volumes (e lution ti me s) and peak width. 14 5 Polymer PSb HPSb Mn 15,000 13,500 HP5b -7 00 Mp 25,800 24,300 I J 8 00 Minut es Mw 26,500 25,000 Mz 39,400 39,000 9 00 obta in ed val u es a r e relative to polystyrene s tandard s. POI 1.8 1.8 Figure 4.14. GPC Chromatograms and curve results of PSb (unsaturated) and HPSb (saturated). Thermal Analysis of Methyl Substituted Polyethylene The incorporation of a co monomer unit into linear polyethylene ca n greatly affect its physical and mechanical properties. Much of the research involved with the design of co polymer sys tems concerns their thermal

PAGE 130

121 beh av ior. For a linear flexible chain polymer, t he thermal behavior in its c hemi cal ly usable range is largely dependent on the size shape, and number of crystalline regions, which in turn are dependent on the chain structure of the poly mer The incorporation of a branch point c auses conformational def o rmation s which depress the crystalline region s melting point. Therefore the melting point and heat of fusion of a polymer obtained by differential sc anning c alorimetry ( DSC ), is indicative of the percent crystallinity of the sample. Thermal studies of methyl branched pol ye thylenes have shown that as the percentage of comonomer ( methyl branches ) increases the melting point transition decreases as well as broadens as a result of more diffuse melting. 146 The amount of theoretical temperature depression for random copolymers was described by Flory 73 and is illustrated in equation 3 in the introduction. Studies using ethylene and propylene copolymer mixtures have demonstrated that the methyl substituted polymers closely follow this relationship and can therefore be fairly represented by the Flory equation.134,147,148 By using copolymers derived from the reaction of diazoamethane and diazoethane, Ke 1 3 4 reports melting points (Tm) of 134 C and 126 C and ranges of ~ l6 C and 20 C for sa mples with 0.6 and 1.7 methyl branches per 100 ca rbons, respectively. These values are close to what would be expected for these low co monomer mixtures. Similarly Grisky and Foster 1 4 9 obtained values of 130, 126, and 121 C for samples with 0.4, 1.1, and 2.6 branches per 100 ca rbons. The broadening of the melting peaks is clearly visible as the feed of the branch co -unit is increased, with the sample containing 2.6 branches disp lay ing a range of approximately 60 C. The crystallinity of these sa mples also decreases as the co-unit i s increased with the values of 88.4, 73.4, and 44 /o as determined by DSC, respectively. 14 9 As the comonomer is increased e ven

PAGE 131

122 further to a value of 3.8 methyl branches per 100 ca rbons, Hosoda et al 150 reports the lack of a detectable melt transition The se results are indicative of samples containing a broad distribution of chemical composition. This type of broadening b e havior was discussed in the introduction and is illustrated in Figure 1 .1 3. To better control the chemical composition of the PE copolymers, different polymerization techniques must be used. It should be considered that when comparing samples form one s tudy to another, that the broadening of the melt transition can be affected by many factors from the sc an rate used to the molecular weight distribution of the samples Better comparisons can be made within a sample group where the same polymerization conditions DSC techniques and thermal history of the polymers are used. Thermal Analysis of ADMET Methyl Substituted Polyethylene Thermal analysis of the polymers HP5a and HP5b was accomplished by the use of differential scanning ca lorimetry (DSC), and was used for a comparison of the melt transitions of these s ubstituted polymers with that of perfectly linear sa mples. Perfectly linear polyethylene samples prepared using ADMET are conducive to a high crystallinity due to their simplistic structure and inherent chain flexibility 100 In order to make reasonable comparisons, the same conditions were used to polymerize and characterize the branched and linear polymers. All samples had s imilar polydispersities and molecular weights. DSC analysis was performed by first scanning the sample at 10-20 C/ min over a wide range, followed by isotherming for 5-10 minutes above the transitions in order to erase the previous thermal history. The samp l e was then cooled a t a rate of 2 C / min to -25 C and iso thermed for 5 minutes

PAGE 132

123 followed by scanning the heating range at 2 C / min and isotherming f or 5 minutes above the transitions before c ycling back. All data were acquired after the second 2 C / min. cy cle with s ample s izes ranging between 5-10 mg. The addition of the methyl branch point along the polymer back bone was expected to reduce the polymers melting point. Polymer HPSa has a methylene run length of 8 between branch points (which corresponding to a degree of branching of 11.1 methyl branches for every 100 backbone carbons ) while polymer HPSb has a run length of 14 (which corresponds to a degree of branching of 6.7 methyl branches for every 100 backbone carbons ). If it is HPSa HPSb Figure 4.15. Maximum lamellae folds for HPSa and HPSb a ssuming methyl branch exclusion and a lamella crystalline structure.

PAGE 133

124 assumed that the branches are excluded from the crystal structure due to steric reasons and the crystallites are ordered in a lamella fashion 61 then the crystallization must occur with maximum run lengths (lamellea folds) of 8 or 14 respectively (Figure 4.15). Lamella-like c rystallites have been shown to be the fundamental structural feature of bulk crystallized homo and co-polymers even with a relatively high co-unit contents.58-60 l5l Further, it has also been s hown that there is a possibility that methyl groups can enter the c rystal lattice as an equilibrium requirement. 1 5 2 Further experimentation must be done in order to determine the true chain morphology. The DSC thermogram for polymer HPSa exhibited one first order transition at -2.01 C during the heating cycle (Figure 4.16) and an equivalent recrystallization curve with a peak at -16.31 C during the cooling cycle. These peaks were reproducible as demonstrated by the repeated cycling .:I: F 3. 0 u_ .., ,0 Ql :t 50 0 ~------------------------, 48 0 46 0 ,l 4 0 Heatin g 4 2 0 40 0 38 0 36 0 -25.0 20 0 H 3 IH3 ? CH 2 H Peak 16.31 C MI 29.61 J ig 15.0 -to o Temp e rature l"Cl Peak -2 .0 1 C CH 3 (\ MI 31.42 J ig n I \ J \ "--5 0 (JV PEA KIN -EL M EA Cooling o o I ' I I I I I I -, I l I 5 0 7 Ser 1e s Thermal Anal ysis System Figure 4.16. Thermal analysis of polymer HPSa at a scan rate of 2 C / min. The displayed scan is consistent with subsequent cycling.

PAGE 134

125 through the designated temperature range. Some initial shifting and narrowing of the curves was observed due to annealing of the sample. No effort was made to obtain maximum c rystallinity by controlling the crystallization co nditions 153 or by slowly approaching Tm during analysis 154 No other transitions were apparent in the range of -70 to 130 C. Polymer HPSb exhibited similar results with a higher melting point peak of 45.61 C (Figure 4.17). As expected, a decrease in the mole percent of the noncrystallizable branch units resulted in an increase in the temperature of the melt transition. This is the apparent result of longer run lengths of the crystallizable linear chain surface creating larger more thermally stable crystals as well as an increased overall ordering. This change is illustrated by the 51 C increase in the melting point that is observed by increasing the run length from 8 (HPSa) (11.1 methylenes per 100 carbons) to that of 14 carbons 50 0------------------, '18 C 46 C 44 0 C LL .., :o 42 C ., :c 40 0 38 C 36 0 Peak 45.61 C L\H 49.16 Jig ------'--------=:; -\ / Cooling 36 0 \ I \ / \ I Peak 37 .80 C .J dH-49.04 Jig 38.0 40 0 I 42 0 44 0 emoeratu re ( :) dv 14 H n 46 .0 48 .0 50 0 t~ IIW t 0 0 C 1 y 2 0 "' P AT~t. T' CIIIIP2: :io o C z o C / nnn PEAKIN-ELMEA 7 S eries Thermal Analysis Syste m T ue Mar 4 21 : 0 9 : 4B 1997 Figure 4.17. Thermal analysis of polymer HPSb at a scan rate of 2C/min. The displayed scan is consistent with subsequent cycling.

PAGE 135

126 (HP Sb) (6.7 methylenes per 100 carbons). The thermal transitions of this sample were also reproducible v ia cycling and showed an equivalent cooling peak at 37.80 C The amount of temperature depression due to random incorporation of a copolymer unit can be calculated by use of the Flory equation (equation 3 Chapter 1). 73 Using the Flory equation and the well-known melting temperatures of orthorhombic n-alkanes from Broadhurst,1 5 5 Gerum et al solved the regression function for l which connects the fusion peak temperature with the number l of C-atoms in the chain of n-alkanes that crystallize in the lamellae as polyethylene does ( equation 6). l = 0.01648/(1/Tp 0.0024096) (6) If it is assumed that methyl branches are excluded from the lamellae, the run lengths for HP5a and HP5b are 8 and 14, respectively. If a run length is 8 and equation 6 is solved for Tp a predicted value of -49 C is obtained. The difference of 47 degrees is assumed to be due to the highly chemically ordered substituted system of HP5a because the Flory equation is based on random copolymerization methods. The calculations using a run length of 14, predicts the peak melting temperature to be at 6 C; a difference of 40 degrees compared to the experimental value from HP5b. The predicted temperature depression by the Flory equation is similar to results obtained in the literature using random copolymerization methods that incorporate a methyl branch point.134,146,14~ This unique ordering phenomenon is exhibited by the sharpness of the thermal transition of HP5a and HP5b. The melting transitions occur over a range of roughly 5 degrees, again the apparent result of the lack of chemical

PAGE 136

127 composition distribution. Endotherms displaying this characteristic indicate a population of crystals with very s imilar t hermodynamic stabilities. 15 6 As mentioned earlier, the branched polymers generated v ia the ADMET condensation are the result of a homopol y merization where the branched co monomer unit is incorporated in the ADMET monomer. Therefore the resulting condensation products do not have homo-monomer run lengths that are free of branching, as do the random Ziegler type systems. This is the essence of the unique control granted over the chemical microstructure ( exact repeat unit structure) provided by the use of the ADMET polymerization mechanism. The existence of linear run lengths along the random copolymer materials (chain addition polymers) is assumed to generate their exhibited crystallinity. The randomness of these branch-free run lengths (linear PE segments), using Ziegler catalysts, forms chains that can form multiple size and types of crystals. This significantly contributes to the broadening of the melt transition as a higher order comonomer is added It has been found that the degree of crystallinity of random copolymers decreases very rapidly with the increase of side group content. 1 5 7 This behavior is a result of a decrease in the length and number of crystallizable sequences as the co-unit content is increased. 73 From the DSC thermograms the enthalpies of fusion were converted to degrees of crystallinity by adopting a relatively simple method of dividing the experimentally determined heat of fusion by 293 J / g, a value determined from long chain alkanes or of pure crystallized polyethylene. 158 As displayed in table 4.3 the degree of crystallinity drops ~50/o with the addition of a regularly spaced methyl branch every 15 carbons followed by an additional 6 o/o with a branch every 9 carbons. The observed trend is similar to those

PAGE 137

128 displayed by copolymerized s amples but a larger sample size is needed before a relationship can be developed. Table 4 3. Analysis of polymers HP Sa, HP Sb and AD MET linear PE. Polymer Mn(GPC) Mw(GPC) Mw/Mn PDI HPSa HP5b Linear PE 30,600 13,500 15,000 41,500 25,000 40,000 1.4 1.8 2.6 ~Hm (J/g) 31.42 49.16 204 O/o crystal DSC 11 17 70 GPC v a l ue s a r e r e lati v e t o p o ly s tyrene standard s. Tm C Branches (per 100 C) -2.01 45.61 133.9 11.1 6.7 0 Number-average molecular weight g / mol ( Mn), weight-average molecular weight g / mol (Mw ), Polydi s persity index ( PDI), Heat of fusion Vs indium and cyclohexene s tandards Mim % crystallinity as determined by DSC ( LlHm / 293) first order melt transition temperature taken at the peak of the curve ( Tm C ), # of branches per 100 backbone carbon atoms (100 / 9 = 11.1 ). Conclusions The use of the ADMET polymerization mechanism has proven to be a viable method o f cleanly producing a series of branched unsaturated polyethylene prepolymers with a well defined structure. Using the Hahn modification of the diimide hydrogenation, the unsaturated groups of the prepolymer can be quantitatively removed ( by NMR analysis) without detectable chain decomposition or loss of backbone integrity. The resulting saturated polymers show a high order of regularity as indicated by spectrometric t e chniques This type of branch distribution regularity in branched PE samples has presentl y not been obtained by any other method. Therefore these samples are ideal with regard to well defined c hain branching and perfectly co ntrolled branch distribution for the investigation

PAGE 138

End o HP5a C H 3 H 3 n-f? -( C H 2 C H 3 H Peak -6 40 c Lili 32 18 J /g 0. 0 25 0 n 5 0 0 HP5b P e ak 4 5 6 c ~ H 4 9. 1 6 J /g 7 5 0 T emperature ( c1 Lin e ar P o l ye th y l e n e n Pe ak 13 3. 9 c ~ H 204 J /g 10 0 .0 \2 5 0 1 29 Figure 4.18. Thermal analysis of perfectly linear polyethylene and the methyl substituted models HP5a and HP5b. All scans were performed at the same rate (2 C / min.) and with equivalent thermal histories. Heat flow scale is arbitrary of the influence of short chain branching on material properties. These samples have a lso displayed unique thermal characteristics that can be attributed to their regular repeat unit s tructure ( Figure 4.18) The addition of the methyl branches results in a significant drop in the melting point of the crystalline regions and the maintenance of a narrow transition range as compared to the similarly prepared linear s ample. These results suggest that polymers prepared by this synthetic approach may allow for the systematic determination o f the effect of branch placement and branch identity on crystallite structure, morphology and the thermodynamics of crystallization processes.

PAGE 139

CHAPTERS THE DIRECT SYNTHESIS OF WELL-DEFINED ALCOHOL FUNCTIONALIZED POLYMERS VIA ACYCLIC DIENE METATHESIS (ADMET) POLYMERIZATION The direct conversion of vinyl alcohol into its corresponding polymer and copolymers is currently not possible because the monomer does not exist, or, at least, not long enough. The equilibrium between vinyl alcohol and its keto form, acetylaldehyde, lies heavily in favor of the aldehyde (K=[enol]/[keto] 3 x lQ-7)159, resulting in a lifetime that is inadequate for addition chemistry (Figure 5.1). Novak and Cederstav 1 60 attempted to extend 0 OH H Slow Figure 5.1. Keto-enol tautomerization of vinyl alcohol. the life of the enol by the generation of the 0-D vinyl alcohol via the hydrolysis of ketene methyl vinyl acetal (Figure 5.2). The deuterated enol appeared metastable at room temperature but did not undergo cationic or radical homo-polymerization. Polyvinyl alcohol and its copolymers (typically ethylene-vinyl alcohol copolymers) are commercially prepared by the free radical polymerization of vinyl acetate, forming the acetate polymer as a prepolymer to polyvinyl alcohol (Figure 5.3). Like the radical polymerization of ethylene, 130

PAGE 140

131 90 % 0~ +MeOH Figure 5.2. The deuterium hydrolysis of methyl vinyl acetal. acetate copolymers undergo branching due to intramolecular chain transfer 21 22 (Figure 1.5), and can be considered to be copolymers of low density PE (LDPE) and vinyl acetate. The prepolymer (acetate homologue) is IH3 H3 C 0 C 0 I I 0 R + CH2 I CH Cl-f2 CH N polyvinyl acetate OH I CH2 CH--+N polyvinyl alcohol Figure 5.3. The radical homopolymerization of vinyl acetate followed by alcoholysis to form polyvinyl alcohol. converted to the secondary alco hol via hydrolysis or alcoholysis of the ester functionality (Figu re 5.3).

PAGE 141

1 32 Alcohol Functionalized Polymers via Metathesis The use of metathe s is to synthesize f unctionalized polymers has grown significantly with the advent o f new c atalyst s ystems that displa y a higher tolerance to fun c tional groups 1 6 1 162 166 Since the c atalysts typically used in these systems are inherently Lewi s a cids, the metathesis of variou s olefins c ontaining Lewi s basic functional groups (specifically alcohol) have been unsuccessful. To avoid this problem, functionalized metathesis polymers have been made by protecting the functional group prior polymeri z ation. An example of thi s was shown by Chung et al who produced alcohol functionalized polymers through the ring opening metathesis polymerization ( ROMP) of boron functionalized monomers followed by the oxidation of the boron containing polymer (Figure 5.4) 1 6 3 This method produced an unsaturated polymer that contained the hydroxy functionality in well-defined s equences. Absolute placement did not occur due to placement isomerism c au s ed by the ROMP o f a n unsymmetrical monomer which resulted in the random head to tail ( HT ), tail to head ( TH) e tc. monomer connections. Hydroxy functionalities can also be placed on hydrocarbon metathesis polymers by taking advantage of the unsaturated group that i s inherently maintained in the repeat unit. Ramakrishnan 1 6 4 reported the synthesis of ethylene-vinyl a lcohol c opolymers by hydroboration o f the metathesis polymer s olefin g roups followed by the oxidation o f the boron moiety to an

PAGE 142

W (ca t ) N OH N Figure 5.4. The indirect synthesis of polyalcohols via the ROMP of boron functionalized monomers. 133 alcohol group. This method also provided considerable control over the polymers sequence distribution, but the random nature of the hydroboration reaction (1,2 or 2,1 addition) causes the hydroxide functionalized product to be somewhat isomerized (Figure 5.5). The interest in synthesizing these types of ROMP X=2,3 ,5 ,6 10 CH X N CH X N OH 9-BBN B Figure 5.5. The indirect synthesis of polyalcohols via the hydroboration of ROMP hydrocarbons. N

PAGE 143

134 polyalcohols is driven by the need to develop models with well-defined composition and distribution of functional g roups for a better understanding of their st ructure-propert y relationships. The selection of the alcoho l functionality is inspired by the co mmercial s ignificance of ethylene-vinyl alcohol copo lymers. 165 Fu et al.166 have recently introduced a well defined ruthenium alkylidene C2 (Fig ure 5.6) that is metathetically active in the presence of a wide va riety of heteroatoms--specifically alcohols. The more recent ruthenium benzylidene ( C3 Figure 5.6) displayed s imilar tolerances to Lewis bases as well as a higher activity. 1 6 7 Specifically, a lkylidene C2 was R3 Ar iR3 Cit,,, Cv,,, ,. ,.R R Ar c~ I c~ I PR3 PR 3 C2 C3 Grubbs Vinyl Alkylidene Grubbs Benzylidene Figure 5.6. Ruth e nium based metathe s is ca talysts. R= cyclohexyl or phenyl. successfully used for both the metathesis cyclization, 166 and the ROMP of alcohol containing monomers. 1 68 Hillmyer,1 68 using catalyst C2 Figure 5.6, demon s trated the direct synthesis of alcohol functionalized polymer s via the ROMP o f 5-hydroxy-1-cyclooctene (F igure 5.7). The limitation of this method for modeling purposes is the lack of complete co ntrol over the substituent locati o n as well as the co nstraints of sy nthesizing ROMP monomer s to va ry polymer structure. The frequency of a substituent in ROMP polymers is limited by the need for s trained cycloolefins as monomers to achieve a high

PAGE 144

135 degree of polymerization. Symmetry of the cyclic olefins i s also important to control the exact location ( regular frequency) of the substituent along the polymer backbone, which is difficult to achieve in ROMP monomers. OH OH Ru cat 65 % N Figure 5.7. The direct ROMP of 5-hydroxy-1-cyclooctene. The Direct Synthesis of Alcohol Functionalized Polymers via ADMET Due to the observed alcohol tolerance and the success in using it to produce polymers via acyclic diene metathesis (ADMET), 9 9 catalyst C3 was selected to study the polymerization of a series of alcohol containing dienes. This method provides a unique pathway toward the direct synthesis of alcohol functionalized polymers with well-defined structures. Alcohol functionalized terminal dienes were exposed to Grubbs ruthenium phenyl alkylidene C3 (Figure 5.6), resulting in their condensation to form their analogous unsaturated polymers (Figure 5.8). These alcohol functionalized polymers can then be l1ydrogenated producing polymer structures that are similar to a poly(ethylene) / poly(vinyl alcohol) copolymers. The advantage of using ADMET as the synthetic method is that various derivatives of alcohol containing polymers can be investigated with precise control over monomer sequence distribution. The range of functional group placement and identity, is only limited by the synthesis of the

PAGE 145

OH Ru ADMET c at n n A R= Hor variable length alkyl goups Figure 5.8. Goal: The direct metathesis condensation of alcohol funcionalized dienes. 136 X + corresponding acyclic terminal diene. This structural control can be used to model the substituent effect (branch and or functional group identity and frequency) on the physical properties of the polymer. The success of this polymerization provides the first means towards the direct synthesis of aliphatic polyalcohols via condensation chemistry. Monomer Design and Synthesis Primary Al, secondary A2 and tertiary A3 (Figure 5.9) monomers were considered for this study. All the monomers were designed to be symmetrical and have a minimum of two methylene spacers between the functional group and the olefin. The former was done to avoid unsymmetrical placement of the alcohol group ( placement isomerization) and the latter done to avoid the possibility of experiencing the negative neighboring group effect (NNGE). The NNGE is promoted by the ability to form a five member ring, or other favorable cyclic co mplex, between the l o ne pairs on the functional group (Lewis base), in this case the alcohol, and the electrophilic metal ce nter (Lewis acid) of the catalyst. Similar intramolecular Lewi s base binding effects have been reported fo r ROMP sys tems ,169,l70 as well as the co nsi ste nt

PAGE 146

137 demonstration of this behavior with ADMET monomers. 98, 171 Wagener and Brzezinska have demonstrated that this poisoning" effect c an be rendered innocuous by sufficient carbon spacing (typically two or more methylene spacers) between the electron rich functional g roup and the metal c enter of the alkylidene. 17 2 The length of the alkyl group (which is 5 for the alcohol monomers) provides an extended diene structure, therefore discouraging the formation of cyclic condensation products. Shorter dienes (7 a nd 8 carbons long) produce favored 5 and 6 membered rings, respectively, v ia ring closing meta thesis (RCM). I H (Al) I H (A2) (A3) Figure 5.9. Primary, secondary, and tertiary symmetrical alcoholic diene monomers for ADMET. Tertiary alcohol A3 (Figure 5.9) was s ynthesized by forming the Grignard reagent of 5-bromo-1-pentene, which was then reacted with ethyl acetate (Figure 5.10). Using ethyl acetate as the ester reagent produces the methyl branched alcohol. The monomer was then distilled under reduced pressure to achieve a purity of > 99 /o in order for it to satisfy the stringent conditions required for ADMET polymerizations. The advantage of this simple monomer synthesis is the ability to design a whole series of tertiary alcohol monomers with precise control over the identity and frequency of the alkyl substituent and the alcohol. For example the alkyl substituent (branch point) can be easily changed by c hanging the identity of the ester ( higher

PAGE 147

138 order), an d s imilarly, the frequency of the substituents can be controlled by c hangin g the identity of the alkenyl bromide (F i g ure 5.10). sr--(c H,.J---' /1/ n + Mg B rM o---tCHrl----' /1/ n 1. 0.5 EtOCOR R = H ( A2) or CH3 ( A3 ) 84% Figure 5.10. Synthesis of secondary and tertiary alcohol functionalized diene monomers. The synthesis of the secondary alcohol A2 was performed in much the same manner as the tertiary A3 (Figure 5.10). Two equivalents of the Grignard reagent from 5-bromo-1-pentene were reacted with one equivalent of ethyl formate resulting in the formation of the secondary alcohol in high yield. As with the tertiary alcohol, the frequency of the functional group can be precisely controlled by varying the length of the Grignard reagent. This monomer was first purified by distilling under reduced pressure until a purity of greater than 99/o was achieved by gas chromatography. Due to the poor observed reactivity of this monomer (A2) with the catalyst, f urther a nalysis of monomer purity was explored using thin layer chromatography and 1 H NMR. Using silica co ated plates with a 5/o ethyl acetate pentane mobile phase, two distinct spo ts were observed. These spots

PAGE 148

139 were isolated by s ilica c olumn c hromatography using 1 /o e thyl acetate / pentan e eluent. This separation resulted in two pure products identified by 1 H NMR and GC / MS as the secondary alcohol (A2) and the aldehyde (mon o ene). Neat s amples of these pure compounds displayed relatively the same retention times using GC. This result suggests that the monoene and the diene, unexpectedly, have very similar boiling points, therefore rendering vacuum distillation and gas chromatography inadequate separation techniques. The incorporation of the monoene in the monomer feed would greatly reduce the number of conversions in this step condensation polymerization v ia stoichiometric imbalance. Preparative column chromatography provided adequate separation and purity for ADMET polymerization. The primary alcohol A 1 was synthesized as described in chapter 3 for compound 3a. The multistep synthesis was pursued (Figure 5.11) involving enolate chemistry on beta-ketoesters, or more specifically, the one pot disubstitution of ethyl acetoacetate forming compound la. These types of dialkylations can be found in the literature using a variety of bases and conditions (see chapter 3 and corresponding references). Compound la was then deacylated using sodium ethoxide (Step 3 Figure 5.11) forming the disubstituted ester 2a. The final step (step 4 Figure 5.11) was the reduction of the ester using LAH to the primary alcohol. The target alcohol Al (which corresponds to 3a from chapter 3) was purified by distillation under reduced pressure ( ~ 102 torr) until a purity greater than 99 o was achieved by GC. Refer to chapter 3 for an extended discussion on all four steps.

PAGE 149

C0 2 Et I CH 2 I C=O I CH3 ethyl acetoacetate H Al (3a) 1. K + t -BuoBr 2. K+ t-BuoBr Steps 1 & 2 1 LiAIH~HF 2. H+, H20 Step4 Step 3 C02Et I + KBr I y=O la CH3 1 EtOH / Na+ -oEt I H 2a Figure 5.11. Four step synthesis of primary alcohol diene Al. ADMET Polymerization of Hydroxy Functionalized Dienes General ADMET Polymerizations 140 The polymerization of the alcohols was accomplished by the unification of a 500:1, 300:1 or 200:1 monomer to catalyst ratio in an inert atmosphere. All monomers were dried and degassed prior to use. Upon the addition of the catalyst, very s low to moderate bubbling was observed in each case as the ethylene side product of productive metathesis was released. Soon after this addition the mixtures were exposed to agitation, vacuum, and a slow increase in temperature (up to 70 C) until a very high viscosity was obtained and bubbling ceased. The reactions of all three monomers typically required two to three day s for co mpletion Higher molecular weight sa mples

PAGE 150

141 were produced using higher catalyst loadin g ( 200:1 monomer / catalyst), which is consistent with results obtained v ia the ADMET polymerization of hydrocarbon monomers with Grubbs' ruthenium benzylidene. 99 Table 5.1. Gel permeation Chromatography ( GPC) results of the alcohol functionalized polymers. Polymer mon: cat Mn (g/mole)G PC Mw(g / mole ) GPC POI Primary ( PAl) 500:1 6,300 8,200 1.3 Primary (P Al) 200:1 24,000 38,000 1.6 Secondary ( P A2 ) 500:1 10,100 13,400 1.3 Secondary ( P A2 ) 200:lb 9,300 12,600 1.4 Secondary ( PA2 ) 200:1 18,600 30,400 1.4 Tertiary ( PA3 ) 300 :1 13,600 18,600 1.4 Tertiary (P A3) 200:1 13,000 17,700 1.4 Note: Samp les run in 80% chloroform 20% methanol against polystyrene standards (b ). Presence of rnonoene impurity discovered Polymerization of 6-(4-Pentene)-l-Heptene-7-ol (Al) The resistance of catalyst C3 to alcohol functionality was displayed via the productive ADMET polymerization of the primary alcohol Al ( Figure 5.12). The primary alcohol diene was polymerized twice using monomer to catalyst ratios of 500:1 and 200:1 (Tab le 5.1) In both cases, the slow evolution of ethylene (bubbling) was observed upon additio n of the catalyst, followed by a color change to yellow orange and a steady increase in viscosity. Initially, the monomer was a colorless liquid and the catalyst a purple solid Intermittent vacuum was app lied for the first 2 hours or until the viscosity noticeably increased. When the increase in viscosity was observed, high vacuum was applied (> 104 torr) resulting in a dramatic increase in the rate of ethylene evolution (bubbling). These reactions proceeded for 48 to 72 hours with a slow increase in temperature until 70 C was obtained. When

PAGE 151

142 bubbling of ethylene had c eased the reactions were c ooled to room temperature and quenched with air. OH I CH2 6~ H Al OH I~ H A2 OH ........ '~ I CH3 A3 Ru Ru OH I CH2 I C H 9H C I H OH I C ...... ,,, I CH3 n PAl PA2 n PA3 n Figure 5.12. The acyclic diene metathesis polymerization of primary, secondary and tertiary alcohol functionalized dienes. The polymeric product from the first reaction (cat:mon = 500:1) was a very viscous liquid. These characteristics can be attributed to the lower observed molecular weight. When the ratio was 200 : 1 the material was a clear tacky, fiber forming solid. All samples were characterized by NMR, GPC and elemental analysis and were found to be consistent with their assigned structure. The lower molecular weight sample, (Table 5.1), was highly soluble in common organic solvents at room temperature, with little to no water solubility observed. The higher molecular weight sample (Table 5.1) displayed poor s olubility in common organic solvents, but displayed limited

PAGE 152

I H OH I CH2 I C I H Al H PAl H H H .. 7 6 7 6 .. 5 4 3 n 5 4 3 2 1 2 l 143 0 pprr 0 p pm Figure 5.13. lH NMR end group analysis showing the high, and clean conversion of the primary alcohol diene Al to the linear unsaturated alcohol functionalized polymer P Al

PAGE 153

144 solubility in chlorinated solvents. A solvent mixture of 80/o chloroform and 20/o methanol was determined to be the best solvent mixture for this system. The productive ADMET condensation proceeds with the conversion of external olefins into internal olefins. When a high degree of polymerization is obtained the concentration of internal olefins with respect to the external olefins drastically increases. The resulting NMR spectra indicates this occurrence by the disappearance of the external olefin resonances and the appearance of a strong internal olefin resonance (Figure 5.13). The multiplets at 4.9 ppm and 5.8 ppm are the external olefin proton resonances of the pure monomer, which are subsequently converted to the single internal olefin resonance at 5.4 for the high molecular weight polymer PAl. The clean nature of this polymerization mechanism is also illustrated by the absence of unexpected resonances in both the 1 H and l3C NMR. The spectrum of polymer PAl (Figure 5.13) was taken from a crude non precipitated aliquot of the quenched reaction mixture. A similar observation can be made by comparing the infrared spectrum of both the monomer and polymer (Figure 5.14). The disappearance of the terminal olefin signal at 1640 cm1 is also a common observation for high molecular weight ADMET polymers. The IR also shows the preservation of the alcohol functionality in the polymer structure (broad signals at 3300 cm-1 range, Figure 5.14). These observations can be further supported by the GPC analysis which showed a Mn value of 24,000 for this sample. The elemental analysis, based only on the repeat unit, provided additional evidence to support the assigned structure of this linear polymer.

PAGE 154

+ CH 2 =CH2 OH C I H Al H PAl H H H 80 % 70 T r a 60 n >--H j t so t a n c e 30 4000 60 % T r a n SO 6 j t t 40 a Il C e 30 145 350 0 3000 2500 2000 ]500 1000 500 llavenumbers 3500 3000 250 0 2000 1500 1000 500 ~avenumbers Figure 5.14. IR transmittance spectra of Al and PAl showing the disappearance of the terminal olefin s ignal at 1640 cm1 and the retention to the alcohol functionality (330 0 c m-1). Polymerization of 6-Methyl-1,10-Undecadiene-6-ol (A3) The tertiary alcohol monomer was co ndensed twice u si ng two monomer to catalyst ratios ; 300 :1 and 200:1 (Ta ble 5.1). The polym e rizations

PAGE 155

146 proceeded in the same manner as those obs erved for A 1. Visually the increase in v isc os ity of the tertiary monomer occurred at a faster rate than that of the primary. This ma y be due to the more s teric crowding of the potential negative neighboring group ( tertiary al cohol ). The resulting polymers were clea r tacky fiber forming so lids that were l-lighly s oluble in organic s olvents with little to no room temperature water solubility. As is the case for ADMET polymers the percent conversion of the polymer can be calculated from the 1 H NMR. The terminal olefin end groups of P A3 were concentrated enough to undergo end group analysis as a result of the lower obtained molecular weight of the polymers. The GPC analysis of the polymer (Table 5 1) was consistent with the NMR spectroscopy, showing a M n of around 13,000 for both runs. As with polymer PAl the 1 H and 13C NMR spectra showed a clean conversion from monomer to polymer. These spectra, as well as elemental analysis based solely on repeat unit, also provided good evidence to support the assigned structure of this polymer. Polymerization of 1,10-Undecadiene-6-ol (A2) The polymerization of the secondary a lcohol A2 provides a unique opportunity to directly synthesize a structurally defined unsaturated version of an ethylene-vinyl alcohol copolymer. As mentioned earlier, due to the nature of the ADMET mechanism (condensation chemistry), the exact location of the alcohol along the polymers backbone is determined by the design of the s ymmetrical acyc lic diene monomer. Furthermore, hydrogenation o f this well-defined unsaturated polymer will produce a s tructurally perfect copolymer of vinyl a lcohol and ethylene. The hydrogenation of this polymer will be discussed in a future section.

PAGE 156

147 The secondary a lcohol diene was a lso polymerized using t wo monomer to catalyst ratios ; 500:1, and 200:1 ( Table 5.1 ) The polymerization proceeded as outlined before. This reaction mixture ( A2 + C3) was more yellow in color than were the primary and the tertiary alcohol monomer catalyst mixtures The resulting product appeared to be considerably different than P Al and P A3 in that it was a non-fiber forming opaque waxy solid. Contrary to the primary and tertiary alcohol polymers, polymer P A2 had little to no room temperature solubility in common organic solvents even though the molecular weights, M n' s GPC for the first two reactions were a modest 10,000 g/mole. It was determined that the monomer feed mixture in the first two runs (Table 5.1) was contaminated with residual aldehyde monoene. This was confirmed by a singlet proton signal at 8.1 ppm in both the monomer and the polymer. This monoene will inherently limit the molecular weight of the polymer by capping off the end of the stepwise growing chain. Consequently, lower molecular weights were observed for the first to polymerizations of monomer A2. The polymers from both these samples showed a significant amount of an aldehyde signal which indicated the presence of the monoene. As an aside, it can be noted that the presence of an aldehyde functional group did not appear to shut down productive metathesis. The possibility that the presence of the monoene limited the molecular weight of the polymer was investigated by the purification of the monomer v ia column chromatography, followed by exposure to the C3 (PA2 Table 5.1). This addition resulted in vigorous bubbling with a noticeable increase in viscosity within 1/2 hour. Visually, this was considerably faster than the first two polymerizations, where a similar change was noticed after 2 hours. Further experimentation must be done to quantify this observation. A new

PAGE 157

148 overall purity may be the result of this observed ac celeration. After 24 hours a 1 H NMR was performed on an aliquot of the sa mple showing greater than 85/o conversion by end group analysis. The polymerization was terminated after 72 hours resulting in undetectable end groups by 1 H NMR This was confirmed by GPC were a number average molecular weight of 18,600 g/ mole was obtained Molecular weights (Mn and Mw) were roughly twice that of the first two contaminated samples. Some limited solubility was detected in chlorinated solvents, and as before an 80 / 20 chloroform/methanol mixture proved to be a good solvent system. Since the ADMET polymerization is the condensing of acyclic dienes to a linear polymer, direct comparison of the monomer and polymer s carbon NMR spectra demonstrates the clean nature of this conversion ( Figure 5.15). The spectrum for polymer P A2 was taken directly from the reaction mixture with no further purification (Figure 5.15). The 1 H NMR peak placement and integrals are what is expected from the repeat unit structure represented in Figure 5 12 and 5.15. Comparison of the 13C and the lH NMRs of PA2 with that of the unsymmetrical ROMP polymers, 1 63 ( Figures 5.4 and 5.8) shows a s impler structure due to the lack of detectable placement isomeri s m. The secondary alcohol is inherently more linear in nature which could encourage better packing as well as an increase in secondary c hain interactions In the case of P A2, its decrease in solubility and its waxy white appearance may be a direct result of this induced order. Therefore this polymer structure may not only be able to undergo secondary a nd tertiary hydrogen bonding, but may also be somewhat crystalline in nature. This is consistent with observations made by Ramakrishnan and Chung of poly(5hydroxyoctenylene) via ROMP (Figure 5 4). 163

PAGE 158

H OH I C I H A2 PA2 149 ,,, ,,-. 140 1 ,0 120 110 100 90 80 7 0 &O SO 40 30 20 10 0 n I Ii/fr" f lliiJiiillii I i ""''I", +1. 1r 11 "(' 111111 1111 iilllf!rllllii'lill! liltjilliiiiiijfritltflfl jlll liifljfliiiillllfll ppm 140 1 )0 120 110 100 to 80 70 to 50 40 30 20 10 -0 Figure 5.15. The 1 3 C spectra of the ADMET polymerization of the secondary a lcoh ol diene A2 to it's corresponding linear polymer PA2.

PAGE 159

150 Hydrogenation of Poly(6-hydroxynonenylene) P A2 PA2 was exposed to the diimide hydrogenation via toluenesulfonyl hydrazide (TSH ) and tripropyl amine (TP A). The same methodology as in chapter 4 was used with the exception of using toluene as the solvent in place of a-xylene. Both the polymer and the TSH dissolved while heating followed by the evolution of gas (N2). TSH and TPA were administered twice in 2eq concentrations (2eq of diimide per mole of olefin) after 3 hour periods. Upon cooling to room temperature there was an almost immediate precipitation of white solid. The solid was recovered by precipitation from a warm solution into cold methanol followed by drying under vacuum at 50 C. A hard white, granular solid was recovered and characterized by high temperature lH and 13c NMR. GPC analysis was not performed on this sample due to the insolubility of the solid at room temperature. It can be assumed via literature results 124 and the soluble hydrogenated polymers in chapter 4 (HPSa and HP Sb), that there was no polymer decomposition due to the diimide hydrogenation. Further characterization and purification of this solid i s required. The conversion from the unsaturated polymer (PA2) to the saturated homologue (HP A2) proceeded as expected and is demonstrated by proton NMR ( Figure 5.16) The spectra demonstrate the disappearance of the internal =CH signal at ~5.6 ppm as well as the allylic C-H signal at ~2.2 ppm. The retention of a single methine proton at ~ 3.7 ppm indicates the absence of isomers as well as the retention of the alcohol functionality The absence of placement isomerism cannot be directly inferred from this spectrum due to the simpler structure of the hydrogenated sample and the distance between the hydroxy substitution. Ramakrishnan 1 6 4 report s that the alcohol segment

PAGE 160

PA2 I ' I I I I I j ,. I I I v' \ u I I I ... I 7 6 5 3 1 -0 ppm HPA2 n * 1 ) Figure 5.16. lH NMR of the unsaturated polymer P A2 and the diimide hydrogenated HP A2. ( *) Residual toluene 1 51 must be le ss than a 1,6 diol for detection of different environments by NMR. An example of this detection was s hown from the h yd roboration / oxidation

PAGE 161

152 of 1,4-polybutadiene, where 3 diol se gment s can be detected ( 1 ,4 1,5, a nd 1 ,6 in a ratio of 1:2:1 ) 1 64 The s ymmetry of the A DMET monomer preve nts placement i so merism therefore resultin g in a be tter defined microstructure. The ca rbon N MR co nfirms this si mple s tru c ture with o nly 5 c arbon environments ( 25.89, 29.83, 29 .9 0 38 00, 7 1 .63 ppm) as well as excellent spectroscopic ag reement with Ramakrishnan s 164 ethylene-vinyl alcohol copolymer in Figure 5 5 were x = 8. Thermal Analysis of the Alcohol Containing Polymers Prior to thermal investigations polymers PA1-PA3 were not precipitated or purified in any manner therefore the presence of impurities (low molecular weight molecules and catalyst) must be considered during PA2 1 0 0 0 -r--=~=====~--------------=~---, ......_,_ ..,._ PA3 9 0 0 BO 0 / 0 0 .. ..., 6 0 0 ...; g, 50 0 .. J: 40 0 ~o o 20 0 10 0 '\ -~ \ :---.. ~ PAI \ \ \ \ \\ \ \ \ \ \ \ \ \ \ \ \ \ 00 ~--,~ -.------= ::=~=!==.,---.--.--J 0 0 20 0 0 4 00 0 60 0 0 T empe r aturt') (' C J a o :> o :1 V OE AK!N EL M ER 1000 0 7 Se r ies Ana l ysis Svste m "4 o n .Nn 9 19 44 10 1 9 97 Figure 5.17. Thermal gravimetric analysis of polymer PA1-PA3 at a scan rate of 10 C/min under N2 purge.

PAGE 162

153 interpretation. Thermal gravimetric analysis (TGA) was used to determine the thermal stability and decomposition pattern for polymers PA1-PA3 (Table 5.2). Figure 5.17 displays the thermograms for a ll three polymers which were scanned at a rate o f 10 Clmin under N2. Table 5.2. Thermal analysis of alcohol containing ADMET polymers. Sample PAl PA2 PA3 HPA2 24,000 18,600 13,000 Tgb (OC) -11 -10 13 NIA Tmc (OC) NIA NIA 120 437 413 234 430 (a) relative to pol ysty rene s tandard s in a 20/80 methanol / chloroform so lvent mixture, (b) 20 C/ min scan rate transition taken from second scan of thermal cycle, (c) peak temperature from first order endotherntic transition with a scan rate of 20 C/ min, ( d) onset temperature at 10 C/ min sc an rate, (e) first order transition detected in first sc an only. The polymers P Al and P A2 thermal decomposition did not begin until temperatures approached 250 C, at which point a slow loss of weight until 410 C was observed followed by a rapid 100 /o weight loss. Polymer P A3 also began a two stage weight loss at 250 C but proceeded at a faster rate than the primary and secondary alcohols. P A2 displayed a rapid weight loss at 413 C which is consistent with that reported by Chung for poly(5-hydroxy-1octenylene)163 (F igure 5.4), where the onset decomposition rate is at 430 C. The same polymer directly synthesized by Hillmyer (Figure 5.8) showed a decomposition temperature of 386 C. It should be noted that the molecular weights of these ROMP sa mples are ~ 10 times that of the ADMET samples.

PAGE 163

154 The two s tage weight loss exhibited by the polymers PA1-PA3 may be the result of a co mbination of the degassing of impurities a nd or the thermal elimination of water. An example of this is that polymer PA2 loses ~ 15 /o of its weight between 200 and 410 C followed by the sharp decomposition to 100 /o weight l os s. This result is co nsistent wi th the thermal elimination of H20 (13/o of the repeat unit weight) in the first stage followed by complete decomposition. The DSC transition temperatures of the polymers are listed in Table 5.2. Samples PAl a nd PA3 only displayed glass transitions at -11 and 13 C, respectively. This transition was observed in both the co oling and heating cycle and was reproducible on subsequent sc ans. The first scan of P A2 displayed both a glass transition at -10 C and a first order melt at 70 C (Figure 5.18). The melt transition was only visible on the first scan, with He at FlJW (m W ) s 2 58 0----------------------, 57 0 56 0 55.0 E :s 0 54 0 .... ..., "' 53 0 52 .0 5 I 0 50 0 H Tm 70 C n I PA2 Tg -10 C d9 0 -<------,----.---~--.-----,------,-----,-----1 I 75 0 -25.0 Te 11pe ra ture i C) 25 0 75 0 a v ~Ef:I( I "-ELME>1 7 S eries TrerraJ Analysis System W ea Jvn ll 23 G7 38 1997 Figure 5.18. DSC heating scan of PA2 a t 20 C / min before thermal history was erased.

PAGE 164

1 55 s ubsequent scans only showing the glass transition. This melt was thought to be due to the a nnealing of the s ample during polymerization at around 70 C, fo ll owed by s low cooling. In order to investigate the possibility of c rystallizable im purities the sa mple was reprecipitated from water, followed by drying in vac ue for 12 hours at 50 C. This repricipitated (also so mewhat annealed) sample displayed the sa me melt transition of 70 C This data s uggests that a sca n rate of 20 C / min is too fast to allow for recrystallization of the sample. Fu rther quenching experiments should help confirm this assumption. The TGA of HP A2 showed a rapid single stage onset of decomposition at 430 C, which is approximately 130 C higher than that of poly (vinyl alcohol) (PV A rapid ly loses weight at 300 C). 1 63 The DSC curve of HP A2 shows a first order transition with a peak temperature of 120 C, which is 70 : ES O., 66 0 j 6 4 0 X e l: 52 0 0 I60 0 ... ., C1J 58 0 r 5-6 0 54 C 52 0 7 50 :l JB C he 25 CC / 'I\Jn ,~ _..., f UCJ T E. 111 : 02: ..0 (l ; OH I C C H I 8 n H HPA2 25 0 50 0 !I 0 ~ 1' ''1 ~ e.1 ~o o Cl ,., '"' He at Flo w l ~ W l s : 6 75.0 Te1110eralure t c, Tm 120 C 10 0 .C 12 5 .0 O v PE'AKI NEL t,i E A 7 S e r ie s Th erma l A nal y sis Sys~e~ T ue J u n 1 0 1 4 41 C7 1997 Figure 5.19. DSC heating scan at 20 C/min for polymer HPA2. The thermogram was taken after the first cycle and was reproducible upon cycling.

PAGE 165

156 reproducible upon cycling at a scan rate of 20 C / min ( Figure 5.19). This melt value of 120 C closely c ompares t o the s tructurally s imilar s ample of Ramakrishnan s (Figure 5.5 where x =8) which has a Tm value of 128 C. 16 4 It should be noted that there was no crystallite annealing o r quenching experiments performed on sample HP A2. Preliminary Dimerization Experiments A series of experiments involving mono and difunctional alcoholic olefins were investigated to empirically assess their ability to undergo metathesis (Figure 5.20). The preliminary study of these reactions can help in the design of alcohol functionalized terminal dienes that will undergo productive ADMET polymerization. The investigation of the functional group proximity (alcohol) and its effect on productive metathesis has been termed the negative neighboring group effect ( NNGE). 1 7 3 This structure / reactivity behavior of alcohol functionalized o lefins with that of Grubbs ruthenium benzylidene C3 ( Figure 5.6 B) was initially investigated v ia the attempted dimerization and or the attempted polymerization of a variety of alcohol functionalized olefins. The productive metathesis of the monoenes was assessed by observing the appearance of an internal olefin by 1 H NMR. Significant o lefin isomerization, which would result in a false reading of internal olefins can be eliminated by the comparison of the chemical shifts before and after the reaction Shifting of the upfield signals would be an indication of o lefin isomerization.

PAGE 166

157 OH NR ( 1 ) 13 OH Ru OH HO ( 2) 12 8 0 % H Ru ( 3 ) 14 O H OH 9 7% Ru internal olefin detected O H 16 ( 4) H Ru NR ( 5 ) OH 15 Figure 5.20. Preliminary model s tudies of the st ructure / reactivity behavior of alcohol functionalized olefins in the presence o f the metathesis active ruthenium benzylidene. All the monomers were e ither purchased o r obtained with a purity greater than 99 /o by GC. The reactants were first degassed via freeze pump thaw cycles in the presence of high vac uum (> 104 torr) followed by the addition of ca t a l ys t to the neat liquid under dry box conditions. Catalyst to reactant ratio s we re the sa me as those used for polymerizations (between 400 : 1 and 200:1, respectively). The reactions we re al lowed to stir a t room

PAGE 167

1 58 te mperature followed by intermittent exposure to low vacuum ( <10 2 torr) for removal of ethylene. The relative percent conve rsion was determined by the ratio between the terminal and internal olef in sig nals. So me error may be introduced using this method due to the possible evaporation of the lower boiling reactants. Investigation of the primary alcohol was done by exposing 2-propene-1ol (a llyl alcohol) ( 13) and 4-pentene-1-ol (12) to the C3 ( Figure 5.19 (1) and ( 2 )). The primary alcohol would be coordinatively more available for catalyst coordination ( Lewis basic Lewis acidic interaction) causing either the slowing of the metathesis reaction or preventing it all together. The addition of the allylic a lcohol resulted in the immediate change of color of the catalyst from a purple to a yellow color, with the absence of detectable bubbling. Internal olefin signals were not detectable by 1 H NMR, although a new aldehyde proton signal was observed at 9.78 ppm. This observation is consistent with the solution isomerization of allylic alcohol to the co rresponding aldehyde by that of the vinylidene ruthenium catalyst174 ( Figure 5.6 C2) and other ruthenium co mplexes .175 The addition of the ruthenium benzylidene to 12 resulted in a purple brown color followed by noticeable degassing upon addition of vacuum. The proton NMR confirmed the existence of a metathesis product with a percent conversion of 80/o after 18 hours. Further experimentation must be done to asses the limitation the reactivity of primary alcohols. It can be assumed from this data, though, that primary unprotected en-ols with only three methylene spacers between the olefin and the alcohol are metathetically active and could possibly be in co rporated in the design of metathesis polymers. This i s confirmed by the synthesis of the ruthenium hydroxy a lkylidene via the reaction of 10 equivalents of 4-pentene-1-ol with catalyst C2 Figure 5.6.176

PAGE 168

159 Reactin g 14 with ca talyst C3 (F igure 5 .1 9 (3)) brings the alcohol functionality one ca rbon c loser in the form of a secondary a lcohol. T his co mbination of monoene and ca talyst produces a dimer with 51 /o co nversion after 1 hour follo wed by 97/o co nversion after 24 hours ( determined as b efo re by 1 H NMR ) This again can be inferred as a syn thetically useful structure for the production of metathesis products. terminal CH= Possible cis/trans from the internal olefin I I I I I terminal =CH 2 t ---r-~ -~~-~.. 6 5 pp m Figure 5.20. Olefin region of the 1 H NMR showing the possible existence of cis/trans isomers from the dimerized diene-ol 16. The alcoholic dienes 15 and 16 were available from Aldrich and presented the o pportunity to further investigate the reactivity of allylic alcohols in the fo rm of secondary diene-ols. As would be expected, the diol 15 showed no change in the proton NMR before a nd after the addition of C3. The lack of cata lyst solubility in this neat highly polar monomer was c onsidered and investigated by the addition o f a small amount of deuterated

PAGE 169

160 chloroform. The resulting 1 H NMR was still consistent with that of the unreacted diene. The sa me investigation was done for the mono functionalized diene (16) demonstrating a small amount of internal olefin at 5.5 ppm after 12 hours ( Figure 5.20). Further investigations are required to determine the identity of the minor constituent. Recent investigations by Wagener et al 173 have demonstrated the production of 1,5-cyclooctadiene by exposure of 1,5-hexadiene and a catalytic amount of catalyst C3. Conclusions The robust nature of Grubbs ruthenium benzylidene RuCl2( =CHPh)(PCy3)2 as a metathesis catalyst, in the presence of alcohol functionalities, promotes the catalytic ADMET polymerization of alcohol functionalized dienes at a reasonable rate. The clean nature of this conversion (as demonstrated by NMR and elemental analysis) provides a method for the direct synthesis of linear unsaturated alcohol functionalized polymers with a well-defined chemical environment. The advantages of using a condensation mechanism was demonstrated by the polymerization of primary se condary and tertiary alcohols. It can be inferred by these results that the structure of the resulting polymer is only limited to the design of the monomer, therefore, accurate models ca n be developed to investigate fundamental structure / property relationships of functional groups as well as branching in the polymer backbone. This was demonstrated by the synthesis of an ethylene-vinyl alcohol copolymer via the hydrogenation of the secondary alcohol functionalized ADMET polymer (prepolymer). The resulting polymer exhibits similar spectra and physical properties to those synthesized via ROMP.

PAGE 170

REFERENCES 1. von Pechmann, H. Ber., 1898, 31, 2643. 2. Bamberger, E.; Tschirner, F. Ber., 1900, 33, 955. 3. Raff, R.; Lyle, E. in Crystalline Olefin Polymers; Interscience Publishers, New York, 1965 Ch 1. 4 Meerwein, H. Angew. Chem., 1948, 60, 78. 5. Buckley, G. D.; Cross, L. H.; Ray, N. H. J. Chem. Soc., 1950, 2714. 6. Kantor, S. W.; Osthoff, R. C. J. Am. Chem. Soc., 1953, 75, 93. 7. Kantor, S. W.; Osthoff, R. C. U. S. 2,749,318 to General Electric, June 5, 1956. 8. Bawn, C. E. H.; Rhodes, T. B. Trans. Faraday Soc. 1954, 50, 934. 9. Davies, A. G.; Hare, D. G.; Khan, 0. R.; Sikora, J. Proc. Chem. Soc. (London) 1961, 172. 10 Carothers, W. H.; Hill, J. W.; Kirby, J. E.; Jacobsen, R. A. J. Am Chem. Soc., 1930, 52, 5279. 11. Koch, H.; Thing, G. Brennstoff-Chem., 1935, 16, 141. 12. Pichler, H.; Buffleb, H. Brennstoff-Chem., 1940, 21, 257 13. Hahn, W.; Muller, W. Makromol. Chem., 1955, 16, 71. 14. Brit. Pat. 2,816,883 (Sept. 6, 1937), E. W. Fawcett, R. 0 Gibson, M. H. Perrin, J. G. Paton, and E. G. Williams (to Imperial Chemical Industries, Ltd.). 15. H. M. Stanley, in Ethylene in Its Industial Derivatives (S. A. Miller, Ed.), Ernest Benn, London, 1969, Chap 1, pp. 28-32. 161

PAGE 171

162 16. U. S. Pat 2,816,883 (Dec. 17, 1957), A. W. Larcher and D. C. Pease (to E. I. du Pont de Nemours & Co., Inc.) 17. James D. E. in Encyclopedia of Polym er Science and Engi neering 2nd ed. Wiley-Interscience New York, 1986 p. 3 84. 18. Jacovic, M. S.; Pollock D.; Porter, R. S. J. Ap pl. Polym. Sci., 1979, 23, 517. 19. Stark, P.; Lindberg, J. J. Angew. Makromol. Chem., 1979, 75, 1. 20. Axelson, D. E.; Levy, G. C.; Mandelkem L. Macromolectles, 1979, 12, 41. 21. Morichima, Y.; Nozakura S. -I. ]. Polym Sci. Polym. Chem. Ed., 1976 14, 1277. 22. Nozakura, S. -I.; Morishima, Y.; Iimura H.; Irie, Y. J. Polym. Sci. Polym. Chem. Ed., 1976 14, 759. 23. Ziegler, K. Kunststoffe 1955, 45, 506. 24. Ziegler K ; Holzkamp, E.; Briel, H.; Martin, H. Angew. CJ,zem., 1955, 67, 426. 25. Ziegler, K.; Holzkamp, E.; Briel, H.; Martin, H. Angew. Chem., 1955, 67, 541. 26. Natta, G. J. Polymer Sci., 1955, 16, 143. 27. Natta G.; Pino, P.; Corradini, P.; Danusso F.; Mantica, E.; Mazzanti, G.; Moraglio G. J. Am. Chem. Soc., 1955, 77, 1708. 28. Ziegler K.; Martin, H. Makromol. Chem., 1956, 18/19, 186. 29. Odian, G., in Principles of Polymerization, 2nd ed. John Wiley & Sons, New York 1970 pp. 594-599. 30. Cossee, P., in The Stereochemistry of Macromoleciles (A.D. Ketley, Ed.) Vol. 1, Ketley, A. D. Ed., Marcel Dekker New York, 1967, Vol. 1, Chap. 3. 31. Arlman E. J.; Cossee, P. J Catalysis, 1964 3, 99 and references therein. 32. Sinn, H. ; Kaminsky, W. in Advanc es i11 Orga11ometallic Chemistry, Academic Press, 1980, Vol. 18, pp. 99-149.

PAGE 172

163 33. Kaminsky W. Macromol. Che m. Phys. 1996, 197, 39 07 and references therein. 34. Wilkinson G ; Birmingham I. M. J. Am. Clze m. Soc., 1954 76, 4281. 35. Fischer, E. 0. Angew. Cl1em 1952 22 620 36. Breslow D. S.; Newburg, N. R. J Am CJ1em. Soc 1957, 79, 5072. 37. Andresen, A.; Cordes, H. G.; Herwig J.; Kaminsky, W.; Merk A.; Mottweil e r, R.; Pein J ; Sinn, H. ; Vollmer H J. An gew Chem. 1976 88, 688; Angew. Chem. Int. Ed. Engl., 1976 15, 630. 38. Kaminsk y, W.; Miri, M .; Sinn, H .; Woldt R. Makromol. Chem. Rapid Commun. 1983 4 417. 3 9 Zucchini U ; Cecchin, G. Adv. Polym. Sc i., 1983, 51, 101. 40. Spaleck, W .; Aulbach M.; Bachmann B.; Kiiber F.; Winter, A. Macrom ol. Symp., 1995, 89, 237. 41. Cozewith, C.; Ver Strate, G. Macromoleci,le s, 1971, 4, 482. 42. Kakugo M.; Naito, Y.; Miyatake, T.; Mizunuma K. Macromol ec ule s 1980 15, 1150. 43. Soga, K .; S hiono, T. ; Doi, Y. Polym Bull. (Be rlin ) 1983, 10, 168. 44. Doi, Y.; Ohnishi, R. ; Soga, K. Makromol. Che m. Rapid Commun., 1983, 4 169 45. Kaminsk y, W.; Schlobohm, M. Makromol. Chem. Symp 1986, 4, 103. 46. Ewen, J. A., in Catalyti c Polymeri za tion of Olefins, (T. Keii, K. Soga, Eds.) Kodansha Tokyo, 1986, p. 271 47. Chien, J. C. W.; He, D. Polym. Sci., Part A: Polym. Chem. 1991 29 1585. 48 Herfert N.; Fink, G. Makr o mol Chem. 1992, 193, 1359. 49. Keller, A. J. Polym Sci. Symp., 1975, 51, 7 and 1977 59, 1 50. Mandelkern, L. Acct. Chem. R es., 1976 9, 81.

PAGE 173

164 51. Odian, G., in Principles of Polymerization, 2nd ed. John Wiley & Sons, New York, 1970. 52. Jaccodine R Natiire ( London) 1955, 176, 301. 53. Till, P. H. J Polym. Sci., 1957, 24, 301. 54. Keller, A. Philos. Mag., 1957, 2, 1171. 55. Fischer, E. W. Z. Naturforsch. Teil A, 1957, 12, 753 56. Mandelkern, L., Price, J. M.; Gopalan, M.; Fatou, J. G. J. Polym. Sci. Part A-2, 1966, 4, 385. 57. Mandelkern, L., Crystallization of Polymers" McGraw-Hill, New York, 1964. 58. Voigt-Martin, I. G.; Mandelkern, L. J Polym. Sci. Polym. Phys. Ed., 1981, 19, 1769. 59. Voigt-Martin, I. G.; Mandelkern, L. J. Polym. Sci., Polym. Phys. Ed., 1984, 22, 1901. 60. Voigt-Martin, I. G.; Alamo, R.; Mandelkern, L. J. Polym. Sci., Polym. Phys. Ed., 1986, 24, 1283. 61. Mandelkern, L., in Crystallization and Melting, Pergamon Press, New York, Vol. 2 1989, pp. 363-413 and references therein. 62. Vaughan A. S ; Bassett, D. C., in Crystallization and Melting, Pergamon Press, New York, 1989, Vol. 2 p. 414 and references therein. 63. Passaglia, E. J. Appl. Polym. Sci., 1963, 17, 119. 64. Kao, Y. H. Polymer, 1986, 27, 1669. 65. Bassett, D. C., Principles of Polymer Morphology Cambridge University Press, New York, 1981. 66. Mandelkern, L., in Physical Properties of Polymers, American Chemical Society, New York, 1984, p. 155. 67. Chiang, R.; Flory, P. J. J Am. Chem. Soc., 1961, 83, 2857.

PAGE 174

16 5 68 Ungar, G.; Stejny, J.; Keller A.; Bidd I .; Whiting, M. C. Scie nc e, 1985 229 p 386. 69 Lee, K. S.; Wegner G. Makromol. Chem., Rapid Commitn. 1985 6, 203. 70. Kloos, F.; Go, S.; Mandelkem L J Polym. Sci Polym. Phys. Ed. 1974 12, 1145. 71. Prasad, A.; Mandelkern, L. Macromolecitles 1989 22, 914. 72. Mandelkern, L.; Prasad, A.; Alamo, R. G.; Stack, G. M. Macromole cit les 1990, 23 3696. 73. Flory P. J. Tran s Faraday Soc., 1955, 51, 848. 74. Clas, S. D. ; Mcfaddin D. C.; Russel K. E.; Scannell-Bullock, M. V.; Peat, I. R. J. Polym. Sci., Polym Chem. Ed. 1987 25, 3105. 75. Alamo, R. G.; Mandelkern L. Macromolecitles, 1989, 22, 1273. 76. Mandelkern, L.; Glotin, M.; Benson, R. A. Macromolecitles, 1981, 14, 22. 77. Lambert, W. S.; Phillips, P. J. Polymer, 1996, 37, 3585. 78. Alamo R. G.; Viers, B. D ; Mandelkern, L. Macromolecules, 1993 26, 5740. 79. Fu, Q.; Chiu, F-C ; McCreight, K. W.; Guo, M.; Tseng W W. ; Cheng, S. Z. D. Joitrnal of Macro Scie n ce -Phy s i cs 1997 VB36 1 41. 80. Gerum, W.; Hahne, G. W. H.; Wilke, W.; Arnold, M.; Wegner T. Macromol C hem. Phys ., 1995, 196, 3797. 81. Gerum, W ; Hohne, G. W. H.; Wilke, W.; Arnold, M.; Wegner T. Macromol. Chem. Phy s ., 1996, 197, 1691. 82. (a) Calderon, N., J Ma c romol. Sci.Rev. Macromol. Chem., 1972, C7(1), 105. ( b) Schrock, R. R.; Krouse, S. H.; Knoll, K.; Feldman, J.; Murdzek, J. S.; Yang, D. C. J Mol. Cat. 1988, 46, 243. (c) Ivin K. J Olefin Metathesis Academic Press: New York, 1983. ( d) Schrock, R. R. The St rem Chemiker 1992, 14(1) 1. 83. Pakuro, N. I .; Gantmakher, A. R.; Dolgoplosk, B. A. Dokl. Akad. N auk SSSR 1975 223, 868.

PAGE 175

166 84. Schaverien, C. J.; Dewan J. C.; Schrock R. R. J Am. Chem. Soc. 1986, 108, 2771. 85. ( a) Schrock, R. R.; Depue, R. T.; Feldman, J.; Schaverien C. J.; Dewan, J. C.; Liu, A. H. J. Am. Chem. Soc. 1988, 110, 1423. (b) Depue, R. T.; Schrock, R. R.; Feldman, J ; Yap, K.; Yang, D. C.; Davis, W. M.; Park, L.; DiMare, M.; Schofield, M.; Anhaus, J.; Walborsky, E.; Evitt, E.; Kruger, C.; Betz, P. Organometallics 1990, 9, 2262. ( c) Schrock, R. R.; Schaverien C. J.; Dewan, J. C. J. Am. Chem. Soc 1986, 108, 2771. 86. ( a) Feldman, J.; Schrock, R. R. Prag, Inorg. Chem. 1991, 39, 1. (b) Feldman, J.; Davis, W M.; Schrock, R. R. Organometallics 1989, 8, 2266. (c) Feldman, J.; Murdzek, J. S.; Davis, W. M.; Schrock, R. R. Organometallics 1989, 8, 2260. 87. Nguyen, S. T.; Johnson, L. K.; Grubbs, R. H.; Siller, J. W. J. Am. Chem. Soc. 1992, 114,3974. 88. Fu, G. C.; Nguyen, S. T.; Grubbs, R.H. J. Am. Chem. Soc. 1993, 115, 9856. 89. Grubbs, R. H.; Ziller, J. W.; Schwab, P. J. Am. Chem. Soc. 1996, 118, 100. 90. Wagener, K B.; Boncella, J. M.; Nel, J. G. Macromoleci,les 1991, 24, 2649. 91. Wagener, K. B ; Puts, R. D.; Smith, D. W. Macromol. Chem. Rapid. Commi,n. 1991, 12, 419. 92. Watson M. D.; Wagener, K. B. Polym. Prepr. 1996, 37(1), 609. 93. Marmo, J. C.; Wagener, K. B. Macromolecules 1995, 28, 2602. 94 Marmo, J. C.; Wagener, K. B. MacromoleciLles 1993, 26, 2137. 95. Wagener, K. B.; Marmo, J. C. Macromol Rapid Commun. 19 9 5, 16, 557. 96. Wagener K. B.; Boncella, J. M.; Nel, J. G.; Duttweiler, R. P.; Hillmyer, M. A. Makromol. Chem. 1990, 191, 365. 97. Konzelman, J.; Wagener, K. B. Polym. Prepr. (Am. Che1n. Soc. Div. Polym. Chem.) 1992, 13, 109. 98. Konzelman, J.; Wagener, K. B. Macromoleci1les 1995, 28, 4686. 99. Brzezinska, K.; Wolfe, P. S.; Watson M. D.; Wagener, K. B. Macromol. Chem. Phy s 1996, 197 2065.

PAGE 176

167 100. O'Gara, J. E.; Wagener K. B.; Hahn, S. F. Makromol. Chem., Ra pid Commun., 1993, 14, 657. 101. ( a) Schrock, R. R.; Murdzek, J S.; Bazan, G. C.; Robbins, J.; Dimare, M.; O'Regan, M. J. Am. Chem. Soc. 1990, 112, 3875. ( b) Bazan, G. C.; Khosravi, E.; Schrock, R. R.; Feast, W J ; Gibson, V. C.; O'Regan, M. B.; Thomas, J. K.; Davis, W. M. J. Am. Cl1em. Soc. 1990, 112, 8378. (c) Bazan, G. C.; Oskam, J. H.; Cho, H. N.; Park, L. Y.; Schrock, R. R. J. Am. Chem. Soc. 1991, 113, 6899. (d) Fox, H. H.; Schrock, R. R. Organometallics 1992, 11, 2763. (e) Feldman, J.; Murdzek, J. S.; Davis, W. M.; Schrock, R. R. Organometallics 1989, 8, 2260. (f) Oskam, J. H.; Schrock, R. R. J. Am. Chem. Soc. 1992, 114, 7588. 102. Perrin, D. D., Purification of Laboratory Chemicals, W. L. F. Amarego, Pergamon Press, New York, 1988. 103. ( a) Carruthers, W. Some Modern Methods of Organic Synthesis, Cambridge University Press, Cambridge 1986. (b) Stowell, J.C., Carbanions in Organic Synthesis, John Wiley & Sons, New York. 1979. (c) Brian, T.; McMurry, H.; Work, A.; and McKenna, B. J. Amer. Chem. Soc. Perkin Trans. 1991 1,. (d) Wallingford, V.H.; Thorpe, M.A.; Homeyer, A.H. J. Amer. Chem. Soc. 1942, 64, 580. 104. Fujita, E.; Fuji, K.; Sai, M.; Node, M.; Watson, W. H.; Zabel, V. j Chem. Soc. Chem. Commun. 1981, 17, 899 105. Aurelie, H.; Prome, J.C. Tetrahedron Lett. 1980, 21, 3277. 106. Bruggink, A.; McKillop, A Tetrahedron 1975, 31, 2607. 107. House, H. 0. Modern Synthetic Reactions, 2 ed. W. A. Benjamin, Menlo Park, CA. 1972. 108. ( a) Renfrow, W. B. J. Am. Chem. Soc., 1944, 66, 144. (b) Renfrow, W B.; Renfrow, A., ibib., 1946, 68, 1801. 109, ( a) Renfrow, W. B.; Walker, G. B. J. Am. Chem. Soc., 1948, 70, 3957. ( b) Fonken, G. S.; Johnson, W. S. J. Am. Chem. Soc., 1952, 74, 831. (c) Lawesson, S. O.; Gron wall, S.; Sandberg, R. ibid., 1962, 42, 28. ( d) Riegel, B.; Lilienfeld, W M. J. Am. Chem. Soc., 1945, 67, 1273. (e) Lalancette J. M.; Lachance, A. Tetrahedron Letters 1970, 45, 3903.

PAGE 177

168 110. (a) Swan, G. A. J Chem. Soc., 1948, 1408. ( b) Gassman, P. G.; Lumb, J. T.; Zalar, F. V. J. Am. Chem. Soc. 1967, 89, 946. 111. March, J.; Plankl, W. J. CJ,zem. Soc 1977 1, 460. 112. Wallingford V. H.; Thorpe M. A.; Homeyer A. H.J. Amer. Chem. Soc 1942, 64, 580. 113. (a) Zaugg, H. E.; Horrom, B. W.; Borgwardt, S. J. Am. Chem. Soc., 1960, 82, 2895. (b) Zaugg, H E. J. Am. Chem. Soc., 1960, 82, 2903. 114. Marchall, F J.; Cannon W. N. J. Org. Chem. 1956, 21, 254. 115. Zaugg, H. E. ; Dunnigan, D. A.; Michaels, R. J.; Swett, L. R.; Wang, T. S.; Sommers, A. H.; DeNET R. W. J. Org. Chem. 1961 26, 644. 116. Oediger, H.; Moller, F., Liebigs Ann. Chem. 1976, 348. 117. Harris, T. M.; Harris, C. M. Organic Reactions, 1969, 17, 155. 118. Gaylord N. G. Reduction with Complex Metal Hydrides, Interscience, New York, 1956. 119. Kabalka, G.W.; Varma, M.; Varma, R.S.; Srivastava, P.C.; and Knapp F.F.Jr. J. Org. Chem. 1986, 51, 2386. 120. Krapcho, A. P. Synthesis, 1982, 805, 893. 121. Sita, L. R. MacromoleciLl es 1995, 28, 656. 122. Konzelman J.; Wagener, K. B. Macromolecules 1995, 28, 4686. 123. March, J. Advanced Organic Chemistry, 3rd ed., Wiley-Interscience, New York, 1985. 124. Hahn, S. F. J. Polym. Sci., Part A, 1992, 30, 397. 125. Carey, F. A.; Tremper, H. S. J. Org. Chem. 1971 36, 758. 126. Ireland, R. E.; Muchmore D. C.; Hengartner U. J. Am. Chem. Soc. 1972, 94, 5098. 127. Olah, G. A.; Balaram Gupta, B. G.; Malhotra R.; Narang, S. C. J. Org. Chem., 1980, 45, 1638.

PAGE 178

169 128. Paynter 0. I.; Simmonds, D. J. ; Whiting, M. C. J. Chem. Soc Che m. Commun. 1982, 1165 129. Bidd I .; Holdup, D. W .; Whiting M. C. J. C hem. Soc Perk in Trans 1987 1, 2455. 130. Wu, Z.; Grubbs, R. H. Macromolecitles 1994 27, 6700. 131. Wu Z; Wheeler, D.R.; Grubbs, R. H.J Am Chem. Soc. 1992, 114, 146. 132. Gopalan, M.; Mandelkern L. Journal of Physical Chemistry, 1967, 71, 3833. 133. Alamo, R. G.; Chan, E. K. M.; Mandelkern L.; Voigt-Martin I G. Macromol ec ules 1992, 25, 6381. 134. Ke B. J. Poly1n. Sci., 1962 61, 47. 135. (a) Wagener, K. B.; Nel, J. G.; Konzelman, J. and Boncella, J. M. Macromolecules 1990, 23, 5155. (b) Konzelman, J. Acyclic Diene Metathesi s Polymerization: A Hydrocarbon Structure Reactivity Study, Ph. D. Dissertation, University of Florida, Gainesville, 1992. 136. Nel, J. G Acyclic Diene Metathesis Polymerization, Ph D Dissertation, University of Florida, Gainesville, 1990. 137. Patton, J T. Acyclic Diene Metathesis (A DMET ) Polymerization: The Polymerization of Carbonyl Containing Diene s, Ph. D. Dissertation University of Florida, Gainesville, 1992. 138. Sanui, K .; MacKnight, W J.; Lenz, R., W J. Polym. Sci. B 1973, 11, 427. 139. Rachapudy, H.; Smith, G. G.; Raju, V. R ; Graessley, W. W. J. Polym. Sci. Polym. Phy s Ed. 1979 17 1211. 140. Storey R. F.; George, S E. Proc. Am. Chem. Soc. Di v Polym. Mat. Sci. Eng. 1988 ,58,9 85. 141. Mango L. A .; Lenz, R. W. Makromol. Chem. 1973, 163 13.

PAGE 179

170 142. Harwood, H.J.; Russell, D. B.; Verthe J. J. A.; Zymonas, J. Makromol. CJ1em. 1973 163, 1. 143. Carman, C J.; Tarpley A. R., Jr.; Goldstein, J. H Macro1noleci,les, 1973, 6, 719. 144. (a) Hsieh E. T.; Randall J C. Macromolecilles 1982, 15, 353. ( b) Hsieh, E T.; Randall, J. C. Macromoleci,l es 1982, 15, 1402. 145 To maintain consistancy, these samples were injected multiple times in sequence of one another. 146. Ke, B. in Differential Thermal Analysis, (N. M. Bikales Ed.), WileyInterscience, New York, 1971 pp. 191-219 147. Wunderlich, B.; Poland, D., J. Polym. Sci. A 1963 1, 357. 148. Wunderlich, B.; Bodily, D., J. Polym. Sci. A-2 1966, 4, 25. 149. Grisky R. G.; Foster, G. N., J Polym. Sci. A-11970, 8, 1623. 150. Hosoda, S.; Nomura, H.; Gotoh, Y.; Kihara H., Polymer 1990, 31, 1999. 151. (a) Mandelkem, L.; Price, J. M.; Gopalan, M.; Fatou, J. G., J. Polym. S ci ., Part A-2 1966, 4, 385. (b) Eppe, R.; Fischer, E. W.; Stuart, H. A., J. Polym. Sci 1959 34, 721. (c) Bassett, D. C.; Hodge, A. H., Proc R. Soc. London Ser. A 1981, 377, 25. (d) Strobl G.; Schneider M.; Voight-Martin, I. G., J Polym. Sci., Polym. Phys. ed. 1980, 18, 1368. (e ) Voigt-Martin, I. G.; Fischer, E. W.; Mandelkem, L J. Polym. Sci., Polym Phys. Ed., 1980, 18, 1368 152. (a) Richardson, M. J ; Flory, P. J ; Jackson, J B. Polymer 1963, 4,221. ( b) Baker, C. H.; Mandelkem, L. Polymer 1966, 7, 7. (c) Bowmer, T. N.; Tonelli, A. E. Polymer 1985, 26, 1195. 153. Wunderlich, B. Macromolecular Physics, Vol. 3, Crystal Melting, Academic Press, New York, 1980. 154. Mandelkem, L.; Hellmann, M.; Brown, D. W.; Roberts, D. E.; Quinn F. A. Jr. J. Am. Chem. Soc. 1953, 75, 4093. 155 Broadhurst M. G., J. Res. Natl. Bur Stand. USA 1962, 66A, 241. 156. Qiang, F.; Chiu, F-C; McCreight K. W ; Guo, M.; Tseng, W. W ; Cheng, S. Z. D ]. Macro. Sci. Phys. 1997, 36, 41.

PAGE 180

171 157. Alamo, R.; Domszy, R.; Mandelkern, L. J. Phys. Chem. 1984, 88, 6587. 158. Mathot, V B. F., Calori1netry a11d TJ1ermal Analysis of Polymers, Hanser Publichers, Munich, 1994. 159. Capon, B.; Zucco, C. J. Am. Chem. Soc. 1982, 104, 7567. 160. Novak, B. M.; Cederstav, A. K. J. Am. Chem. Soc. 1994, 116, 4073. 161. Fu, G. C.; Grubbs, R. H.J. Am. Chem. Soc. 1992, 114, 7324. 162. Fu, G. C.; Grubbs, R. H.J. Am. Chem. Soc. 1992, 114, 5426. 163. Chung, T. C.; Ramakrishnan, S. Macromolecules 1990, 23, 4519. 164. Ramakrishnan, S. Macromolecules 1991, 24, 3753. 165. (a) Foster, R. H. Polym. News 1986, 11, 264. (b) Moroi, H. Br. Polym. J. 1988, 20, 335. 166. Fu, G. C.; Nguyen, S. T.; Grubbs, R. H.J. Am. Chem. Soc. 1993, 115, 9856. 167. Schwab, P. F.; Marcia, B.; Ziller, J. W.; Grubbs, R. H. Angew. Chem. Int. Ed. Eng.1995,34,2039. 168. Hillmyer, M. A. The Preparation of Functionalized Polymers by RingOpening Metathesis Polymerization, Dissertation, California Institute of Technology, 1995. 169 Oskam, J. H.; Schrock, R. R. J. Am. Chem. Soc. 1992, 114, 7588. 170. Wu, Z.; Wheeler, D. R.; Grubbs, R. H.J. Am. Chem. Soc. 1992, 114, 146. 171. Wagener, K. B.; Smith, D. W., Jr. Macromoleci,les 1991, 24, 6073. 172. Brzezinska, K.; Wagener, K. B. Unpitblished results Department of Chemistry, University of Florida, Gainesville 32611, 1991. 173. Wagener, K. B.; Brzezinska, K.; Anderson, J. D.; Younkin, T. R.; DeBoer, W. A Macromoleci,les and references therein, accepted with revisions 1997. 174. Wu, Z; Nguyen, S. T.; Grubbs, R. H.; Ziller, J. W. J. Am Chem Soc. 1995, 117, 5503.

PAGE 181

175. (a) McGrath D. V.; Grubbs, R.H. Organometallics 1994 13, 224. (b) Karlen T.; Ludi, A. J Am. Chem Soc. 1994 116, 11375. 17 2 176 Schwab, P.; Grubbs, R H ; Ziller, J. W. J. Am. Chem. Soc 1996 118, 100.

PAGE 182

BIOGRAPHICAL SKETCH Dominick J. Valenti was born on April 4, 1969, in Melbourne, Florida, as the second of two children. His interest in science and technology was fostered by life on the space coast as well as his family. He attended the University of Central Florida in Orlando where he was introduced to polymer chemistry by Dr. Guy Mattson. Upon receiving his Bachelor of Science degree in chemistry and a minor in business administration (1992), Dominick began his graduate career in organic/polymer chemistry at the University of Florida under the advisement of Dr. Ken Wagener Dominick received his Ph.D. in August of 1997, and joined Dr. Gerhard Wegner at the Max-Planck Institute in Germany for the exploration of new and interesting scientific challenges and the never-ending pursuit of happiness. 173

PAGE 183

I certify that I have read this study and that in my opinion it confo1111s to acceptable standards of scholarly presentation and is fully adequate, in scope and quality as a dissertation for the degree of Doctor of Philosophy. Kenneth B. Wagener, Chairm"'.l.l Professor of Chemistry I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Jo n Re nolds Pro ssor of Chemistry I certify that I have read this study and that in my opinion it conforrns to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. William R. Dolbier, Jr. Professor of Chemistry / I certify that I have read this study and that in my opinion it confor111s to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ames M. Boncella Associate Professor of Chemistry I certify that I have read this study and that in my opinion it confor1ns to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Anthony B Brennan Associate rof essor of Materials Science and Engineering

PAGE 184

This dissertation was submitted to the Graduate Faculty of the Department of Chemistry in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy Dean, Graduate School August 1997

PAGE 185

LD 17 8 :~ 199 7 t ... -ea ,V14 / UNIVERSITY OF FLORIDA II I II IIII II Ill Il l ll ll l l ll l II II I II I II II I II I Il l 111 11 111 11 1 11 1111 1 3 1262 08554 9128