Unsaturated polysulfides and polyamines via acyclic diene metathesis (admet) polymerization

MISSING IMAGE

Material Information

Title:
Unsaturated polysulfides and polyamines via acyclic diene metathesis (admet) polymerization
Physical Description:
ix, 131 leaves : ill. ; 29 cm.
Language:
English
Creator:
Portmess, Jason D., 1968-
Publication Date:

Subjects

Subjects / Keywords:
Chemistry thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Chemistry -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1996.
Bibliography:
Includes bibliographical references (leaves 122-130).
Statement of Responsibility:
by Jason D. Portmess.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002215747
oclc - 36701594
notis - ALF5897
System ID:
AA00009012:00001

Full Text














UNSATURATED POLYSULFIDES AND POLYAMINES
VIA ACYCLIC DIENE METATHESIS (ADMET) POLYMERIZATION














By

JASON D. PORTMESS


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


1996





























This dissertation is dedicated to the people
who have always given me the love, trust and support
to chase life's pursuits.

-To My Family-














ACKNOWLEDGEMENTS


The title of Ph.D. would be worthless without the friendships and respect of the

people that I have worked with during my experience. Eternal gratitude must be extended

to Dr. Jim Konzelman and Dr. Dennis Smith for showing me two completely different yet

effective ways of approaching the art of chemistry. Sincere thanks are also extended to the

past and present members of the Wagener group: Dr. Chris Marmo, Dr. Kathleen Novak,

Dr. John O'Gara, Dr. Jasson Patton, Dr. Dehui Tao, Dr. Fabio Zuluaga Sophia

Cummings, Tammy Davidson, Fernando Gomez, Lauri Jenkins, Debra Tindall, Dominick

Valenti, Mark Watson, and Shane Wolfe.

The contributions and support of these people cannot be measured, but without

their presence the realization of my scientific goals could not have been possible. Also

special thanks go to the past and present members of the Duran and Reynolds group for

providing a unique multidisciplinary environment which was essential in my scientific and

personal development.

Respects go to Don Cameron, Mike Cruskie, Peter Balanda, Troy Bergstedt, Brian

Hauser, Tim Herod, Scott Whipps, and Paul Whitley for their special friendships and

sharing of philosophical ideas.

Special recognition is given to Lorraine Williams for her consistent warm

personality, willingness to help disorganized students, and impeccable proficiency.

Eternal love and respect is extended to Tammy Davidson for being my best friend.

Without her support, understanding, and patience the completion of this degree and

manuscript would have been tremendously more difficult.









Finally, the most heart felt regards and respect are gratefully extended to Professor

Ken Wagener for his philosophy of learning without restriction, for possessing the

confidence in me that I didn't have for myself, and for being a friend and mentor.















TABLE OF CONTENTS



Page
A CKN OW LEDG EM ENTS .............................................................................................. iii

A B ST R A C T ......................................................................................................................viii

CHAPTERS

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


History of Sulfur-Containing Polymers...................................... ...... 2
Synthesis of Polymonosulfides................ ...... ................. 3
Ring Opening Polymerization..................... ........................... 3
Anionic Polymerization........................ .............................3
Cationic Polymerization.......................................... 5
Coordinative Ionic Polymerization................................. ..6
Condensation Polymerization........................... 7
Addition of Thiol to Olefinm... .....7
Addition of Thiol to Dihalide..................... 8
History of Amine-Containing Polymers............................... .................9. 9
Synthesis of Main Chain Polyamines....................... .... ... .... 9
Ring Opening Polymerization................................... 9
Aziridine Polymerization ......... ..................... 10
Azetidine Polymerization ..................................... 13
Condensation Polymerization.............................. ........... 14
Development of Olefin Metathesis and Catalyst Systems............... 16
Traditional Catalysts for Olefin Metathesis............. 16
Elucidation of the Olefmin Metathesis Mechanism............................. 18
Development of Well-Defined Metathesis Catalysts.....................22
Olefin M etathesis Polymerization...................................................28
Ring Opening Metathesis Polymerization (ROMP) ..........29
Acyclic Diene Metathesis Polymerization (ADMET)........... 32

2 EXPERIM ENTAL ................................................................................ 43

Instrumentation and Analysis....................................... ..... .. ..43
M materials and Techniques........................................................... ..... 44
Synthesis and Characterization ................................ .. ........... ........ 46
Sulfur Containing Monomers........................................... 46
Synthesis of Bis(3-butenyl)sulfide (5)................................... 46
Synthesis of Bis(4-pentenyl)sulfide (6) ......................47
Synthesis of Bis(5-hexenyl)sulfide (7).................... ..... 47
Bulk Cyclization of Diallysulfide (4) .............................. 47









Solution Cyclization of Diallysulfide (4)
to 2,5-Dihydrothiophene (8)...................................... 48
Synthesis of Poly(thio-3-hexene- 1,6-diyl) (9)................................48
Synthesis of Poly(thio-4-octene-1,8-diyl) (10) ............. 48
Synthesis of Poly(thio-5-decene-1,10-diyl) (11).... ................ 49
Synthesis of Poly[(thio-5-decene-1,10-diyl)-co-
(1-octenylene)] (12)...................................... ........ 49
Attempted Metathesis of Monomers 4-7
with Ruthenium Catalyst .......................... ... 50
Bulk Cyclization of Diallydisulfide (13).................................... 50
Solution Cyclization of Diallydisulfide (13) ........................... 51
Synthesis of Bis(5-hexenyl)disulfide (15)......... ...................... 51
Attempted metathesis of Bis(5-hexenyl)disulfide (15)................. 51
Synthesis of 2-Propenylthiophene (16)........................................ 52
Attempted Metathesis of 2-Propenylthiophene (16) .................. 52
Amine Containing Model Compounds, Monomers, and Polymers.............53
Attempted Bulk Metathesis of Diallylamine (17)
w ith Catalyst I and 2 .................................................................... 53
NMR Solution Reaction of Diallylamine (17) with
M olybdenum Catalyst 2....................................................... 53
Attempted Copolymerization of 17 and 1,9-Decadiene
w ith C atalyst 2 ............................................... ........................ 53
Synthesis of 4-Dodecylaminobenzaldehyde (18).................. 53
Synthesis of 4-(Prop-l-ene)-N-Dodecylaniline (19) ....................54
Synthesis of 4-(Prop-1-ene)-N,N-Dimethylaniline (20) .............. 55
Synthesis of 4,4'-Bis(dodecylamino)stilbene (21) ....................... 55
Synthesis of 4,4'-Bis(dimethylamino)stilbene (22) ................ 55
Synthesis of N,N-Diallylaniline (23).......................................... 56
Synthesis of N,N-Dibutenylaniline (24)......................... .... 56
Synthesis of N,N-Dipentenylaniline (25).................... ...... 57
Synthesis of 1-Phenyl-3-pyrroline (26)....................................... 57
Synthesis of Poly(N-phenylamino-3-hexene-1,6-diyl) (27)........... 57
Synthesis of Poly(N-phenylamino-4-octene-1,8-diyl) (28)........... 58
Synthesis of Poly[(N-phenylamino-4-octene-1,8-diyl)-
co-(l-octenylene)] (29)...................................... ........... 59
Synthesis of N-methyl-2-propenylpyrrole (30)......................... 59
Attempted metathesis of N-methyl-2-propenylpyrrole (30) ............60
Attempted Polymerization of 1,9-Decadiene in the
Presence of Aniline ............................... 61
Formation of Cyclic M onomer (31)................................................ 61
Formation of Cyclic Monomer (32) and Cyclic
D im er (33a-c)......................................................................... 6 1
Attempted Polymerizations of Monomers 23-25
with Ruthenium Catalyst 3............................................... .. .. 61
Synthesis of N,N-Diallylmethylamine (34).................................. 62
Synthesis of N,N-Dibutenylmethylamine (35).............................62
Attempted Bulk Cyclization of N,N-Diallylmethylamine (34).........62
Attempted Solution Cyclization of N,N-Diallylmethylamine (34)... 62
Attempted Polymerization of N,N-Dibutenylmethylamine (35)...... 63
Synthesis of N,N-Diallyl-tert-butylamine (36)............................ 63
Attempted Bulk Cyclization of N,N-Diallyl-tert-butylamine (36)....63
Solution Cyclization of N,N-Diallyl-tert-butylamine (36)..............63
Synthesis of N,N-Diallylbenzylamine (37) .............................. 64
Synthesis of N,N-Dipentenylbenzylamine (38) ............. 64









Attempted Bulk Cyclization of N,N-Diallylbenzylamine (37).........64
Attempted Solution Cyclization of N,N-Diallylbenzylamine (37)... 64
Attempted Polymerization of N,N-Dipentenylbenzylamine (38)..... 65
Attempted Metathesis of Monomers 34-38
with Ruthenium Catalyst 3.......................................... ...... 65

3 DESIGN AND SYNTHESIS OF UNSATURATED
PO LY SU LFID ES........................................... .......................................... 67

Historical Review of Sulfide Metathesis ....................... 69
Monosulfide Structure and Reactivity Relationships
for AD M ET ............................................................... 71
Sulfide ADM ET Cyclization ........................................................... ............73
ADMET Polymerization of Polymonosulfides...................................... 75
Polysulfide Thermal Properties ................................................. ....... 80
Attempted ADMET Cyclization and Polymerization
of Disulfide Monomers .............................................. 84
Model Studies for Polythiophenes via ADMET ........................ ....... 87
C conclusions .............................................................................................. 88

4 UNSATURATED AROMATIC AND ALIPHATIC AMINES
FOR ACYCLIC DIENE METATHESIS (ADMET)
POLYM ERIZATION ........... ....... ...................................................... 89

Historical Review of Amine Metathesis ........................89
Aromatic Amine Model Compound Structure
and Reactivity Relationships............................... ............ ... ... 92
Aromatic Amine Monomer Synthesis .......................................................98
Metathesis of Aromatic Monomers..................................................... 99
Ring Closing Metathesis Cyclization...........................99...................
Synthesis of Unsaturated Main Chain Polyamines............. .. 101
Polymer Analysis and Characterization.................................................. 102
Polyamine Thermal Properties ................................................ ............ 107
Postcondensation Cyclization Reactions................................................ 110
Aliphatic Amine Monomer Structure Reactivity
Relationships ............................................... ...... 113
Conclusions .................. ........................................... ........ 121

R E FE R E N C E S ................................................................................................................. 122

BIOGRAPHICAL SKETCH........................................................ .......131














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


UNSATURATED POLYSUFIDES AND POLYAMINES
VIA ACYCLIC DIENE METATHESIS (ADMET) POLYMERIZATION


By

Jason D. Portmess


December 1996

Chairman: Professor Ken Wagener
Major Department: Chemistry


The design and synthesis of unsaturated main chain polysulfides and polyamines

via acyclic diene metathesis (ADMET) polymerization is presented. A series of sulfur and

aminodienes were designed to delineate the specific monomer structural requirements

necessary for ADMET polymerization using Schrock's molybdenum alkylidene,

Mo=CHR(N-2,6-C6H3-i-Pr2)[OCMe(CF3)212 (M=W or Mo), and Grubbs' ruthenium

alkylidene, RuCI2(=CHPh)(PCy3)2, as catalytic initiators.

Diallylsulfide showed resistance to bulk metathesis, but underwent quantitative

solution cyclization when exposed to Schrock's catalyst. Extended methylene spacing

resulted in clean metathesis polymerization yielding high molecular weight unsaturated

polysulfides with good thermal stability. Efforts to polymerize disulfide monomers proved

ineffective under bulk and solution conditions. Further, sulfide and disulfide monomers

also did not demonstrate significant productive metathesis with the above ruthenium

alkylidene.









N-phenyl derivatized aminodienes showed no resistance to homometathesis and in

certain cases yielded high molecular weight, perfectly linear main chain polyamines. These

polymers are the first example of a direct synthetic route to high molecular weight, main

chain polyamines. These polymers possessed glass transition temperatures below room

temperature and exhibited good thermal stability in nitrogen. Postcondensation cyclics

were also observed at elevated temperatures as a consequence of intramolecular backbiting

reactions. As demonstrated for sulfur dienes, the ruthenium alkylidene did not produce

unsaturated polysulfides or polydisulfides.

Electronic considerations were also examined for a series of various aliphatic

substituted aminodienes. Only limited metathesis condensation was observed in these

cases, with no high yielding cyclization or polymerization observed under bulk or solution

conditions.

Complete synthesis, characterization and thermal behavior for the derived

unsaturated polysulfides and polyamines are discussed.














CHAPTER 1
INTRODUCTION


Since the inception of the concept of the macromolecule in the early 1920s,

polymer chemistry has played an integral role in almost every modem day technological

advancement. Investigations in the field of polymer science have made major

contributions to the scientific boom of the last 50 years. Intensive research activities in

several laboratories around the world has succeeded in elucidating polymerization

mechanisms and developing new synthetic strategies for engineering polymeric materials

with specifically desired properties. Polymer chemists have personified the ideals of

change and adaptation, constantly searching for a new scientific methodology or

approach to meet the demands and needs of our changing world. The quest to understand

the application of natural and synthetic macromolecules began over a century ago and

continues to be pursued vigorously today.

This dissertation is a product of the ongoing research in acyclic diene metathesis

(ADMET) polymerization as a viable route for producing multifunctionalized linear

unsaturated polymers. The work presented herein describes the reactivity of amine and

sulfur functionalized dienes toward ADMET polymerization. The specific monomer

structural requirements, in addition to the synthesis and characterization of their novel

unsaturated polymers, will be discussed in detail. Due to the diverse nature of this study,

it is appropriate to describe the historical and continuing role with which amine and

sulfur compounds play in polymer chemistry. Although the area of ADMET chemistry is

rather unique, several amine and sulfur-containing polymers have been synthesized by

traditional methods which are fairly analogous to the subject of this work, and thus will









be the principal focus of this introduction. In addition to the following brief historical

perspective of amine and sulfur-containing polymers, a brief discussion of metathesis

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


History of Sulfur-Containing Polymers


Since 1839, when Charles Goodyear used elemental sulfur to crosslink or
'vulcanize' natural rubber (polyisoprene), sulfur and sulfur-containing polymers have held

a special position in the field of polymer chemistry. The most fundamental sulfur-

containing polymer is obtained directly from elemental sulfur. Elemental sulfur exists at

room temperature primarily as a mixture of eight-membered rings, a-sulfur (rhombic),

and 13-sulfur (monoclinic) forms. As sulfur is heated to temperatures exceeding 1600C, a

viscous melt is obtained which consists of an equilibrium mixture of linear polymeric

chains and eight-membered rings (Figure 1.1).1


S- 1600C -(S 8)r
S' S S room temp.



Figure 1.1. Ring-chain equilibrium of polymeric sulfur.


Upon gradual cooling the heat established equilibrium is shifted toward monomeric or

elemental sulfur (rhombic form only), and complete depolymerization occurs.
Alternatively, rapid cooling and suppression of the temperature below the Tg of the

polymeric material maintains its integrity. Depending on the initial melt temperature,

polymeric sulfur can have molecular weights as high as 2,000,000.2,3 This

thermodynamically driven process is controlled by the initial melt temperature and the

subsequent cooling rate. The polymer properties obtained are intimately dependent upon









these parameters. This type of processing versatility has garnered considerable attention

toward a multitude of industrial applications. The chemistry and properties of elemental

and polymeric sulfur have been reviewed extensively.4

While sulfur-containing polymers make up one of the many broad fields of

synthetic polymer chemistry, this introduction will focus only on polymonosulfides,

which are closely related to the experimental chemistry discussed in this dissertation.


Synthesis of Polymonosulfides

Polymonosulfides represent an important class of sulfur-containing polymers and

are typically produced via a ring opening polymerization process or by condensation

chemistry. Since both polymerization methods are related to the experimental work

which follows, it is prudent to introduce these topics at this time.


Ring Opening Polymerization


The three and four-membered cyclic sulfides, thiiranes and thietanes respectively,

possess enough inherent ring strain to polymerize when properly initiated.

Polymerization of these systems can occur by an anionic, cationic or coordinated

polymerization mechanism. Each mechanism will be discussed briefly in the following

sections.


Anionic Polymerization


The first thiirane based high polymer was obtained in 1961 by Boileau and

Sigwalt by using various anionic initiators such as sodium, potassium hydroxide, and

sodium amide.5 They studied the effects of various initiators, techniques and purity, with

respect to reactivity and properties.6 The polymers obtained were typically high

molecular weight, highly crystalline, and soluble in high boiling solvents such as

hexylmethylphosphoric triamide.7 In contrast, the polymers of methylthiirane tend to be









amorphous materials that are readily soluble in common organic solvents. Interestingly,

when napthylsodium is used as an initiator in tetrahydrofuran solution, no termination or

transfer reactions occur under conditions of optimal purity.8 Without the contention of

chain transfer and termination reactions to consider, this monomer system has been

wisely utilized in the construction of several block and multiblock copolymers.9-13 As

previously mentioned, several anionic initiators have been successful in polymerizing

various substituted thiiranes, and depending on the employed initiator, different initiation

steps have been proposed to predominate.14,15 Classical anionic polymerization of

methylthiirane proceeds with the use of organolithium reagents. As described in Figure

1.2, the monomer reacts with an organolithium reagent to form a lithiumalkylthiolate and



s S + C2H5-Li C2H5S- Li++ CH2=CH-CH3

H3 )



S HP
CH3 H3


n
Figure 1.2. Alkyllithium initiation and propagation of methylthiirane
polymerization via alkylthiolate.


propene. Propagation then proceeds through the thiolate anion, which is also believed to

be the active initiating species for polymerization.16

Similarly thietane, the four-membered cyclic, has been polymerized using

napthylsodiuml7 in highly polar solvents to obtain controlled high molecular weight

polymers, via a similar reaction mechanism. In contrast to methylthiirane, which

polymerizes through a thiolate intermediate with alkyllithium initiators, 2-methylthietane









can also be polymerized by n-butyllithium,18 yet proceeds via active carbanion chain

ends (Figure 1.3). These carbanion chain ends are readily active and can initiate the

polymerization of several vinyl monomers, such as styrene (at -780C), to construct well

defined block copolymers.19,20



S + C4H9-Li H9C4-SCHCH2CH2" Li-

CH3
CH3

Figure 1.3. Carbanion initiation of 2-methylthietane by alkyllithium.


Cationic Polymerization


Cyclic sulfides are also commonly polymerized via cationic methods. Strong

acids, Lewis acids or strong alkylating agents are routinely chosen as appropriate

initiators. Kinetic studies of the polymerization of methylthiirane with triethyloxonium

tetrafluoroborate as the initiator have revealed that initiation is believed to go through the

formation of a three-membered sulfonium salt with propagation occurring at the a-carbon

(Figure 1.4).21

Contrary to the anionic mechanism, cationic initiators do not produce polymers

void of termination or chain transfer events. Intramolecular backbiting cyclization is the

principal cause of termination, where a sulfur atom in the polymer main-chain attacks the

sulfonium cyclic. This observation has been further supported by the presence of twelve-

membered rings, which implies that these polymers are terminated by macrocyclic

sulfonium salts.22 These macrocyclic terminated polymers are known to be slowly

reinitiated as the macrocyclic chain end is cleaved from the polymer chain and reforms

the true propagating species.








Initiation

Hz S + [(C2H5)30]+BF4- ki S-C2H5 + (C2H5)20
H3C H3C

Propagation


CA- + SS k_ P S HS-CH-CHS
H3 CH3 CH3

Termination
CH3 CH3
CH3 CH3 CH3

( SktS-CH-C 43,k S-CH2-C S

H3C H3C

Figure 1.4. Cationic polymerization of methylthiirane.


Coordinated Ionic Polymerization

This area of polysulfide chemistry has not been heavily investigated but the
polymers produced by this process have utility and potential unmatched by the previously
described ionic methods. Substituted thiiranes have not only been polymerized to
controlled high molecular weight materials, but the application of various zinc and
cadmium reagents have imparted high degrees of crystallinity unattainable by their ionic
counterparts. In addition, polymethylthiirane has been synthesized with high degrees of
stereoregularity using various zinc and cadmium reagents, in some cases without
termination events.23,24 The lack of competing termination presents some obvious
advantages for the synthesis of various phase separated block copolymers. Further, the
utilization of several cadmium alkyl initiators has provided a unique synthetic route to









highly crystalline polysulfide materials in aqueous media.25 This methodology provides

a distinct advantage over the more often utilized, yet purity dependent, anionic route.

Condensation Polymerization

Condensation polymerization to yield polymonosulfides can be generalized into

two basic categories: Thiol addition to olefins and thiol addition to halides. Polymers

obtained by thiol or mercaptan addition to an olefin has been loosely defined in the

literature as a class of condensation polymer. The polymers of this class are referred to as

condensation polymers based on the fact that the final polymer obtained has the same

general backbone structure of the polymer derived from the classical condensation of a

dithiol and dihalide.26 As flawed as this historical classification may be, it will not be

challenged here.

Addition of Thiol to Olefin

Marvel and coworkers have been the principal investigators behind the study of

polyaddition of dithiol and diolefin monomers to produce polymonosulfides.27-38 It also

has been shown that various unsaturated thiols have the ability to spontaneously

polymerize39 while others simply need the addition of a peroxide initiator to induce

polymerization to generate the linear polymonosulfide.40 Both methods have been

effective in producing high molecular weight materials (up to 60,000) under bulk and

solution conditions, although the most promising results, as shown by Marvel, have been

obtained from emulsion systems.30 In general, the polymerization is believed to occur


+ H SH -- S-(CH2)6]-



Figure 1.5. Polyaddition of dithiol and diolefin monomers to produce
polymonosulfide.









via a free radical mechanism, generating only the purely linear saturated polymer in a

step process (Figure 1.5).28 The addition of mercapto monomers to a wide variety of

derivatized diolefins has also been successful in producing polymers with

ketone,37ester,27 amide,27 aromatic,41 and sily135 functionalities. The diversity of this

approach has made this an attractive route to a variety of functionalized

polymonosulfides, yet high degrees of conversion and concomitant high molecular

weights are limited. This is primarily due to the ease in which the reactive thiol moiety

can be oxidized to the nonreactive disulfide linkage. Side reactions of this nature have a

positive effect on increasing molecular weight, but the structural purity of the material

becomes inconsistent and unreliable.


Addition of Thiol to Dihalide


This process is a true polycondensation reaction where a dithiol reacts with a

dihalide to produce predominantly polymer and a small degree of cyclic dithioethers, as

described in Figure 1.6. Polymers produced by this method typically have molecular

weights less than 3000.42 Due to the step nature of this reaction process, purity, thiol

oxidation, and other factors which hinder the high degrees of conversion required to make

high polymer, limit this process significantly. The unsaturated polymonosulfides

developed by ADMET chemistry, which will be described in Chapter 3, are most

analogous to this classification of polysulfides.


n NaS-(CH2)x-SNa + n Br-(CH2)x-Br N -[ S-(CH2)x--- + n NaBr
L *'n


Polycondensation of a dithiol and dihalide.


Figure 1.6.









History of Amine-Containing Polymers


Similar to the area presented on sulfur-containing polymers, polymeric amines

and ammonium salts have generated substantial research efforts in both academic and

industrial institutions. Polyamines have been practically applied as chelating polymers,

ion exchange resins, purification of waste waters, flocculation, biomedical and

pharmaceutical applications, emulsifiers, polymeric catalysts, and photographic

applications.43 Due to the vast nature of this subject, this introduction will only focus on

main-chain polyamines, which are analogous to those which will be described in this

dissertation. The area of polyammonium salts will not be discussed here, but has been

reviewed extensively by the father of cyclopolymerization, George B. Butler.44


Synthesis of Main-Chain Polyamines

Main-chain polyamines represent an important class of polymeric materials that

typically are produced via a ring opening process or by condensation polymerization.

Since both synthetic methods have been well documented as effective in generating

polymers of a similar structure to the polymers of this study, it is meaningful to briefly

discuss both methods here.


Ring Opening Polymerization


The two most general and direct methods for synthesizing high molecular weight

linear main-chain polyamines have been derived from substituted and unsubstituted

aziridine and azetidine aminocyclic monomers (three and four-membered rings

respectively). These monomer units are comparable to the previously described cyclic

sulfides and possess tremendous versatility in regard to chemical modification, which can

be translated to a wide range of potential polymer structures.









Aziridine Polymerization


Aziridines are typically divided into three classes of compounds on the basis of

chemical reactivity. This classification has been particularly useful for modeling

polymerization reactions. These groups have been labeled as 1-unsubstituted aziridines,

1-substituted aziridines, and activated aziridines as described in Figure 1.7. Common to

each group is the ability to undergo ring opening reactions readily which leads to

susceptability to ring opening polymerization. The compounds of each group, however,


0
N-H N-R jN-C-X

1 2 3

Figure 1.7. Aziridine Classes: 1) 1-Unsubstituted. 2) 1-Substituted.
3) Activated.


have unique reaction features which are characteristic of that group. For example, 1-

unsubstituted aziridines, like secondary amines, may undergo typical amine substitution

reactions which can specifically lead to the other classes of aziridine compounds. 1-

Substituted aziridines, a type of tertiary amine, are similar in that they are basic and

undergo conventional tertiary amine reactions such as quatemization by alkylating

agents.

For more than 50 years the most studied and well understood polymerizable

aziridine monomer has been ethylenimine. Polymerization is initiated, typically at room

temperature, by the addition of a small amount of acid catalyst which converts the

ethylenimine to an alkylating agent which is believed to be the initiating species (Figure

1.8). Although a different alkylating agent may be used, the greater basicity of the

unsubstituted aziridine soon converts the ammonium salt to the acid which creates the









Initiation
H
N-H + H N-H
or H

N- H + RX N-H N-R + N--H
R N-H

Initial Propagation
H H
N-H + N-H H >-/CH2-CH2-NH2- + N-CH2-CH2-NH3


Figure 1.8. Initiation and initial propagation in ethylenimine polymerization.


same initiating species. The initial propagation reaction has been interpreted as a simple

ring opening reaction of the protonated aziridine by ethylenimine where the driving force

of the reaction is derived from the release of ring strain. Polymer propagation continues,

but the picture becomes much more complex. Subsequent reactions of the aziridine ring

with primary and secondary amine groups promote both linear growth and competitive

branching reactions (Figure 1.9). A significant amount of research has been conducted in

the area of measuring the degree of branching in polyethylenimine and methods to reduce

or increase the extent of branching.45,46 Finally, chain termination of the growing

polymer chain has been described as a probable intramolecular cyclization reaction to

form large macrocyclic rings (Figure 1.10).45,47 Polymers produced by this process

typically have been hindered by high degrees of branching, although purely linear

polyethylenimines of low molecular weights (< 5000) have been prepared by alternative

methods.48,49

Unlike the polymerization of 1-unsubstituted aziridines, the two possible initiating

species dictate the subsequent propagation steps in 1-substituted aziridine polymerization.

When acid is used to catalyze the polymerization the initial propagating reaction creates









N H+
S[N CH2-CH2- H
N-CH2-CH2-NH2 3


H
SN--CH2-CH2-NH2 + hN-H


- N CH2-CH2- H
2


N-CH2-CH2-NH2


CH2-CH2-NH2


Figure 1.9.


Continuing propagation in ethylenimine polymerization.


NH2

/CH2-CH2-NH2


Branching

N-"v"NH + N-^- .NH2



Cyclization
NH2

/ /
H CH2-CH2-NH
N'^^ NH


Figure 1.10. Termination steps in ethylenimine polymerization.


the quaternized aziridinium ion and is simultaneously converted into a secondary amine

endgroup. Reaction of the secondary amine group, however, can result in the initiation of

new polymer chains. Alternatively, when an alkylating agent is employed to catalyze the

polymerization the initiating species is the quaternized aziridinium ion. When the ion is









formed, the initial propagating reaction creates a tertiary amine endgroup which is not

capable of transferring a cationic group to monomer to initiate new polymer chains.

Thus, the aziridinium ion endgroup is continuously created as the polymerization

proceeds until either all monomer is consumed or a non-propagating reaction destroys the

aziridinium endgroup. This reaction pathway has been claimed to produce highly 'living'

polymers with controllable low and high molecular weights (>30000, PDI < 1.3) when t-

butylaziridine is initiated with methyltriflate.50 This monomer specific synthetic method

has opened the way to synthesizing new block and graft copolymers with high degrees of

polyamine character.51 This contrasts typical cationic polymerizations of 1-substituted

aziridines, which generally lead to low molecular weight oligomers at low conversions

and cyclics in high yield.47,52 In general, 1-substituted aziridines undergo chain

termination via intramolecular cyclization (similar to their unsubstituted analogs) and

potential intermolecular branching.

Activated aziridines have been reported to be more reactive to anionic

polymerization, but they will not be discussed here since the products obtained are not

main-chain polyamines but are composed of complex main-chain mixtures of amide

nitrogen and nitrogen substituted polymeric amines.53


Azetidine Polymerization

Compared with aziridines, azetidines have received considerably less attention as

monomers for the synthesis of main-chain polyamines. This has been principally due to

the fact that the four-membered cyclic amines require long and difficult synthetic

procedures. Although they have a distinct advantage over their three-membered

counterparts with regards to lower toxicity, they are hindered by lower reactivity. Several

synthetic methods for the preparation of azetidine54-56 and 1-substituted azetidines57,58

have been developed and are reported elsewhere.









The polymerization dynamics of azetidine (poly(trimethylenimine)) are very

similar to that observed for the polymerization of aziridine (poly(ethylenimine)).

Mechanistically the reaction is cationic in nature with the thermodynamic driving force

being derived from the release of ring strain in the four-membered cyclic. Initiation can

be induced by acid catalysis or via an appropriate alkylating agent. Although the

polymerization mechanism is very similar to that presented for aziridines, the decreased

reactivity of the four-membered cyclics usually requires higher temperatures (> 600C) to

initiate polymerization. During the course of propagation, several reaction paths are

possible, due to the presence of reactive tertiary amino and primary amino groups. This

nonselective reactivity results in high degrees of branching, as presented for aziridines.

Pure linear poly(trimethylenimine) has not been produced via this method but has been

produced by the isomerization-polymerization of oxazine, followed by alkaline

hydrolysis of the obtained polyamide.48 The resulting hydrolyzed polymer is a white

crystalline solid as opposed to the viscous oil obtained from the cationic ring opening of

azetidine. The physical difference is most likely due to the absence of side-chain

branching which prevents polymer crystallization.

Consistent with the mode of polymerization of 1-substituted aziridines,

substituted azetidines are polymerized employing similar cationic initiators at higher

temperatures. These systems produce high yielding, moderate molecular weight main-

chain polyamines with highly 'living' character and limited side-chain branching.59,60

These polymers have also been utilized to prepare block copolymers and graft

copolymers containing polymeric tertiary amino segments. Polyionene type polymers

with various counterions have also been prepared using substituted azetidine salts .61


Condensation Polymerization


Synthetic procedures for linear, main-chain polyamines have been well

documented but have proven to be quite complicated and in many cases, limiting. The









ionic polymerization of substituted and unsubstituted aziridines and azetidines, though

conceptually simple and straightforward, seldom produces linear polyamines due to the

propensity for side-chain branching. An alternative method for producing linear main-
chain polyamines has been accomplished via polycondensation reactions of primary a, co-

alkanediamines with a, ot-dihalo-alkanes. This method has proven to be quite effective,

yet suffers from undesirable cyclization, branching, and/or crosslinking reactions which

limit its utility as a viable approach to linear polyamines possessing controlled

structure.62,63

In every case reviewed thus far, the main-chain skeletons are more or less

predetermined by the constraints imposed by the nature of the particular monomer and

reaction method employed. Several attempts have been made to synthesize well defined

linear polyamines via a polycondensation approach due to the ease and versatility of the

procedure. The most successful polycondensation approach uses tertiary diamines and

dihalogenoid compounds via the Menshutkin reaction (Figure 1.11).64,65 This method

produces ionic polymers which contain quaternary ammonium groups in the backbone,


H3 CH3 CH3 CH3 B
N CH2 N + Br-(CH2 -Br C -
H3C" x CH3 y
CH3 CH3 .n

Figure 1.11. Polycondensation to linear main-chain polyamines.


which reduces the potential of crosslinking and side-chain branching within the polymer

main-chain, although cyclized ring structures are often problematic.66 This procedure

has also been effective in producing tertiary amine-containing polymers of moderate

molecular weights (< 2000) by selective N-dealkylation of the polyammonium main-

chain with various nucleophiles.67,68









Development of Olefin Metathesis and Catalyst Systems


In order to gain an appreciation of the research presented within, it is necessary to

possess an understanding of the art of olefin metathesis and the historical development of

the accepted mechanism of today. As described by chemical definition, metathesis refers

to the interchange of atoms by the movement of bonding electrons to form new bonds.

Olefin metathesis, a term first coined by Calderon69 in 1967, is used to express the

interchange of carbon atoms between two olefins. Olefin metathesis can be classified

into three basic categories: 1) exchange reactions, both productive and degenerative; 2)

polymerization reactions, ring opening and condensation types; and 3) degradation

processes via cyclization (Figure 1.12).

The industrial and academic impacts of these reactions have been significant over

the past 30-40 years. Exchange reactions have been the cornerstone for the production of

a-olefins for Ziegler/Natta polymerization since its discovery in the early 1960s.70,71

Degradation metathesis has realized considerable success in the area of depolymerization

chemistry72,73 and the synthesis of well defined telechelics via tandem exchange

reactions using functionalized internal olefins.74-77 Without question, the olefin

metathesis reaction has long been considered the pawn of polymer chemistry, and it has

been in this area where the true utility of this reaction has blossomed.78,79


Traditional Catalysts for Olefin Metathesis


As described in Figure 1.12, olefin metathesis requires a metathesis catalyst and a

structurally appropriate olefin, both of which have been a model of continuing chemical

evolution. Traditional catalyst systems were typically composed of various transition

metal halides and oxohalides accompanied by a wide range of metal alkyl and

organohalide Lewis acid cocatalysts. The most successful systems of this type employ












M,


Me


Me


Me


17






catalyst M

Me
1) Productive Metathesis


catalyst


Me


2) Degenerative Metathesis


S catalyst.n
1) Ring Opening Metathesis (ROMP)
1) Ring Opening Metathesis (ROMP)


catalyst


I R n + Ethylene
l *'n


2) Condensation Metathesis


catalyst


Et


Et


Figure 1.12


Olefin metathesis classifications. A) Exchange reactions;
1) productive and 2) degenerative. B) Polymerization reactions;
1) ring opening (ROMP) and 2) condensation. C) Degradation.


+ II


Me

+ r,


R









tungsten and molybdenum high oxidation state transition metal complexes with Lewis

acid cocatalysts comprised primarily of aluminum and tin.80,81 WC16/EtAICl282 and

WOCl4/SnMe470 are among many examples of traditional catalyst systems used for

olefin metathesis. Truett et al.83 and others were the early pioneers in olefin

metathesis and were able to demonstrate these classical systems to be effective in

polymerizing strained cyclic olefins, such as norbornene (Figure 1.12.B), but their

catalytic activity was low at moderate temperatures and they showed only marginal

tolerance to functionality. In addition to these restraints, the olefin metathesis mechanism

was not well understood, principally in part to the unknown nature of the true catalytic

entity involved during metathesis. In order to overcome these inhibiting limitations, the

academic and industrial community spawned an intense multidisciplinary research effort,

aimed at elucidating a convincing metathesis mechanism and increasing the tolerance of

practical catalyst systems to functional groups. This constantly evolving pursuit has

continued for more than four decades.


Elucidation of the Olefin Metathesis Mechanism


The quest to understand the mechanism of olefin metathesis lasted for more than

20 years, until the advent of well defined catalyst systems were able to decipher the true

mechanistic nature of the reaction. During this two-decade search, new catalyst systems

were specifically designed to uncover and elucidate mechanistic pathways. Olefin

metathesis appeared to be an untapped well of potential, but mechanistic uncertainties

prevented the construction of optimal catalytic systems to meet the needs of the polymer

community. The following section will briefly recount the historical developments

leading up to the accepted metathesis mechanism of today.

Natta was one of the initial investigators studying the olefin metathesis reaction,

proposing a mechanism which involved the splitting of the a-carbon-carbon s-bonds,

similar to that of the accepted mechanism of Ziegler/Natta polymerization.84 In 1967,








Bradshaw and coworkers confirmed some previously published work on the reversibility
of the olefin metathesis reaction, and proposed a 'quasicyclobutane' intermediate (Figure
1.13) as a vehicle to determine the mechanism.85 Unfortunately, this did not provide any
insight into the key role of the transition metal, but did lay foundation for later theories.


R R'. RR' .- RR


R R' R R' R R'

Figure 1.13. Quasicyclobutane intermediate proposed by Bradshaw.


Concurrent with Bradshaw's studies, Calderon and Ofstead were the first to report
a detailed mechanistic investigation into olefin metathesis, providing a model for metal
assisted metathesis for which Bradshaw admittedly had no answers. Their initial
experiments studied the olefin exchange reaction of d8-2-butene and 2-butene. The


R R'


j--iM] s, ;[M] -. [M]

R R' R R' R
R R'
Figure 1.14. Pairwise mechanism of olefin metathesis. [M] = catalyst complex.


product of this reaction, d4-2-butene, proved that the carbon-carbon double bond was
completely broken and reformed during the course of metathesis.86 This experiment
gave rise to what became known as the pairwise mechanism (Figure 1.14). This was the
first theory that provided specific insight into how the d-orbitals of the metal could
overlap with the p-orbitals of the olefin to form a cyclobutane type complex.82,87








In 1970, Herisson and Chauvin proposed an alternative mechanism simply based
on the statistical product distribution they obtained from the metathesis exchange of
cyclopentene and 2-pentene at essentially 'zero' reaction time.88 The initial product
distribution (at t=0), as shown in Figure 1.15, can only be explained by a dissimilar
mechanism to that proposed by Calderon, as the pairwise mechanism mandates that only
the exchange product could be produced (at t=0). This led to what was originally
referred to as the 'nonpairwise mechanism', but eventually yielded to the present day


Et

0(
Me

Observed Products Predicted Product


Me Et
1
Me ^^- Me
+
Et
2
KZMe
+
-Et

-EEt

Figure 1.15. Exchange reaction which led to the proposal of the metal carbene
mechanism.


accepted 'metal carbene' mechanism (Figure 1.16). Although the experiments of Chauvin
seemingly usurped the pairwise mechanism as the prevailing mechanism of olefin
metathesis, the debate created feverish research efforts for another decade.70,89-91












IM] I r [M],_
R' R1 R1

Figure 1.16. Chauvin's proposed metal carbene mechanism.


Ironically polymer chemistry, the principal driving force in the evolution of

metathesis chemistry as well as the greatest benefactor, provided the final proving ground

in determining this elusive sought after mechanism. The product of ring opening

metathesis polymerization (ROMP) generally consists of a linear high molecular weight

fraction, and a low molecular weight fraction comprised primarily of cyclic oligomers.

This behavior has been observed for several cyclic olefins.70 According to the pairwise

mechanism, molecular weight increases in a stepwise fashion via a growing macrocyclic

polymer. These fundamental principles were tested and proven to be instrumental to

supporting the metal carbene mechanism proposed by Chauvin. In 1976, Katz et al. used

a well defined metathesis active metal carbene, first synthesized by Casey and

Burkhardt,92 to initiate the ROMP of cyclooctene.93 This experiment was pivotal and

proved fatal to the pairwise mechanism as the initial alkylidene complex was

incorporated into the linear polymer chain, an event impossible by the mechanism

proposed by Calderon (Figure 1.17).

It had taken nearly 25 years since the first reports on olefin metathesis appeared to

acquire enough concrete evidence to substantiate an acceptable operating mechanism for

olefin metathesis. The final pieces of the catalytic cycle puzzle were pursued

passionately for the next decade, with various researchers contributing to the

development of the metathesis mechanism.92,94-98










(CO)5W=CPh2 Ph2

n

Figure 1.17. First ROMP with a well defined carbene catalyst.


Development of Well-Defined Metathesis Catalysts


The primary impetus behind developing well-defined metathesis catalysts was to

elucidate the olefin metathesis mechanism. The metal carbenes developed by Casey,92

Tebbe,95 Grubbs96 and others were vital to this goal, but these systems demonstrated

poor catalytic activity as compared to their ill-defined brethren. As described previously,

several classical catalytic systems existed which exhibited high activity, but required the

addition of a Lewis acid cocatalyst. The following section will focus on the recent

developments of highly active, well-defined catalytic initiators which do not require the

support of a Lewis acid cocatalyst, and as a consequence are more tolerant to

functionality. The work of Schrock and Grubbs will be the principal focus of this section,

as the evolution of their catalyst systems directly pertains to the work discussed in

Chapters 3-4.

In 1986, Schrock developed the first Lewis acid free (no cocatalyst required),

highly active four coordinate dO tungsten alkylidene complex capable of catalytic

metathesis (Figure 1.18).99 These alkylidene catalysts have been at the epicenter to

which all modern day metathesis chemistry has been forged and compared. These

'Schrock alkylidenes' have found unparalled utility in the mechanistic study of the

metathesis reaction. Prior to the invention of these catalysts, many of the key

intermediates in the metathesis cycle, various substituted and unsubstituted

metallacyclobutanes, were unknown. With their discovery, structure elucidation and

mechanistic pathways were finally delineated.100-104 It was believed that the imido, N-









2,6-C6H3-i-Pr2, ligand was essential to provide the necessary steric bulk to prevent

intermolecular dimerization. 105


R

(CH3)3C

N A(CH3)(CF3)2C
,II M=W,Mo
RO/ M C3F7(CF3)2C


Figure 1.18. Highly active Lewis acid free (cocatalyst) Schrock alkylidenes.


Schrock was also able to demonstrate that the metathesis activity could be

strategically controlled via the influence of the electronic and steric factors of the

associated alkoxide ligands (Figure 1.18).100,105 Experimental data indicated that the

greater the electron withdrawing power of the alkoxide ligand (via fluorine substituents),

the more electrophilic the metal became, and consequently the greater the stability of the

resulting metallacyclobutane intermediate. For example, unfluorinated alkoxide ligands
such as, OR = OC(CH3)3 show only marginal metathesis activity with internal acyclic

olefins (2 turnovers/h for 2-pentene), but are quite active as ROMP initiators.106
Alternatively, perfluorinated alkoxides such as OR = O(CF3)2C3F7 produce a

metallacyclobutane too stable for useful metathesis reactions. However, when OR =
OC(CF3)2CH3, a perfect balance of electronic and steric factors is provided which

results in optimal performance, particularly for acyclic dienes (1000 turnovers/min for 2-
pentene).

Although these highly active tungsten based alkylidenes became quite useful for
olefin metathesis, they presented many of the same limitations as their classical

counterparts. It had already been reported in the literature that polar functional groups,

such as esters, readily reacted with metathesis catalysts based on Ziegler/Natta metals








(Ta, Ti, and Zr).107-109 These tungsten based systems proved to be similar in this

regard, terminating reactivity with ester functionalities in a Wittig type fashion (Figure

1.19.A). The inability of these alkylidenes to metathesize functionalized olefins spawned

a new era of functionally tolerant metathesis catalysts.

In 1987, Schrock realized that the molybdenum analog of the tungsten alkylidene
would possess a less electrophillic metal center, and therefore may be tolerant to certain

carbonyl functionality. The molybdenum alkylidene did in fact demonstrate a higher

degree of specificity for metathesis by polymerizing ester containing olefins110 (Figure

1.19.B), but Wittig-like termination continued with more reactive carbonyl containing

compounds (i.e. aldehydes and acetone). These highly selective, functionally tolerant

molybdenum alkylidenes have allowed the incorporation of multiple functional groups

into the backbone of ROMP polymers: ether, thioether, ester, nitrile, halogen, and metal

containing polymers have all been polymerized by ROMP methods.105 In addition, Fu

and Grubbs have utilized this catalyst system to efficiently synthesize a series of

oxygen111 and nitrogen112 containing heterocycles via ring closing metathesis (RCM)

of symmetrical and unsymmetrical dienes.


[W] 0 H R
A) + II 0 [W]=O +
A) 'RRCOCH3 'R OCH3





CO2CH3 [Mo]
B) & .CO2CH3 0 n

H3CO2C CO2CH3

Figure 1.19. Reactivity of Schrock's tungsten alkylidene and molybdenum
alkylidene toward ester functionality, a) Wittig-type deactivation
of [W] alkylidene. b) ROMP of ester functionalized norbomene via
[Mo] alkylidene.









As will be detailed in following sections, including the main body of this work,

Schrock alkylidenes have proven to be an integral part of the recent success of ring

opening metathesis polymerization (ROMP) and the key ingredient to the realization of

acyclic diene metathesis (ADMET) polymerization as a viable route to polyalkenylenes.

While the modem day explosion of research endeavors in the area of olefin metathesis

and metathesis polymerization can be attributed to the 'Schrock alkylidenes', the recent

advances made by Grubbs and coworkers may have created a reality for olefin metathesis

never dreamt possible. 113-119

The multidisciplinary efforts to develop more rugged and functionally tolerant

catalyst systems shifted focus from the success of early transition metal complexes (Ti,

Ta, W, Mo, etc.) toward synthesizing highly active late transition metal alkylidenes. It

was believed that late transition metal alkylidenes should exhibit increased metathesis

activity in the presence of highly functionalized olefins, and show resistance to protic

media. These observations had been precedented for some rhenium complexes,120 in

addition to an aquaruthenium complex, which initiated ROMP of a 7-oxonorbomene

derivative in an aqueous environment.114

In 1992, a landmark publication by Grubbs presented the synthesis of a well-

defined ruthenium alkylidene which initiated ROMP in both organic and protic/aqueous

solvents.115 In both solvent systems, a stable propagating species was observed and

produced polynorbornene in a 'living manner.' This pentacoordinate ruthenium

vinylcarbene (Figure 1.20) was stable for several minutes in air and stable for several




PR3
ClU, I ,o R= Ph, Cy (cyclohexyl)
Ru
Cl I /
PR3


Figure 1.20. Grubbs' first generation ruthenium alkylidene.









days in organic media contaminated with protic solvents (water, alcohol, HC1 in diethyl

ether). In an analogous study that was initially conducted with Schrock's molybdenum

alkylidene, Fu and Grubbs showed this catalyst system also to be highly effective for ring

closing metathesis (RCM). In addition to the products in their prior study, they were able

to synthesize unsaturated heterocycles and cycloalkenes with functionalities aldehydess,

alcohols, carboxylic acids and quaternary amines Figure 1.21) unattainable with

previous metathesis catalyst systems.121 This complex represented the first generation

of highly robust, well-defined late transition metal alkylidene catalysts for olefin

metathesis. Although this system showed unprecedented promise due to its remarkable

tolerance to functionality and resistance to protic media, limited reactivity (primarily with

acyclic olefins), slow initiation times (presumed to be due to intramolecular coordination

of the conjugated ligand) and synthetic difficulty limited its utility.116



[Ru]


X = CO2H, CH2OH, CHO

Figure 1.21. Highly functionalized ring closing metathesis (RCM) with [Ru].


The diphenylvinyl ruthenium alkylidenes described in Figure 1.20 are prepared
from the ring opening of diphenylcyclopropene with RuCl2(PPh3)3. The

RuCl2(=CHCH=CPh2)(PCy3)2 analog is the product of PPh3 exchange with PCy3

(cyclohexyl). The low yield and synthetic difficulties of producing diphenylcyclopropene

made the synthesis of these complexes far from trivial. Grubbs and coworkers began

designing new, more highly active catalyst systems that were more easily accessible, but

did not sacrifice the functional group stability and tolerance to protic media of its first

generation predecessor.








As is so often the case with Robert H. Grubbs, he can make the monumental look
mundane. This tremendous strategic/synthetic challenge was accomplished with

revolutionary results which may forever change the way the chemical community views
olefin metathesis. Due to the synthetic diversity and ease of synthesis, a diazoalkane
approach to a series of alkylidene complexes was conducted to test the influence of the
alkylidene moiety on reactivity (Figure 1.22).119


PPh3 PR"3
R= Me, Et CI,. I| h H I H
C iu 2 PR"3 Cl",,u ==<

N2 1 -N2 PPh3 PR"3
RuCl2(PPh3)3 + 80C -PPh3 3 PR3
-PPh3 PR"3
R H Cl.. R nH 2 PR" CIC lu H
Ru= 2 PR" "Ru
Cl R' I R'
SPPh3 PR"3
X R" = Cy, Cp, i-Pr
X = H, NMe2, OMe, Me, F, CI, NO2

Figure 1.22. General synthesis of next generation ruthenium alkylidenes.


This elegant study showed that ruthenium alkylidene complexes of the general
type RuCl2(=CHR)(PPh3)2(R = alkyl, aryl), could be synthesized via a one pot

procedure, display enhanced metathesis activity (up to 103 faster), and produce low

polydispersity polynorbornene in a living manner, as compared to their laborious
diphenylvinyl alkylidene analogs. Despite increased metathesis activity, these systems
exhibited similar limitations toward the ROMP of less strained cyclic olefins (>10-15
kcal/mol) and metathesis exchange of acyclic olefins. 119 Alternatively, as demonstrated
in the first generation ruthenium alkylidenes, the RuCl2(=CHR)(PCy3)2 system obtained

from phosphine exchange showed a marked increase in ROMP activity of less strained
cyclic olefins and became the first ruthenium alklidene to show high levels of metathesis
exchange for acyclic olefins.









Comparative kinetic studies for the metathesis exchange of 1-hexene with
RuCl2(=CH-p-C6H4-X)(PCy3)2 have shown that the electronic influence of X on the

initiation rate to be relatively small, with X = H giving the optimum results.
RuCl2(=CHPh)(PCy3)2 has been fully characterized and demonstrates long term air

stability in the solid state and in solution (CH2C12 or C6H6) contaminated with alcohols,

amines, and water. The high stability of RuCl2(=CHPh)(PCy3)2 toward functional

groups was further demonstrated to be of great utility as functionalilzed olefins (acetate,

halogens, and alcohols) were incorporated into the alkylidene moiety. 119 These

alkylidene complexes can therefore give precise control to the end groups of polymer

structures, which could have tremendous application in the area of telechelics. Further,

this catalyst system has also metathesized conjugated and cumulated olefins to generate

the corresponding vinylalkylidene and vinylidene complexes respectively.119

This highly active ruthenium catalyst has also had a significant impact on olefin
metathesis reaction from a mechanistic standpoint. When exposed to 1 atm of ethylene,
RuCl2(=CHPh)(PCy3)2 quantitatively produces RuC12(=CH2)(PCy3)2. This has proved

to be the first isolated and fully characterized metathesis active methylidene complex.119

This discovery, as will be described in a following section, has indirectly provided

tremendous support to the proposed mechanism of acyclic diene metathesis (ADMET)

polymerization.


Olefin Metathesis Polymerization


For nearly three decades ring opening metathesis polymerization (ROMP) had
been the cornerstone of olefin metathesis and provided the incentive for most of its

research endeavors. Recently, olefin metathesis polymerization has been extended to

include a new method which utilizes an equilibrium condensation exchange reaction to

produce high molecular weight polymers. Both methods will be discussed in detail, as it

directly pertains to the scope of the work described herein.











Ring Opening Metathesis Polymerization (ROMP)


Ring opening metathesispolymerization (ROMP), as described in Figure 1.12.B 1,

occurs via a chain growth process to produce linear high molecular weight polymers. As

for all olefin metathesis reactions, ROMP is governed by competing equilibria. The

thermodynamics of the ring-chain equilibria dictate the polymerizability of cyclic

olefins:70


AG =AH TAS (1)

Polymerization is typically enthalpy (AH) driven as a consequence of a release in ring

strain of the monomer unit as the reaction proceeds to polymer. The bond angle strain in

3, 4 and 8-membered rings, in addition to bicyclic monomers such as norbomene, supply

the necessary energetic for polymerization. The inherent ring strain of these monomer

systems allow the equilibrium to be shifted from cyclic monomer toward linear polymer.
The polymerizability of strain free (AH=0) macrocyclic olefins is an entropically (AS)

driven process as a result of the formation of linear polymer from discrete monomer

units. The ROMP of 5, 6, and 7-membered rings presents a thermodynamic dilemma.
Possessing competing enthalpy and entropic terms (AH-AS ~ 0), these stable cyclic

olefins can be polymerized only when the temperature variable is controlled.122,123
This phenomenon is referred as the polymerization ceiling temperature (Tc), the

temperature above which polymerization of any cyclic monomer is prohibited. For

example, Patton and McCarthy have demonstrated that by lowering the temperature of

polymerization to -23C, it even became possible to polymerize cyclohexene. 123

The mechanism of ROMP chemistry is outlined in Figure 1.23 for the

polymerization of norbornene. The catalyst, [M], is a transition metal carbene complex

obtained via a preformed well-defined initiator, or generated in situ from a classical

catalyst system, as described earlier. The first stage of polymerization occurs when the









olefin in the monomer unit coordinates to the metal to form an initial 7n complex (A). The

monomer then undergoes an insertion process to form a metallacyclobutane intermediate

(B). At this stage of the equilibrium process, productive cleavage of the
metallacyclobutane must take place for propagation to occur, thereby generating a chain
extended 7t complex (C). Finally, disassociation of the n complex resets the catalytic

cycle (D).



_(A) (B)

HH
+ H ( H [M]
[M] [MN]

P P




[M][ (C)

(D) [M H
P


Figure 1.23. Mechanism for the ROMP of norbornene. [M] = catalyst, P =
polymer.


Ring opening metathesis polymerization, as the above mechanism implies, is
controlled by a chain growth mechanism. Polymerization or chain propagation continues
at a reactive, growing chain end until secondary metathesis reactions (chain transfer or
cyclization) become significant. When these secondary reactions become prominent the
thermodynamics of the polymerization is then controlled by ring-chain equilibria.124
Classical catalysts impart standard chain growth characteristics, such that the molecular
weight and polydispersity change only marginally after initial reaction. Under these









conditions, the equilibrium molecular weight of the polymer is essentially independent of

the extent of monomer conversion.

It has always been a goal of polymer chemists to synthesize polymers of precisely

controlled structure and narrow molecular weight distributions. This ideal state usually

can only be achieved if the rate of initiation is both faster than the rate at which the

polymer chain propagates and much faster than the rates of termination or chain transfer.

Polymerizations which conform to these criteria are referred to as 'living' and result in

polymeric materials with narrow molecular weight distributions (Mw/Mn < 1.1) and

expected degrees of polymerization dictated by the initial monomer to initiator (catalyst)

ratio.125 Living polymerization is a special case of chain growth where molecular

weight increases linearly to the consumption of monomer; i.e. molecular weight increases

linearly with time. Living polymerization systems are particularly useful due to the ease

in which molecular weight, polydispersity, and structure can be controlled. Due to the

persistence of the active chain end, the addition of another suitable monomer upon

consumption of the original monomer can provide an excellent route to versatile block

copolymers .126

Classical ROMP catalysts typically polymerize via a classical chain growth

process, although some systems have shown to produce polymers of a 'living' type for

short reaction times, finally succumbing to secondary metathesis reactions which broaden

the molecular weight distribution.70,127 Until the recent advent of well-defined ROMP

catalysts, 'living' polymers were only accessible via anionic,128 cationic,129 and group

transfer polymerizations.130 Grubbs and Gilliom reported the first 'living' ring opening

metathesis polymerization of a cyclic olefin, using a thermal generated titanocyclobutane

to polymerize norbornene.131 The polymers obtained from this catalyst have narrow and

controllable molecular weight distributions (<1.1) and possessed a characterizable 'living'

chain end which were eventually utilized to synthesize a wide variety of well-defined

block copolymers.132-135 As discussed in the previous section, the development of









well-defined catalytic initiators which do not require a Lewis acid cocatalyst have proven

instrumental in the area of synthesizing high molecular weight functionalized polymers

and block copolymers in a 'living' manner. The recent advances in catalyst design,

primarily due in part to Schrock and Grubbs, have given polymer chemists new ways of

determining and controlling overall polymer structure.


Acyclic Diene Metathesis (ADMET) Polymerization


The advent of the well-defined alkylidene catalysts put forth by Schrock and

Grubbs continue to have profound impact on the viability of ring opening metathesis

polymerization (ROMP). However, it was the early contribution made specifically by

Schrock which helped to forge a new metathesis polymerization reaction which had

appeared too problematic to overcome. Acyclic diene metathesis (ADMET)

polymerization has been an area of intermittent study for the past 25 years, and only since

the discovery of Schrock alkylidenes has this class of polymerization become a practical

reality. This section will attempt to focus the importance of the work described in this

dissertation in addition to providing a cursory overview of ADMET polymerization.

Acyclic diene metathesis (ADMET) polymerization is an equilibrium, step

propagation condensation reaction where the production and removal of a small alkene,

typically ethylene, drives the reaction to form high polymer (Figure 1.24). This

mechanism varies significantly from that of ROMP, where the driving force of the

reaction is derived from the release of ring strain in the monomer unit and the polymer

forms via a chain growth process.


R catalyst + Ethylene


Figure 1.24. General reaction scheme for ADMET polymerization.









Early investigations into the potential application of the metathesis polymerization

of acyclic olefins garnered only marginal success, with low molecular weight oligomers

and competing side reactions providing the principal limitations. 136,137 These initial

studies were conducted using classical metathesis type catalyst systems. The prevailing

belief was that competing vinyl addition chemistry, due to the presence of the Lewis acid

cocatalyst, prevented the formation of high polymer.

In 1987, Lindmark-Hamberg and Wagener demonstrated in a detailed
reinvestigation of the ADMET polymerization of 1, 5-hexadiene with WCl6/EtAICl2 that

even though metathesis products were predominantly obtained, insoluble materials were

also observed and believed to be the product of vinyl addition crosslinking. Wagener and

coworkers conclusively determined that cationic vinyl addition chemistry was the

limiting factor to polymerization as demonstrated in an irrefutible model study using

styrene, a well documented cationic polymerizable monomer. The results of the study

showed that competing vinyl addition reactions occurred to give polystyrene exclusively,

rather than the expected metathesis product, stilbene (Figure 1.25).138

The development of Schrock's well-defined tungsten alkylidene complexes

concurrent with this study led Wagener and coworkers to employ these systems toward

the metathesis of acyclic dienes in order to eliminate competing vinyl addition reactions.

This approach proved successful, as evidenced by the quantitative conversion of styrene

into stilbene (Figure 1.25). This result fostered an in depth investigation of the dynamics

and potential of ADMET polymerization, which still continues today.

As described earlier, ADMET polymerization is a step condensation process

which must abide by strict statistical requirements for high molecular weight polymer to

form.125 As the name implies, step polymers exhibit increasing molecular weight in a

stepwise fashion (i.e. monomer forms dimer; dimer forms tetramer etc.) For high

molecular weight polymers to be obtained by this method, complete conversion (>99%)of

the reactive functional groups (terminal olefins) is required. This is contrary to that of








ROMP, where molecular weight grows in a chain addition fashion. For ROMP,
propagation occurs at the end of the growing polymer chain, generating high molecular
weights at low monomer conversions.


K-


WC16/EtAlCl2


NAr
0CW OC(CH3)(CF3)2
OC(CH3)(CF3)2






+ Ethylene


Vinyl Addition Product Metathesis Products
(Polystyrene) (Stilbene)

Figure 1.25. Effect of Lewis acidity on ADMET.


In order to achieve these high levels of conversion necessary for polymer
formation, the stoichiometry of the functional groups present must be meticulously
balanced and all side reactions must be eliminated. The average degree of polymerization
(DP) for a step process is controlled by the measure of converted functional groups (p)
and the balance of stoichiometry (r):


DP = (l+r) / (l+r-2rp)


The stoichiometric imbalance ratio, r, for pure acyclic dienes is equal to unity and
therefore ADMET polymerization or any step polymerization which has perfect
stoichiometric balance of functional groups can be predicted by the Carothers' equation:











DP = 1 / (1-p)


Figure 1.26 graphically depicts Carothers' equation and shows how the degree of

polymerization, DP, is highly dependent upon large functional group conversions

(>99%).125 These strict statistical guidelines for step polymerization in part explain the


200


150


DP = 1 /(1-p)


DP
5
10
20
100
200


P
0.8
0.9
0.95
0.99
0.995


0 0.5


Figure 1.26.


Stepwise molecular weight growth.


early failures of ADMET polymerization due to the competing vinyl addition reactions

which served to limit functional group conversions and alter the original inherent

stoichiometric balance. With the development of Schrock's well-defined alkylidenes,

which eliminated Lewis acid competition, and the work put forth by Wagener and

coworkers, ADMET became a viable polymerization process.

With the concern for competing side reactions being eliminated, Wagener et al.
used Schrock's tungsten alkylidene, W(CH-t-Bu)(NAr)(OC(CH3)(CF3))2, to synthesize

the first ADMET polymer by condensing 1,9-decadiene to form polyoctenamer (Figure

1.27).138 In addition, perfectly linear 1,4-polybutadiene was obtained from 1,5-









hexadiene and various random copolymers were soon discovered. 139,140 This catalyst

system was used quite extensively and established many of the basic mechanistic and

synthetic rules which govern ADMET. Also at this time, the structure-reactivity

relationships for successful ADMET were beginning to develop.

NAr
c: W OC(CH,)(CF3,)
(CH2) x (CH2)6- + Ethylene
n
Figure 1.27. The ADMET of 1, 9-decadiene to produce polyoctenamer.



The metathesis polymerization of 1, 9-decadiene proved instrumental in trying to

determine the operating mechanism which has come to be known as the ADMET cycle

(Figure 1.28). The mechansim, as proposed by Wagener, Boncella, Nel, and Hillmyer

was fundamentally based upon theoretical and experimental observations.139 The

mechanism evolved from the fact that the polymerization reaction was consistent with a

true step polycondensation process: molecular weight distributions approached two (as

predicted); there was a continuous production of ethylene; and high molecular weight

polymer was formed only at high conversions.

The mechanism for ADMET polymerization, as described above, contains several

key intermediates which are consistent with ROMP chemistry (Figure 1.28). The

fundamental difference between the two reactions is that ADMET is an equilibrium

process which is driven entropically from the release of a small olefin, whereas ROMP

typically propagates in an irreversible manner due to the release of ring strain in the

cyclic olefin. First, olefin coordination to the alkylidene forms a nt complex which

undergoes an insertion reaction to form a productive (1) and nonproductive

metallacyclobutane intermediate. The nonproductive intermediate collapses to regenerate

the original nt complex, where (1) undergoes productive metathesis to eliminate the









/C (CH2)J
(CH2)6
+ --- L + L
LnNM=CR2 LnLi=CR2 R R
(1)

///\)R2C::

H2C- CH I(CH2)6- mon
2 LnlM=--q (orpo
H
(2)


monomer


Ln-M=CH2
(4)


(CH2)6
nR

R R


=CH2

lomer
'lymer)


-n"v "(CH2)6

S(3)



(CH26 (CH2 6
L -In


Figure 1.28. The ADMET polymerization cycle.


precatalyst fragment and a new alkylidene at the end of a monomer unit (2). The
polymerization cycle continues with coordination of another monomer unit to the metal
containing monomer to form the productive (nonproductive not shown) a, Oj-disubstituted
metallacycle (3). This new metallacyclobutane (3) collapses to form an ADMET dimer
(initially growing polymer in subsequent cycles) and a methylidene (4), the latter of
which is believed to be the true active catalyst species for this system. Although an
unstabilized methylidene (4) has never been observed using Schrock type alkylidenes,
Grubbs has recently isolated and characterized the first metathesis active ruthenium









methylidene,119 which provided further support to the proposed ADMET cycle. The

methylene alkylidene continues the cycle by reacting with another monomer (or polymer

chain end) to form metallacycle (5) which is the precursor to the formation of ethylene.

Ethylene is continuously removed which forces the equilibrium in a clockwise manner

towards generating high molecular weight. Due to these competing equilibria, ADMET

polymerizations, unlike for ROMP chemistry, are generally conducted under bulk

conditions to favor the release of ethylene and minimize cyclization. This true

equilibrium cycle also applies to the reverse process where high molecular weight

unsaturated polymers can be depolymerized in solution with excess ethylene 141,142 and

various functionalized monoenes to produce mass exact telechelics.74,75,143

With a general understanding of the ADMET cycle in hand, attentions soon

turned to determining the role of substituent effects and aggressive functionalities on

monomer reactivity. Wagener and Konzelman were able to demonstrate that substituent

effects play an important role in the metathesizability of a, co-hydrocarbodienes. It was

shown in a series of model and monomer compounds that when an alkyl substituent is

positioned at either the 2-carbon or a to the olefin the compound is rendered metathesis

inactive (Figure 1.29).144 This is most probably due to steric interactions and the

nonfavored nature of the tetrasubstituted metallacyclobutane intermediate and its

concomitant tetrasubstituted product. This metathetic nonreactivity created an

opportunity to synthesize perfectly alternating butadiene-isoprene copolymers from

strategically designed monomers (Figure 1.29).145

In addition to the hydrocarbon polymers and copolymers synthesized using

Schrock's tungsten alkylidene; polyferrocenes146 and poly-p-phenylenevinylene

copolymers147 were also investigated. Concurrent with these studies was the exploration

into the area of synthesizing polymers with highly functionalized characteristics.

Wagener and Brzezinska were able to report that highly functionalized ether containing










Reactant


Product i



No Reaction





No Reaction







n
perfectly alternating
butadiene-isoprene copolymer


Figure 1.29. Hydrocarbon model and monomer studies to determine reactivity.


dienes could also be polymerized using Schrock's tungsten alkylidenel48 if certain

monomer-structure rules are applied. When less than three methylene spacers separated

the reactive olefin from the Lewis basic oxygen, metathesis was sluggish or completely

prevented (for divinyl ether and diallyl ether). However, when the olefin was separated

by three or more spacers, polymerization occurred cleanly yielding unsaturated

polyethers. As was observed for ROMP, Schrock's molybdenum alkylidene

demonstrated higher reactivity in the presence of functionalized cycloalkenes and the

focus of Wagener and coworkers, as well as other researchers, began to exclusively

utilize this more functionally tolerant catalyst system. Wagener and Brzezinska showed

the molybdenum alkylidene, Mo=[CH(C(CH3)2(Ph))][N-2,6-C6H3-i-Pr2]

[OCCH3(CF3)212, to be even more effective for the polymerization of ether containing

dienes.149









The general synthetic rules for ADMET were established with an understanding

of the role of substituent effects in tandem with the effects of carbon spacing between a

reactive olefin and a potentially poisoning functionality. In light of these developments, a

novel polycondensation approach was created in an effort to synthesize a wide array of

functionalized linear unsaturated polymers and copolymers using Schrock's molybdenum

alkylidene: polyketones,150 polyesters,151 polycarbonates,152 polycarbosiloxanes,153

and polycarbosilanes.154

A well-defined mechanistic interpretation of these observations is not presently

available, but this phenomenon has been denoted as the "negative neighboring group

effect" (NNGE) and continues to be an area of ongoing mechanistic investigations. The

NNGE provides a deleterious competing equilibrium process to the ADMET cycle which

helps to retard or completely inhibit ADMET polymerization. A strong Lewis basic

functionality, such as an ether-oxygen or carbonyl-oxygen, can potentially undergo

intermolecular or intramolecular complexation with the high oxidation state electrophilic

metal center. If a stable complex is formed as a consequence of coordination, then the

catalyst will be rendered metathesis inactive and the formation of the catalytic

methylidene (Figure 1.30) will be prevented. Intramolecular Lewis base binding effects

of this type have been reported previously for similar ROMP systems,155,156 in addition

to the formation of stable metallacyclobutane intermediates during the metathesis of ester

containing olefins.157 As previously described and demonstrated by Wagener and

Brzezinska, these intramolecular coordinative effects can be overcome by methylene

carbon spacing, which disfavors backbiting coordination into the metal center versus

polymerization.

The development of Grubbs' well-defined ruthenium alkylidene, as described

earlier, has demonstrated significant potential for ADMET polymerization due to its

rugged nature and tolerance to high degrees of functionality. Although the











NAr


ROMX

Complexation/


1N Ar
II
ROW M
R"



A "Stable" Complex


00 0
II =I II
X = O, C, CO, OCO


V


NAr X

RO o XMo
RO


Precursor to Elimination
of Methylidene


Figure 1.30. The negative neighboring group effect (NNGE).


polymerization of 1,9-decadiene with Grubbs' ruthenium catalyst is much slower than

Schrock's catalyst, it has created an opportunity to study the kinetics of ADMET

polymerization and to what degree the negative neighboring group effect alters those

kinetics. Preliminary studies have shown that the polymerization of 1,9-decadiene with
Grubbs' ruthenium alkylidene, RuCl2(=CHPh)(PCy3)2, to display classic second order

kinetics as predicted for step polymerization, such as that observed for

polyesterification. 125 While typical step polymerizations possess relatively high

activation energies, only modest activation energies are required for ADMET, with the

overall process being only slightly exothermic, as originally predicted.158 This kinetic

data has provided more verifying evidence to the already generally accepted mechanism

of acyclic diene metathesis polymerization.

The desire to expand the viability and potential of ADMET polymerization has
been the driving force of recent studies in the area of polychlorosilanes,159,160

polyphosphines, polyboronates,161 and polycarbostannanes.162 In an effort to further






42

explore the negative neighboring group effect and extend the range of functional groups
which may be incorporated into polymers made via ADMET polymerization, a series of

amine and sulfur functionalized dienes and model compounds have been synthesized to

demonstrate their metathetic activity with Schrock's molybdenum neophylidene,
Mo=[CH(C(CH3)2(Ph))][N-2,6-C6H3-i-Pr2][OCCH3(CF3)212, and Grubbs' well-
defined ruthenium catalyst, RuC12(=CHPh)(PCy3)2. The results of this study form the

basis of this dissertation.














CHAPTER 2
EXPERIMENTAL

Instrumentation and Analysis


1H NMR 300 MHz and 13C NMR 75 MHz spectra were recorded on a Gemini-

Series NMR Superconducting Spectrometer system. All NMR spectra were recorded in
chloroform-d(CDC13) or benzene-d6 with 0.03% or 1% v/v tetramethylsilane as an

internal reference. Resonances are reported in 8 units downfield from TMS at 0.00 ppm.

Heteronuclear decoupled quantitative 13C NMR spectra were acquired over 8-12 hours
with a pulse delay of 10-15s. Number average molecular weights (Mn) obtained were

determined by integration of terminal olefins versus internal olefin carbon signals.

Infrared spectra were recorded on neat oils monomerss) or solution casted

(polymers) from an appropriate solvent, between NaCl plates with a Perkin-Elmer model

1600 FTIR spectrometer. UV data were collected on a Cary 5E-UV-Vis-NIR

spectrophotometer. Elemental analyses were performed by Atlantic Microlab, Inc.,

Norcross, Georgia. HRMS data were recorded on a Finnigan MAT 95Q or Finnigan

MAT GCQ Gas Chromatograph/Mass Spectrometer under CI or El conditions.

Differential Scanning Calorimetry (DSC) and Thermogravimetric Analysis (TGA)

data were recorded on a DuPont 2000 Thermal Analysis System interfaced to a HIRES

TGA 2950 Thermogravimetric Analyzer and DSC 2910 Differential Scanning

Calorimeter respectively. DSC samples (5-10 mg) were analyzed with liquid nitrogen as

coolant under a helium flow at a rate of 30 mL/min. All polymer samples were cycled

from -100 to 100 C from scan rates of 10-200C/min. All TGA samples were performed

under air and nitrogen with a flow rate of 40 mL/min and program heating from 25-

6000C at 100C/min.









Gel Permeation Chromatography (GPC) data were recorded using a Waters

Associates Liquid Chromatograph U6K system, 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 THF or CHC13 (1% w/w), passed through a 50 mm filter, and

injected (20mL) successively through 5 x 103 A and 5 x 104 A Phenogel columns

(crosslinked polstyrene gel) at a flow rate of 1 mL/min. Retention times were calibrated

against polystyrene standards (Scientific Polymer Products, Inc.) or polybutadiene

standards (Polysciences, Inc.) The molecular weights of the polystyrene and
polybutadiene standards used were the following: PS(Mp) = 1900, 7700, 12000, 30000,

48900, 59000, 79000, 139400, 650000; PB(Mw) = 439, 982, 2760, 24000, 110000. All

polydispersities (Mw/Mn) of the polymer standards used were less than 1.07.



Materials and Techniques


The metathesis catalysts employed during the course of this work have been
previously developed by Schrock et al. and Grubbs et al. Both the tungsten and

molybdenum versions of Schrock's catalyst, M(CHR')(NAr)(OR)2 where M = W (1)101

or Mo (2)163, Ar = 2,6-(i-Pr)2-C6H3, R' = CMe2Ph, and R = CMe(CF3)2 were
synthesized by published methods. Grubb's ruthenium catalyst, RuCl2(=CHR)(PCy3)2

(3) where Cy = cyclohexyl, was generously provided by Mark Watson, Shane Wolfe, Dr.

John D. 0. Anderson, and Lauri Jenkins via literature procedures 19 and personal

communications with the Grubbs group at the California Institute of Technology,
Berkley, CA. All catalyst systems employed in this study are graphically depicted in

Figure 2.1.












C, o.PCy3
N Ru
II Cl* Ph
F3C OM- PCy3
3 0 IPh (3) Cy = cyclohexyl


F37 CF3 M = W (1)
= Mo (2)

Figure 2.1. Schrock alkylidenes 1 and 2 and Grubbs' ruthenium catalyst 3.


All monomer and catalyst manipulations were conducted in an argon glove box or

under high vacuum conditions maintained by a silicone oil diffusion pump (>10-5

mmHg). All volatile monomers were fractionally distilled or purified via column

chromatography to ensure a purity greater than 99.5%, as determined by GC analysis,

prior to polymerization. Once pure, all volatile monomers were dried under high vacuum
conditions on CaH2 (when chemically appropriate), degassed several times by freeze-

pump-thaw cycles using liquid nitrogen, and transferred to a potassium mirror until no

visible compromise of the mirror surface could be detected. The monomers were finally

transferred, under high vacuum conditions, into a storage flask or reaction flask equipped

with a high vacuum TeflonT stopcock valve. High boiling monomers were treated in a

similar manner, but due to their high boiling nature and difficulty in transferring, these
monomers were distilled directly from CaH2 into their appropriate storage or reaction

flask. All solid model compounds studied were purified by repeated recrystallizations

and subsequent sublimation.

Diallylsulfide (4) and 1,9-decadiene were purchased from Aldrich and purified in

the manner described previously. Diallylamine (16), thiophene, and aniline were

purchased from Aldrich and distilled prior to use.









Polymerizations of liquid ADMET monomers were conducted in a 25-100 mL

round bottom flask equipped with a high vacuum TeflonTM valve and a magnetic stir bar.

All ROMP reactions discussed were conducted in TeflonT capped vials containing a

magnetic stir bar in an argon dry box. All monomers (0.2 -2.0 g) were delivered to the

reaction flasks in the dry box where catalyst (2-20 mg) was introduced. Exact monomer

to catalyst ratios are discussed below. Once the ADMET polymerizations were initiated,

the reaction flasks were removed from the dry box, placed on a high vacuum manifold,

and allowed to stir for several days (2-5 days). Intermittent vacuum was applied until a

viscous material was observed (1-3 hours), at which time the reaction was subjected to

continuous dynamic vacuum. All polymerizations were initially conducted at 250C until

the increased viscosity of the reaction hindered stirring, at which time the temperature

was increased 15-200C/12 hours until no further evolution of ethylene was observed.

Polymerizations initiated with Schrock alkylidenes were quenched by exposure to air,

while Grubbs' catalyst was terminated by addition (~ ImL) of ethyl vinyl ether. Polymers

were isolated in high yield (>90 %), and NMR, GPC, and elemental data were collected

using the crude polymers prior to purification by precipitation from toluene/methanol.

The catalyst residue was removed prior to collecting thermal data by passing a toluene

solution of the polymer mixture through a short column of alumina.

All volatiles were removed from failed metathesis reactions, leaving behind only

the catalyst residue. The residue was then tested with 1,9-decadiene to determine catalyst

activity, which would be confirmed by ethylene evolution upon addition.


Synthesis and Characterization
Sulfur-Containing Monomers and Polymers


Synthesis of bis(3-butenyl)sulfide (5). Monomer 5 was prepared and purified as

described in the literature and had physical constants in agreement with those previously









recorded.192 Monomer 5 had the following spectral characteristics (previously not

provided):
1H NMR (CDC13): 8 2.29 (m, 2H), 2.54 (t, 2H, S-CH2), 5.05 (m, 2H, CH=CH2),

5.79 (m, 1H, CIH=CH2); 13C NMR:8 31.4, 33.9, 115.6, 136.6; IR (NaCi, thin film):

2917, 1640, 1439, 991, 910 cm-1; HRMS: 142.0810, Calcd for C8H15S 142.0994; Yield

67%.

Synthesis of bis(4-pentenyl)sulfide (6). Monomer 6 was prepared and purified as
described in the literature and had physical constants in agreement with those previously

recorded.192 Diene 6 had the following spectral characteristics (previously not

provided):
1H NMR (CDC13): 8 1.66 (m, 2H), 2.15 (m, 2H), 2.49 (t, 2H, S-CH2), 4.99 (m,
2H, CH=CH2), 5.78 (m, 1H, CH=CH2); 13C NMR: 8 28.6, 31.2,32.6, 114.8, 137.6; IR

(NaCl, thin film): 2928, 1640, 1439, 991, 910 cm-1; HRMS: 170.1053, Calcd for
C10H19S 170.1208; Yield 85%.


Synthesis of bis(5-hexenyl)sulfide (7). Monomer 7 was prepared and purified as
described in the literature and had physical constants in agreement with those previously

recorded. 192 Monomer 7 had the following spectral characteristics (previously not

provided):
1H NMR (CDC13): 8 1.4-1.7 (m, 4H), 2.05 (m, 2H), 2.50 (t, 2H, S-CH2), 4.97 (m,

2H, CH=CH2), 5.79 (m, 1H, CH=CH2); 13C NMR: 8 27.9, 28.9, 31.8, 33.2, 114.4,

138.3; IR (NaCI, thin film): 2927, 1640, 1439, 991, 910 cm-1; HRMS: (M+H)
199.1525, Calcd for C12H23S 199.1436; Yield 76%.

Bulk cyclization of diallylsulfide (4). Monomer 4 (Aldrich, 97%) was fractionally
distilled prior to use, then vacuum transferred from CaH2 (stirred for 3 d) to an

appropriate storage flask equipped with a TeflonTm vacuum stopcock. This reaction was

conducted as previously described using catalyst 2. Molar ratio of 4:2 @ 2000:1; 250C,









20 h. Upon addition of catalyst 2, a temporary evolution of ethylene was observed. 1H

NMR revealed a small internal olefin spike at 5.43 ppm (7 mol%), indicating cyclization

had occurred to produce 2,5-dihydrothiophene (8). A second addition of catalyst 2

(1500:1) resulted in an increase in the cyclic content (10.5 mol%). Another reaction was

conducted in a similar manner with a molar ratio of 4:2 @ 250:1; 250C, 3 h. This

resulted again in a rapid evolution of ethylene producing a cyclic to monomer content of

4:1. In each case, all volatiles were removed from the reaction flask and the remaining

catalyst residue was tested with 1,9-decadiene for catalyst activity. No activity was

observed.

Solution cyclization of diallylsulfide (4) to 2.5-dihydrothiophene (8). Monomer
26 was prepared as above. 2 mL of a 0.36 M solution of monomer 4 in C6D6 was

prepared in a dry box and added to a TeflonTm capped vial with catalyst 2 at 190:1 (4:2)

molar ratio. The golden yellow reaction was allowed to stir at 250C for 3 h. IH, 13C

NMR and CG/MS confirmed the assignment of 8 in >99% yield.164


Synthesis of poly(thio-3-hexene-1,.6-diyl) (9). Monomer 5 was synthesized as
previously described. 5 was vacuum transferred from CaH2 (stirred for 3-5 days) to an

appropriate storage flask equipped with a TeflonTm vacuum stopcock. This reaction was

conducted in a similar manner to that of 4. Molar ratio of 5:2, @ 2000:1; 25C, 45 min;

45 650C, 5 d. Upon addition of 2, immediate evolution of ethylene was observed which

was accompained by increased viscosity within 40 min. Polymerization was terminated

by exposure to air. Polymer 2 had the following characteristics:

1H NMR (CDC13): 8 2.29, 2.55 (S-CH2), 5.50 (CH=CH); 13C NMR: 8 27.5 (cis

allyl, 16%) 31.8, 32.6 (trans allyl, 84%), 128.7 (cis C=C, 16%), 129.6 (trans C=C, 84%);

IR (NaCl, thin film): 2916, 1441, 971 cm-1; Elemental: Calcd: C, 63.36; H, 8.82; S,

27.82; Found: C, 63.13; H, 8.82; S, 27.45; Yield: 89%; GPC (PB): 3100 g/mol









(Mw/Mn= 1.47); Mn(13C NMR): 3100; TGA(100C decomp N2): 2950C; DSC

(100C/min): Tm 460C.

Synthesis of poly(thio-4-octene-1.8-diyl) (10). Monomer 6 was synthesized as
previously described. 6 was vacuum transferred from CaH2 (stirred for 3-5 days) to an

appropriate storage flask equipped with a TeflonTM vacuum stopcock. This reaction was
conducted in a similar manner to that of 5. Molar ratio of 6:2 @ 1000:1; 250C, 72 h; 45

-700C, 4 d. Upon addition of 2, immediate evolution of ethylene was observed which

was accompained by increased viscosity within 30 min. Polymerization was terminated
by exposure to air. Polymer 10 had the following characteristics:
1H NMR (CDCl3): 8 1.63, 2.13, 2.49 (S-CH2), 5.41 (CH--CH); 13C NMR: 8 26.3

(cis allyl, 19%), 29.2, 31.3 (trans allyl, 81%), 31.5, 129.4 (cis C=C, 20%), 129.8 (trans

C=C, 80%); IR (NaCl, thin film): 2923, 1436, 967 cm-1; Elemental: Calcd: C, 67.61; H,

9.86; S, 22.53; Found: C, 67.54; H, 9.86; S, 22.57; Yield: 93%; GPC (PB): 16330 g/mol
(Mw/Mn= 1.62); Mn(13C NMR): 17400; TGA(100C decomp N2): 3700C; DSC
(100C/min): Tm VC.

Synthesis of poly(thio-5-decene-1,10-diyl) (11). Monomer 7 was synthesized as
previously described. 7 was vacuum transferred from CaH2 (stirred for 3-5 days) to an

appropriate storage flask equipped with a TeflonT vacuum stopcock. This reaction was

conducted in a similar manner to that of 5. Molar ratio of 7:2 @ 2000:1; 25C, 12 h; 45

-700C, 4 d. Upon addition of 2, immediate evolution of ethylene was observed which
was accompained by increased viscosity within 15 min. Polymerization was terminated
by exposure to air. Polymer 11 had the following characteristics:
1H NMR (CDC13): 8 1.48, 1.97, 2.47 (S-CH2), 5.38 (CH=CH); 13C NMR: 8

26.7 (cis allyl, 19%), 28.7, 29.1, 31.9 (trans allyl, 81%), 32.0, 129.6 (cis C=C, 19%),
130.1 (trans C=C, 81%); IR (NaCl, thin film): 2925, 1437, 967 cm-1; Elemental: Calcd:
C, 70.59; H, 10.59; S, 18.82; Found: C, 70.51; H, 10.60; S, 18.77; Yield: 94%; GPC









(PB): 9410 g/mol (Mw/Mn= 1.86); Mn(13C NMR): 10200; TGA(100C decomp N2):

3550C; DSC (100C/min): Tm 230C.


Synthesis of poly [(thio-5-decene- 1.10-diyl)-co-(1-octenylene)] (12). Copolymer

12 was prepared by addition of catalyst 2 to equimolar amounts of 1,9-decadiene and 7.

Molar ratio of 7:1,9-decadiene:2 @ 1000:1000:1; 25C, 72 h, then increased 15C/day to

700C. Upon addition of 2, immediate evolution of ethylene was observed which was

accompained by increased viscosity within 15 min. Once ethylene evolution had ceased,

the reaction was terminated by exposure to air. Copolymer 12 had the following spectral

properties:
1H NMR (CDC13): 8 1.27, 1.52, 1.98, 2.48 (S-CH2), 5.35 (CH=CH); 13C NMR:

8 26.7, 26.8, 27.2, 28.7, 28.8, 28.89, 29.0, 29.1, 29.3, 29.5, 29.6, 29.7, 32.0, 32.1, 32.5,

124.38, 129.2, 129.6, 129.8, 130.2, 130.3, 130.8; IR (NaC1, thin film): 2925, 1438, 966

cm-1; Elemental: Calcd: C, 76.19; H, 11.11; S, 12.70; Found: C, 76.23; H, 11.14; S,
12.61; Yield 90%; GPC (PB): 35000 g/mol (Mw/Mn= 1.31); TGA(100C decomp N2):

3810C; DSC (10C/min): Tm 130C.


Attempted metathesis of monomers 4-7 with ruthenium catalyst (3). All

metathesis reactions were performed under bulk and solution conditions as reported

above. Monomer to catalyst concentrations ranged from 400:1 1000:1. In each case,

ethylene evolution was never observed and no signs of productive metathesis were

detected via 1H NMR or GC/MS. Although no detection of homometathesis was

observed, color changes to the reaction solution did occur. Initial stages of reaction went

from purple to red-orange (1 min) to yellow (10 min) finally to orange (>1 h). The

catalyst residue from each reaction was charged with a minimal amount of 1,9-decadiene

to test catalyst activity. No catalyst activity was observed.








Bulk cyclization of diallyldisulfide (13). Monomer 13 (Aldrich, 80%) was
fractionally distilled prior to use, then vacuum transferred from CaH2 (stirred for 3 d) to

an appropriate storage flask equipped with a TeflonTm vacuum stopcock. This reaction

was conducted in a similar manner to that of 4 using catalyst 2 and 3. Molar ratio of 11:2

and 13:3 @ 150:1-500:1; 250C, 24 h. Upon addition of catalyst 2, a temporary evolution

of ethylene was observed. No evolution of ethylene was observed upon addition of

catalyst 3. 1H NMR did not reveal the presence of an internal olefin in either case. All

volatiles were removed from the reaction flask and the remaining catalyst residue was

tested with 1,9-decadiene for catalyst activity. No activity was observed.

Solution cyclization of diallyldisulfide (13). This reaction was performed in an

analogous manner to that of 4 using catalyst 2 and 3. 2 mL of a 0.36 M solution of
monomer 13 in CDC13 was prepared in a dry box and added to a TeflonT capped vial

with catalyst 2 at 200:1 (11:2., 11:) molar ratio. No ethylene was observed, but

cyclization results were obtained that were comparable to the bulk experiment.

Synthesis of bis(5-hexenyl)disulfide (15). Monomer 15 was synthesized from the
oxidation of 5-hexenylthiol (14). 5-hexenylthiol (14) was synthesized from NaSH*xH20

(Aldrich, 6.55 g, 0.059 moles) and 5-hexenylbromide (Aldrich, 9.30 g, 0.057 moles).
The alkenylbromide was added via syringe(1 h addition) to a 250 mL three-neck flask
with 50 mL of a 1:1 THFIH20 mixture held @ 100C. The reaction was permitted to stir

for 24 h at 25C. 9.12 g of NaOH (Fisher, pellets) was then added slowly as to keep the
temperature of the reaction below 150C. 7.10 g of a H202 (30 wt%) solution was then

added via syringe at a rate which the temperature did not exceed 300C. The organic layer
was separated, extracted with diethyl ether (3 x 40 mL), dried over MgSO4, reduced via

rotoevaporation, and distilled by short path distillation (980C @ 1 mm Hg) to give 5.57 g
(85 % yield) of clear liquid 15. 15 had the following spectral properties:









IH NMR (CDC13): 8 1.44 (m, 2H), 1.65 (m, 2H), 2.03 (q, 2H, allyl), 2.62 (t, 2H,
S-S-CH2), 4.94 (m, 2H, CH=CH2), 5.74 (m, 1H, CHj=CH2); 13C NMR: 8 27.54, 28.68,

33.37, 38.94, 114.77, 138.31; HRMS: 230.1166, Calcd for C8H15S2 230.1163.

Attempted metathesis of bis(5-hexenyl)disulfide (15).. Monomer 15 was
synthesized as previously described. 15 was vacuum transferred from CaH2 (stirred for

3-5 days) to an appropriate storage flask equipped with a TeflonT vacuum stopcock.

This reaction was conducted in a similar manner to that of 7. Molar ratio of 15:2, @
500:1; 250C, 3 d. Slight bubbling was observed upon addition of catalyst 2. 1H or 13C

NMR did not reveal linear internal olefin or cyclic formation, but dimer formation was
confirmed by GC/HRMS : HRMS (M+l): 433.1960 Calcd for C22H41S2 433.2091.


Synthesis of 2-propenylthiophene (16). The synthesis of 16 was performed using
standard Schlenk line techniques with argon as the inert gas. To a 250 mL, 3-neck round

bottom flask, freshly grounded magnesium turnings (1.63 g, 0.068 moles) were added

with a catalytic amount of iodine, 150 mL of dry THF (distilled from NaK amalgam) was

cannula transferred, and 2-bromothiophene (Aldrich, 10.90 g, 0.0669 moles) was

dropwise added via addition funnel over a 2 h period. The reaction was allowed to stir
for 3 h at 250C. 65 mg (0.2 mol%) of [Ni(dppp)C12] (Aldrich) was then added to the

stirring solution followed by the addition of 1-bromo-2-propene (Aldrich, 8.15 g, 0.0673

moles) over a two hour period. The reaction was heated at reflux for 12 h and quenched
with dilute IM HC1. The dark brown THF layer was separated and extracted with
NaHCO3 (3 x 25 mL) and dried over MgSO4. The ether layer was removed by

rotoevaporation and yielded a red oil. Additional freshly grounded magenesium was
introduced in an effort to quench residual alkylbromides. 4.6 g (56 %) of clear liquid 16
was obtained via static high vacuum transfer.
1H NMR (CDC13): 8 1.82 (dd, 3H, =CH-CH3, trans), 1.96 (dd, 3H, =CH-CH3,
cis), 5.66-6.10 (m, 1H, Ar-CH-CH3), 6.80-7.05 (m, 3H, ArH); 13C NMR: 8 18.34,









123.03, 124.01, 124.35, 124.82, 124.99, 125.80, 126.73, 126.97, 127.22; HRMS:
124.0347, Calcd for C7H8S 124.0339. Elemental: Calcd: C, 67.74; H, 6.45; S, 25.81.

Found: C, 66.65; H, 6.25; S, 25.50.


Attempted metathesis of 2-propenylthiophene (16). Model compound 16 was
synthesized as previously described. 16 was vacuum transferred from CaH2 (stirred for

3-5 days) to an appropriate storage flask equipped with a TeflonTm vacuum stopcock.

Molar ratio of 16:2, @ 500:1; 250C, 24 h. No ethylene evolution was observed and 1H

NMR did not reveal internal olefin formation. This negative finding was supported by

GC/MS. All volatiles were removed in vacuo and the catalyst residue was tested with

1,9-decadiene. No metathesis condensation was observed.


Attempted polymerization of 1.9-decadiene in the presence of thiophene. 10

equivalents of freshly distilled thiophene was added to 1 molar equivalent of catalyst 2.

The solution was allowed to stir 5 min prior to the addition of 500 equivalents of 1,9-

decadiene. Ethylene evolution was not observed upon addition of 1,9-decadiene nor was

internal olefin formation detected via 1H NMR.


Amine-Containing Model Compounds. Monomers, and Polymers

Attempted bulk metathesis of diallylamine (17) with catalyst 1 and 2. Monomer

17 was purchased from Aldrich Chemical Co. as 99% pure and was purified via fractional

distillation. In each case, a ratio of 500:1 (diallylamine:catalyst) was employed to study

the metathesis condensation activity of monomer 17. Reactions were performed in a 50

mL round bottom flask equipped with a TeflonTM vacuum stopcock and stir bar. Catalyst

I and 2 (separate reactions) were added to the monomer in an argon dry box, the reaction

flask was sealed, removed from the dry box, and placed on a high vacuum line as

described previously. Once placed on the high vacuum line, the reaction flask was









degassed and allowed to stir for 24 hours at room temperature with intermittet vacuum

applied. The reaction mixture was then heated to 450C and allowed to stir for 12 hours.

In both cases, only pure starting material 17 was detected via 1H NMR with no visible

evolution of ethylene observed. Upon removal of all volatiles, the catalyst residue was

charged with a minimal amount of 1,9-decadiene to test catalyst activity. No catalyst

activity was observed.

NMR solution reaction of diallylamine (17) with molybdenum catalyst 2. An
approximate 5:1 ratio of 17:2 was dissolved in 1 mL of d6-benzene and added to an NMR

tube equipped with a TeflonTm vacuum stopcock. The dark red solution was monitored

via 1H NMR for 12 hours.

Attempted copolymerization of 4 and 1.9-decadiene with catalyst 2. The reaction

of a 500:500:1 ratio of 17:1,9-decadiene:2 was performed in a similar manner described

above for the bulk metathesis of diallylamine (17). Only starting materials were detected

via 1H NMR, with no visible detection of ethylene observed during the reaction.


Synthesis of 4-dodecylaminobenzaldehyde (18). Compound 18 was prepared in a

similar procedure published by Bader by the addition of n-dodecylamine (Aldrich, 20.52

g, 0.1108 moles) to 4-fluorobenzaldehyde (Fluka, 21.6 mL, 0.2014 moles) and 27.85 g of
K2CO3 in 200 mL of DMF.165 The solution was heated at 950C for 30 hours then an

additional 17 hours at 1500C. The reaction was allowed to slowly cool to room

temperature and was quenched in 1000 mL of ice water, producing a yellow solid that

was filtered and collected. Th e filtered material was extracted with diethyl ether, dried
with MgSO4, evaporated via rotoevaporation, and recrystallized in heptane to give gold

crystals in a 42% yield (mp 75.0-75.5C). Compound 18 had the following spectral

properties:

1H NMR (CDC13): 8 0.85 (t, 3H), 1.25 (s, 18H), 1.61 (q, 2H), 3.18 (t, 2H, N-

CH2), 4.30 (s, 1H, RN-H), 6.58 (d, 2H, ArH), 7.68 (d, 2H, ArH), 9.72 (s, 1H, aldehyde);









13C NMR: 8 14.10, 22.25, 27.73, 29.24, 29.85, 31.00, 32.26, 61.87, 115.88, 116.03,

129.97, 159.22, Elemental: Calcd: C, 78.82; H, 10.82; N, 4.84. Found: C, 78.38; H,

10.71; N, 4.78.

Synthesis of 4-(prop-l-ene)-N-dodecylaniline (19). Compound 19 was

synthesized by a Wittig reaction with 18. (Ethyl)triphenylphosphonium bromide

(Aldrich, 1.73 g, 0.00465 moles) was added to an argon purged 250 mL 3-neck flask in

25 mL of dry THF. The mixture was cooled to -780C then 10 mL of 1.0 M n-butylithium

was added, at which point the reaction mixture was allowed to slowly warm to room

temperature. The solution stirred for 20 min at room temperature and was cooled in a salt

bath to -150C. A 0.1 M THF solution of 18 was dropwise added via an addition funnel

over a period of 65 min then refluxed for 16.5 h. The reaction was quenched with
aqueous NaHCO3, extracted with diethyl ether, and filtered through a silica gel column.

The ether layer was dried over MgSO4, filtered, evaporated, and the crude product was

recrystallized from ethanol/water (9:1). 19 was further purified by sublimation at 60C
@ 10-2 mmHg to give a white crystalline solid (mp 51.0-51.50C) in a 48% pure isolated

yield. The following spectral properties were observed:
1H NMR (CDC13): 8 0.85 (t, 3H), 1.25 (s, 18H), 1.60 (q, 2H), 1.82 (trans)(d, 3H),

1.90 (cis)(d, 3H), 3.10 (t, 2H, N-CH2), 3.60 (s, br, 1H, N-H), (6.00 (m, 1H, CH=CH-

CH3), 6.25 (trans) 6.35 (cis) (d, lH, CH=CH-CH3), 6.52 (d, 2H, ArH), 7.15 (d, 2H,

ArH); 13C NMR: 8 14.00, 14.80, 18.20, 22.80, 27.00, 29.50, 32.00, 44.10, 112.5, 113.0,

121.0, 123.5, 127.0, 127.5, 130.0, 131.0, 147.0, 148.0; Elemental: Calcd: C, 83.73; H,

11.62; N, 4.65. Found: C, 83.49; H, 11.64; N, 4.59.

Synthesis of 4-(prop-l-ene)-N. N-dimethylaniline (20). Model compound 20 was

prepared in a similar manner to that of 19. Dimethylaminobenzaldehyde (Aldrich) was

purchased and used without further purification. The reaction was quenched as reported

above, then evaporated to a concentrated mixture which was extracted with pentane (3 x








30 mL). This procedure isolated a white crystalline material which was recrystallized in
ethanol/water (9:1) in an 80% yield (mp 50.5-51.5C). 20 was further purified by
sublimation at 600C @ 10-2 mmHg. Spectral characterization is as follows:
1H NMR (CDC13): 8 1.89 (d, 3H), 3.08 (s, 6H, N-CH3), 6.05 (m, 1H, CH=CH-
CH3), 6.28 (trans) 6.35 (cis) (d, 1H, CH=CH-CH3), 6.68 (d, 2H, ArH), 7.22 (d, 2H,

ArH); 13C NMR: 5 18.43, 40.65, 112.71, 121.41, 126.68, 126.89, 130.78, 149.63;

Elemental: Calcd: C, 81.99; H, 9.32; N, 8.69. Found: C, 82.01; H, 9.37; N, 8.73.

Synthesis of 4. 4'-bis(dodecylamino)stilbene (21). Compound 21 was produced
from the neat (no solvent) metathesis reaction of 19 with catalyst 1. Molar ratio of 21:1

@ 500:1; 600C, 75 min. After 15 min of heating, solid 19 began to melt with

concomitant evolution of gas (2-butene) upon liquification. After 75 min at 600C, the

reaction mixture turned to a brown solid. The product mixture was quenched by
exposure to air, dissolved in chloroform, filtered, evaporated, and recrystallized from
ethanol to give 21 as a tan powder in 60% yield (>99:<1 E:Z, mp 235-2360C). 21 had the

following spectral properties:
1H NMR (CDC13): 8 0.88 (t, 6H), 1.29 (s, 36H), 1.62 (q, 4H), 3.12 (t, 4H, N-
CH2), 3.55 (s, br, 2H, N-H), 6.58 (d, 4H, ArH), 6.82 (s, 2H, CH=CH trans) 7.32 (d, 4H,

ArH). Elemental: Calcd: C, 82.76; H, 9.81; N, 7.43. Found: C, 82.66; H, 9.65; N, 7.69.

Synthesis of 4. 4'-bis(dimethylamino)stilbene (22). Compound 22 was
synthesized in a similar manner as 21. Upon liquification, gas evolution occurred which
continued for 30 min, at which time the reaction mixture turned solid. The reaction was
purified by subliming the starting material from the mixture. The product was isolated in
a 62% yield (>99:<1 E:Z, mp 240-240.5C). 22 had the following spectral properties:
IH NMR (CDC13): 8 3.01 (s, br, 12H, N-CH3), 6.77 (d, 4H, ArH), 6.91 (s, 2H,
CH=CH), 7.43 (d, 4H, ArH); 13C NMR: 8 40.45, 112.80, 249.79, 126.88; Elemental:

Calcd: C, 81.20; H, 8.27; N, 10.53. Found: C, 80.95; H, 8.34; N, 10.41.








Synthesis of N. N-diallylaniline (23). Monomer 23 was prepared by the
simultaneous dropwise addition of allyl bromide (Aldrich, 16.13 g, 0.1333 moles) and 10
M NaOH (70 mL) to stirring aniline (Fisher, 6.20 g, 0.0667 moles) over a period of 3 h.
The stirring solution was refluxed for 40 h then allowed to cool to room temperature.
The reaction mixture was extracted (4 x 35 mL) with diethyl ether, dried over MgSO4,

and concentrated by rotoryevaporation. The light yellow liquid was then fractionally
distilled (122-124C @16 mm Hg), yielding 9.48 g (83 % yield) of clear liquid 23.
Monomer 23 had the following spectral properties:
HRMS: 173.1219, Calcd for Cl2H15N 173.1204; Elemental: Calcd: C, 83.24; H,

8.67; N, 8.09. Found: C, 83.05; H, 8.71; N, 8.04. Further details and complete
characterization of monomer 10 is given elsewhere. 166

Synthesis of N. N-dibutenylaniline (24). Monomer 24 was prepared by a similar
procedure to that used for 23 using 4-bromobutene (Aldrich, 5.20g, 0.0382 moles). After
repeated fractional vacuum distillations (116-1170C @ 1 mm Hg), 5.20 g (74 % yield) of
clear liquid 24 was recovered. Monomer 24 had the following spectral properties:
1H NMR (CDC13): 8 2.27 (m, 4H), 3.28 (t, 4H, N-CH2), 5.02 (m, 4H, CH=CH2),
5.76 (m, 2H, CH=CH2), 6.50-7.30 (ArH, 5H); 13C NMR: 8 31.80, 50.12, 112.05, 115.96,
116.20, 129.10, 135.34, 147.42; HRMS: 201.1539, Calcd for Cl4H19N 201.1517;

Elemental: Calcd: C, 83.58; H, 9.45; N, 6.97. Found: C, 83.56; H, 9.49; N, 7.05.

Synthesis of N. N-dipentenylaniline (25). Monomer 25 was prepared by a similar
procedure to that used for 23 using 5-bromopentene (Aldrich, 6.15g, 0.0411 moles).
After multiple fractional vacuum distillations (135 C, 1 mm Hg), 7.60 g (62% yield) of
clear liquid 25 was recovered. Monomer 25 had the following spectral properties:
1H NMR (CDC13): 8 1.63 (m, 4H), 2.03 (m, 4H), 3.21 (t, 4H, N-CH2), 5.22 (m,
4H, CH=CH2), 5.75 (m, 2H, CH=CH2), 6.50-7.30 (ArH, 5H); 13C NMR: 8 26.11, 31.05,

50.21, 111.71, 114.76, 115.28, 129.00, 137.86, 147.77; HRMS: 229.1827, Calcd for









C16H23N 229.1832; Elemental: Calcd: C, 83.84; H, 10.04; N, 6.12. Found: C, 83.79; H,
10.08; N, 6.13.


Synthesis of 1-phenyl-3-pyrroline (26). Monomer 26 was synthesized as
previously described. 23 was vacuum transferred from CaH2 to a potassium mirror and

allowed to stir for 12 h. 23 was then vacuum transferred to an appropriate storage flask

equipped with a TeflonTm vacuum stopcock. This reaction was conducted in a similar

manner to that of 16. Molar ratio of 23:2 @ 500:1; 250C, 24 h. Upon addition of 2,

immediate evolution of ethylene was observed and continued for 1 min until the reaction

mixture turned solid. The reaction was allowed to stir at 450C for 24h on the high

vacuum line, after which time the reaction was quenched by exposure to air. The light

brown product was recrystallized in methanol to give a white crystalline solid in an 85%

yield. 26 had the following spectral properties:

1H NMR (CDC13): 8 4.12 (s, 4H, N-CH2), 5.97 (s, 2H, CH=CH), 6.62-7.28 (Ar-

H, 5H); 13C NMR: 854.38, 111.12, 115.57, 126.37, 129.29; HRMS (EI): 145.0999,

Calcd for C10HI iN 145.0892.


Synthesis of poly (N-phenylamino-3-hexene-1. 6-diyl) (27). Monomer 24 was
synthesized as previously described. 24 was vacuum transferred from CaH2 to a

potassium mirror and allowed to stir for 12 h. 24 was then vacuum transferred to an

appropriate storage flask equipped with a TeflonT vacuum stopcock. This reaction was

initially conducted in a similar manner to that of 26. Molar ratio of 24:2 @ 1000:1; 250C,

12 h, then increased 150C/day to 700C. Upon addition of 2, immediate evolution of

ethylene was observed resulting in an increase in viscosity within 1 h. When increasing

viscosity began to hinder stirring, the temperature was gradually elevated to facilitate

stirring. This procedure was repeated until the evolution of ethylene was no longer

observed. After the temperature exceeded 600C, a trace amount of liquid volatiles were








collected and upon analysis were shown to be the cyclic monomer 31. Polymerization

was terminated by exposure to air. Polymer 27 had the following spectral properties:
1H NMR (CDC13): 8 2.26, 3.25 (N-CH2), 5.47 (CH=CH), 6.62 (ArH), 7.19
(ArH); 13C NMR: 8 25.43 (cis allyl, 20%), 30.52 (trans allyl, 80%), 50.75, 111.72,

115.52, 128.22 (cis C=C), 129.13, 129.21 (trans C=C), 147.48; UV: lmax(CHCl3) = 261

nm, e = 16,900 (@10-4 M); Elemental: Calcd: C, 83.17; H, 8.73; N, 8.06. Found: C,
83.18; H, 8.68; N, 8.06; Yield: 92%; GPC (PS): 9200 g/mol (Mw/Mn= 1.7); TGA(10C
decomp N2): 4000C; DSC (100C/min): Tg -30C.

Synthesis of poly (N-phenylamino-4-octene-1,. 8-diyl) (28). Monomer 25. was
synthesized as previously described. 25 was vacuum transferred from CaH2 to a

potassium mirror and allowed to stir for 12 h. 25 was then vacuum transferred to an
appropriate storage flask equipped with a TeflonT vacuum stopcock. This reaction was

conducted in a similar manner to that of 27. Molar ratio of 25:2 @ 1000:1; 250C, 6 h,
then increased 15C/day to 700C. Upon addition of 2, immediate evolution of ethylene
was observed which was accompained by an increase in viscosity within 15 min. After

the temperature exceeded 600C, a trace amount of a solid white film was collected and
upon analysis were shown to be the cyclic monomer 32 and cyclic dimers 33a-c.

Polymerization was terminated by exposure to air. Polymer 28 had the following spectral
properties:
1H NMR (CDC13): 6 1.61, 2.01, 3.22 (N-CH2), 5.43 (CH=CH), 6.61 (ArH), 7.16
(ArH); 13C NMR: 8 24.64 (cis allyl, 26%), 26.87, 29.94 (trans allyl, 74%), 50.36,

111.74, 115.22, 129.04 (cis C=C), 129.53, 129.97 (trans C=C), 147.83; UV:

lmax(CHCl3) = 261 nm, e = 15,350 (@10-4 M); Elemental: Calcd: C, 83.58; H, 9.49; N,
6.91; Found: C, 83.56; H, 9.46; N, 6.95; Yield: 94%; GPC (PS): 14100 g/mol
(Mw/Mn= 2.1); TGA(100C decomp N2): 4250C; DSC (100C/min): Tg -25 C.








Synthesis of poly[(N-phenylamino-4-octene-1.1 0-diyl)-co-(1-octenylene)1 (29).

Copolymer 29 was prepared by addition of catalyst 2 to equimolar amounts of 1,9-

decadiene and 25. Molar ratio of 25:1,9-decadiene:2 @ 1000:1000:1; 250C, 3 h, then

increased 150C/day to 700C. Upon addition of 2, immediate evolution of ethylene was

observed which was accompained by a qualitative increase in viscosity within 10 min.

No cyclic byproducts were observed. Once ethylene evolution had ceased, the reaction

was terminated by exposure to air. Copolymer 29 had the following spectral properties:

1H NMR (CDC13): 8 1.28, 1.62, 1.97, 3.22 (N-CH2), 5.33-5.42 (CH=CH), 6.53-
7.18 (Ar-H).; 13C NMR: 8 24.69, 26.86, 26.90, 27.12, 27.21, 28.95, 29.07, 29.48, 29.53,

29.63, 29.95, 30.00, 32.50, 50.34, 111.75, 115.20, 128.70, 129.05, 129.20, 129.55,

130.01, 130.17, 130.58, 131.04, 147.93; Elemental: Calcd: C, 84.82; H, 10.68; N, 4.50.
Found: C, 84.56; H, 10.72; N, 4.60; Yield: 89%; GPC (PS): 23400 g/mol (Mw/Mn=

1.5); TGA (100C decomp N2): 4400C; DSC (100C/min): Tg -90C.


Synthesis of N-methyl-2-propenylpyrrole (30). Compound 3Q was synthesized by
a Wittig reaction with N-methyl-2-pyrrolecarboxaldehyde (Aldrich, 10.00 g, 0.0916

moles), which was used as received. (Ethyl)triphenylphosphonium bromide (Aldrich,

35.08 g, 0.0945 moles) was added to an argon purged 250 mL 3-neck flask in 150 mL of

dry-distilled THF. The mixture was cooled to -780C then 35.0 mL of 2.5 M n-

butylithium (in hexanes) was added, at which point the reaction mixture was allowed to

slowly warm to -300C. N-Methyl-2-pyrrolecarboxaldehyde was dropwise added via an

addition funnel over a period of 30 min then refluxed for 14 h. The reaction was
quenched with aqueous NaHCO3, extracted with pentane (5 x 50 mL), and filtered

through a silica gel column. The pentane was dried over MgSO4, filtered, and

evaporated leaving a golden yellow crude liquid 30 in 93% yield. 30 was further purified

by short path distillation (1360C) to give a clear liquid (66:33 trans:cis by 1H NMR) in a
88% isolated yield. The following spectral properties were observed:









1H NMR (CDC13): 8 1.82, 1.90 (dd, 3H, cis and trans CH3), 3.50 (s, 3H, N-

CH3), 5.65, 5.95 (m, 1H, =CH-CH3), 6.05, 6.25 (dd, 1H, Ar-CHI=CH); 13C NMR: 8

14.98, 18.65, 33.92 (N-CH3), 105.13, 107.39, 107.53, 109.42, 118.33, 119.92, 122.08,

123.92, 124.08, 129.94, 132.22; HRMS (EI): 121.0899, Calcd for C10H IN 121.0891.

Elemental: Calcd: C, 79.29; H, 9.15; N, 11.56. Found: C, 78.48; H, 9.18; N, 11.34.


Attempted metathesis of N-methyl-2-propenylpyrrole (30). Model compound 3Q
was synthesized as previously described. 30 was vacuum transferred from CaH2 (stirred

for 3-5 days) to an appropriate storage flask equipped with a TeflonTm vacuum stopcock.

Molar ratio of 30:2, @ 250:1; 250C, 24 h. No ethylene evolution was observed and 1H

NMR did not reveal internal olefin formation. This negative finding was supported by

GC/MS. All volatiles were removed in vacuo and the catalyst residue was tested with

1,9-decadiene. No metathesis condensation was observed.

Attempted polymerization of 1.9-decadiene in the presence of aniline. 10

equivalents of freshly distilled aniline was added to 1 molar equivalent of catalyst 2 in a

procedure similar to that reported for 17. The solution was allowed to stir 5 min prior to

the addition of 500 equivalents of 1,9-decadiene. Ethylene evolution was not observed

upon addition of 1,9-decadiene nor was internal olefin formation detected via 1H NMR.

Formation of cyclic monomer (31). The pure cis cyclic byproduct 31 was isolated

as a clear liquid by distillation directly from the active reaction mixture of polymer 27 in

< 1% yield. The identity of the product was determined by GC/HRMS. HRMS (EI):
173.2506, Calcd for C12H15N, 173.2515


Formation of cyclic monomer (32) and cyclic dimer (33a-c). The pure cis cyclic

byproduct 32 and the isomeric mixture of cis-cis, cis-trans, and trans-trans cyclic

byproducts 33a-c were isolated as a white solid by sublimation directly from the active

reaction mixture of polymer 28 in < 1% yield. Cyclic 32: HRMS (EI): 201.1553, Calcd








for C14H19N, 201.1517; Cyclics 33a-c: HRMS (El): 402.3029, Calcd for C28H36N,

402.3035.

Attempted polymerizations of monomers 23-25 with ruthenium catalyst (3). All
polymerization reactions were performed in a similar manner as reported above.
Monomer to catalyst concentrations ranged from 400:1 1000:1. In each case, ethylene

evolution was never observed and no signs of productive metathesis were detected via 1H

NMR. The catalyst residue was charged with a minimal amount of 1,9-decadiene to test

catalyst activity. No catalyst activity was observed.

Synthesis of N. N-diallylmethylamine (34). Monomer 34 was synthesized in the
manner as reported by Beckwith (see experimental procedure for 23), conforming to all

published spectral observations.166

Synthesis of N. N-dibutenylmethylamine (35). Monomer 35 was prepared by a
similar procedure to that used for 34 using 4-bromobutene (Aldrich, 10.14 g, 0.0761

moles). After repeated fractional vacuum distillations (158-161 C), 2.30 g (49 % yield)
of clear liquid 35 was recovered. Monomer 35 had the following spectral properties:
1H NMR (CDC13): 8 2.20-2.30 (m, 4H; s, 3H, N-CH3), 2.42 (t, 4H), 5.02 (m, 4H,
CH=CI.2), 5.82 (m, 2H, CH=CH2); 13C NMR: 8 29.80, 31.90, 41.85, 115.60, 135.90;

HRMS: 139.1360, Calcd for C9H17N 139.1365; Elemental: Calcd: C, 77.70; H, 12.23;

N, 10.07. Found: C, 76.91; H, 12.19; N, 9.88.

Attempted bulk cyclization of N. N-diallylmethylamine (34). Monomer 34 was
vacuum transferred from CaH2 (stirred for 24 h) to a potassium mirror and allowed to stir

for 12 h. 34 was then vacuum transferred (static transfer) to an appropriate storage flask
equipped with a TeflonT vacuum stopcock. This reaction was conducted in a similar

manner to that of 26 using catalyst 2 with monomer to catalyst ratios ranging from 150:1-
450:1. Only pure starting material 34 was detected via 1H NMR with no visible









evolution of ethylene observed. Upon removal of all volatiles, the catalyst residue was

charged with a minimal amount of 1,9-decadiene to test catalyst activity. No catalyst

activity was observed.

Attempted solution cyclization of N. N-diallylmethylamine (34). Monomer 34
was prepared as above. 2 mL of a 0.30 M solution of monomer 34 in CDC13 was

prepared in a dry box and added to a TeflonTm capped vial with catalyst 2 at 150:1-450:1

(34:2) molar ratio. The golden yellow reaction was allowed to stir for 24 h. 1H NMR

analysis did not reveal the formation of an internal olefin which was confirmed by

GC/MS.

Attempted polymerization of N. N-dibutenylmethylamine (35). Monomer 35 was
synthesized as previously described. 35 was vacuum transferred from CaH2 (potassium

mirror isomerized diene) to an appropriate storage flask equipped with a TeflonTm

vacuum stopcock. This reaction was conducted in a similar manner to that of 27. Molar

ratio of 35:2 @ 400:1; 250C, 24 h, then 500C for 24 h. No visible evolution of ethylene

was observed upon addition of 2, and 1H NMR only revealed original starting material.

Upon removal of all volatiles, the catalyst residue was charged with a minimal amount of

1,9-decadiene to test catalyst activity. No catalyst activity was observed.

Synthesis of N. N-diallyl-tert-butylamine (36). Monomer 36 was prepared and

purified as described in the literature and had physical constants in agreement with those

previously recorded, and NMR spectra were in accord with the assigned structure.167

Attempted bulk cyclization of N. N-diallyl-tert-butylamine (36). Monomer 36
was vacuum transferred from CaH2 (stirred for 24 h) to a potassium mirror and allowed

to stir for 12 h. 36 was then vacuum transferred (static transfer) to an appropriate storage

flask equipped with a TeflonTM vacuum stopcock. This reaction was conducted in a

similar manner to that of 34 using catalyst 2. Molar ratio of 36:2 @ 450:1; 250C, 24 h.









Upon addition of catalyst 2, a temporary evolution of ethylene was observed. 1H NMR

revealed a small internal olefin spike at 5.32 ppm (0.4 mol% conversion), indicating

cyclization had occurred. This observation was confirmed by GC/HRMS (M/Z =

125.1240). Upon removal of all volatiles, the catalyst residue was charged with a

minimal amount of 1,9-decadiene to test catalyst activity. No catalyst activity was

observed.

Solution cyclization of N. N-diallyl-tert-butylamine (36). Monomer 36 was
prepared as above. 2 mL of a 0.30 M solution of monomer 36 in CDC13 was prepared in

a dry box and added to a TeflonTm capped vial with catalyst 2 at 450:1 (3.:2) molar ratio.

The golden yellow reaction was allowed to stir for 24 h. 1H NMR analysis revealed an

internal olefin at 5.39 ppm at 1.7 mol% conversion.

Synthesis of N. N-diallylbenzylamine (37) Monomer 31 was prepared and

purified as described in the literature and had physical constants in agreement with those

previously recorded, and NMR spectra are in accord with the assigned structure. 167

Synthesis of N. N-dipentenylbenzylamine (38). Monomer 38 was prepared by a

similar procedure to that used for 37 using 5-bromopentene (Aldrich, 4.80 g, 0.0322

moles). After repeated fractional vacuum distillations (91-94C @ 1 mm Hg), 2.95 g (79

% yield) of clear liquid 38 was recovered. Monomer 3.8 had the following spectral

properties:
1H NMR (CDC13): 8 1.57 (m, 4H); 2.10 (q, 4H), 2.46 (t, 4H), 3.58 (s, 2H, N-

CH2-Ar), 4.96 (m, 4H, CH=CH2), 5.81 (m, 2H, CH=CH2), 7.10-7.40 (m, 5ArH); 13C

NMR: 8 26.48, 31.52, 53.37, 58.76, 114.30, 114.35, 126.59, 128.01, 128.71, 138.75;

HRMS: 244.2076; Calcd for C17H26N 244.2065; Elemental: Calcd: C, 83.61; H, 10.66;

N, 5.73; Found: C, 83.72; H, 10.61; N, 5.67.









Attempted bulk cyclization of N. N-diallylbenzylamine (37) Monomer 32 was
vacuum transferred from CaH2 (stirred for 24 h) to an appropriate storage flask equipped

with a TeflonT vacuum stopcock. This reaction was conducted in a similar manner to

that of 23 using catalyst 2. Molar ratio of 31:2 @ 170:1-450:1; 250C, 24 h. Upon

addition of catalyst 2, a temporary evolution of ethylene was observed, but 1H NMR did

not reveal an internal olefin spike nor was a cyclic product detected by GC/HRMS. Upon

removal of all volatiles, the catalyst residue was charged with a minimal amount of 1,9-

decadiene to test catalyst activity. No catalyst activity was observed.

Attempted solution cyclization of N. N-diallylbenzylamine (37). Monomer 37
was prepared as above. 2 mL of a 0.36 M solution of monomer 37 in CDC13 was

prepared in a dry box and added to a TeflonTm capped vial with catalyst 2 at 150:1-500:1

(37:2) molar ratio. The golden yellow reaction was allowed to stand for 24 h. 1H NMR

analysis did not reveal the formation of an internal olefin which was confirmed by

GC/MS.

Attempted polymerization of N. N-dipetenylbenzylamine (38). Monomer 38 was
synthesized as previously described. 38 was vacuum transferred from CaH2 (potassium

mirror isomerized diene) to an appropriate storage flask equipped with a TeflonTM

vacuum stopcock. This reaction was conducted in a similar manner to that of 28. Molar

ratio of 38:2 @ 500:1; 250C, 24 h, then 50C for 24 h. No visible evolution of ethylene

was observed upon addition of 2, but IH NMR revealed a newly formed internal olefin

signal at 5.36 ppm. The terminal olefins condensed at 6 mol% conversion to form dimer,
which was confirmed by GC/HRMS: HRMS (M+I): 459.3628; Calcd for C32H47N2

459.3739.

Attempted metathesis of monomers 34-38 with ruthenium catalyst (3). All
metathesis reactions were performed under bulk and solution conditions as reported

above. Monomer to catalyst concentrations ranged from 150:1 1000:1. In each case,






66

ethylene evolution was never observed and no signs of productive metathesis were

detected via 1H NMR or GC/MS. The catalyst residue from each reaction was charged

with a minimal amount of 1,9-decadiene to test catalyst activity. No catalyst activity was

observed.














CHAPTER 3
DESIGN AND SYNTHESIS OF UNSATURATED POLYSULFIDES


Polymonosulfides, as introduced in Chapter 1, represent a classification of

polymeric materials whose limitations have been fundamentally rooted in noncontrollable

synthetic approaches and limited range of monomeric starting materials. Difficulties in

monomer synthesis, in addition to competing oxidative side reactions have prevented the

full utility and viability of these systems for widespread practical application. The work

presented in this chapter describes the direct synthesis of a series of monosulfide dienes

and disulfide dienes to study their metathetic activity towards Schrock's molybdenum
neophylidene, Mo(CHCMe2Ph)(N-2,6-C6H3-i-Pr2)(OCMe(CF3)2)2 (2) and Grubbs'

ruthenium alkylidene, Ru=CHPh(PCy3)2CI2 (3) in the effort of producing linear

unsaturated polysulfides.

The viability of acyclic diene metathesis (ADMET) polymerization in the

synthesis of unsaturated polymers is now well established. Acyclic diene metathesis

polymerization is an equilibrium, step propagation condensation reaction where the

production and removal of a small alkene, typically ethylene, drives the reaction forward

(Figure 3.1).139 In this manner, unsaturated hydrocarbon polymersl4 as well as


Rcat^ R N N4. + CH2=CH2

n
Figure 3.1. Acyclic diene metathesis (ADMET) polymerization.








polymers and oligomers containing ether,148 carbonyl,151,152 silyl,153,168,169 and
ferrocenyl146 moieties have been prepared.
The typical catalysts employed for ADMET polymerization are the Lewis acid-
free Schrock's catalyst of the type M(CHR')(NAr)(OR)2 where M = W105 (1) or Mo,163
(2) Ar = 2,6-(i-Pr)2-C6H3, R' = CMe2Ph, and R = CMe(CF3)2. More recently,
warranted investigations using Grubbs' ruthenium alkylidene, RuCl2(=CHPh)(PCy3) (3),

have been performed due to the tolerance of the catalyst to functionality and resistance to
protonic media. Catalyst systems 2 and 3 (Figure 3.2) will be the initiators of choice in
this study. These classes of alkylidene complexes are perhaps better known for their



r PCy3

Cl, I
N Ru==\
II Cl I Ph
F3C O- M- PCy3
3p IPh (3) Cy = cyclohexyl

F3C CF3 M = Mo (2)

Figure 3.2. Schrock alkylidene 2 and Grubbs ruthenium catalyst 3.


applications in ring opening metathesis polymerization (ROMP) where they act as
initiators in the metathesis of cyclic olefins.105,117 ROMP is a chemically different
process from ADMET polymerization in that the thermodynamic driving force for ROMP
is derived from the release of ring strain in the monomer unit, and the polymer forms via
a chain growth process.70
Common to both processes is the ability of the molybdenum complexes to initiate
polymerizations with monomers possessing a range of functional groups. For the
ADMET studies mentioned above, a general rule of monomer structure has evolved









where if a functional group is separated from the terminal alkene by three or more

methylene spacers, the reaction will proceed cleanly to generate linear, high molecular

weight polymers.148,151-153,168,169 In some cases, two or fewer methylene spacers

will also provide polymer or oligomer, but the result varies depending upon the

functionality involved. A well-defined mechanistic interpretation of these observations is

not presently available, but it has been denoted by Wagener, Boncella and coworkers as

the 'negative neighboring group effect.'

In an attempt to further explore the 'negative neighboring group effect' and extend

the range of functional groups which may be incorporated into polymers made via

ADMET, a series of symmetrical ac, co-unsaturated thioethers were studied under bulk

ADMET conditions. The present studies demonstrate that ADMET is equally productive,

as compared to previous ADMET systems, when sulfur is located directly in the main-

chain. Presently, these polymers appear to be the first linear, unsaturated polythioethers

made in this way, and this novel reaction suggests that the catalytic molybdenum

complexes involved in the ADMET polymerization mechanism are more stable to Lewis

bases than was previously believed. In contrast, the ruthenium based catalyst systems

developed by Grubbs and coworkers are ineffective in producing metathesis condensation

products due to unknown decomposition pathways.


Historical Review of Sulfide Metathesis


The potential application of olefin metathesis to cyclic and acyclic olefins

possessing highly functionalized Lewis basic substitutents comprises one of the most

encouraging uses of this reaction since its discovery. The broad range versatility of this

process has created new avenues to synthesizing natural organic products and highly

functionalized polymers which are difficult or impossible to produce by conventional

organic transformations. Until recently, there were no reports of successful metathesis of








cyclic or acyclic sulfur-containing olefins. The limited research in this area was primarily

due to the ability of the sulfur atom to coordinate and/or poison the active catalyst. These

deleterious effects have been reported for a limited number of classical heterogeneous
and homogeneous catalyst systems. 170,171

Schrock and coworkers recently reported the first successful metathesis reaction
of a sulfur-containing cyclic olefin by producing a series of multiblock copolymer
systems based on a thioether-substituted norbornenyl monomer via ring opening
methathesis polymerization.172 The t-butoxy version of catalyst 2. (Figure 3.3) was

employed which yielded high molecular weight, living polymers which were utilized to

create metal cluster containing semiconducting materials that exhibited microphase
separation. These reactions went to further demonstrate the versatility and unprecedented
tolerance to functionality of Schrock-type alkylidenes.

Basset and coworkers provided further insight into the metathesis puzzle by
demonstrating the ring opening metathesis polymerization of a series of 5-
alkylthiocyclooctenes173 and the ring closing metathesis cyclization of a series of

strategically substituted diallylsulfides.174 In both cases, a homogeneous
aryloxo(chloro)neopentylidenetungsten complex was utilized to produce high molecular
weight alkyl-thioether substituted polymers (Figure 3.3) and high yielding





Rw I
SR ArOl
Et20
SR n
Mn= 39700
R = Et, n-Bu, t-Bu, n-Hex Mw/Mn = 3.5

Figure 3.3. ROMP of 5-alkylthiocyclooctenes using Basset's catalyst.









solution cyclization products respectively. The experimentally observed high levels of

activity and functional group tolerance can be attributed in part to the completing Et20

molecule (promotes coordinative saturation) and the rigid aryloxo ligand which facilitates

steric shielding of the tungsten metal center from nucleophilic coordination of the thio

moiety. The extremely rigid nature of the metallacyclic structure also contributes to the

sterically controlled nature of the metal coordination sphere.


Monosulfide Structure and Reactivity Relationships for ADMET


The broad technical applications of sulfur-containing polymers, as described in

Chapter 1, created a desire to investigate potential approaches via ADMET

polymerization in their synthesis.26 In keeping with previous ADMET studies for

investigating functional group compatibility with Schrock's metathesis alkylidenes, a

series of four a, co-unsaturated thioethers were synthesized in a manner described by

Butler and Price.192 These potential monomer compounds were strategically

synthesized where the terminal diene unit was sequentially separated from the thio moiety

with one to four methylene spacers (compounds 4-2, Figure 3.4).


1/2 Na2S*9H20 + Br EtfOH uS
x x x
x=2, 67%
x=3, 85%
x=4, 76%
Figure 3.4. Synthesis of a, o-unsaturated thioethers sulfidess) 5-2.


An initial concern in these investigations was the potential ability of the sulfur

group to act as a Lewis base and bind to a molybdenum or ruthenium complex at some

stage in the metathesis cycle (Figure 1.28). In this manner, the catalyst could become

poisoned and decompose, or an intermediate in the metathesis cycle could become









relatively stabilized and slow or stop the propagation rate needed to form high polymers.

Lewis base binding effects have been reported previously for similar ROMP

systems155,156 as have the formation of stable metallacyclobutanes in the metathesis of

ester containing olefins. 104

Despite these concerns, previous successful ADMET bulk polymerizations of

acyclic dienes possessing Lewis basic moieties warranted the investigation of sulfur-

containing monomers. As described earlier, Schrock and Basset have reported the

successful ROMP preparation of sulfur-containing polymers from a methylthioether

substituted norbornene,172 and 5-alkylthiocyclooctene monomer173 respectively, further

suggesting that the sulfur moiety would be tolerated in the ADMET reaction mechanism.
Herein, it will be described that simple a, co-unsaturated thioethers can be successfully

polymerized under standard ADMET conditions (Figure 3.5). The rules and conditions

of reactivity versus monomer structure have been found to match closely with those of

the previously reported oxygen ethers with the tungsten analog (1) of catalyst 2.


,_S__2 No polymerizationa
4



5 2



61Q n




2 + (0

12

Figure 3.5. Bulk ADMET chemistry of a, co-unsaturated sulfide monomers.
aSee Table 3.1 and cyclization discussion.











Sulfide ADMET Cyclization


When one equivalent of catalyst 2 was added to 2000 molar excess of diallyl

sulfide (4) in the bulk, the instantaneous evolution of ethylene was observed which

subsided shortly thereafter. Analysis of the mixture by 1H NMR showed three new

resonances assigned to ethylene (5.28 ppm) and cyclization product 2,5-dihydrothiophene

(8) (5.43 and 3.49 ppm in C6D6). Subsequent analysis by 13C NMR and GC/MS

confirmed the assignment of 8.175 The ratio of acyclic diene 4 to cyclic 8 was 14:1 as

determined by 1H NMR integration. This ratio did not change when the

monomer/catalyst mixture was diluted with either dry benzene-d6 in an inert atmosphere

or with wet benzene-d6 after exposure to air. Furthermore, no change in the ratio of 4 to

8 was observed over the course of 20 hours while the neat mixture was stirred under an

argon atmosphere.

When additional catalyst was added to the one day old mixture (ca. 33% increase

in 2), a second brief evolution of ethylene was observed, and the ratio of 4 to 8 was

lowered to 9.5:1 with no further change after several hours of stirring. Subsequent

removal of volatile products by high vacuum (< 10-4 Torr) distillation at room

temperature left reacted catalyst 2 in the vacuum flask. Addition of 1,9-decadiene to the

catalyst afforded no reaction, indicative of catalyst decomposition or deactivation.

When a new 250:1 ratio mixture of monomer to catalyst was prepared, the same

reaction was observed. In this instance, however, cyclization product 8 was the major

component of the mixture in a 4:1 ratio to diallylsulfide (4). When a similar

monomer/catalyst reaction was prepared and diluted with dry benzene-d6 immediately

prior to catalyst addition, 2,5-dihydrothiophene (8) was formed in a 99% NMR yield with

no diallyl sulfide resonances observed in the respective NMR spectra. Table 3.1

summarizes the results of the metathesis condensation, cyclization reactions.









These observations are similar to the reactivity of diallylether and catalyst 2,

where in the bulk an equilibrium mixture of oligomer and 2,5-dihydrofuran was

formed.149 Why in the present case oligomers do not form under bulk conditions

remains open to question, although the sulfur heterocyclic should be less reactive to

ROMP chemistry than the oxygen analog due to its lower ring strain and higher

coordinative character. The unsaturated five-membered ring product 8 is apparently

thermodynamically favored due to its comparatively low ring strain.

In solution, where cyclization would be kinetically favored, the catalyst appears to

remain sufficiently reactive in the presence of 4 to complete one intramolecular

metathesis cycle to form 8. This cyclization result is consistent with Fu and Grubbs'

metathesis ring-closing of a, 0)-unsaturated ethers, amines, and amides with catalyst
2.111,112,149 This cyclization result also agrees with Basset's report of the quantitative

bulk cyclization of diallylsulfide using an aryloxotungsten catalyst.173 It is interesting to

speculate that in the bulk, the Lewis basic dihydrothiophene & may be reacting to poison

the catalyst after its formation, although in the Basset case the rigid nature of the catalyst

center appears to deter this problem. This hypothesis would explain the increased

conversion of diallylsulfide (4) to 2,5-dihydrothiophene (8) as a function of increased

catalyst addition and solvent concentration. Similar tests under identical conditions

employing Grubbs ruthenium 3 did not yield the cyclic product 8 or any qualitative

detection of ethylene upon addition.

The potential poisoning nature of 2,5-dihydrothiophene (8) upon cyclization by

catalyst 2 was subjected to investigation via a control experiment. Cyclic B was produced

in an analogous fashion as described above (Table 3.1) for the solution cyclization

experiment of diallylsulfide (4). Upon completion, the cyclic product was purified by

distillation and 10 mole equivalents was exposed neat to catalyst 2 and allowed to stir

prior to addition of an inert diene in efforts to test catalytic activity. The exposure of 1,9-

decadiene to the stirring mixture did not generate ethylene gas nor was condensation









observed in the 1H NMR. This evidence helped to provide further support of the

deactivation of catalyst 2 by direct coordination of cyclic 8 during bulk cyclization.

Table 3.1. Cyclization reactivity of diallylsulfide (4) to form 2,5
dihydrothiophene (8).

Relative Ratio Relative Ratio

a,G)-diene catalyst a,Go-diene cyclic


^^S [Mo] 0

4 2 4 8

Reaction
Conditions

2000 1 Bulk, rt, 0-20h 14 1

1500 1 Bulk, rt, 0-20h 9.5 L1


250 1 Bulk, rt, 0-20h 1 4



190 1 0.36M (C6D6), rt, 3h 1 4


ADMET Polymerization of Polymonosulfides


When a catalytic amount (< 0.1%) of 21 was added to acyclic sulfur-dienes 5-27
where the number of methylene spacers between the terminal double bond and sulfur

atom are two, three, and four respectively, the evolution of ethylene ensued. Upon an

increase in viscosity of the polymeric oils, high vacuum was applied to the mixtures to

remove ethylene and drive the equilibrium to high polymer. Polysulfides 9-12 were









stirred at room temperature until the viscosity prohibited magnetic agitation at which time

the polymeric products were heated to 50 C where stirring was continued. When

ethylene evolution was visually observed to have subsided or when the polymers again

became too viscous to stir, the reactions were quenched by exposure to air.

Characterization using NMR, IR, GPC, and elemental analysis found products 2-

12 to be consistent with the assigned structures and typical of ADMET polymers. Figure

3.6 displays a classic ADMET 13C NMR spectra of a pure acyclic diene monomer

(monomer 5) and its resulting unpurified condensation metathesis polymer (polymer 9_.

As demonstrated in the spectra, the terminal C=C olefin resonances (116 and 136 ppm) of

monomer 5 diminish in intensity while two new signals, which represent internal C=C

olefins, appear as a consequence of metathesis condensation. These internal signals (129

and 130 ppm) represent the cis and trans geometric isomers respectively, of the olefins

along the unsaturated polymer backbone. As can be observed in the spectrum of polymer

9, residual terminal olefin resonances are still observed after reaction completion, as

evidenced by cessation of ethylene evolution. The quantitative 13C NMR of polymer 2

provided reliable, absolute data for determining number average molecular weights based

on endgroup analysis. 1H and 13C NMR spectroscopy also showed the polymers to be

perfectly linear unsaturated polysulfides (polythioethers) with a trans:cis internal double

bond ratio approaching 4:1 (Table 3.2). Relative molecular weight values determined by

gel permeation chromatography (GPC) were consistent with those determined by

quantitative endgroup NMR integration, and polydispersities (GPC) approached values of

1.5-2.0, consistent with the ADMET step condensation mechanism. Combustion analysis

of the polymers, using NMR determined molecular weights, clearly showed the expected

sulfur content for the assigned polymer structures. As reported for the diallylsulfide (4)

case, Grubbs ruthenium complex 3. did not afford the corresponding condensation

metathesis polymers when exposed to sulfide-dienes 5-7. The above characterization is

summarized in Table 3.2.












4
3

2
1 3

2 4









140l 120 100 80 60 40 2li'j]ilppm 01 T
140 120 100 80 60 40 20 ppm 0


7 (trans)


1 3 6
2 4 5 7 n


7 (cis)


6 (trans)


6 (cis)


40 120 100


80 60 40 20 ppm 0


Figure 3.6. 13C NMR of a) monomer 5 and b) polymer 2.








Comparison of number average molecular weights for polysulfides 9-12 indicate a
significant weight increase once three or four methylene spacers separate the terminal

diene from the sulfur atom. The transition in reactivity is in accord with other

functionalized ADMET polymers where the 'negative neighboring group effect' is not

observed when three or more methylene spacers are present. An explanation for the

slightly higher molecular weight of polymer 10 as compared to polymer 11 is not


Table 3.2. Unsaturated polysulfides prepared by ADMET polymerization.


Polymer Mna Mnb Mw/Mna Tmc TGAd Elemental
(GPC) (NMR) (oC) N2(oC)

36,43, C 63.13 (63.36)
9 3100 3100 1.47 47 295 H 8.82 (8.82)
S 27.45 (27.82)

C 67.54 (67.61)
10 16330 17410 1.62 1,7 370 H 9.86 (9.86)
S 22.57 (22.53)

C 70.51 (70.59)
11 9410 10200 1.86 23 355 H 10.60 (10.59)
S 18.77 (18.82)

C 76.23 (76.19)
12 34940 1.31 13 381 H 11.14 (11.11)
N 12.61 (12.70)
a Relative to polybutadiene standards. b Measured by quantitative 13C NMR.
c 100C/min scan rate. d Temperature at 10% weight loss (100C/min).


known at this time, however, the presence of impurities resultant from the more difficult

distillation procedure of high boiling monomer 7 may be responsible. An alternative
explanation prescribes the higher molecular weight of polymer 10 to its physical
properties. Polymer 10 remains in a 'fluid-like' state throughout the course of its
formation which allow its reactive ends the mobility necessary to make connections









leading to higher degrees of polymerization. In the case of il, however, the polymer has

a higher degree of crystallinity which restricts terminal olefin condensation in the bulk

(See Thermal Properties below).
When a catalytic amount of 2 (< 0.12%) was added to a 1:1 mole mixture of a, (0-

unsaturated thioether 7 and 1,9-decadiene, ethylene evolution ensued leading ultimately

to copolymer 12. Full characterization in the manner outlined above once again found

the polymer to display properties consistent with the assigned structure. NMR

spectroscopy showed the polymer to be perfectly linear with a ratio of ether and

octenamer repeat units approximating 50:50. In the 13C NMR spectrum, eight

unsaturated carbon-carbon double bond resonances were observed as expected for the

trans and cis (approaching 4:1 for each) connections of thioether-thioether (two carbons),

octenamer-octenamer (two carbons), and thioether-octenamer (4 carbons) repeat unit

combinations. Gel permeation chromatography (GPC) further confirms the synthesis of

one copolymer as opposed to a blend of two homopolymers (Figure 3.7).


10
Elution Volume (mL)


Figure 3.7. Gel permeation chromatogram (GPC) of copolymer 12.









The above observations suggest that the catalytic molybdenum intermediates

involved in polymer formation do not distinguish chemically between the terminal olefins

of the thioether and 1,9-decadiene monomers. Finally, copolymer 12 had the highest

molecular weight of the series (Table 3.2). Once again monomer purity may be

responsible, but the 'fluid-like' state of the copolymer, even at high molecular weights,

may facilitate continued terminal olefin metathesis.


Polysulfide Thermal Properties


Thermogravimetric analysis (TGA) of polymers 2-12 showed all macromolecules
to be stable at elevated temperatures (Figure 3.8 and Table 3.2). In comparison to the

oxygen ether analogs,148 ten percent weight loss was found to occur at slightly lower


50 150 250 350 450
Temperature (C)

Figure 3.8. Thermogravimetric analyses of unsaturated polysulfides 9-11.









temperatures. This was particularly pleasing in view of the anticipated lower thermal

stability due to the relatively weaker carbon-sulfur covalent bond energy.176 Secondly,

stability was found to be directly related to molecular weight, in accord with the oxygen

ether system.

In comparison to the saturated analogs of these polymers, 177 the presence of an

olefin in the repeat unit apparently leads to no significant change in the thermal stability

of the polymer. While differences in molecular weight and analysis technique make

comparison tenuous, values between the saturated and unsaturated polythioethers fell

within 50 C of each other. Finally, the sharp weight loss of the TGA curves for 2-11

indicated a clean decomposition pathway as expected for these linear polymers.

Differential scanning calorimetry (DSC) showed some significant differences in

the melting properties of polymers 2-11 as a function of methylene spacers between the

sulfur atom and olefin of the repeat unit (Figure 3.9 and Table 3.2). Unsaturated polymer

9 showed a three phase melting transition at 36, 43, and 47 C. Polymer 10 displayed a

two phase transition at 1 and 7 C, and polymer H displayed only one melting transition

at 23 C. All endotherms remained unchanged after at least three recrystallization-

melting cycles. In comparing the melting point versus the number of methylene spacers

between the repeat unit sulfur atom and double bond for each polymer, it is observed that
the two methylene spacer polymer 2 had the highest Tm, four methylene spacer polymer

11 was lower, and finally three methylene spacer polymer 10 had the lowest Tm of the

series.

Previous ADMET synthesized polymers have displayed similar melting point

variations as the number of methylene (CH2) spacers between the functional group and

olefin changed from two, to three, and/or four. For example, in the case of unsaturated

polyesters,178 melting points were observed at 163, 40, and 101 C for the symmetrical

two, three, and four CH2 spacer polymers respectively. Symmetrical unsaturated

ADMET polyethers and carbonates with three CH2 spacers were found to have no









melting transitions at all.148,152 Similar changes in Tm as a function of methylene

spacer have been reported for the saturated versions of the sulfur polymers where
poly(methylene sulfide) and poly(ethylene sulfide) have high Tm values, the Tm of

poly(trimethylene sulfide) is considerably lower, and the Tm for the higher homologues

increase again as the number of CH2 spacers increases. 177,179,180 In fact, a variety of

other polymers possessing a linear chain of methylenes and one functional group per

repeat unit also show these trends in melting temperature as the number of CH2 spacers

change.181

Based on the limited data from ADMET synthesized polymers, a similar trend in
Tm as a function of CH2 spacers between the functional group and olefin appears to be

developing. That is, for the appropriate ADMET polymers made from symmetrical a, co-

unsaturated dienes, the three CH2 spacer polymer of the series can be expected to have

the lowest Tm. It is noted, however, that ADMET polymers possess two different

groupsin the backbone (an olefin and in the present case a sulfur atom), as compared to
the monofunctional polymers for which the CH2 spacer versus Tm relationship was

originally observed. Secondly, differences in polymer molecular weights,

polydispersities, and structures resulting from the synthetic method and degree of
polymerization are also expected to affect Tm values. For these reasons, a relationship

between the physical properties of ADMET polymers and the number of methylene

spacers per functional group and olefin is discussed with caution.

Finally, DSC analysis of copolymer 12 showed a single endothermic transition at
130C which remained unchanged after repeated temperature cycles. As expected, the Tm

value was below that of either of the two homopolymers, polymer 11 or 77% trans

polyoctenamer (52 C).139 This value was also lower than that reported for the oxygen
analog of the copolymer which had a Tm of 31 C for an approximate 2:1 octenamer:ether

repeat unit ratio. 148
















S" n


-o
V






12 n










10 n


-20 0 20 40 60
Temperature (C)

Figure 3.9. Differential scanning calorimetry (DSC) analysis for polysulfides 2-
It.









Attempted ADMET Cyclization and Polymerization of Disulfide Monomers


The successful metathesis condensation of thioether monomers 5-7 proved that an

effective synthetic methodology could be developed to circumvent the 'negative

neighboring group effect' for the historically challenging sulfide functionality. The next

logical extension of this study was to investigate the metathesis reactivity of disulfide

monomers toward Schrock's alkylidene 2 and Grubbs' ruthenium alkylidene 3. This

progressive study into disulfide monomers was more than just an intellectual curiosity,

but rather an exercise in developing a polymer system with two built in functional groups

for depolymerization (Figure 3.10). The disulfide linkages can be cleaved

stoichiometrically under reducing conditions to produce thiol telechelics (HS-R-SH) or

the unsaturated polymer can be reversed by ethylene depolymerization to low molecular





n

[H] / ethylene


HSH SSH S-

thiol telechelic disulfide diene

Figure 3.10. Unsaturated polydisulfide depolymerization to thiol telechelics and
disulfide dienes.


weight disulfide dienes. Obviously, the novel approach to synthesizing low molecular

weight thiol telechelics provided the initial thrust into this investigation.

Initial disulfide studies involved the attempted metathesis of the commercially

available (80%, Aldrich) diallyldisulfide (13). It was demonstrated that monomer 13

showed qualitative resistance (no ethylene evolution) to metathesis when exposed to

catalyst 3 under standard ADMET bulk or similar solution conditions described for









diallylsulfide (4), but did show marginal ethylene evolution when treated with catalyst 2.

Irregardless of catalyst concentration (150:1-500:1), no cyclized nor linear internal olefin

product was detected via 1H NMR or GC/MS. Although diallyldisulfide (13) did not

produce an observable metathesis product, exchange with the initial precatalyst 2 was

predicted based on previous observations with diallylsulfide and visible appearance of

ethylene. The failure to produce a detectable metathesis product was most likely due to

intramolecular coordination and/or intermolecular catalyst deactivation. In order to prove

that catalyst deactivation had occurred rather than the catalyst system simply being inert

to the disulfide-diene, all volatile components were removed in vacuo to test the

remaining catalyst residue with a bench test diene (1,9-decadiene). Metathesis

condensation of 1,9-decadiene did not occur indicating that catalyst deactivation and/or

coordination was more favorable than condensation metathesis.

Bis(5-hexenyl)disulfide (15) was synthesized from the oxidation of 5-hexenylthiol

in a manner described in Figure 3.11. Due to the resistance of diallyldisulfide to

quantitative metathesis, an alternative approach was initiated to investigate the

structure/reactivity relationship for disulfide monomers. By developing a monomer

system with four methylene spacers between the disulfide moiety and the reactive olefin

termini, linear propagation should be favored over the process of intramolecular metal-

cyclized coordination which would produce a pseudo seven or eight-membered ring.


1. NaSH*xH0
Br 2. 30 wt%H202 I


Figure 3.11. Synthesis of bis(5-hexenyl)disulfide (15).


Upon addition of catalyst 2 to 15 a marginal, temporary evolution of ethylene was

observed under bulk conditions. The polymerization reaction was allowed to stir at 25C

for 24 hours prior to analysis. Figure 3.12 displays the 13C NMR spectrum of the









metathesis reaction of monomer 15, where no internal olefin signals can be observed to

indicate appreciable condensation metathesis has occurred. In spite of the negative

results provided by NMR spectroscopy, a trace amount of the metathesis dimer was

confirmed by GC/HRMS (M/Z+1 = 433.1960). The initial indication of ethylene

evolution upon addition of catalyst 2 to 15 demonstrates that metathesis is possible, but

that competing coordination chemistry whether it be via intramolecular or intermolecular

pathways prevents this from being a viable route to a unique class of polymers. As

presented for diallyldisulfide (13), catalyst 3 demonstrated no catalyst activity toward

monomer 15, as confirmed by NMR and GC/HRMS.


[Mo], 2


140 120 100 80 60 40 20 ppm

Figure 3.12. 13C NMR of monomer 15 in the presence of catalyst 2.








Model Studies for Polythiophenes via ADMET


The field of conducting and electroactive polymers has been at the forefront of
polymer science for the past decade, with unique and novel approaches to these
interesting polymeric systems saturating the literature. Polythiophenes, in particular,
have garnered considerable interest in recent years, but many of the synthetic methods
employed often yield insoluble, intractable materials that are rendered noncharacterizable
by conventional methods of analysis. Acyclic diene metathesis (ADMET)
polymerization may offer a synthetic alternative to obtain these elusive structures via well
defined, clean condensation chemistry no side reactions, no mysteries.
Initial investigations were centered around the durability of the catalyst systems
used in this study to the thiophene molecule itself. Ten mole equivalents of freshly
distilled thiophene was introduced to catalyst 2 and 3 and allowed to thoroughly stir prior
to addition of 1,9-decadiene. In each case studied, ethylene evolution was not observed
and no detectable condensation metathesis products were determined by NMR
spectroscopy. In spite of these results, 2-propenylthiophene (16) was synthesized by a Ni
catalyzed Grignard coupling reaction as described in Figure 3.13. The propenyl version
of this model compound was synthesized to avoid possible vinyl polymerization during
purification or upon standing in ambient light.


S Br Ni(dppp)C12 -'C-
BrB

56%
Figure 3.13. Synthesis of 2-propenylthiophene (16).


Upon purification of 16, catalyst 2. and 3 were introduced, but both proved
ineffective in generating a metathesis condensation dimer. All volatiles were removed









from the reaction vessel and charged with neat 1,9-decadiene to test the catalyst residue

for activity. No activity (absence of ethylene) was observed indicating that both catalyst

systems had become deactivated as in the case for neat thiophene. Although the ultimate

path of decomposition or nonreactivity remains unanswered, possible explanations are

direct coordination of the thiophene unit or decreased reactivity of both catalyst systems

toward the aromatic substituted internal olefin.


Conclusions


A series of unsaturated polysulfides have been synthesized under standard acyclic
diene metathesis (ADMET) conditions. Symmetrical a, co-dienes (15-) where the

internal sulfur atom is separated from the terminal olefin by two, three, and four

methylene spacers all react cleanly to form linear polymers when catalyst 2 is utilized.

Catalyst 3 proved ineffective as a viable metathesis catalyst for the condensation of

sulfide olefins. A lower molecular weight for the two methylene spacer polymer 2 may

suggest that some form of intramolecular deactivation of the catalyst metal center is

occurring. When one methylene spacer separates the olefin from the sulfur atom, partial

cyclization occurs in the bulk, but quantitative cyclization only occurs in solution. All of

the polymers synthesized displayed a high degree of thermal stability. Melting transitions

were also observed for the polymers 9-12 which vary with the number of methylene

spacers in the polymer repeat unit. Further, disulfide monomers 13 and 15 proved to be

resistant to metathesis with catalyst 2 and 3 as well as the thiophene based olefin 16.














CHAPTER 4
UNSATURATED AROMATIC AND ALIPHATIC AMINES FOR ACYCLIC
DIENE METATHESIS (ADMET) POLYMERIZATION


Main-chain polyamines, as introduced in Chapter 1, are very unique polymers that

have garnered considerable interest in both academic and industrial arenas. Difficulties

in monomer synthesis, in addition to the inherent complications of competing side

reactions, whether it be via a ring opening process or condensation polymerization, have

prevented polyamines from becoming an even more highly investigated area of polymer

chemistry. It has been the focus of this work to not only expand the range of

functionalities available to acyclic diene metathesis (ADMET) polymeriztion, but to also

provide a unique and unprecedented approach to perfectly linear, main-chain polyamines.

This chapter describes the synthesis of a series of aromatic and aliphatic amines to study

their metathetic activity towards Schrock's molybdenum (2) and Grubbs' ruthenium (3)

alkylidenes in the effort of generating a new class of well defined, main-chain

polyamines.


Historical Review of Amine Metathesis


The metathesis of amine-containing olefins with classical metathesis catalyst

systems has been severely limited by the highly coordinating ability of the amine

functionality with high oxidation state transition metals. In spite of the aggressive nature

of the amino-group, detailed accounts of olefin metathesis with amino-functionalized

olefins have been presented, albeit in small yields with bulky amine substituents

providing the best results.70,182,183 It has been postulated that the highly coordinative,









Lewis basic amino-group establishes an equilibrium between intramolecular coordination

(cyclization favored) and the linear extended alkylidene metalizeded' olefin), the

intermediate required for final metathesis product formation (Figure 4.1).





N
CH3
H3C CH3

Figure 4.1. Intramolecular coordination Extended alkylidene equilibrium.


Cycloamines of considerable ring strain have also been investigated to determine

if the more kinetically and thermodynamicaly favored ring opening metathesis

polymerization would allow these aggressive functionalities to be more metathetically

active with classical catalyst systems.184 It was demonstrated that when a N-

benzylnorbornenyl monomer was exposed to WCl6-t-butylphenol/Et3AI, only broad

polydispersed oligomers were recovered in yields of less than 10%. It was postulated, as

in the case for linear olefins, that intramolecular coordination drastically reduced the

reactivity of the strained olefin system, and prevented high degrees of conversion and

substantial molecular weight growth (Figure 4.2). Further, when a less bulky methyl


Figure 4.2. Intramolecular coordination during the ROMP of a cycloamine.








substituent was employed, no polymer formation was observed, possibly giving way to
direct coordination of the monomer system to the Lewis acidic metal center.
For more than 15 years after the initial reports on amine metathesis were
published, only marginal success was achieved. With the advent of Schrock's well-
defined Lewis-acid free (cocatalyst) alkylidenes, chemists were now able to utilize these
catalyst systems to synthesize organic compounds and polymeric materials with various
functionalities in a relatively facile manner. Wudl and coworkers were able to
cyclopolymerize various amino-diacetylenes by metathesis cyclopolymerization using
Schrock's molybdenum catalyst (2) to construct self-doped polyacetylenes.185

Table 4.1. Catalytic ring-closing metathesis of aminodienes with Schrock's [Mo]
catalyst (4 mol % catalyst, C6H6, 200C).112


Substrate Product Time Yield
(min) (%)
Ph Ph

| K R=H 40 86
N N Me 180 85


RP

Ph Ph
N N60 86


Me
Et
PhP


60 73
MMe

Me