SYNTHESIS AND CHARACTERIZATION OF LATENT REACTIVE
OLIGO(OXYETHYLENE)/CARBOSILANE AND
CARBOSILOXANE/CARBOSILANE COPOLYMERS VIA ACYCLIC DIENE
METATHESIS POLYMERIZATION (ADMET) FOR THERMOSET MATERIALS
By
PIOTR PAWEL MATLOKA
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
2006
Copyright 2006
by
Piotr Pawel Mattoka
Mojej ukochanej zonie Komelii i moim rodzicom za ich bezgraniczn4 miloC6,
wyrozumialtoS i przyjaili.
To my lovely wife Komelia and my parents for their eternal love, understanding, and
friendship.
ACKNOWLEDGMENTS
There are a great number of people who have given me their support,
encouragement, and support during my graduate career. First and foremost, I must thank
my wife Komelia. She has been everything for me during the course of my dissertation
work. She has been my rock, my motivation and my friend. Without her I would be
incomplete.
I would like to thank my parents for their endless love, devotion, support and
sacrifices they have made that allowed me to become a person that I am today. They
taught me at a young age what was important in life, and showed me how to work hard to
achieve my goals.
I need to acknowledge the people that molded my career as a student of chemistry
before going to collage and finally to graduate school. I would like to thank my
secondary school chemistry teacher Mrs. Krystyna Swiqch and my high school chemistry
teacher Mrs. Danuta Bolifiska for being wonderful teachers.
At Adam Mickiewicz University I would like to express my sincere thanks to Dr.
Wioleta Patroniak for her friendship and help in making right decision at my
undergraduate career. Sincere gratitude also goes to Prof. Bogdan Marciniec, my
undergraduate advisor whose encouragement and support made my undergraduate
research career extremely rewarding experience. I cannot leave out my best friend that I
had pleasure to study with Dr. Piotr Pawlu6, whose friendship I value very much.
During my career at University of Florida I have had pleasure to meet and work
among extraordinary people, such Dr. John Sworen, who I would like to thank for taking
the time to teach me good laboratory techniques, for introducing me to my project early
in my graduate career, and for being a true friend. I extend my thanks to my
classmates/friends Dr. Florence Courchay (aka BMB) and Dr. Travis Baughman for all
wonderful time we have spend together, especially marvelous time we had during our trip
to Europe. I cannot omit my "Polish fellow" Dr. Ryan Walczak who I thank for all
scientific and non-scientific discussion that we had. There have been many people that
have enriched my life, both inside and outside of the laboratory sincere thanks goes to
all of the past and current members of the Wagener research group for their helpful
discussions and friendship including Dr. Jason Smith, Dr. Tim Hopkins, Dr. Violeta
Petkovska, Dr. Garrett Oakley, Giovanni Rojas, Erik Berda, James Leonard, Meghan
Greenfield and Zach Kean. I have to thank also other members of the floor who I had
pleasure to meet. My experience at the University of Florida would not have been
complete without making the acquaintances of Dr. Carl Gaupp, Dr. C.J. DuBois, Dr. Ben
Reeves, Tim Steckler, Bob Brookins, Merve "Marwa" Ertas, Sophie Bernard, and
Victoria Broadstreet.
Enough thanks cannot be expressed to Lorraine Williams and Sara Klossner in the
polymer office. Without them, the little things that make everything run smoothly, but no
one wants to think about, would never get done. I also extend my thanks to Lori Clark for
her help in the graduate office.
Many thanks are extended to my committee members, Dr. John R. Reynolds, Dr.
Michael J. Scott, Dr. Daniel R. Talham, and Dr. Anthony B. Brennan for their help
throughout my journey at the University of Florida
Finally, I would like to express my sincere appreciation to my research advisor
Professor Ken Wagener. First, I would like to thank him for his encouragement,
guidance, patience, and understanding during the course of my graduate career. Second,
I gratefully value his advice and his confidence in my abilities as a chemist. Most
importantly I would like to thank him for having always time to sit and talk about not
only chemistry but simply everyday life. I will forever be thankful for his presence in my
life.
TABLE OF CONTENTS
page
A C K N O W LED G M EN TS ............................................................ .................................. iv
L IST O F T A B L E S ............................................................................... .......................... xi
LIST O F FIG U R ES .............................................. ............ ......... ......................... xii
A B S T R A C T ..................................................................................................................... xv i
CHAPTER
1 ACYCLIC DIENE METATHESIS POLYMERIZATION (ADMET)
APPROACH TO SILICON-CONTAINING MATERIALS.....................................1
1.1 Introduction ....................................................... ... ............... ......................1
1.2 Silicon C hem istry .......................................................... ................................ 3
1.3 O lefin M etathesis............................. ................................................................. 4
1.3.1 Brief History of Olefin Metathesis...........................................................6
1.3.2 Introduction to ADMET Polymerization................................................8
1.4 Olefin Metathesis Chemistry of Compounds Containing Silicon ........................12
1.4.1 Acyclic Diene Oligomerization and Polymerization of Containing
S silicon D ien es ..................................................................................................14
1.4.2 ADMET Catalyzed by "Ill-Defined" Catalytic Systems.......................... 15
1.4.3 ADMET Catalyzed by "Well-Defined" Catalytic Systems....................... 18
1.4.3.1 Polycarbosilanes.............................................................................. 18
1.4.3.2 Polycarbosiloxanes........................................................................ 21
1.4.3.3 Telechelic oligom ers .....................................................................24
1.4.3.4 Functionalized polycarbosilanes .................................................26
1.4.3.5 Latent reactive elastomers................................................................27
1.5 Silylative Coupling (SC) Polycondensation.................................................28
1.6 Crosslinking Methods for Silicon-Containing Materials................................... 31
1.7 Purpose of Study .................................................................................................33
2 SYNTHETIC MODIFICATION OF CROSSLINK DENSITY AND SOFT
SEGMENT IN POLYCARBOSILANE/POLYCARBOSILOXANE
E L A ST O M E R S ....................................................................................................34
2 .1 Introdu action ....................................................................... ....................................34
2.2 Synthetically Modified Crosslink Density..........................................................36
2.3 Synthetic Soft Segment Modification.................................................................39
2 .4 C on clusion s............. ................ ..... ..... ........ ...... ...... .............. ............43
2.5 Experim mental M ethods.................................................................................... 44
2.5.1 G general Considerations ................................... ......................................44
2.5.2 M materials. .............................................. ..... ................... ............ 45
2.5.3 M onom er Synthesis................................................................................. 46
2.5.3.1 Tris(4-pentenyl)methylsilane (80).................................................46
2.5.3.2 Diundecenyltriethylene glycol (83)..................................................47
2.5.4. General Metathesis Conditions...............................................................47
2.5.4.1 Polymerization of diundecenyltriethylane glycol (83)...................48
2.5.4.2 Terpolymerization of monomers 75, 76, and 80 to produce
terpolym er 81 ........................................... .................. .........................48
2.5.4.3 Terpolymerization of monomers 75, 83, and 80 to produce
terpolym er 84.................................................................... .................... ..49
3 SYNTHESIS AND CHARACTERIZATION OF
OLIGO(OXYETHYLENE)/CARBOSILANE COPOLYMERS AND
T H E R M O SE T S .................................................................... ............................. 50
3.1 Introduction..........................................................................................................50
3.2 Synthesis of M onom ers .................................................................................... 51
3.2.1 Monomers for Soft Phase Modification ..................................................51
3.2.2 Monomers for Crosslink Density Modification .......................................53
3.3 ADMET Polymerization.................................................... .....................54
3.3.1 Synthesis of Oligo(oxyethylene)/Oligo(ethylene) Copolymers .................54
3.3.1.1 Structural characterization................................................................ 56
3.3.1.2 Thermal behavior of unsaturated polymers....................................59
3.3.1.3 Thermal stability of unsaturated polymers.....................................64
3.3.2 Polymerization of "Chain-Internal" Crosslinker 3sp20Me .......................65
3.3.3 Synthesis and Characterization of "Latent Reactive" Copolymers...........66
3.4 Preparation and Characterization of Thermosets.................................................69
3.4.1 Thermal Behavior and Degradation of Thermosets ..................................71
3.4.2 M echanical Properties ................................... .........................................74
3.5 C onclusions.................................... .......................................... ................... .. 79
3.6 Experim mental M ethods................................. ...............................................80
3.6.1 M materials .......................... .................. ................................................ 80
3.6.2 Instrumentation and Analysis..................................................................80
3.6.3 M onom er Synthesis.................................................................................. 82
3.6.3.1 Synthesis of di(oxyethylene) a,o-di-4-pentenyl ether (3sp2gl)
(8 6 )............. ............... .................. ...................................................... ... 82
3.6.3.2 Synthesis oftri(oxyethylene) a,a-di-4-pentenyl ether (3sp3gl)
(8 7 ).............. ............... ................................................................ ... ..... 8 2
3.6.3.3 Synthesis oftetra(oxyethylene) a,co-di-4-pentenyl ether (3sp4gl)
(88)............................... ...................... .............. ................... ..... 83
3.6.3.4 Synthesis of tri(oxyethylene) a,co-di-7-octenyl ether (6sp3gl) (89).83
3.6.3.5 Synthesis oftetra(oxyethylene) a,o-di-7-octenyl ether (6sp4gl)
(90) .................................................... ............ ................................. 83
3.6.3.6 Synthesis of di(oxyethylene) a,o-di-10-undecenyl ether (9sp2gl)
(9 1).............................. ........... .......................................... ........... ... 8 4
3.6.3.7 Synthesis of tri(oxyethylene) a,o-di-10-undecenyl ether (9sp3gl)
(9 2 )................................. ....................................... ...... .................. ...... 8 5
3.6.3.8 Synthesis of tetra(oxyethylene) a,o-di- 10-undecenyl ether
(9sp4gl) (93) ............................................................... .................... 85
3.6.3.9 Synthesis of 10-undecenyldimethoxymethylsilane 9sp2OMe (97) .85
3.6.3.10 Synthesis of di(4-pentenyl)dimethoxysilane 3sp20Me (100) .......86
3.6.4 General Metathesis Conditions.............................................................. 87
3.6.4.1 Synthesis of poly3sp2gl ................................................... .............88
3.6.4.2 Synthesis of poly3sp3gl .................................................... ...........88
3.6.4.3 Synthesis of poly3sp4gl ..................................................................88
3.6.4.4 Synthesis of poly6sp3gl .................................................... ...........89
3.6.4.5 Synthesis of poly6sp4gl ..................................................................89
3.6.4.6 Synthesis of poly9sp2gl ..................................................................89
3.6.4.7 Synthesis of poly9sp3gl .................................................... ...........90
3.6.4.8 Synthesis of poly9sp4gl .................................................... ...........90
3.6.5 Preparation of Copolymers and Thermosets ............................................91
4 SYNTHESIS AND CHARACTERIZATION OF
CARBOSILOXANE/CARBOSILANE COPOLYMERS AND THERMOSETS.....93
4.1 Introduction............................. ...........................................................................93
4.2 M onom er Synthesis ................................................................... ................... 94
4.3 A D M ET Polym erization..................................... .......................................... 97
4.3.1 Synthesis of Oligo(siloxane)/Oligo(ethylene) Copolymers .......................97
4.3.1.1 Structural characterization.............................................................. 98
4.3.1.2 Therm al analysis .......................................................................101
4.3.2 Synthesis of "Latent Reactive" Carbosilane/Carbosiloxane Copolymers 104
4.4 Preparation and Characterization of Thermosets................................................105
4 .5 C conclusion s............................ ... ... .............................................................108
4.6 Experim ental M ethods..................................................... ................................. 109
4 .6.1 M materials ................................................. ............................................... 109
4.6.2 Instrum entation and Analysis................................................................. 109
4.6.3 M onom er Synthesis ......... ............................................ ........................109
4.6.3.1 Synthesis of 1,5-bis(hexenyl)-1,1,3,3,5,5-hexamethyltrisiloxane
4sp3Si (192)............................... ............. ...........................109
4.6.3.2 Synthesis of 1,7-bis(hexenyl)-1,1,3,3,5,5,7,7-
octamethyltetrasiloxane 4sp4Si (193)................................................ 110
4.6.3.3 Synthesis of 1,5-bis(octenyl)-1,1,3,3,5,5-hexamethyltrisiloxane
6sp3Si (194)..................................................... ............ ...................... 111
4.6.3.4 Synthesis of 1,7-bis(octenyl)-1,1,3,3,5,5,7,7-
octaamethyltrisiloxane 6sp4Si (195) .................................................111
4.6.4 G general M etathesis Conditions................................................................111
4.6.4.1 Synthesis of poly4sp3Si ................................................................112
4.6.4.2 Synthesis of poly4sp4Si .................................................................113
4.6.4.3 Synthesis of poly6sp3Si .............................................................. 13
4.6.4.4 Synthesis of poly6sp4Si .................................................................113
4.6.5 Preparation of Copolymers and Thermosets ............. ..........................1. 14
APPENDIX.................................. ....... .... ....................................................... .....115
LIST OF REFERENCES................................................... ...........................................118
BIOGRAPHICAL SKETCH ........................................................................................127
LIST OF TABLES
Table page
1-1 Comparison of various silicon and carbon bond lengths and energies ..................3
2-1 Mechanical properties of the unsaturated ADMET copolymers and a
com m ercially available elastom er. ................................... ....................................35
2-2 Synthetic research objectives. ................................................................ ................36
3-1 GPC data of unsaturated polym ers.............................................. ...................... 55
3-2 Thermal analysis of the unsaturated polymers ....................................................60
3-3 Physical properties of selected thermosets compositions.................................77
4-1 GPC data for unsaturated polycarbosiloxanes. .....................................................97
4-2 Thermal analysis of the unsaturated polymers.....................................................102
4-3 Mechanical properties of carbosiloxane/carbosilane thermosets...........................106
4-4 Thermal degradation data of carbosiloxane thermosets.........................................108
A-1 Results of mechanical tests on dicomponent thermosets xlink 1-8......................1115
A-2 Results of mechanical tests on dicomponent thermosets xlink 9-16......................115
LIST OF FIGURES
Figure page
1-1 Types ofolefin metathesis: a) simple exchange reactions; b, c) metathesis
polymerization and ring-closing metathesis; c) metathesis degradation ..................
1-2 M etallacyclobutane m echanism ................................................... .....................6
1-3 W ell-defined m etathesis catalysts. ................................... ................ .................7
1-4 Acyclic diene metathesis (ADMET) and ring-opening metathesis (ROMP)
polym erizations. .. .................... ................. ........................... ............................... 8
1-5 Theoretical degree of polymerization vs. conversion in condensation type
polym erization ............................................... ....................................................... 10
1-6 ADMET polymerization of 1,9-decadiene................................................... ............10
1-7 The catalytic cycle of ADMET polymerization.....................................................11
1-8 Ring-opening metathesis polymerization of 1,1-dimethyl-l-silacyclopent -3-ene..13
1-9 ROMP of silylsubstituted norbornens................................................................. 13
1-10 ROMP using Casey's carbene complex............................................................... 13
1-11 Polymerization of diallyl- and allyl-butenyl derivatives of disubstituted silanes.... 15
1-12 Polymerization of diallyltetramethyldisiloxane .................................................. 16
1-13 M etathesis of diallyltetramethyldisilane. ........................................... ....................17
1-14 M etathesis of dialkenylsilacyclobutane. ..................................................................17
1-15 Copolymerization of divinyldimethylsilane (32) with 1,9-decadiene (10).............. 18
1-16 Productive and non-productive metathesis. ........................................................19
1-17 Polycarbosilanes via ADM ET................................................................................20
1-18 Polycarbosilanes with the disilacyclobutane rings.................................................21
1-19 Non-productive ADMET polymerization. .......................................................21
1-20 ADMET polymerization of siloxane containing dienes.........................................22
1-21 Controlled diene insertion via ADMET. .............................................................23
1-22 ADMET copolymerization of divinyltetraethoxydisiloxane...................................24
1-23 Telechelic polym ers via AD M ET. ................................... .....................................25
1-24 ADMET polymerization of a,co-telechelic diene macromonomer. .........................25
1-25 Chloro- and dichloro-functionalized polycarbosilanes via ADMET.....................26
1-26 Two routes to polym er m odification...................................................................... 26
1-27 Latent reactive carbosiloxane/carbosilane copolymer. ..........................................27
1-28 The latent reactivity concept to attain diverse behavior of materials....................28
1-29 Catalytic cycle of silylative coupling. ..................................................................29
1-30 SC polycondensation of divinylsubstituted silicon monomers ..............................30
1-31 Starburst oligom er via SC ................................................................................31
2-1 Atmospherically crosslinked ADMET copolymers ..................................... ..34
2-2 Chain-end crosslinking molecules. ............................................................. 37
2-3 Chain-end crosslinking in the polycarbosilanes via ADMET..................................38
2-4 1H NMR spectra of monomer 83.................................................................................40
2-5 Glass transition temperatures for various lengths of polyoxyethylene ....................41
2-6 High elastic material containing increased soft phase and chain-end crosslinker. ..43
3-1 G lycol-based m onom ers..........................................................................................52
3-2 Synthesis of 9sp4gl (93)........................................................................................ 52
3-3 Synthesis of end9sp20Me and 3sp20Me. .............................................................54
3-4 H NMR spectra of monomer 6sp3gl and polymer poly6sp3gl .............................56
3-5 13C NMR spectra of monomer 6sp3gl and polymer poly6sp3gl ............................57
3-6 FT-IR spectra of 6sp3gl and poly6sp3gl. .............................................................58
3-7 DSC traces of unsaturated 3sp microblock copolymers...................................62
3-8 DSC traces of unsaturated 6sp microblock copolymers...........................................62
3-9 DSC traces of unsaturated 9sp microblock copolymers...........................................63
3-10 TG A traces of poly6sp4gl..................................................................................... 64
3-11 1H NMR spectra of 3sp20Me and poly3sp20Me................................................65
3-12 Compositions of "latent reactive" unsaturated copolymers .................................67
3-13 H (A) and 13C (B) NMR spectra of copolymer 38 (147). ....................................68
3-14 Composition and structure of crosslinked materials ..............................................70
3-15 Moisture cured xlink 39 at room temperature. ...................................................71
3-16 DSC curves of poly6sp4gl, copolymer 29 and xlink 29........................................72
3-17 TGA traces in an oxidative atmosphere. .............................................................73
3-18 TGA traces in an inert atmosphere........................................................................74
3-19a Stress/strain curves. ............................................................................. ............... 75
3-19b Stress/strain curves. .......................... ........................................ ....................... 75
3-19c Stress/strain curves........................... ..................................................................... 76
3-20 Stress/strain curves and picture ofthermosets with good elastic properties...........79
4-1 Synthesis of siloxane-based a, o dienes via hydrosilylation ..................................95
4-2 Siloxane-based a, o diene monomers. ....................................................................96
4-3 1H NMR spectra of 6sp4Si (top) and poly6sp4Si (bottom) .....................................98
4-4 3C NMR spectra of 6sp4Si (top) and poly6sp4Si (bottom)...................................99
4-5 FT-IR spectra of 6sp4Si (blue) and poly6sp4Si (red)...........................................100
4-6 DSC traces of polycarbosiloxanes.................................................................102
4-7 Thermogravimetric analysis of polymer (198) in nitrogen and air ......................103
4-8 "Latent reactive" carbosilane/carbosiloxane copolymers. ...................................104
4-9 1H (top) and 13C (bottom) NMR spectra for copolymer 48 (208). ......................105
4-10 Siloxane-based therm osets. ...............................................................................106
4-11 Stress/strain curves of carbosiloxane/carbosilane thermosets ..............................108
A-1 Stress/strain curves of dicomponent thermosets xlink 1-8.....................................116
A-2 Stress/strain curves of dicomponent thermosets xlink 9-16...................................117
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
SYNTHESIS AND CHARACTERIZATION OF LATENT REACTIVE
OLIGO(OXYETHYLENE)/CARBOSILANE AND
CARBOSILOXANE/CARBOSILANE COPOLYMERS VIA ACYCLIC DIENE
METATHESIS POLYMERIZATION (ADMET) FOR THERMOSET MATERIALS
By
Piotr Pawel Mattoka
May 2006
Chair: Kenneth B. Wagener
Major Department: Chemistry
Elastomeric materials can be found in many aspects of our life and the spectrum of
their use is broad. From shoe soles, cable insulators, and automotive parts to advanced
medical equipment, these polymeric materials are based on natural or synthetic rubbers
such as butyl rubber or silicones. Although many of these products have been used for
years, the development of novel synthetic methods for the improvement and creation of
new classes of elastomers are continuously being sought.
Acyclic diene metathesis (ADMET) has been proven to be an effective tool in the
synthesis of latent reactive carbosilane/carbosiloxane copolymers bearing a methoxy-
functional group on each repeat unit. The polymerization results in a linear thermoplastic
polymer. The latent reactive methoxy groups remain inert during polymerization;
however, exposure to moisture triggers hydrolysis and the formation of a chemically
crosslinked thermoset. The thermoset's properties can be modified by varying the ratio
of carbosilane and carbosiloxane repeat units in the final material.
Synthetic modification of the crosslink density and run length in the soft phase in
polycarbosilane/polycarbosiloxane elastomers is discussed. The incorporation of "chain-
end" crosslinks has improved mechanical performance by limiting the formation of
dangling chains. In addition, increasing the run length of the soft phase, to achieve a
better balance between soft phase concentration and crosslink density, generates a
material showing significantly improved mechanical performance.
Various oligo(oxyethylene) and carbosiloxane a, o-dienes with different lengths of
oxyethylene or siloxane units and alkenyl chains have been synthesized to investigate
their influence on the properties of the resultant materials.
Consequently, numerous thermoset materials that show physical properties from
elastic to flexible plastic behavior were prepared and analyzed. The results obtained from
mechanical testing and thermal analysis can be directly related to the structure of the
thermoset.
xvii
CHAPTER 1
ACYCLIC DIENE METATHESIS POLYMERIZATION (ADMET) APPROACH TO
SILICON-CONTAINING MATERIALS
1.1 Introduction
As an inherent part of organometallic chemistry, organosilicon chemistry is one of
the rapidly developing fields of science. Throughout nature, many well-known inorganic
and polymeric compounds contain silicon; however, there are no naturally occurring
organosilanes. Organosilicon chemistry has blossomed since Kipping's pioneering work
at the beginning of the 20th century. Numerous applications1-4 have been reported in the
manufacture of synthetic reagents and intermediates, bioactive molecules, biomedical
materials, electronic devices, specialty plastics, elastomers, coatings, and fibers.
Throughout the years, researchers have focused on silicon-carbon hybrids compounds
rather than purely inorganic structures.
Silicon/carbon hybrid polymeric materials have garnered considerable attention for
their enhanced properties (some that cannot be attained by using polymers based only on
carbon). The uniqueness of the silicon-carbon copolymers comes from their high thermal
stability, good electrical resistance, low surface tension, release and lubrication
properties, high hydrophobicity, low glass transition, and low toxicity for the natural
environment.4 The siloxane unit or the silicon atom can be incorporated into carbon-
based polymer backbone to tailor desired properties of the resultant material.2'' Several
synthetic methods involving anionic ring-opening polymerization, thermal
cyclopolymerization, condensation, and coupling reactions have been applied to
synthesize such hybrids.1,4
The simplest and most widely known organosilicon polymers are based on
silicones or polysiloxanes. This general class, well known for over 100 years, has many
valuable and unique properties. Silicones in particular, like polydimethylsiloxanes
(PDMS), exhibit hydrophobicity, thermal and oxidative stability, electrical resistance,
low glass-transition temperature, low environmental toxicity, gas permeability, and
antiadhesive properties. However, silicone elastomers based on polydimethylsiloxane
(PDMS) alone have relatively weak mechanical properties.2,'46,7 To improve their
mechanical performance (thereby extending their commercial use), researchers have
modified these silicon-based materials. The most economically viable method has been
to incorporate silica-based fillers in the polymer's matrix. For example, fillers have been
shown to impart elongation and overall mechanical properties of the resultant material.
These silicon-based materials (PDMS) can also be enhanced by incorporating carbon
segments resulting in silicon/carbon copolymers, which can be achieved via block or
grafted copolymers. There are numerous examples of these hybrid materials based on
PDMS-block-polysulfone,9 PDMS-block-poly(methyl methacrylate),'0 PDMS-block-
polyamide-6,11 PDMS-block-poly( -methylstyrene),12 PDMS-block-(bisphenol A
polycarbonate),1314 PDMS-block-polystyrene,1 poly(vinyl alcohol)-graft-PDMS,16 and
segmented polyurethane-graft-PDMS.17
Silicon-based hybrids are not the only example of organosilicon copolymers.
Progress in organometallic chemistry and in the chemistry of novel materials led to
production of new silicon-containing materials such as polycarbosilanes and
polycarbosiloxanes. These saturated or unsaturated materials offer many potential
industry applications, and produce new challenges for organosilicon chemists. Recently,
these materials have found uses in gas-separation membranes or as ceramics precursors.
1.2 Silicon Chemistry
Silicon is a second-row element, directly under carbon in the periodic table,
suggesting similar chemistry and reactivity to that of carbon. Silanes when compared to
their analogous carbon compounds (particularly in nucleophilic substitution) show
enhanced reactivity and relative rates. This trend relates to silicon's larger size and lower
electronegativity, and the atom's availability to impart low-energy orbital bonding (Table
1-1).1s
Table 1-1. Comparison of various silicon and carbon bond lengths and energies
Bond Bond Length Bond Energy
(A) (kJ/mol)
Si-C 1.90 318
C-C 1.54 334
Si-O 1.63 531
C-O 1.44 340
Si-Cl 2.11 471
C-C1 1.81 335
Correlations to silicon's reactivity or stability can be made and explained using
Table 1-1. For example, silicon-halogen bonds are easily cleaved by nucleophiles while
resistant to homolytic fission. These observation are realized by the large
electronegativity differences and high polarizability of both atoms producing highly ionic
bond in character.19 Also, silicon-oxygen bonds (siloxanes and silanols) are not
analogous to carbon-oxygen sigma bonds (ethers and alcohols). The lone pairs on
oxygen can overlap with an empty d-orbital on silicon, resulting in a highly polarized
bond. Recently, it was shown that o* orbitals can participate in back-bonding because of
their lower energy versus empty d-orbitals.1 This interaction justifies the fact that
silanols are stronger proton donors than aliphatic alcohols and that the decreased basicity
of oxygen is observed in the following order: C-O-C > Si-O-C > Si-O-Si.
1.3 Olefin Metathesis
Olefin metathesis has been investigated for over 50 years as one of the most
motivating discoveries in organometallic chemistry, widely used in both academic and
industrial settings.2022 In 2005, professors Yves Chauvin, Robert H. Grubbs, and
Richard R. Schrock were awarded the Nobel Prize in Chemistry "for the development of
the metathesis method in organic synthesis." The name, proposed by Calderon, Chen,
and Scott23 in 1967, originates from the Greek word metathesiss" meaning changing or
exchanging positions. The metathesis process involves reaction of two carbon-carbon
double bonds that yields two new olefins (Figure 1-1) and is described as an
intramolecular and intermolecular exchange of substituents on 1, 2 di-, tri-, or
tetrafunctional double bonds and "can be compared to a dance in which the couples
change partners." 24 This process can be divided into four categories:
* Simple exchange reactions
> Metathesis reaction involving two equal olefins
>Cross-metathesis (CM) reaction involving two different olefins
* Metathesis polymerizations
> Ring-opening metathesis polymerization (ROMP)
> Acyclic diene metathesis polymerization (ADMET)
* Ring-closing metathesis (RCM)
* Metathesis degradation of unsaturated polymers
a)
R R
b, c)
RCM
d)
Degradative R'
Metathesis of
SR' Unsaturated
Polymers +
Figure 1-1. Types of olefin metathesis: a) simple exchange reactions; b, c) metathesis
polymerization and ring-closing metathesis; c) metathesis degradation.
Many applications benefit from olefin metathesis, making this mild method for
creating a new carbon-carbon double bond a convenient tool for synthesis of small
molecules and for macromolecular synthesis. While the pharmaceutical industry has
embraced ring-closing metathesis to synthesize complex cyclic structures, acyclic diene
metathesis and ring-opening polymerizations are widely used by polymer chemists to
produce a vast number of functionalized hydrocarbon polymers and silicon-carbon based
hybrids.
In the next sections some background information on the ADMET approach to
silicon-containing polymers and copolymers catalyzed by "well-defined" catalyst systems
is described. Complementary method to ADMET polymerization such as silylative
coupling (SC) polycondensation of divinyl organosilicon derivatives (which shows low
metathesis activity) is also discussed.
1.3.1 Brief History of Olefin Metathesis
The first research on olefin metathesis and its industrial application that led to
publication was by Banks and Bailey25 from Philips Petroleum in 1964. The observed
reaction was called olefin disproportionate and was performed in the presence of the
heterogeneous catalyst. The publication, however, was foreshadowed by Eleuterio's26
and Peters'27 patent disclosures few years earlier.
Several mechanisms were initially proposed to account for all observed exchange
reaction products. The process consistently involves subsequential bond breaking and
reformation after a transition metal 2+2 cycloaddition/retroaddition mechanism. The
Chauvin mechanism (Figure 1-2) was initially ignored, only to become the most widely
accepted mechanism with both metalla-cycle and metalla-carbene intermediates.28
R R' R R' R R'
M] [M] [M]
R" R" R"
Figure 1-2. Metallacyclobutane mechanism.
Olefin metathesis is a catalytically initiated reaction with numerous active catalytic
systems. These catalysts, either homogeneous or heterogeneous, contain transition
metals such as molybdenum, tungsten, rhenium, tantalum, ruthenium, rhodium, or
titanium.20 Common to industrial applications, the catalyst systems are typically
heterogeneous versions of either Mo, W, or Rh oxides or carbonyls supported on highly
developed A1203 or Si02. The presence of co-catalyst is often required in these systems,
usually a Lewis acid (a main group tri- or tetravalent metal). On the other hand,
homogeneous systems (coordination complexes) can be active with or without the
presence of the co-catalyst.
In the early days of olefin metathesis, organometallic chemists raced to synthesize
metal carbene complexes that might improve understanding of the metathesis process.29-
33 Further investigations during the mid-1980s led Schrock and coworkers to the
development of neutral carbene complexes based on tungsten34-36 and molybdenum37-39
(Figure 1-3). These catalysts were active in metathesis transformations and were called
"well-defined" because the catalytically active species were easy too identify by
spectroscopic methods. Because of the oxophyllic nature and low-tolerance to
polar-functional groups, there were some limitations in the utility of these catalysts.
Efforts were made to develop catalyst systems based on late transition metals that in
complexes form catalysts more tolerant to air, moisture and functional groups.
H3C N CH3 H3C N CH3
F3C O-W-O- CF3 F3C -0Mo-Ok CF3
F3C CHR CF3 F3C CHR" CF3
R'= t-Bu; R"= -C(Me)2Ph
[W] [Mo]
(1) (2)
PCy Ph PCy3Ph
ClRu j Ph Ru C1=\ Y
PCyz
PCy3Ph
[Ru] [Ru]* [Ru]**
(3) (4) (5)
Figure 1-3. Well-defined metathesis catalysts.
In 1992 Grubbs and coworkers40 introduced the first well-defined ruthenium based
catalyst [Ru] (3) (Figure 1-3), where their increased stability allowed for manipulations
outside an inert atmosphere. The ligation sphere of the first-generation ruthenium-based
catalyst [Ru]* (4) was soon modified by substitution of one phosphine ligand with a
N-heterocyclic carbene ligand.41'42 In some cases this change achieved catalytic activity
comparable to one of the molybdenum Schrock's catalyst41 [Mo] (2), and a higher
tolerance of functional groups was observed than in the first-generation Grubbs' catalyst
[Ru]* (4).42 Structural isomerization 43,44 of the olefin can accompany olefin metathesis,
an event that influences the synthesis of the exact structures through RCM and ADMET.
Despite of vast number of "well-defined" catalysts that have been synthesized, none of
them have proven to be applicable for all substrates or transformations.
1.3.2 Introduction to ADMET Polymerization
The discoveries and developments in the field of olefin metathesis and
organometallic chemistry have made olefin metathesis a convenient synthetic tool for the
synthesis of functionalized and non-functionalized polyolefins.21 As previously
mentioned there are two modes of olefin metathesis polymerization: ring-opening
metathesis polymerization (ROMP)21 and acyclic diene metathesis polymerization
(ADMET)21 (Figure 1-4).
ADMET
cat.
-CH4
(6) (7)
ROMP
cat.
(8) (9)
Figure 1-4. Acyclic diene metathesis (ADMET) and ring-opening metathesis (ROMP)
polymerizations.
ROMP falls into chain-addition type polymerization category. It is an exothermic
reaction, thermodynamically driven by the release of ring strain in a cyclic olefin
monomer. ADMET is step-growth condensation type polymerization and thermally
neutral. It is important to understand requirements and consequences for these
classifications.
Acyclic diene metathesis is a step-growth condensation type polymerization that
normally employs an a,wo-diene in conjunction with a metathesis catalyst to produce high
molecular unsaturated polymer. The reaction is driven by removal of a small condensate,
usually ethylene. ADMET follows the same rules as any other polycondensation
reaction, meaning that high conversions (>99%) with no side reactions are necessary to
produce high molecular weight polymers.45 This reaction, as in the case any
condensation polymerization, is an equilibrium process, which proceeds in a stepwise
fashion to form dimer, trimer, tetramer, and so on to obtain high molecular polymer. The
size of the polymer increases continuously with time (conversion) at relatively slow rate.
Growth of molecular weight is described the Carother's equation that relates the number-
average degree of polymerization Xn (also known as the degree of polymerization, DP)
to the extent of reaction p or conversion of the diene monomer.
1
Xn = -
l-p
Figure 1-5 shows a typical plot of theoretical X" versus conversion of the
monomer in any condensation type polymerization. High molecular weight polymer is
obtained at the end of polymerization when the conversion of monomer is larger than that
of 99%. Thus, a highly active catalytic system is required throughout the polymerization;
most catalysts however, show higher activity towards terminal olefins for steric hindrance
reasons, and so a,oo-diene monomers are often used in acyclic diene metathesis
polymerization.
200
S150
o 100
I-1
50
0 20 40 60 80 100
% Conversion
Figure 1-5. Theoretical degree of polymerization vs. conversion in condensation type
polymerization.
Early investigations of polymerization of 1,5-hexadine (6) and 1,9-decadine (10)
(Figure 6), by Wagener and co-workers in the mid-1980s, demonstrated that "classical"
catalysts such as WC16/EtAlC12 can lead to both cationic vinyl addition chemistry and in
addition to polycondensation chemistry,20'46 thereby defining the need for active and
well-defined catalyst for metathesis. Fortunately, in 1986, Schrock34 and co-workers
successfully prepared the first well-defined, single site catalyst, a breakthrough in this
field. This breakthrough led to more stable ruthenium based catalytic systems discovered
by Grubbs.
cat.
'' CA- C2H4
(10) (11)
Figure 1-6. ADMET polymerization of 1,9-decadiene.
Like many other polycondensation reactions, ADMET polymerization is best
performed in the bulk (neat) to maximize the molar concentration of monomer (terminal
olefins) and to promote a shift in the equilibrium of the reaction from monomer towards
unsaturated polymer formation. Bulk polymerization also minimizes undesired
cyclization, a well known concurrent reaction in polycondensation chemistry.45
Moreover, high vacuum is applied to remove the condensate ethylene, which further
shifts reaction towards polymer formation.
[n1-,R R
H2C=CH LnM R
LnMn- Ln R- R
I I
R LnM'CH2 'R
Figure 1-7. The catalytic cycle of ADMET polymerization.
The mechanism of the ADMET polymerization cycle, Figure 1-7, has been well
documented.46'47 Each step in this mechanism is in true chemical equilibrium, and the
ultimate driving force for the reaction is removal of ethylene. The catalytic cycle starts
with coordination of an olefin, followed by formation of a metallacyclobutane
intermediate and productive cleavage that leads to the formation of the metathesis active
alkylidene complex. These subsequent reactions produce a methylidene complex
followed by the coordination of another monomer. Productive cleavage of another
metallacyclobutane results in release of one ethylene molecule per each catalytic cycle.
The developments in the field of well-defined catalysts along with the clean
mechanism of ADMET have become instrumental in the synthesis of novel polymers
with controlled architectures. The design starts with the preparation of a,c-diene bearing
desired type of functionality or specific type of branching. The resultant polymer retains
this functionality and a specific backbone sequence. ADMET has produced unsaturated
polymers where the functionalities have been ethers,48 esters,49 carbonates,50 sulfides,51
and many more functionalities.52'53 Consequently, ADMET bypass typical synthetic
obstacles in the synthesis of polymers and copolymers, and allows creating polymer
structures that cannot be prepared via other routes.
1.4 Olefin Metathesis Chemistry of Compounds Containing Silicon
In the early seventies, research on metathesis was predominantly focused on
producing materials from simple silicon functionalized olefins. In 1971, Friedman
described the first example of these materials by the metathesis of monoalkenylsilanes in
the presence of MoO3/A1203.54 In addition to these simple materials, a variety of
organosilicon compounds can be produced using metathesis and can be divided into three
categories based on choice of polymerization method and structure of monomer:
* ROMP of unsaturated silacycloalkenes
* ROMP of silylsubstituted cycloalkenes
* ADMET of a,o-dienes containing silicon
First, the ROMP of the unsaturated silacycloalkenes leads to polymers containing
silicon atoms incorporated in the carbon backbone, as exemplified in Figure 8. The
metathesis of 1,1- dimethyl- 1-silacyclopent- 3-ene (12) proceeds in the presence of
tungsten55 or alumina-rhenium56 and offers a broad scope for all silacycloalkenes, except
for the thermodynamically stable silacyclohexene derivatives.
CH3
n cat IN
Si Si
H3C 'CH3 CH3 Jn
(12) (13)
Figure 1-8. Ring-opening metathesis polymerization of 1,1-dimethyl- 1-silacyclopent
-3-ene.
Second, the ROMP of silylsubstituted olefins leads to unsaturated polymers
(Figure 1-9) containing pendant silyl groups used to modify the properties of the
polymer's backbone. One such example is the ROMP of silylsubstituted norbornenes in
the presence of either a homogeneous tungsten catalyst or a heterogeneous alumina-
rhenium catalyst.57
I R cat.
R
(12) (13)
R = SiMe3, SiMe2CH2SiMe3, SiC13, Si(OEt)3
Figure 1-9. ROMP of silylsubstituted norbornens.
An interesting example of ROMP is the reaction of 1-(trimethylsilyl)cyclobut- -ene
(16) in the presence of Casey's carbene complex (Figure 1-10).58
SiMe3 cSiMe
i-n
(16) (17)
cat. = (CO)5W=CPh2
Figure 1-10. ROMP using Casey's carbene complex.
The resultant polymer shows a repeating head-to-tail structure containing only the cis
double bond configuration. Moreover, this trimethylsilyl groups can be easily converted
into different functionality making them a surrogate for the functional group before
metathesis.
Third, the ADMET polymerization of a,co-dienes containing silicon leads to
formation of unsaturated polymers which, depending on the structure of monomer, can
either include silicon functionality incorporated into polymer backbone or as a pendant
groups along the polymer chain. Contrary to ROMP which can be limited to the use of
strained cyclic monomers, ADMET offers wide spectrum of polymer structures. Thus,
utility of ADMET polymerization will be discussed in the next section.
1.4.1 Acyclic Diene Oligomerization and Polymerization
Acyclic diene metathesis polymerization of silicon-containing compounds is the
most universal method amenable for the synthesis of unsaturated organosilicon oligomers
and polymers. The clean and efficient chemistry of ADMET combine desired material
properties of both carbon and silicon based polymers in the random linear copolymer.
Moreover, the high selectivity of olefin metathesis facilitates an introduction of a wider
range of possible "functional handles" to manipulate the overall material properties.
Metathesis of dialkenylsilicon derivatives can be achieved either by an
intramolecular or an intermolecular approach. The intermolecular ADMET reaction
between a,o-dienes (with exceptions of divinyl derivatives) produces linear, unsaturated
polycarbosilane or polycarbosiloxane oligomers or polymers. Unsaturated heterocyclic
organosilicons can also be produced by an intramolecular reaction if preferred. The
mechanism of reaction is strictly dependent on the structure of the initial a,o-diene and
substituents on silicon; note that in some cases a six membered ring compound will be
preferentially formed as the most thermodynamically stable product. Since all possible
pathways are in equilibrium, both the seven and five membered rings are rarely observed.
Product formation can be influenced to some extent by the presence or absence of solvent
in the reaction environment.
As mentioned, the divinyl derivatives of silicon compounds do not undergo
homometathesis, mainly due to the steric and electronic effects of two silyl groups
present on the adjacent carbon atoms in the intermediate metallacyclobutane.35 However,
while studying the hydrosilative activity of simple Ru(II) and Ru(III) complexes,
Marciniec and coworkers59-62 have shown that catalytic metathesis of alkoxyvinylsilanes
is a competing reaction. However, soon after was found that this process does not
operate through metallacyclobutane mechanism. Thus, application of this silylative
coupling (SC) polycondensation in the formation of macromolecular structures
containing silicon will be described in the later in separate section.
1.4.2 ADMET Catalyzed by "Ill-Defined" Catalytic Systems.
Ill-defined catalyst systems do not easily lend themselves to metathesis
polymerization chemistry, thus the practical utility and the scope of olefin metathesis
have been limited toward functional groups, such the ones based on silicon.
R ,R2 [Re] R R R, R2
S1 +
S -C2H4,35 0C I<- n
F2 XR
(18) (19) (20)
R1 RI
+ 4 ySi.,,. -^ A
R2 2 X
(21)
[Re] = Re207/A1203 + Sn(or Pb)R4
R1 = R2 = Me, Et or R1 = Me, R2 = Ph where: n = 1; R = R2 = Me where n = 2; x = 1-4
Figure 1-11. Polymerization of diallyl- and allyl-butenyl derivatives of disubstituted
silanes.
The first information regarding metathesis and co-metathesis reactions in the
presence of these aforementioned catalytic systems were published in early eighties. The
reactions reported by Vdovin et al.63 utilized the heterogeneous Re207/A1203 catalyst
along with a tetraalkyltin or tetraalkyllead co-catalysts. Vdovin's and Finkelshtein's6366
research interests were focused on polymerization of diallyl and the allyl-butenyl
derivatives of methyl, ethyl, and phenyl substituted silanes, as presented in Figure 1-11.
Predominantly, cyclic products (20) were formed along with some traces of oligomeric
compounds.
The attempted polymerization of diallyltetramethyldisiloxane (22)67, Figure 1-12,
performed in the presence of Re207/Al203+Sn(Pb)R4, was accompanied by partial
cleavage of the Si-Callyl and Si-O bonds, resulting in dissproportionation of the Si-O
bonds and formation ofpolysiloxane chains.
Me Me Me 1 Me
cat. I
Si-O-Si^ _HSi-O -Si +Me S
Me Me 4Me JMe Met O Me
(22) (23) (24)
SMe Me
+ Si-O-Si
Me Me
n
(25)
Figure 1-12. Polymerization of diallyltetramethyldisiloxane.
Similarly to what is observed during attempted polymerization of diallyl and the
allyl-butenyl derivatives of disubstituted silanes, polymerization of
diallyltetramethyldisilane (26) resulted in formation of both linear and cyclic products,
where the former predominates (Figure 1-13). In this case, cyclized product is formed
over the linear structure, what one can be attributed to the higher stability of the six
membered disilane product.
Me Me Me Me
I I I Me S-Me
f Si-Si t Si + iMe
Me Me Me Me
(26) (27) (28)
Figure 1-13. Metathesis of diallyltetramethyldisilane.
Many attempts have been taken to introduce a secondary functionality ("latent
reactive sites") to silicon-based monomer other than those used by ADMET. An
interesting example shows ADMET polymerization of dialkenysilanes containing the
highly reactive silacyclobutane 68,69 (Figure 1-14). Hence, Finkelshtein observed that
ADMET chemistry occurs in the same manner as for any other dialkenylsilanes, while
maintaining the highly reactive silacyclobutane in every repeat unit. The resultant
oligomer's expansion, when heated, cleavage of Si-C bonds in the four-membered ring
yielding a crosslinked, waxy material.
cat + + /Qn
-m
.C2H4"-
(29) (30) (31)
n=l,2
m=l-3
Figure 1-14. Metathesis of dialkenylsilacyclobutane.
In spite of the lack of metathetical conversion for divinyl organosilicon derivatives,
Kawai and coworkers70'71 have recently reported successful synthesis of O-2t conjugated
oligo(silylene-vinylene)s from tetravinylsilanes and divinylsilanes in the presence of
Re203/A1203 promoted with SnBu4. These polymers are important class of partly
conjugated optical materials, which based on the ab initio molecular calculation, should
exhibit high transparency and relatively high second and third hyperpolarizability. It was
found that the addition oftetrabutyltin is essential for ADMET to occur. The reactivity
of Schrock's and Grubbs' catalysts was also examined; however the catalysts did not
succeeded, for reasons described previously. All products were linear oligomers, and
cyclic products were not detected.
1.4.3 ADMET Catalyzed by "Well-Defined" Catalytic Systems
In the early 90's Wagener and co-workers72-75 opened a new chapter in the
synthesis of unsaturated polymers and copolymers containing silicon by preparing
polycarbosilanes and polycarbosiloxanes with acyclic diene metathesis polymerization
using well-defined ADMET catalysts (Figure 1-2).
1.4.3.1 Polycarbosilanes
Polycarbosilanes are an interesting class of polymers due to the presence of
covalent bond between silicon and carbon atoms in the backbone. Potential applications
of these materials in the broad range of fields have triggered considerable interest in their
-2
preparation.2
r l] I
Si ~+ #- % L~4. J Si
CH3 6 -C2H4
CH3 LCH3 m -- m
(32) (10) (33)
Figure 1-15. Copolymerization of divinyldimethylsilane (32) with 1,9-decadiene (10).
The first silane diene, dimethyldivinylsilane (32),72 subjected to ADMET
polymerization did not polymerize in the presence of the tungsten based Schrock's
catalyst [W] (1), where the lack of productive metathesis was attributed to the steric
influence of the trisubstituted silicon atom adjacent to the double bond. Similar results
were obtained by Schrock and coworkers36 in the attempted metathesis of
vinyltrimethylsilane. The steric influence was further demonstrated via copolymerization
of divinylsilane (32) with 1,9-decadiene (10) where the divinyl repeat unit was found
to insert in a singular manner along the backbone, Figure 15.
Metathesis
CH CH13
CH3
LnW=CH2 + -'iHi ,i
CH I CH3
L (35) CH3
LW Si (36)
H3C, CH3
H3C -Si LnW IH3 CH3
S L + Si
(34) I
CH3 CH3
(37) (32)
Co-Metathesis
CH
SLnW=CH2 + 1 Si"
LnW H i (39) CH3
1 (40)
SCH3
R
CH3
(38) LnW R +
(41) CH3
(32)
Figure 1-16. Productive and non-productive metathesis.
Finkelshtein 66 provides a different explanation for the lack of reactivity of
divinyldimethylsubstituted silanes in the ADMET polymerization. His rationalization is
based on the formation of the highly stable silylated carbene complex, Figure 1-16, and
its reactivity under copolymerization conditions. During the homopolymerization of
divinylsilanes, formation of the metathesis active methylidene complex (35) competes
with an inactive silylated carbene (37). The reaction goes through the
thermodynamically more favorable process that leads to formation of the stronger
stabilized silylated carbene, and thus the metathesis product is not observed. In the case
of copolymerization, the stability of both carbene complexes does not differ significantly
and two pathways are possible. The productive copolymer formation is realized along
with the unproductive formation of the starting divinylsilanes.
Further investigations have shown that monomers with the methylene spacers
incorporated between the silicon atom and the olefin are able to undergo productive
metathesis polymerization (Figure 1-17) 72. Successful ADMET polymerization
produces unsaturated polymers containing predominately (80%) trans double bonds.
Exceptions are observed while polymerizing stiffer carbon segments; for example, the
polymerization of 1,4-bis-(dimethylallylsilyl)benzene results in a 47% cis configuration
along the unsaturated backbone. 72
CH3 CH3
I [W] {
CH3 4 .CH3
(42) (43)
CH3 CH3
CH3 ]. CH3
H CH3 -C2H4 CHzi I
CH, CH3 n
(44) (45)
CH CH3 CH3 CH3
CH3 CH3 _CH3 CH3 .
(46) (47)
Figure 1-17. Polycarbosilanes via ADMET.
As shown in Figure 1-14 Finkelshtein68'69 has successfully polymerized a,co-diene
bearing highly reactive silacyclobutane using ADMET methodology. Similar study using
"well-defined" catalytic system was conducted by Interrante and coworkers.76 The
carbosilane polymers and copolymers were synthesized with the disilacyclobutane rings
imbedded into the copolymer's backbone (Figure 1-18).
Me Me Me
Si Si [Ru]** Si SiTPA SSi
Me -CH2CH2 Me 2 Tn T Me
(48) (49) (50)
TPA: tripropylamine
TSH: p-toluenesulfonydrizide
Figure 1-18. Polycarbosilanes with the disilacyclobutane rings.
Unsaturated materials were hydrogenated using methodology developed by Hahn77
which does not require the use of high-pressure H2, and the disilacyclobutane rings were
thermally induced to open resulting in a transparent, crosslinked solid material.
1.4.3.2 Polycarbosiloxanes
A similar to carbosilanes, investigation of ADMET has shown that siloxanes are
also polymerizable with the Schrock's molybdenum catalyst [Mo] (2) (Figure 1-19 and
Figure 1-20). Polymerization leads to formation of unsaturated polycarbosiloxanes.73'74
CH3 CH3
3 3 [Mo]
Si-O-Si no metathesis
I I
CH3 CH3
(51)
CH3 CH3 H3C CH3
e3 e3 [Mo]i SiN. -0
Si-O-Si o H3 C-S CH3
I I -CH4
CH3 CH3 -
(52) (53)
Figure 1-19. Non-productive ADMET polymerization.
As in the case of divinyldimethylsilane, sterics has a predominant effect on both the
productive and non-productive metathesis rates of vinylcarbosiloxane (51) (Figure 1-19).
On the other hand, the reaction of 1,3-diallyl-l,1,3,3-tetramethyldisiloxane (52) where
siloxane group is "spaced out" by one methylene unit, leads to the ring formation (53)78
and essentially quantitative conversion for cyclization of diallyldisiloxane was observed.
This exclusive and fast cyclization with no oligomerization indicates great
thermodynamic stability of this seven membered ring (Figure 1-19). Thus, an increased
flexibility of siloxanes requires longer distances between the olefin and silicon to produce
ADMET linear polymers, as shown in Figure 1-20.
CH3 CH3 CH3 rCH3 CH3 CH3
I I I [M o] I I I
,-Si-O-Si-O-Si [S. IN Si-O-Si-O-Si1
CH3 CH3 CH3 4 3 CH3 CH3
(54) (55)
CH3 CH3 CH3 CH3 CH3 CH3
I |I 3 CMo] iO "S N
Si-O-Si-O-Si [ i-O- i-O-i
SCH3 CH3 CH3"3 -C2H4 CH3 CH3 CH3 .n
(56) (57)
CH3 CH-a CH3 CH3
Si-O-S i---r ---i-O-Si %
CH3 CH3 CH3 CH3
(58) [Mo]
CMH3 CH32 CH3 CHi
(59)
Figure 1-20. ADMET polymerization of siloxane containing dienes.
Another noteworthy example of ADMET copolymerization is an insertion of an
aromatic diene repeat unit into an unsaturated polycarbosiloxane.74 Smith et al. have
shown that slow addition of small diene monomer to unsaturated oligomer offers more
than statistical control of copolymer by kinetically inhibiting homometathesis of the
added diene (Figure 1-21).
As previously shown in Figure 1-19 divinyl siloxane derivatives are reluctant to
homometathesis; however, like divinyl silane they undergo copolymerization. The first
successful copolymerization of divinyltetraethoxydisiloxane (63) with 1,9-decadiene (10)
in the presence of the Grubbs' catalyst [Ru]** (5) was reported by Malecka et al. Figure
1-22.79 No consecutive linkage of divinyltetraethoxydisiloxane was observed.
Surprisingly, in the similar experiment where divinyltetraethoxydisiloxane (63) was
substituted with divinyltetramethyldisiloxane (51), copolymerization failed and no
polymeric or oligomeric products were detected. The lack of metathetical activity of
divinyltetramethyldisiloxane (51) in copolymerization was attributed to the
decomposition of the Grubbs' catalyst [Ru] ** (5) due to the p-transfer of silyl group to
ruthenium in the metallacyclobutane, previously observed by Pietraszuk et al.80
CH3 CH3 CH3
I I I CH3 CH3 CH
Si-O-Si-O-Si I I I
I I I Si-O-Si-O-Si
CH3 CH3 CH 3 I I I
CH3 CH3 CH3
(60)
Toluene
[w] / \/ \/
(61)
Si-O-Si-O-Si I I I C /
CH3 CH3 CH3 I ,C
SCH3 CH3 CH3 Y
(62)
Figure 1-21. Controlled diene insertion via ADMET.
Marciniec and Majchrzak81 have subjected divinyltetradisiloxane to ADMET
copolymerization conditions in the presence of 1,4-divinylbenzene (65) (Figure 1-22).
Analysis of this copolymer showed the structure to be perfectly linear and trans(stereo)-
regular.
OEt OEt OEt OEt
I I ,, [Ru]
Si-O-Si- + i-O- '
6I CH2CH2
OEt OEt OEt OEt
(63) (10) (64)
OEt OEt [Ru]* OEt OEt /
i"Si-O-sOSi + / -- -
I I/1 CH2CH2 I i
OEt OEt OEt OEt
(63) (65) (66)
Figure 1-22. ADMET copolymerization of divinyltetraethoxydisiloxane.
1.4.3.3 Telechelic oligomers
Telechelic polymers are macromolecules which contain one or more functional end
groups that can form bonds with another molecule. They offer potential use e.g. in the
formulation of model networks82 or in the preparation of thermoplastic elastomers.83
Preparation of this type of macromolecule is not trivial, especially those possessing
functionality of 2.0 essential for the stoichiometric balance in the polycondensation
reactions.84
ADMET polymerization is a convenient route to prepare an a,o-telechelic
polyoctenamers with the functionality close to 2.0.85'86 1,9-decadiene (10) was
polymerized in the presence of the Schrock's catalyst [Mo] (2) using dimers (71) of
5-hexenyl-methoxydimethylsilane or 5-hexenyl-chlorodimethylsilane as chain limiters
Figure 1-23. As expected, the number average molecular weights for these telechelic
oligomers were proportional to the ratio of the monomer to the chain limiter. Further,
these highly reactive macromonomers were exploited to synthesize ABA block
copolymer of polydimethylsiloxane-polyoctenamer-polydimethylsiloxane.
CH CH3 CH3
X-S CH + [Ru]* or [Mo] X-Si Si-
i+ # i3S -CH2CH2 CH3L3 JY CH3
6H3 IiH3 kIn3
CH3
(71) (10) (72)
where: X= -Cl, -OCH3
Figure 1-23. Telechelic polymers via ADMET.
Smith and Wagener73 synthesized a,o-oligodiene via hydrosilation of
1,4-pentadiene (68) with 1,1,3,3-tetramethyldisiloxane (67), which then was exposed to
ADMET conditions as shown in Figure 1-24. This nonvolatile, quick flowing liquid is an
ideal monomer for ADMET polymerization due to the fact that continuous high vacuum
can be applied during polymerization.
CH3 CH3
H-Si-O-Si-H
I I
CH3 CH3
(67)
+ s
I IN
X
(69)
[Mo]
- CH2CH2
i
[Pt] = Karstedt's catalyst
Figure 1-24. ADMET polymerization of a,o-telechelic diene macromonomer.
1.4.3.4 Functionalized polycarbosilanes
The polycarbosilanes presented in Section 1.3.1.2.1 possess a simple silicon-carbon
backbone. In order to introduce functionality along the polymer's backbone, Cummings
et al. 8788 synthesized chloro- and dichlorosilane a,co-dienes (73), that undergo successful
ADMET polymerization (Figure 1-25) in the presence of the Schrock's catalyst [Mo] (2).
Cl Cl
I I
S ^ [Mo] 4 *Si N >
I CHzCH I ^ '"x
SJ-n
(73) (74)
where: R= Cl, CH3
Figure 1-25. Chloro- and dichloro-functionalized polycarbosilanes via ADMET.
These chlorofunctionalized thermoplastic materials were exposed to different
nucleophilic organolithium salts yielding polymers with various physical properties
depending on the bulkiness of the nucleophiles. This concept was previously used by
Allcock4 to tailor desired properties ofpolyphosphazanes. Allcock has also showed that
cyclolinear phosphazane containing polymers that can be obtained via ADMET
polymerization.89,90
Nuc
Si
I N
Nuc [Cl
N I
X Nucx Cl x
Cl
Figure 1-2. Two routes to polymer modification.
Figure 1-26. Two routes to polymer modification.
Subsequently, Church et al.9192 demonstrated the utility and flexibility of ADMET
in the one-pot synthesis of alkoxy-functionalized carbosilanes. They demonstrated that
functionalization can be facilitated either via macromolecular substitution or substitution
on the a,o-diene monomer (Figure 1-26). As noted above both pendant groups on the
carbosilane polymer backbone have a significant effect on the physical properties of
resultant polymer. These alkoxy-functionalized polymers can be further crosslinked to
yield a polymer network.
1.4.3.5 Latent reactive elastomers
The functionalized polycarbosilanes via ADMET described above offer a new
approach to the synthesis of materials exhibiting variable physical properties. This
research has led to a new type of synthesis involving "latent reactive"
carbosilane/carbosiloxane copolymers by Brzezinska75 (Figure 1-27), illustrating the
random incorporation of "latent reactivity" throughout polycarbosiloxane backbone in the
form of the alkoxy-substituted carbosilane.
S CH3 CH3 CH3
OCH I I
CH OCH3 CH3 CH3
OCH3
(75) (76)
Latent Reactive Carbosilane HARD [Mo] Carbosiloxane SOFT
OCH3 CH3 CH3
OCH3 I I I
OCH OCH3 X CH 3 CH3 y
OCH3
(77)
Figure 1-27. Latent reactive carbosiloxane/carbosilane copolymer.
It is well known that the methyl silyl ether bond, which inert during ADMET
chemistry, easily hydrolyzes in the presence of moisture and condenses to form stable
siloxane linkage.7 Consequently, the physical properties of ADMET
polycarbosilane/polycarbosiloxane copolymers can be tuned by selecting the appropriate
molar concentrations of each monomer. Figure 1-28 illustrates this modular approach
and shows its extensive application in the generation a large number of materials
exhibiting physical properties from elastic materials to plastics.
OCH3
S Latent Reactive Carbosilane HarC + VVSof
OCH3
Low (10%) High (> 90%)
Hard Segment Hard Segment
Content Equal RatioE Content
(50/50)
|m0--- pwvwwF --- I JA
Hard Soft Hard Soft Hard Soft
Elastomer Intermediate Properties Plastic
Figure 1-28. The latent reactivity concept to attain diverse behavior of materials.
1.5 Silylative Coupling (SC) Polycondensation
Previously mentioned, ADMET polymerization is a universal method to obtain
unsaturated organosilicon polymers. A vast number of a,co-dienes containing silicon
functionality tend to be active in this process, when catalyzed by well-defined carbene
complexes. On the other hand divinyl organosilicon compounds are an exception, since
they remain inert under conditions of acyclic diene metathesis polymerization. It
happens that the sterics of silyl groups stimulate a nonproductive cleavage of
disilylmetallacyclobutane.35
The Marciniec research group has found an important way to solve this problem.
Divergent to well-defined olefin metathesis catalysts, numerous ruthenium and rhodium
complexes that contain or generate in situ metal-hydrogen or metal-silicon bonds catalyze
intermolecular polycondensation of divinyldiorganosilicon derivatives.
SiR3 SiR3 [Ru]-H
SiR3 R3 Si
Ru]S SiR [Ru] SiR3 [Ru] SiR
H H H SiR3 H
SiR3
]Ru]-SiR3
[Ru] = RuHC1(CO)(PPh3)n
Figure 1-29. Catalytic cycle of silylative coupling.
At first, Marciniec et al.60 proposed a mechanism in which carbene is formed by
rearrangement of a metal hydride complex. Soon after it was found that catalysis
operates via non-metallacarbene mechanism;93'94 where series of o-complexes are formed
via insertion-elimination, as depicted in Figure 1-29. These transformations were termed
silylative coupling polycondensation (SC) (or as referred by Finkelshtein66 may as well
be called Marciniec's metathesis). Complexation of the olefin to electrophilic metal
center is followed by metal insertion into the C-H bond. Elimination of ethylene yields
metal-silyl complex that coordinates another vinylsilane, which inserts into M-Si bond
followed by reductive elimination of homometathesis product while regenerating hydride
complex. Formation of the 1,1-bis(silyl)ethene as product along with cis/trans 1,2-
bis(silyl)ethenes distinguish this reaction from olefin metathesis where only former two
are formed.
Early attempts at polymer synthesis via SC polycondensation of divinylsilanes and
bisvinylsiloxanes (Figure 1-30) catalyzed by RuCl2(PPh3)3 and RuHCl(CO)(PPh3)3
yielded polymers in moderate conversions, too low for step-growth polymer
chemistry.
S cat. Si
-CHCH2 Si Si
n m
where:
Si = silane, siloxane
cat.= RuCl2(PPh3)3 or RuHCl(CO)(PPh3)3
Figure 1-30. SC polycondensation of divinylsubstituted silicon monomers.
While rhodium-based catalysts were used two monomers of
divinyltetramethyldisiloxane have undergone ring closure which subjected to X-ray
spectroscopy revealed a boat conformation.97
Further studies in the synthesis of polycarbosilanes and polycarbosiloxanes from
divinyl organosilicon derivatives were conducted by Malecka et al.9-'100 who
demonstrated the dependence of catalyst identity on polymer's morphology. SC
polycondensation of divinyl organosilicon compounds catalyzed by ruthenium-phosphine
complexes, such as shown in Figure 1-30, yielded linear polycarbosilanes or
polycarbosiloxanes. Both trans-1,2- and gem-1,1- fragments were present in the
polymer's backbone, while highly stereoselective trans-tactic structures were formed
when [RuCl2(CO)3]2 was used. Moderate molecular weights of polymers were obtained
as well.
Majchrzak et al.81,01-103 have successfully synthesized numerous phenylene-
silylene-vinylene and siloxylene-vinylene-phenylene polymers and copolymers via SC
polycondensation, producing highly stereo- and regioselective structures with values of
Mw from 13,100 to 34,800 g/mol.103 Alternate ruthenium complexes were screened in
order to find a catalyst that promotes these regular structures. Furthermore, they have
extended the utility of SC polycondensation to the synthesis of organosilicon starburst
materials that contains the silicon-bridged x-conjugated structure,104'105 as shown in
Figure 1-31.
Figure 1-31. Starburst oligomer via SC.
1.6 Crosslinking Methods for Silicon-Containing Materials.
Elastomers are described as ,,materials that can be stretched many times its original
length without breaking, and will snap back to its original size when it is released."106 To
prevent polymer chains from slipping past each other during elongation, either physical
or chemical crosslinks have to be present.
Elastomers containing physical crosslinks e.g. glassy polymer (with Tg above room
temperature) or semi-crystalline polymer (with Tm above room temperature) are termed
thermoplastic elastomers, because they can be reshaped upon heating above Tg or Tm. On
the other hand, elastomers that contain chemical crosslinks are called thermoset
elastomers, and polymer chains are hold together through covalent bonds forming a
single networked molecule. The degree of crosslinking is very important for the final
properties of materials. Too many crosslinks make material stronger, but brittle, and only
lightly crosslinked materials are effective elastomers.
Applications of silicon-based materials in elastomers, coatings, and resins require
that the materials do not flow; therefore crosslinking of these materials is necessary.
In general, crosslinking requires presence oftri- or tetrafunctional silanes that can
react under ionic conditions, or organic groups that are pendant or are on the end of the
silicone polymer. Most silicone elastomers are prepared via one of the three crosslinking
routes: moisture cure, heat-activated radical cure, or addition cure.1,4
First, moisture cure crosslinking, very often referred to as room-temperature
vulcanization (RTV), takes advantage of the susceptibility of the silicon atom to
nucleophilic substitution. Very often hydroxy-terminated polysiloxanes are crosslinked
by reacting with tri- or tetrafunctional silanes at room temperature. The rate of the cure is
strictly dependent on the rate of hydrolysis of the leaving groups, thus very often these
reactions are catalyzed by acids, bases, and by tin or titanium alkoxy/acyloxy derivatives.
Second, heat-activated radical cure (high-temperature vulcanization HTV) involves
formation of free radicals from peroxides e.g. m-chlorobenzoyl peroxide or
di-t-butyl peroxide. Two main reactions are observed while using these systems: the
hydrogen radical abstraction from the methyl group attached to the silicon and the radical
addition to vinyl groups. The reactivity of the latter reaction is higher than the hydrogen
radical abstraction; therefore lower temperatures can be used to crosslink vinyl-
substituted silicones. The distribution of vinyl groups along polymer backbone plays
important role in the characteristics of the cured materials. The formation of the polymer
network is more efficient in the systems where vinyl groups are distributed
homogenously, rather than the blocky systems where vinyl groups tend to react with each
other. In the latter case this reactions do not contribute to formation of the network.
Third, addition cure requires presence of the vinyl-terminated and hydrogen-
terminated polymers in the system. The addition reaction occurs under mild conditions
and in most cases is catalyzed by platinum based catalysts. Since this is an addition
reaction no by products are formed, however the residual catalyst remains trapped in the
network.
1.7 Purpose of Study.
This document reports the progress in the investigation of ADMET application in
production of solvent resistant elastomeric materials and plastics. The objective is to
create a "common denominator" in materials (shown in Figure 1-27), wherein mechanical
performance can be varied from highly elastomeric to flexible plastic, depending on the
molar ratio of monomers used during polymerization. Improvements of processability,
durability, and elasticity are sought.
The study has focused on the improvement of previously synthesized materials
through variation of the length of the soft phase, and introduction of the labile silicon-
alkoxy end group. In result numerous new siloxane or oxyethylene based materials have
been synthesized and characterized.
CHAPTER 2
SYNTHETIC MODIFICATION OF CROSSLINK DENSITY AND SOFT SEGMENT
IN POLYCARBOSILANE/POLYCARBOSILOXANE ELASTOMERS
2.1 Introduction
The clean and efficient chemistry afforded by ADMET, mentioned in the previous
chapter, allows for the combination of the best material properties of both carbon and
silicon into a precise random linear copolymer. Moreover, the high selectivity of olefin
metathesis allows for a wider range of possible functional handles to manipulate the
overall material properties. The behavior of ADMET carbosilane/carbosiloxane
copolymers can be achieved by selecting the appropriate molar concentrations of each
monomer, as demonstrated in Figure 1-27, to generate a large number of materials.
OCH3 CH3 CH3
OCH3 + C HH
CH I I
(75) [Mo] (76)
OCH3
101 (77)
10 '/. 1/
H2O
Single Elastic Link
Chain-Internal Crosslinks in Backbone
/ \ ri3 3cm
CH3 CH3
I H3 CH3 n CH30H
OH3 L,
0 (78) Single Elastic Link
I in Backbone
Chain-Internal Crosslinks
Figure 2-1. Atmospherically crosslinked ADMET copolymers.
The production of these previously obtained hybrid systems using ADMET
chemistry is described in Figure 2-1.75,107 These carbon/silicon copolymers contain
"chain-internal" crosslinks only, where the ratio of 75 to 76 co-monomers determines its
fundamental mechanical performance, varying from a rigid to an elastic behavior.
However co-monomer identity, functionality, and ratio are predominant factors in
controlling the elastic properties of the material. While our initial silicon thermoset
copolymers of this type are freestanding films, they are relatively weak (low tensile
strength) (Table 2-1).
Table 2-1. Mechanical properties of the unsaturated ADMET copolymers and a
commercially available elastomer.
Materis Modulus Tensile Strength Elongation
Materials
(MPa) (MPa) (%)
Polymer 78 0.16 0.2 20
Polymer 82 0.7 0.4 84
Polymer 85 6.0 0.6 400
Gelest Zipcone fSeriesa 2.0 0.7 150
a) Filler free Fast Cure pure silicon-based elastomer purchased from Gelest
Since the goal of the research is to produce solvent resistant and durable thermosets
based on polycarbosilanes, we focused on the factors influencing tensile strength and
other properties in order to improve our initial material (Figure 2-1). The successful
production of silicon hybrid polymers via ADMET requires that our methodology
conform to certain thermoplastic material criteria. First, crosslink density in such
materials is important and must be tailored precisely to balance mechanical strength and
brittleness while maintaining chemical and solvent resistance. Second, the soft phase
should be amorphous and sufficient in length to observe elastomeric behavior; moreover,
the run length of the soft phase between crosslink sites plays an important role in
elasticity allowing for low crosslink density, yet still enabling solvent resistance. While
our initial materials are interesting (Figure 2-1, material 78) they lack sufficient tensile
strength, most likely from an imbalance in crosslink density and soft phase morphology
generated from the presence of dangling-chains in the three dimensional network.'08
Thus, in Table 2-2 we have outlined synthetic modification to our initial materials that in
our opinion should improve their overall performance
Table 2-2. Synthetic research objectives.
Research Objectives Schematic Representation Benefit
Synthetically modified H3CO Will enhance tensile
crosslink density H3CO-Si^ 1 ^ strength and mechanical
H3CO UIx durability.
Chain-end latent crosslinking
Synthetic soft segment H3 Will significantly improve
modification s-o si degree of elasticity, degree
i~HO 4H3 CHi -- of elastic recovery, and the
Y propensity for phase
Increased run length in soft segment (y>1) separation.
2.2 Synthetically Modified Crosslink Density
The architecture of our initial hybrid copolymers are composed of a carbosiloxane
soft phase containing a siloxane bond (Si-O-Si), giving flexibility and mobility, while the
carbosilane hard monomer lacks this motional freedom since the Si-O-Si link is pendant
to and therefore not part of the linear polymer backbone. Thus, as illustrated in
Figure 2-1, these crosslinks, which we term "chain-internal", are found exclusively in the
hard phase of the copolymer, where their presence is critical since placing them in the
soft segment would limit elasticity. As a result, the latent reactive sites are placed in the
hybrid material by the carbosilane monomer alone, in which its content does not exceed
10% by weight. These low monomer concentrations are necessary to produce flexible
materials. As was shown by Brzezinska75 this chain-internal crosslinker is randomly
incorporated into polymer backbone, therefore crosslinking will occur at random points
to the neighboring chains. This will result in formation of uncapped chain-ends
(dangling-chains). Dangling chains are described as chains that are connected to the
junction at only one end point. These mobile chains, which are elastically ineffective,
represent network imperfections diluting the ultimate properties of the ADMET
elastomer. In fact Andrady et al., has shown that both the ultimate strength and
maximum extensibility of a material are lower for networks containing dangling-
chains.109 In an effort to circumvent this problem, we have introduced the concept of
"chain-end" crosslinking to complement "chain-internal" crosslinking in attempt to
produce more perfect networks. Figure 2-2 shows two different molecules that can be
used as chain-end crosslinkers.
OCH3
B Si-OCH li
xOCH3 CH3
(79) (80)
Figure 2-2. Chain-end crosslinking molecules.
This approach to increase the crosslink density comprises two distinct concepts.
First, note that a polymer synthesized using compound 79 (Figure 2-2) would produce a
latent reactive, end-capped linear polymer. The mention of compound 61 is only for
comparison and introduction of our new latent concept and its use will be discussed in
Chapter 3. On the other hand, compound 80, synthesized by a Grignard reaction of 5-
bromo-1-pentene with trichloromethylsilane, contains three ADMET active terminal
olefins (f = 3), therefore, using chain-end crosslinker 80 will result in a crosslinked
network via the ADMET polymerization mechanism. Initially, we focused on proving
the chain-end crosslinking concept using monomer 80 by combining it with our
previously reported75 the hard and soft carbosiloxane and carbosilane monomers
described in Figure 2-1. As mentioned above, introduction of monomer 80 should
produce better-defined networks in material 82, and an observable improvement in its
mechanical response when compared to material 78..
ADMET polymerization is a step-growth type of polycondensation, where the
initial monomer molar ratios are directly transferred to the polymer produced. Also, due
to the nature of the substrates polymerized, all monomers are incorporated randomly in
the polymer.
0 OCH 3 I I
^ l -OH + CH3I I +YI
OCH3 H 3 x0
OCH, (76)
(75) ) (80)
cat
H O
Chain-End
I CCH3 CH3
1CH3 OCH 3H
H20
Chain-End
Single Elastic Link Crosslinks
Chain-Internal Crosslinks in Backbone
Chain-Internal Crosslinks
Figure 2-3. Chain-end crosslinking in the polycarbosilanes via ADMET.
Figure 2-3 illustrates the synthesis of a copolymer possessing both types of
crosslinks using an initial soft/hard/chain-end crosslinker molar ratio of 20:1:1 ratio,
respectively. The use of metathesis (ADMET) generates polymers in which all
monomers are randomly incorporated. Also, the high conversions needed in metathesis
chemistry allows for a direct transfer of monomer feed ratios to the growing polymer
chains. Both Schrock's molybdenum catalyst [Mo] (2) (at room temperature) and 2nd
generation Grubbs' catalyst [Ru]** (5) (at 67-700 C) were examined in this study.
Polymerizations were initiated using a 250:1 monomer: catalyst ratio regardless of
catalyst chosen. Following initiation the reaction mixture was poured onto a Teflon
plate and placed either in a room temperature vacuum desiccator (Schrock's catalyst), or
in a high temperature vacuum oven for second-generation Grubbs' catalyst. Both catalyst
systems produce free standing films, where solvent resistance is rendered by the chain-
end crosslinker 82. These freestanding films were then exposed to atmospheric moisture
to activate the "chain-internal" latent crosslinks. The resultant material (82) containing
chain-end crosslinks shows significant improvement in tensile strength (Table 2-1), when
compared with our initial material (78) without chain-end crosslinks. Preliminary
mechanical data recorded a modulus of 0.7 MPa and 84% extensibility for material 82.
When compared to material 78 (modulus 0.16 MPa, 20% extension) the importance of
chain-end crosslinking concept is evident.
2.3 Synthetic Soft Segment Modification
The minimization of network defects by using "chain-end" crosslinks has a
profound enhancement over the material's bulk properties. In an attempt to obtain better
elongation, further modification was investigated by the incorporation of a different soft
phase monomer. Throughout our initial study, we have solely employed the carbosilane
monomer 76, a monomer possessing an inherently low Tg reflective of the Si-O-Si base
unit. The homopolymer of monomer 76 has an observed glass transition onset of -91 C
making it a desirable soft phase.75 However, when this monomer is employed in a
random copolymer (material 82) array with the hard phase monomer 75, overall
elongations are limited to approximately 80%.
The most efficient way to enhance elasticity would be to increase the run length of
the siloxane soft phase. However, due to obstacles encountered during the synthesis and
purification of such monomers, we initially focused on an easily synthesizable soft phase.
Diene versions of polyoxyethylene are well known and have found wide application as
soft phase segments, 110-117 while this soft phase is not siloxane based the synthesis of
these dienes are trivial and can be produced in large quantities. Treatment of a
representative bromoalkene with a glycolic alkoxide yields analytically pure monomer 83
after purification.11s Other method used in the synthesis of family of glycol-based dienes
will be discussed in Chapter 3.
S8 f d d d c a
di "b
Figure 2-4. NMR spectra of monomer 83.
Figure 2-4. H NMR spectra of monomer 83.
For example, the NMR spectrum for monomer 83 is shown in Figure 2-4. The
characteristic integration of the polyoxyethylene segment at 3.6 ppm, as well as, the
observed resonances for the diene can be seen at 5.8 and 5.0 ppm.
While polyoxyethylene are excellent soft segments; their glass transition is
dependent on the number of repeating ethylene oxide units. An increase in Tg is
observed with larger numbers of ethylene oxide, Figure 2-5.119 For example, the small
molecule ethylene glycol (n=l) exhibits a Tg = 930C while an increase to -500C is
observed when n=9. Initially, monomer 83 was chosen as the model soft phase due to its
low glass transition (-75C) relative to monomer 76 (-91C). Hopefully, the close
proximity of the two Tgs would reduce the effects caused by the monomers' molecular
differences. The goal in the future is to produce soft phases based solely on siloxane, in
effect utilizing the purely amorphous nature of silicon; in fact, the best-known
elastomers, polydimethylsiloxane (PDMS), exhibit Tgs as low as -1230C.
n Tg (oC)
1 -93
2 -83
3 -75
4 -70
9 -50
Figure 2-5. Glass transition temperatures for various lengths of polyoxyethylene
The homopolymer synthesized with monomer 83 under standard bulk ADMET
polycondensation conditions generated a linear polymer having an M = 28,000 g/mol.
Thermal analysis of this oxyethylene-based ADMET homopolymer yielded a melting
temperature of 35.7 C and a recrystallization temperature of 16.6 C revealing semi-
crystalline properties typical for pristine polyoxyethylene. Repeated DSC scans led to
identical enthalpies of melting and recrystallization for this macromolecule.
Given the viability of this oxyethylene-based monomer in homometathesis, it was
employed in the ADMET terpolymerization scheme describing earlier (Figure 2-3),
substituting the new oxyethylene diene for the carbosiloxane monomer (Figure 2-6).
Identical monomer ratios (20:1:1) and catalyst loadings (250:1) were employed
comparable to the siloxane based terpolymer. A 10 fold increase in elongation-to-break
was observed, from less than 80% to 400% (Figure 2-6). It is apparent that 4 ethylene
oxide units are sufficient to generate highly elastic materials, and this data suggests that
fewer ethylene oxide units (perhaps 2 or 3) may prove of value.
The data in Table 2-1 also illustrates a clear enhancement in modulus and tenacity
for polymer 85, as compared with polymer 78 and 82. The modulus, tenacity, and
elongation increase along with presence of chain-end crosslinking and with a longer soft
phase, as expected. Moreover, the tenacity matches that of commercial silicone
elastomers, combined with a significantly higher modulus and elongation. It is evident
that "chain-end" crosslinking combined with the soft phase modification offer enhanced
elasticity and performance over our initial materials. In order to produce materials that
are more UV stable and offer the lowest possible Tgs, investigations to generate all-
siloxane copolymers possessing fully latent chain-internal and chain-end crosslink sites,
including the use of siloxane chain-end crosslinkers such as 79 (Figure 2-2), are
described in following chapters.
43
OCH, |
I OCH, X I
(75) (83)
[Rul (80)
\ I IH /\ I 9 i\ I /
OCH x \ y\ CH, /
(84)
Atmospheic
Hp
Chain-End
Chain-Internal Crosslinks
Crosshlinks
9 II
Longer Length Soft Phase + CH30H
Chain-End
Chain-Internal (85) Crosslinks
Crosslinks
Apply Tension
Release Tension
Figure 2-6. High elastic material containing increased soft phase and chain-end
crosslinker.
2.4 Conclusions
ADMET polymerization of carbosilane/carbosiloxane copolymers bearing "latent
reactive sites" offers a new approach to the synthesis of materials exhibiting variable
physical properties. Latent reactive methoxy groups remain inert during metathesis
polymerization; exposure of the copolymer to moisture triggers hydrolysis (latent
crosslinking). The incorporation of "chain-end" crosslinks has improved mechanical
performance by limiting the formation of dangling chains. In addition increasing the run
length of the soft phase, in order to achieve a better balance between soft phase
concentration and crosslink density, generates a material exhibiting significant
mechanical performance. The material where the soft phase was changed from the
siloxane unit to the polyoxyethylene glycol leads to formation of crosslinked film 70 that
exhibits good material properties (modulus 6 MPa, elongation 500%). We are pursuing
this research with the goal of generating all-siloxane copolymers possessing fully latent
chain-internal and chain-end crosslink sites.
2.5 Experimental Methods
2.5.1 General Considerations
IH NMR (300 MHz) and 13C NMR (75 Hz) spectra of the ADMET polymers and
monomers were recorded in CDC13 or C6D6 on either a Mercury series or Varian VXR-
300 NMR super-conducting spectrometer. Chemical shifts were referenced to residual
CHC13 (7.25 for IH and 77.23 for 1C) and to residual C6D6 (7.15 for H and 128.39 for
13C). High-resolution mass spectral (HRMS) data were obtained on a Finnegan 4500 gas
chromatograph/mass spectrometer using either the chemical ionization (CI) or
electrospray ionization (ESI) mode. Elemental analysis was carried out by Atlantic
Microlab Inc. (Norcross, GA).
Gel permeation chromatography (GPC) of the unsaturated ADMET polymer was
performed using two 300 mm Polymer Laboratories gel 5pm mixed-C columns. The
instrument consisted of a Rainin SD-300 pump, Hewlett-Packard 1047-A RI detector
(254 nm), TC-45 Eppendorf column heater set to 35 C, and Waters U6K injector. The
polymer samples were dissolved in HPLC grade THF (approximately 0.1% w/v), filtered
before injection, and collected at a flow rate of 1.0 mL/min. Retention times were
calibrated to polystyrene standards purchased from Polymer Laboratories (Amherst,
MA).
Differential scanning calorimetry (DSC) was performed using a Perkin-Elmer DSC
7 at a heating rate of 10 C/min. Thermal calibrations were made using indium and
freshly distilled n-octane as references. Heats of fusion were referenced against indium.
All samples were prepared in hermetically sealed pans (4-7 mg/sample) and were run
using an empty pan as a reference. The samples were scanned for multiple cycles to
remove recrystallization differences between samples, and the results reported from the
second scan cycle.
Mechanical tests were performed at the University of Florida facility by courtesy of
Dr. Brenan's research group, using an Instron model 1122 load frame upgraded with an
MTS ReNew system running MTS TestWorks 4 software and a 500 g load cell for the
experiments.
Tensile specimens were cut from the cured freestanding films using an ASTM
D1822-68 type L dogbone die. Each specimen was individually measured to account for
thickness variability in calculations. The grip separation for the samples was set at 25.4
mm and the width of the testing portion was set at 3.1 mm. Tensile measurements were
made according to ASTM D412-97 at rate of 2 inch/min.
2.5.2 Materials.
All reagent grade solvents were freshly distilled over a Na/K alloy, except for THF
which was dried using K-benzophenone ketyl. 5-bromo-1-pentene and 11-bromo-1-
undecene were purchased from Aldrich and dried over activated 4A sieves. Magnesium
turnings (Aldrich) were activated by vacuum drying at 100 OC prior to use. Triethylene
glycol (Aldrich) was dried by azeotropic distillation using toluene and
bis(trimethoxysilyl)ethane (Gelest) was distilled over CaH2. Any necessary chlorosilanes
were purchased from Gelest and used as received. Deuterated solvents (Cambridge
Isotope Laboratories) were stored over activated 4A sieves. Filler Free fast Cure pure
Silicon based elastomer was purchased from Gelest. The metathesis catalysts
Cl2Ru(IMes)(PCy3)[=CHPh]42 and [Mo=CHCMe2Ph(=N-C6H3-2,6-i-Pr2-)
(OCMe(CF3)2)2]38,120 were synthesized according to literature procedure. All other
compounds were used as received.
2.5.3 Monomer Synthesis
Monomer 75 and 76 were synthesized according to a literature procedure.75
2.5.3.1 Tris(4-pentenyl)methylsilane (80)
A 250 mL three-necked round bottom flask equipped with an additional funnel,
condenser, and stir bar was flame-dried under vacuum, and then flushed with argon.
Magnesium turnings (2.1 g, 0.087 mol) and dry diethyl ether (50 mL) were added. A
solution of 5-bromo- -pentene (11.8 g, 0.0800 mol) and 50 mL of diethyl ether was
added drop-wise to maintain a constant reflux. The solution was refluxed for an
additional hour, cooled to room temperature, and a mixture oftrichloromethylsilane (3.32
g, 0.0221 mol) in Et20 (10 mL) was slowly added. Upon addition, the solution was
refluxed (2 hours), cooled and the product was diluted with 50 mL of dry pentane and
separated via cannula. The organic were evaporated under reduced pressure yielding 6.5
g of a clear, colorless liquid. The crude product was purified by column chromatography
using straight hexanes. Monomer 80 was collected in 80% and the following spectral
properties were observed: 1H NMR (CDCl3): 8 = -0.05 (s, H3), 0.5 (m, br, H6), 1.4 (m, br,
H6), 2.05 (m, br, H6), 5.0 (m, br, H6), 5.8 (m, br, H3). 13C NMR (CDCl3): 6 = -5.25,
13.36, 23.40, 37.79, 114.43, 138.99. MS (70 eV): m/z = 181.1412 (M+-CsH9).
(C16H3oSi) (250.49): Calcd. C 76.72, H 12.07; Found C 76.69, H 12.18.
2.5.3.2 Diundecenyltriethylene glycol (83).
The synthetic procedure was modified from published procedure.'18 Sodium
hydride (6.5 g, 0.27 mol, 60% dispersion) was placed in a flame-dried, Ar-purged, three-
necked 1000 mL round-bottom flask equipped with a stir bar, condenser, and an
additional funnel. Dried triethylene glycol (8.1 g, 0.068 mol) and 260 mL of dry THF
were combined in a flame-dried 500 mL Schlenk flask. The solution was transferred to
the additional funnel and the mixture was added drop-wise under constant stirring. After
24 h, 11-bromo-1-undecene (40 g, 0.17 mol) in 240 mL of THF was added and stirred for
an additional 72 h at reflux. Upon cooling, 50 mL of water was added, stirred for 15
minutes, and extracted using ether. The combined organic extracts were washed with a
saturated NaCl solution, dried over MgSO4, filtered, and evaporated under reduced
pressure. The crude product was then purified by column chromatography using
hexanes/diethyl ether (80:20%). The fractions were concentrated yielding 15 g of
monomer 83. The following spectral properties were observed: 1H NMR (CDCl3): 8 =
1.25 (m, br, H24), 1.60 (m, br, H4), 2.10 (m, br, H4), 3.53 (m, br, H4), 3.60 (m, br, H12),
4.95 (m, br, H4), 5.85 (m, br, H2). 13C NMR (CDC13): 6 = 26.10, 28.94, 29.14, 29.45,
29.49, 29.51, 29.55, 29.62, 29.66, 33.81, 70.04, 70.64, 71.53, 114.06, 139.18. MS (70
eV): m/z = 455.4121 (M+). (C2sH5504) (454.72): Calcd. C 73.96, H 11.97; Found C
73.91, H 12.18.
2.5.4. General Metathesis Conditions
All monomers used in polymerization were purified and degassed prior to
polymerization. All glassware was thoroughly cleaned and dried under vacuum before
use. The polymerizations were initiated in an argon-filled glove box by placing the
appropriate amount of monomer followed by either metathesis catalyst and slowly stirred
for 2 minutes. In all cases the monomer: catalyst ratios were 250:1.
2.5.4.1 Polymerization of diundecenyltriethylane glycol (83)
The monomer was polymerized using the above procedure. Monomer 83 (0.51 g,
0.0021 mol) was added to the reaction tube charged with Grubbs' 2nd generation catalyst
(7.30 mg, 8.58x10-6 mol). The reaction tube was sealed using a Kontes Teflon valve and
placed under an intermediate vacuum until the mixture became viscous. The reaction
flask was then placed in a 40 C oil bath and high vacuum (<0.01 mm Hg) was applied.
The temperature was gradually raised to 70 oC over 72 hours. After three days, the
polymerization was quenched via dissolution in toluene followed by precipitation in
acetone. The following properties were observed: 'H NMR (CDCl3): 6 = 1.25 (m, br,
H24), 1.60 (m, br, H4), 2.10 (m, br, H4), 3.53 (m, br, H4), 3.60 (m, br, H12), 5.56 (br, H2);
"C NMR (CDC13): 6 = 25.91, 26.06, 27.17, 28.96, 29.10, 29.16, 29.31, 29.43, 29.46,
29.60, 29.73, 32.49, 32.57, 70.01, 70.57, 70.58, 71.50, 130.30. GPC data (THF vs.
polystyrene standards): M, = 27 770 g/mol; PDI (MA/n) = 1.55. DSC Results: T,
(melting peak) = 35.7 C, Ah,,, = 86.8 J/g; Recrystallization Tc (recrystallization peak)
20.1 oC, Ah, = -88.8 J/g
2.5.4.2 Terpolymerization of monomers 75, 76, and 80 to produce terpolymer 81
Terpolymer 81 was prepared at 1: 20: 1 monomeric ratio using monomer 75 (0.17
g, 5.0 x 10-4 mol), monomer 76 (3.0 g, 0.010 mol) and monomer 80 (0.13 g, 5.2 x 10-4
mol). The monomers were stirred for 5 minutes to ensure a homogeneous mixture before
the addition of either 2nd generation Grubbs' catalyst (0.0372 g, 4.41 x 10-5 mol), or
Schrock's catalyst (0.0341 g, 4.41 x 10-4 mol). The reaction mixture was sealed,
removed from the glove box, and stirred under vacuum for 10 minutes. The
polymerization mixture was then poured out on a Teflon plate and placed in a vacuum
oven at 700 C for 72 h. In the case of Schrock's catalyst, the reaction mixture was cast in
the glove box and placed inside a vacuum desiccator. The desiccator was removed from
the glove box, and placed on a high vacuum line at room temperature for 72 h. Vigorous
bubbling of the ethylene gas byproduct was observed, followed by formation of a thin
film. After 72 h the Teflon plate was removed from a vacuum oven and exposed to 80%
moist atmosphere having a pH 6.5-7.0, for a period of 3 weeks.
2.5.4.3 Terpolymerization of monomers 75, 83, and 80 to produce terpolymer 84
Terpolymer 84 was prepared by mixing monomer 75 (0.11 g, 3.3 x 10-4 mol),
monomer 83 (3.1 g, 6.6 x 10-3 mol), and monomer 80 (0.0862 g, 3.46 x 10-4 mol)
resulting in a 1: 20: 1 monomer ratio. Following the same procedure as for terpolymer
81, all monomers were mixed and stirred prior to the addition of the catalyst. In this case,
2nd generation Grubbs' catalyst (0.035 g, 4.1 x 10-5 mol) was added and the reaction
mixture was stirred, sealed, removed from the glove box, and placed for 10 minutes on a
high vacuum line. The mixture was then transferred to the Teflon plate and placed in a
vacuum oven at 70 C. After 72 h, the Teflon plate was removed and the thin film was
exposed to 80% moist atmosphere having pH 6.5-7.0 for a period of 3 weeks.
CHAPTER 3
SYNTHESIS AND CHARACTERIZATION OF
OLIGO(OXYETHYLENE)/CARBOSILANE COPOLYMERS AND THERMOSETS
3.1 Introduction
Encouraged by results obtained from the previous investigation,121 we synthesized
fully latent reactive processable and durable materials. Although our ultimate goal is to
produce carbosiloxane/carbosilane based materials, our focus shifted to materials based
on the oxyethylene segments. The previous studyl21 has proven that the incorporation of
these segments can result in a thermoset that exhibits good physical properties.
As mentioned in the previous chapter, polyethylene glycol has found wide
application as a soft phase segment.110,112,122 This hydrophilic polyether received
considerable attention for use in biomaterials due to its low interfacial energy with water,
relative structural stability, lack of binding sites for reactive proteins, and high chain
mobility.123 It has been used extensively in the preparation of polymer electrolytes,124-126
crosslinked hydrogels,127-129 membranes130-132 and amphiphilic copolymers.118,133,134
In the preparation of membranes for biological applications, control over the pore
size and good mechanical properties are a necessity in order to enhance the transport of
hydrophilic entities. Thus the preparation of such membranes requires highly
monodisperse polyethylene glycol chains, which are further terminated with allyl groups
susceptible to hydrosilylation. The synthesis involves the random cohydrosilylation of
a,o-diallyl-PEG and a, o-diallyl-PIB segments with pentamethylcyclopentasiloxane131
resulting in the formation of a network that restrains post reaction shaping.
Silylated polyethers have found applications in the sealant industry. For example
MS Polymer developed by Kaneka Corporation in Japan, composed of high molecular
weight polypropylene oxide endcapped with silyl groups, has been used in formulation of
sealants and caulks.135 The synthesis of silyl terminated polyethers involves reaction of
polypropylene glycol with allyl chloride in the presence of alkali metals which produces
allyl terminated polyether, followed by hydrosilylation with methyldimethoxysilane in
the presence of the platinum chloride catalyst. The resultant silyl terminated polyether is
subsequently mixed with fillers, plasticizers, and a curing catalyst to obtain a room
temperature curing composition.
This chapter describes the design and synthesis of unsaturated "latent reactive"
oligo(oxyethylene)/carbosilane copolymers prepared via acyclic diene metathesis
polymerization. Methoxy groups attached to silicon atoms have been chosen as the
"latent reactive" sites that have previously been proven to be inert during
metathesis.75'85'121'136 Exposure to moisture triggers the hydrolysis of methoxy groups,
which undergo condensation reaction to yield crosslinked materials. The synthesis and
characterization of monomers, polymers, and both linear and crosslinked copolymers is
reported.
3.2 Synthesis of Monomers
3.2.1 Monomers for Soft Phase Modification
A family of monomers (Figure 3-1), with different lengths of oligo(oxyethylene)
and methylene spacers between the carbon-carbon double bond and the
oligo(oxyethylene) unit, has been synthesized in order to investigate the influence of the
structural modifications on the properties of the resulting materials. For clarity, the
nomenclature shown in Figure 3-1 is used instead of the full names of the monomers.
The first number corresponds to the number of methylene units ("sp") between oxygen
and the vinyl group; the second number corresponds to a glycol ("gl") used in the
monomer synthesis (e.g. 2 for diethylene glycol). Consequently, abbreviations used for
the polymers are "poly" followed by the monomer abbreviation.
3sp2gl 6 9 9sp2gl
(86) 6sp3gl (91)
O^{4A) (89)
3sp3gl O 9sp3gl
(87) 4 (92)
O^ O^ 606sp4gl O4
3 (90) 9
3sp4gl 9sp4gl
(88) (93)
Figure 3-1. Glycol-based monomers.
Williamson's ether synthesis, described in Chapter 2, was used to prepare the 3sp
and 6sp monomers. The 9sp monomers were synthesized via another synthetic route that
involves reaction of 1-undecene-11-ol and an appropriate ditosylated oligo(oxyethylene)
glycol in the presence of sodium hydride (example presented in Figure 3-2). The 1-
undecene-11-ol used in this reaction was synthesized according to literature procedure.137
OH + Ts O OT DMF, oOC 0
9 4
(94) (95) (93)
Figure 3-2. Synthesis of 9sp4gl (93).
The synthesis, workup, and product purification via column chromatography was
followed by vacuum distillation over calcium hydride to yield analytically pure
monomers, which were stored in an inert atmosphere of a glove box. Common
spectroscopic methods such as 1H NMR, 13C NMR, and FT-IR were used to characterize
all monomers.
3.2.2 Monomers for Crosslink Density Modification
As described in Chapter 2 our initial materials were composed of two monomers:
the "soft phase" monomer 76 and the latent reactive "chain-intemal" crosslinker 75.
From previous studies75'121 we have learned that the ratio of used monomers dictates
properties of the final thermoset. Moreover, we have found that the crosslink density
modification via an incorporation of the "chain-end" crosslinker as the third component
enhances properties of our materials by limiting the formation of dangling chains.121
However, due to the nature of this "chain-end" crosslinker 80 used in this study, a
thermoset is formed already during the polymerization restraining further processing of
the polymer.
Since the ultimate goal of this research is the synthesis of fully "latent reactive",
linear processable materials for thermoset elastomers, we have decided to synthesize a
latent reactive "chain-end" crosslinker. The synthesis of such monomer shown in Figure
3-3, involves the Grignard coupling of 11-chloro-1-undecene with
trimethoxymethylsilane (96). The use of this monofunctional metathesis active monomer
97 in the synthesis of latent reactive materials has a dual purpose. First, the placement of
97 on the end of chains will ensure linearity of the resultant polymer and facilitate further
processing. Second, presence of labile methoxy groups on the silicon atom that
participate in the crosslinking process will limit the formation of dangling chains in the
final thermoset.
The crosslink density of our materials can be further modified using "chain-
internal" crosslinker 100 (Figure 3-3) bearing only two methoxy groups on the silicon
atom instead of four as used in previous studies.121 Using the same synthetic
methodology as in the case of 97, the Grignard coupling of 5-magnesiumbromo-1-
pentene (98) and tetramethylorthosilicate (99) was applied.
OCH3 OCH3
I Et2O
MgC1 + H3CO-Si-CH3 -CH3
OCH3 OCH3
OCH3 end9sp20Me OC
(95) (96) (97)
OCH3 OCH3
I EtzO I
./ MgBr + H3CO-Si-OCH3 2' 3
OCH3 OCH3
3sp20Me
(98) (99) (100)
Figure 3-3. Synthesis of end9sp20Me and 3sp20Me.
In both cases after workup and distillation an analytically pure monomer was obtained as
proven by 1H NMR, 13C NMR, and FT-IR.
Once again short names were assigned for clarity, as shown in Figure 3-3. The first
part of the name depicts number of methylene groups between silicon atom and the vinyl
group, whereas the second part describes number of the latent reactive methoxy groups
present in the monomer. Consequently, the short name for monomer 75, described in
Chapter 2, will be 3sp40Me. In order to distinguish the a,co-diene monomers from a
monofunctional "chain-end" monomer, the word "end" was placed before an
abbreviation.
3.3 ADMET Polymerization
3.3.1 Synthesis of Oligo(oxyethylene)/Oligo(ethylene) Copolymers
Polymers were synthesized from monomers presented in Figure 3-1 under standard
bulk ADMET polycondensation conditions using the 1st generation Grubbs' catalyst
[Ru]* (4). Although our initial polymerization of monomer 9sp3gl was catalyzed by 2nd
generation Grubbs' catalyst [Ru]** (5), the [Ru]* (4) was used to conduct metathesis in
mild conditions and to prevent olefin isomerization during polymerization.43'138
Resulting linear unsaturated polymers can be described as alternating (AB)n microblock
copolymers of oligo(oxyethylene) and oligo(ethylene). All polymers are viscous liquids
except for ones made from 9sp monomers.
Table 3-1. GPC data of unsaturated polymers.
Polymer GPC (g/mol)a
Polymer
Mn Mw PDI
poly3sp2gl (101) 35,766 63,306 1.77
poly3sp3gl (102) 34,565 65,328 1.89
poly3sp4gl (103) 34,890 66,634 1.91
poly6sp3gl (104) 36,193 62,976 1.74
poly6sp4gl (105) 30,734 49,174 1.60
poly9sp2gl (106) 51,750 106,605 2.06
poly9sp3gl (107) 40,103 72,185 1.80
poly9sp4gl (108) 46,840 96,490 2.06
a) referenced to PS standards in THF at 40'C;
The standard spectroscopic methods (1H NMR, 13C NMR, FT-IR), the gel-
permeation chromatography (GPC) and the thermal analysis (DSC, TGA) were used for
the structural characterization and verification of the monomers' conversion to high
molecular weight polymers. Table 3-1 summarizes the results obtained from the gel-
permeation chromatography analysis on unsaturated polymers, while a more detailed
discussion regarding 1H NMR, 13C NMR, FT-IR and the thermal analysis is described in
the following sections.
3.3.1.1 Structural characterization
The spectra of monomers with the same alkenyl group length or the same number
ofoligo(oxyethylene) units are homologous. Thus the spectra of polymers obtained from
these monomers are very similar as well.
c l g g .
I I
H NMR spectra of monomer 6sp3g and polymer poly6sp3gl.
In Figure 3-4, a representative H NMR spectra of monomer 6sp3g (89) and
2, 3 6 6 I \
S 6 4 I /n
U 6
4 5
1 1 ElF T 1 1 1 T V ,T f F I
6.0 5.0 4.0 3.0 2.0 1.0
Figure 3-4. 'H NMR spectra of monomer 6sp3gl and polymer poly6sp3gl.
In Figure 3-4, a representative Hc NMR spectra of monomer 6sp3gl (89) and
polymer poly6sp3gl (104) is presented. The top of the figure shows 'H NMR spectra
with characteristic peaks between 3.53-3.63 ppm that correspond to protons on the
oxyethylene segment. As shown in Figure 3-4, the chemical shifts for the monomer and
the unsaturated polymer remain the same, but the intensity of signals depends on the
length of the oxyethylene segment.
The 'H NMR spectra clearly illustrates quantitative conversion of the monomer to
the high molecular weight unsaturated polymer. This can be observed by the complete
disappearance of the vinyl groups in the monomer, (peaks at 4.9 and 5.7 ppm in the top
spectra), and by the newly formed peak at 5.3 ppm of the internal olefin that appears in
the spectrum of polymer (bottom).
[ c d h g a
6* 3 f g C b
S3 4 8 7 2
V 6 7 5 1 I
5 trans
130 120 110 100 90 80 70 60 50 40 30 20 10
Figure 3-5. 3C NMR spectra of monomer 6sp3gl and polymer poly6sp3gl.
The chemical structures of the monomer and the polymer were characterized by 3C
NMR and are shown in Figure 3-5. The spectra of the monomer and the polymer show
characteristic carbon signals of oxyethylene segments present at 71.3 and 70.8 ppm. As
in the olefinic region of 6sp3gl (89) and poly6sp3gl (104) of the 1H NMR spectra, the
13C NMR spectral changes are also observed. Two signals of vinyl carbons at 114.1 and
139.8 ppm present in the monomer disappear while two new peaks at 130.3 ppm (major)
and 129.5ppm (minor) are formed in the polymer. These two new signals correspond to
trans and cis isomers of the 1,2-disubstituted double bonds along the polymer backbone,
further confirming high conversion of a monomer to the unsaturated polymer. Moreover,
the resonance of the allylic carbon in the monomer spectrum at 33.6 ppm splits into two
new signals (32.5 and 27.1 ppm) in the spectrum of polymer due to the trans and cis
geometries around this carbon. The relative proportion of cis and trans carbons by 3C
NMR show approximately 80% content of the later, which is in an agreement with
previous reports on ADMET polymerization.52'53 The high content of the trans olefin in
AMDET products is attributed to the most stable conformation of the metallacyclobutane
intermediate.5
-- 6sp4gl
-- poly6sp4gl
3080
914
S* 1121
SI I *\ 968
-968
1121\
3500 3000 2500 2000 1500 1000 500
Wavenumber (cm")
Figure 3-6. FT-IR spectra of 6sp3gl and poly6sp3gl.
Chemical structures of the monomer and the polymer were also verified by FT-IR
spectroscopy. For example FT-IR spectra of 6sp3gl and poly6sp3gl shown in Figure 3-6
contain a strong band at 1121 cm-1 that corresponds to the asymmetric stretch of the C-O-
C ether bonds from the oxyethylene segment and is present in both the monomer and the
polymer. Characteristic wavenumber bands that correspond to the vinyl group in the
monomer are observed at 3080, 1642, 994 and 914 cm-1. These peaks were not observed
in the poly6sp3gl, instead a new sharp absorbance assigned to the trans carbon-carbon
double bond appeared at 968 cm-1. Due to the low concentration of the cis conformation
in the polymer, the expected band at 730-665 cm1 was not detected. These results
further demonstrate high conversion of monomer to corresponding polymer.
3.3.1.2 Thermal behavior of unsaturated polymers
Thermal analysis of the family of alternating (AB)n microblock copolymers was
performed by differential scanning calorimetry (DSC). Detailed characteristics of
thermal transitions for unsaturated polymers are summarized in Table 3-2. Values of
glass transition temperatures (Tg) were taken at the mid-point of the transition and the
melting transition temperatures (Tm) were taken as the temperature at the peak of the
melting transition in the second heating scan. Similarly, the crystallization temperature
(Tc) was taken as the temperature at the peak of crystallization transition in the second
cooling scans. The data was collected by scanning samples from -150 to 80 C at 10
C/min and reported from second heating and cooling scans. All polymers described
below exhibit unique thermal properties. The combination of melting, crystallization,
and glass transition, three phenomena that are important with respect to the design and
processing of polymeric materials, have been observed for this family of unsaturated
(AB)n microblock copolymers.
Table 3-2. Thermal analysis of the unsaturated polymers.
Polymer DSC TGA (oC)e
Polym er-------------------- -
Tg Cp Tm Tc AH g/T TN2 Tda
(oC)a (J/(g OC) (OC)b (oC)c (J/g) g/m
poly3sp2gl -79.0 -0.78 242 250
poly3sp3gl -76.2 -0.83 -- 245 245
poly3sp4gl -74.1 -0.88 -35.4 4.8 0.80 252 251
poly6sp3gl -75.0 -0.29 12.9 -12.4 53.7 0.69 231 235
poly6sp4gl -76.0 -0.33 13.3 -16.2 46.6 0.68 239 227
poly9sp2gl -50.4 -0.39 57.1 35.8 82.4 0.67 314 295
poly9sp3gl -55.2 -044 45.9 26.1 70.6 0.68 282 248
poly9sp4gl -56.4 -0.34 42.5 19.4 55.8 0.68 276 258
a) values taken from the mid-point; b) values assign to the top of the melt; c) values
assign to the bottom of the recrystallization; d) calculated from obsolete temperatures;
e) recorded at 5% total mass loss under nitrogen gas or air, 20 oC/min
The glass transition temperatures reported in Table 3-2 for poly3sp2gl, poly3sp3gl,
and poly3sp4gl polymers were essentially identical with the values published by Baker et
al.124 for the same polymers obtained from ADMET polycondensation catalyzed by [Mo]
(2). The actual DSC thermograms of unsaturated polymers are presented in Figures 3-7 -
3-9 and are grouped into three graphs based on the number of methylene units in the
polymer. Thermograms of poly3sp2gl and poly3sp3gl display glass transition
temperatures at -79 and -76 OC respectively, revealing the purely amorphous character of
these polymers. In the case of poly3sp4gl after relaxation at -74 C, a cold crystallization
at -50 OC is observed, followed by melting peak at -35 oC. Comparison of Tg's in the
series of 3sp polymers shows their dependence on the length of the ethylene oxide
segment. The glass transition values increase as the content of the ethylene oxide
increases, a trend observed previously by Faucher119 in a series of ethylene oxide
polymers.
The thermal behavior of polymers poly6sp3gl and poly6sp4gl (Figure 3-8) is
basically the same. In the heating scan a low relaxation temperature can be seen at-75
and -76 'C for poly6sp3gl and poly6sp4gl respectively, followed by a sharp melt at -13
C with the heats of fusion of 53.7 J/g for poly6sp3gl and of 46.6 J/g for poly6sp4gl.
Upon cooling both thermograms display sharp crystallization peaks at -12 and -16 C and
both the enthalpy of melting and the enthalpy ofrecrystallization exhibit the same energy
for both polymers. Moreover, similar AH values for the melt and recrystallization
suggest a similar crystal packing of these polymers.
Analysis of thermograms for poly9sp3gl and poly9sp4gl exhibit similar trends as
described above for the 6sp polymers. Heating scans reveal very weak glass transition
temperatures at -55 and -56 C for poly9sp3gl and poly9sp4gl correspondingly,
presumably due to the semicrystalline nature of these materials. Sharp melting peaks are
present in both thermograms at 46 OC for poly9sp3gl and 42 C for poly9sp4gl. After
repeated scanning, both the heat of fusion of the melt and the enthalpy of recrystallization
exhibit the same energy. Thermal analysis performed on poly9sp2gl showed relaxation
at -50 C followed by a melt, cold crystallization and another melt at 57 C and a single
sharp recrystallization peak upon cooling scan at 35 C. Overall heats of fusion for the
heating run and the cooling run are virtually identical varying only by 0.08 J/g. Similar
to the series of 3sp polymers, dependence of Tg and Tm values on the number of ethylene
oxide segments present in the repeat unit is evident.
0.2-
0.0-
-0.2-
-0.4-
-0.6-
- poly3sp3gl
p013 o '1p:P4g
-0.81
-155 -125 -95 65 -35 -5 10 25 40 55 70 85
E 'I" Temperature (OC)
Figure 3-7. DSC traces of unsaturated 3sp microblock copolymers.
poly6sp3'l1
poly6sp4gl
0-
,-1
-155 -125 -95 -65 -35 -5 10 25 40 55 70 85
IL-o Dn Temperature (OC)
Figure 3-8. DSC traces of unsaturated 6sp microblock copolymers.
.- poly9sp3gl
1.5
Spoly9sp4gl
0.5-
-0.5-
-1.5
-2.5
-3.5
-155 -125 -95 -65 -35 -5 10 25 40 55 70 85
xo U n Temperature (OC)
Figure 3-9. DSC traces of unsaturated 9sp microblock copolymers.
However, reverse trends are observed when compared to the 3sp polymers, as the
number of oxyethylene units increase the values of melting temperature and glass
transition temperature decrease. The oxygen atoms of the ethylene oxide segment can be
considered as irregularities in the carbon chain of polyolefins. Thus short ethylene oxide
segments and long hydrocarbon segments will contribute to higher melting points and
higher relaxation temperatures as observed for poly9sp2gl, where there are only three
oxygen atoms per repeat unit in the polymer. Hence, by increasing the content of the
oxygen atoms in the polymers' repeat unit (five oxygen atoms present in the poly9sp4gl)
decreases glass transition and melting temperatures accordingly.
As suggested by the Boyer-Beaman rule for nonsymmetrical crystalline polymers
the ratio of Tg/Tm should be close to 2/3. Table 3-2 shows the calculated Tg/Tm ratio
64
values for the unsaturated polymers fall between 0.67 and 0.80, which is in close
agreement with the Boyer-Beaman rule.
3.3.1.3 Thermal stability of unsaturated polymers
The thermal-degradation behavior of unsaturated polymers has been investigated
using thermo-gravimetric analysis (TGA). The analysis was performed for each polymer
both in inert (nitrogen) and oxidative (air) atmospheres. The TGA experiments were
conducted at a scan rate of 20 C/min from 50 to 800 C and summarized results are
presented in Table 3-2.
95. poly6sp4gl N2
poly6sp4gl Air
75
55
35
15
-5 ....
50 150 250 350 450 550 650 750
Temperature (C)
Figure 3-10. TGA traces of poly6sp4gl.
Figure 3-10 illustrates typical TGA curves for unsaturated polymers plotted as
weight-loss versus temperature for poly6sp4gl (105) in both atmospheres. The polymer
exhibits good thermal stability up to 250 OC, after which rapid decomposition follows.
As shown in Figure 3-8 there is no significant difference in thermal stability for
poly6sp4gl in nitrogen and in air. Essentially complete volatilization of the sample is
achieved at 460 OC under both mediums. This rapid and complete elimination of mass is
indicative of the structural purity exhibited by these regular polymers. Moreover, due to
the fact that these polymers completely decompose thermally in a linear fashion over a
narrow temperature range implies that only one mechanism of decomposition, more
likely ether breakdown, is responsible for their decomposition.
3.3.2 Polymerization of "Chain-Internal" Crosslinker 3sp20Me
To ensure metathesis compatibility the monomer was homopolymerized in the
presence of 1st generation Grubbs' catalyst (4). The 'H NMR spectra for the monomer
and resulting ADMET polymer (poly3sp20Me (109)) are shown in Figure 3-11. Indeed,
metathesis does occur and can be easily observed by the disappearance of both the
internal and external olefin protons at 5.8 and 5.0 ppm, respectively. This unsaturated
carbosilane has been previously synthesized by Church91 et al. during investigation of
nucleophilic substitution on silicon-containing am(-dienes; however, 2nd generation
Grubbs' ruthenium complex (5) was used to promote ADMET polymerization.
C C
)OC 13
-
2 2
1 3 4 5
6.0 5.0 4.0 3.0 2.0 1.0
Figure 3-11. 'H NMR spectra of 3sp20Me and poly3sp20Me.
The presence of reactive groups in each repeat unit of the polymer makes the
characterization and handling of polymer challenging. Hydrolytic sensitivity makes GPC
analysis unavailable, instead 1H NMR end-group analysis was performed indicating a Mn
of 45,000 g/mol, four times higher than previously reported. Differential scanning
calorimetry was performed yielding in the same glass transition (Tg= -800C) as observed
by Church et al.
3.3.3 Synthesis and Characterization of "Latent Reactive" Copolymers
Given the compatibility of monomers mentioned in previous section and ADMET
viability to produce high molecular weight unsaturated polymers, their copolymerization
under bulk reaction conditions at mild temperatures is relatively trivial. Our previous
studies of carbosilane or carbosiloxane-based monomers75 have shown that their
copolymerization yields completely random copolymers. Copolymerization of the
glycol-based monomers with the "chain-internal" crosslinkers and/or "chain-end"
crosslinker produced random linear copolymers as well. Three general formulations have
been made where glycol-based monomers have been copolymerized with 10% by weight
of either "chain-internal" crosslinker (75) or (100), or with only "chain-end" crosslinker
to yield endcapped oligomers with approximate degree of polymerization equal to five
(DP-5), or with both "chain-internal" (5% by weight) and "chain-end" (2.5% by mol)
crosslinkers. Figure 3-12 shows general structures of each composition.
Figure 3-13 displays an example of the 'H NMR and 13C NMR spectra for
copolymer 38 (147). Characteristic presence of the internal olefin signal at 5.34 ppm for
all the polymers prepared by ADMET polymerization is visible in the H NMR spectrum.
Three additional peaks are present in the 1H NMR spectrum of copolymer (147), when
compared to the 'H NMR spectrum of poly9sp2gl i.e. at 0.1, 0.6, 3.5 ppm that
67
corresponds to proton signals of Si-CH3, Si-CH2-, Si-OCH3 groups respectively. The
comonomers feed ratio dictates the ratio of monomers present in the copolymer and can
be verified by 1H NMR spectroscopy by relating signals of Si-CH3 groups either to Si-
CH2- or CH2-CH2-O- groups.
10% wt content
OMe
9 OMe3 m
n m
10% wt content
OMe OMe
OM OMe
S- OMe I m
OMe
MeO I /
MOkO\ Si-Me
Me-Si 9 -o O -
Me/ g 9 OMe
MeO n
n-5
5% wt content
OMe
MeO OMe
Me-Si 0 O M'^ SiMeSi-Me
MeO0 OMe O Me
5% wt content
OOMe Me OMe
O\ Si-Me
Me-Si Si 3
/ x 3 Me OMe
MeO OMe
h m
copolymer 1 (110): g
2 (111): g
3 (112): g
4 (113): g
5 (114): g
6 (115): g
7 (116): g
8 (117): g
copolymer 9 (118): g
10 (119): g
11(120): g
12 (121): g
13 (122): g
14 (123): g
15 (124): g
16 (125): g
copolymer 17(126): g
18 (127): g
19 (128): g
20 (129): g
21 (130): g
22 (131): g
23 (132): g
24 (133): g
copolymer 25 (134): g
26 (135): g
27(136): g
28 (137): g
29 (138): g
30 (139): g
31(140): g
32 (141): g
copolymer 33 (142): g:
34 (143): g:
35 (144): gy
36 (145): g:
37(146): g=
38 (147): g
39 (148): g:
40 (149): g
Figure 3-12. Compositions of "latent reactive" unsaturated copolymers.
'3C NMR spectroscopy further confirms formation of copolymer by presence of the
internal olefin carbons at 129.5-131.2 ppm. However, multiple signals are present in this
region rather than only two signals (trans and cis) that are observed in the 3C NMR
spectrum of homopolymers. These multiple signals correspond to presence of additional
linkages between different repeat units in the copolymer. Signals from Si-CH3 at -5.8
ppm, Si-CH2- at 2.5 ppm, and Si-OCH3 at 50.3 ppm groups are also observed; however,
the intensity of these peaks is low due to the low content of both "chain-internal" and
"chain-end" crosslinkers in the copolymer.
5% wt content
Me S i oOMe -OMC
Me-Si--W1 Sii-'
1/ 9 2 OMe I 9 IOMC
MeO OMe
n im
Si-OCH,
I Ik f -CH,-Si
A L .,1t 1 : '.,o'o|oSi-CH
CH,=CH, S -CH3
3.6 3.5 3. 0 6 0.2 0.0
S' I i '' I I
5.0 4.0 3.0 2.0 1.0 0.0
5.0 4.0 3.0 2.0 1.0 0.0
CH,=CH,
-CH,-Si Si-CH,
r 131 130 |51 -- 50 49 -
125 100 75 50 25 0
Figure 3-13. 'H (A) and 3C (B) NMR spectra of copolymer 38 (147).
"' ^""
3.4 Preparation and Characterization of Thermosets
It is well known fact that silicon-alkoxy bonds as silicon-halide bonds are
susceptible to hydrolysis leading to formation of silanol groups (Si-OH), which can
undergo condensation at appropriate conditions. In result a thermodynamically stable
silicon-oxygen-silicon bond is formed and a byproduct such as alcohol or HX is released.
Although good efficiency and availability of compounds containing the silicon-halide
bond, due to formation of acids during crosslinking process the alkoxy groups are
preferred over the later. The process just described is well known as room-temperature
vulcanization (RTV) and has been used in sealant industry for over 30 years. This
opportunity exists for copolymers shown in Figure 3-12 that contain chain-internal and
chain-end crosslink sites. Consequently, when these copolymers are exposed to moisture
the latent reactivity is triggered yielding in thermoset materials. The composition and
short name assigned for simplicity are presented in Figure 3-14.
The first attempt to produce thermoset was performed on xlink 39, whose
composition with exception of latent reactive chain-end crosslinker showed good material
properties in previous study, as described in Chapter 2. An attempt was made to cast a
film of copolymer 39. The polymer was dissolved in toluene, poured into a Teflon mold,
and placed on the lab bench to trigger crosslinking. Unfortunately after one day the
resultant film started to crack yielding after two days a very brittle material as shown in
Figure 3-15.
10% wt content
10% wt content
0A
o1 ASi
J 0 1 xJ
-4 0 i
O\ Si-Me
Me-Si S-e
O n-n
5% wt content
5% wt content
xlink 1 (150): g
2 (151): g
3 (152): g
4 (153): g
5 (154): g
6 (155): g
7 (156): g
8 (157): g
xlink 9(158): g
10 (159): g
11 (160): g
12 (161): g
13 (162): g
14 (163): g
15 (164): g
16 (165): g
xlink 17(166): g
18(167): g
19 (168): g
20 (169): g
21 (170): g
22 (171): g
23 (172): g
24 (173): g
xlink 25 (174): g
26(175): g
27(176): g
28(177): g
29 (178): g
30 (179): g
31 (180): g
32(181): g
xlink 33 (182): g
34 (183): g
35(184): g
36 (185): g
37(186): g
38(187): g
39 (188): g
40 (189): g
Figure 3-14. Composition and structure of crosslinked materials.
The formation of this brittle material can be justify by information obtained from
DSC analysis ofpoly9sp3gl that showed semi-crystalline nature (Tm=46 C) of this
polymer. This kinetic phenomenon, namely crystallization takes place before chemical
crosslinking occurs yielding in physical crosslinks. It is well known fact that crystallinity
makes material strong, but it also makes it brittle. Thus we have decided to increase
2; x=
:3; x=
:4; x=
:3; x=
:4; x
:2; x=
:3; x=
4; x
2; x
3; x=
:4; x=
:3; x=
:4; x=
:2; x=
3; x=
:4; x=
:2; x=
3; x
4; x
=3; x
4; x
2; x
3; x
4; x
:2; x
=3; x
4; x
=3; x
=4; x
=2; x
=3; x
=4; x
home~lymmicaewle~m mrnhuimanepwearances aon Trw
UmieslMN'nawanan llmi Irasseniiemeumlue anvemenumamnr
tiinotewammemewavnt
rlnmemalesrorl..amnns ueaw anurva wedyourereunn il
cameui~lrih seemeart,=pota~ymanr awateme causemanmomr
ereasect ur-a Ihuxwanwurlrarneprlognmmcomognxng nomer
capab wuurmain msenasvamm rigueiV~ atl~pja usn iif a
sN1IYw~ crossineerrmaomeniwimbrassquencorwrauesamn
Rmouslysearamei lita~) in~mli eswesoeowt-scepnym*
29 and to -6 OC for xlink 29. In the case of xlink 29 the presence of chemical crosslinks
limits mobility of the polymer chains to form highly ordered arrangements; as a result the
extent of crystallization is suppressed even more. But not only decreasing changes in
melting temperatures are observed. An increase in values of the glass temperature
transition is evident in the series of poly6sp4gl (Tg= -77 OC), copolymer 29 (Tg= -76 OC)
and xlink 29 (Tg= -73 'C); however, the changes are by far more discrete than for the
melts.
1.5-
copol mer 29
1.0 xlink 29
0.5 -
g 0.0
i
-0.5
-1.0.
-150 -100 -50 0 50
Exo Down Temperature (C)
Figure 3-16. DSC curves ofpoly6sp4gl, copolymer 29 and xlink 29.
The thermal degradation behavior ofthermosets has been also investigated. Figure
3-17 and 3-18 shows traces of a few selected crosslinked materials in both an oxidative
and inert atmosphere respectively as percentage weight loss versus temperature in the 50-
800 OC range. Good thermal stability for the series ofthermosets presented in Figure 3-
17 and 3-18 is observed where essentially zero percent weight is lost upon heating from
50-300 OC, except for those based on 3sp glycols. The linear decomposition of these
materials implies that only one mechanism is responsible for their decomposition.
Thermal stability of these selected thermosets has been improved in both medium
when compared with the TGA results for homopolymers, which is expected for
crosslinked materials.
Thermal analysis results for the remaining thermosets are collected in Appendix.
loo. Air
100-
xlink 35
xlink 36
0xlink 38
xlink 39
S xlink 40
60
toi
50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800
Temperature (C)
Figure 3-17. TGA traces in an oxidative atmosphere.
Nitrogen
xlink 35
xlink 36
xlink 38
xlink 39
xlink 40
40
20-
0
50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800
Temperature (C)
Figure 3-18. TGA traces in an inert atmosphere.
3.4.2 Mechanical Properties
The mechanical properties of moisture cured thermosets were examined. As shown
in Figure 3-14 couple series of different composition networks were prepared and their
stress/strain profiles were analyzed in regard of their composition; Table 3-3 summarizes
data for these compositions.
In order to gain insight into the effect of the relative chain-internal/chain-end/soft
phase composition on mechanical properties a series ofbicomponent (chain-internal/soft
phase) thermosets were prepared and their properties investigated. Results from these
tests and stress/strain curves for these bicomponent thermosets have been placed in
appendix. A decrease in modulus (E) and tensile strength (TS) is observed as the length
of ethylene glycol increases in the series of thermosets with the same number of
methylene groups.
/
.Si-Me
0
n-5
--xlink 17
- link 18
- xlink 19
- xlink 20
12-
10-
8-
6-
4-
0.0 I 0.2 0.3 0.4 0.5 0.6 0.7
Strain (mmimm)
Figure 3-19a. Stress/strain curves.
xlink 22
--xlink 23
xlink 24
in /
1 1 2 3 4 5 6 7 8
Strain (mm/mm)
5% wt content
\
Me-Si'
-S
0.8
0.6-
0.4-
0.2
0.0
0.0 0.2 0.4 0.6 0.8
Strain (mm/mm)
Figure 3-19b. Stress/strain curves.
6-
4-
xlink 30
-- xlink 31
link 32
0 .i I I I
1.0 1.2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Strain (mm/mm)
1.0-
0.8-
S0.6-
" 0.4-
0.2-
I:
i
-Si-Me
0
xlink 36
xlink 37
xlink 39
xlink 40
Strain (m/
Strain (mm/mm)
4 5
Strain (mm/mm)
Figure 3-19c. Stress/strain curves.
Table 3-3. Physical properties of selected thermosets compositions.
Tensile Modulus Elongation Tensile Modulus Elongation
Strength (MPa) (%) Strength (MPa) (%)
(MPa) (MPa)
xlink 17 0.16 0.5 49 xlink 25 0.86 1.8 80
xlink 18 0.36 1.9 24 xlink 26 0.75 2.1 52
xlink 19 0.54 1.7 45 xlink 27 0.32 0.5 100
xlink 20 0.91 2.3 63 xlink 28 1.05 2.6 68
xlink 21 xlink 29 0.88 1.6 105
xlink 22 8.72 30.5 422 xlink 30 7.79 59.5 346
xlink 23 10.42 74.5 710 xlink 31 5.52 115.3 170
xlink 24 10.34 19.1 710 xlink 32 8.36 59.4 285
xlink 33
xlink 34
xlink 35
xlink 36
xlink 37
xlink 38
xlink 39
xlink 40
Tensile Modulus Elongation
Strength (MPa) (%)
(MPa)
0.67 2.0 49
0.55 1.7 47
0.38 0.8 72
4.79 1.4 507
1.86 1.4 298
8.60
6.82
54.6
33.3
335
258
I
A reversed trend is observed in the series ofthermosets where the number of
methylene groups increases and the length of oxyethylene units remains the same. As
was shown above thermal properties of homopolymers varies from totally amorphous to
semicrystalline as the length of carbon chains increases. These crystalline regions are
effective physical crosslinks that makes material stronger but brittle and observed trends
are typical for these kinds of structures. Moreover, these results once again confirmed
that introducing chain-end crosslinker evidently improves mechanical properties e.g.
xlink 12 containing only chain-internal crosslinker shows TS= 0.9 MPa and E= 4.3 MPa,
while for tricomponent xlink 36 TS= 4.8 MPa and E= 1.8 MPa. Although there is an
obvious decrease in the modulus, with value in the range typical for elastomers, the
augmentation of tensile strength is evident and higher values are desired since it
corresponds to the force at which material fractures.
Subsequently, a series of tricomponent networks were prepared and their
structure/property relationship was investigated. Ttricomponent materials shows similar
trends to bicomponent thermosets as presented by the stress/strain curves shown in
Figures 3-19. Results obtained for series of xlink 22, xlink 23, and xlink 24
unequivocally show how small changes in the composition such as one oxyethylene unit
can produce materials with flexible plastic or elastic behavior. Thermosets that contain
longer carbon chains between oxyethylene segments are tougher and shows typical
plastic behavior, while the ones with shorter carbon chains reveal their true elastic nature.
Figure 3-20 shows stress/strain curves of xlink 24, xlink 36, and xlink 37 with
exceptionally good properties for elastomers when compared to other thermosets.
Observed values of tensile strength for these three materials range from 2 to 10 MPa,
while elongation ranges from 300 to 700 %. The other thermosets show either elastic
behavior with low elongation, or plastic behavior with plastic plateau in the stress/strain
curves. In addition, fractured specimen of xlink 24 and xlink 36 presented in Figure 3-
20 show good elastic recovery of these materials. The 1 inch length of each specimen
used in the test did not change after the rapture.
1 -- xlink 24
lxlink 36
9-`
7-
Strain (mm/mm)
5-`
3-`
Figure 3-20. Stress'strain curves and picture of thermosets with good elastic properties.
3.5 Conclusions
Family of oligo(oxyethylene) c, co dienes monomers has been synthesized,
characterized, and subjected to olefin metathesis polymerization. Characterization of the
resultant polymers revealed a wide spectrum of properties dependant on both the length
of oxyethylene segments and also the length of carbon chains in between them.
Consequently, these monomers were copolymerized to obtain targeted linear "latent
reactive" elastomers, exposure to atmospheric moisture triggered crosslinking that led to
the formation of thermoset.
Thermal analysis and mechanical testing were performed on crosslinked materials
to asses their properties. Among fifty different compositions that were prepared many of
them revealed purely elastic behavior while some exhibit properties of flexible plastics.
These thermoset materials may be suitable for biological applications such as
semipermeable membranes or can be used in sealant or adhesive industry.
3.6 Experimental Methods
3.6.1 Materials
All reactions were conducted in flame-dried glassware under an argon atmosphere
unless otherwise noted. Reagents were purchased from Fischer or Aldrich and were used
as received unless otherwise noted. The first-generation Grubbs' catalyst used in this
study was graciously provided by Materia, Inc. 5-bromo-1-pentene, 8-bromo-1-octene,
1-undecene-11-ol, 11-chloro- 1-undecene, 11-bromo-1-undecene were synthesized
according to literature procedures.37
3.6.2 Instrumentation and Analysis
All 1H NMR (300 MHz) and 13C NMR (75 MHz) spectra were recorded on a
Varian Associates Mercury 300 spectrometer. Chemical shifts for 1H and 13C NMR were
referenced to residual signals from CDCl3 (1H = 7.26 ppm and 13C = 77.00 ppm) with
0.03% v/v TMS as an internal reference. Reaction conversions and relative purity of
crude products were monitored by thin layer chromatography (TLC) performed on EMD
silica gel coated (250 pm thickness) glass plates and 1H NMR. Developed TLC plates
were stained with iodine absorbed on silica or vanillin solution (5 wt% in 50mL ethanol
with 2mL H2S04) to produce a visible signature. Low and high-resolution mass spectral
(LRMS and HRMS) data were obtained on a Finnegan 4500 gas chromatograph/mass
spectrometer using the chemical ionization (EI) mode.
Fourier transform infrared (FT-IR) spectroscopy was performed using a Bio-Rad
FTS-40A spectrometer. Monomer was prepared by droplet deposition and sandwiched
between two KBr salt plates. Unsaturated polymer samples were prepared by solution
casting a thin film from tetrahydrofuran onto a KBr salt plate.
Differential scanning calorimetry (DSC) analysis was performed using a TA
Instruments Q1000 series equipped with a controlled cooling accessory (LNCS) at a
heating rate of 10 "C/min. Calibrations were made using indium and freshly distilled n-
octane as the standards for peak temperature transitions and indium for the enthalpy
standard. All samples were prepared in hermetically sealed pans (6-15 mg/sample) and
were run using an empty pan as a reference.
Thermogravimetric analysis (TGA) was performed either on a Perkin-Elmer TGA-
7 at 20 C/min or on TA Instruments TGA Q500 Series using dynamic scans under an
inert (nitrogen) and oxidative (air) atmospheres.
Gel permeation chromatography (GPC) was performed at 40 oC using a Waters
Associates GPCV2000 liquid chromatography system with an internal differential
refractive index detector (DRI) and two Waters Styragel HR-5E columns (10 microns
PD, 7.8 mm ID, 300 mm length) in HPLC grade tetrahydrofuran as the mobile phase at a
flow rate of 1.0 mL/minute. Injections were made at 0.05-0.07 % w/v sample
concentration using a 220.5 tl injection volume. Retention times were calibrated against
narrow molecular weight polystyrene standards (Polymer Laboratories; Amherst, MA)
selected to produce Mp and Mw values below and above the expected copolymer
molecular weight.
Mechanical tests were performed at the University of Florida facility by courtesy of
Dr. Brenan's research group, using an Instron model 1122 load frame upgraded with an
MTS ReNew system running MTS TestWorks 4 software and a 500 g load cell for the
experiments.
Tensile specimens were cut from the cured freestanding films using an ASTM
D1822-68 type L dogbone die. Each specimen was individually measured to account for
thickness variability in calculations. The grip separation for the samples was set at 25.4
mm and the width of the testing portion was set at 3.1 mm. Tensile measurements were
made according to ASTM D412-97 at rate of 2 inch/min.
3.6.3 Monomer Synthesis
Monomer 1 and 2 were synthesized according to a literature procedure75.
3.6.3.1 Synthesis of di(oxyethylene) a,o-di-4-pentenyl ether (3sp2gl) (86)
Monomer 3sp2gl was synthesized according to the procedure described in Chapter
2 for monomer 83. Monomer was purified using column chromatography (1:1 ethyl
acetate:hexane), followed by vacuum distillation over calcium hydride yielded 9.6 g
(58.7% yield) colorless liquid. The following spectral properties were observed: 1H
NMR (CDCl3): 6 = 1.64 (m, br, 4H), 2.10 (m, br, 4H), 3.47 (m, br, 4H), 3.59 (m, br, 8H),
4.95 (m, br, 4H), 5.77 (m, br, 2H). 13C NMR (CDC13): 6 = 28.82, 30.23, 70.15, 70.66,
70.71, 114.64, 138.30. FT-IR: (cm-1) 3077, 2938, 2865, 1641, 1448, 1415, 1350, 1323,
1296, 1247, 1120, 993, 911, 873. MS (70 eV): m/z = 243.1957 (M+H)+. (C14H2603)
(242.35): Calcd. C 69.38, H 10.81; Found C 69.23, H 10.89.
3.6.3.2 Synthesis of tri(oxyethylene) a,o-di-4-pentenyl ether (3sp3gl) (87)
Monomer 3sp3gl was prepare and purify using the same methodology as described
for 3sp2gl, yielded 8.5 g (65% yield) of colorless liquid. The following spectral
properties were observed: 1H NMR (CDCl3): 6 = 1.63 (m, br, H4), 2.07 (m, br, H4), 3.37
(m, br, H4), 3.57 (m, br, H12), 4.87 (m, br, H4), 5.74 (m, br, H2). 13C NMR (CDC13): 6 =
28.77, 30.22, 70.12, 70.62, 70.70, 114.65, 138.28. FT-IR: (cm-1) 3077, 2865, 1641,
1449, 1416, 1350, 1323, 1296, 1248, 1119, 993, 912, 872. MS (70 eV): m/z= 287.2230
(M+H)+. (C16H3004) (286.40): Calcd. C 67.10, H 10.56; Found C 67.05, H 10.61.
3.6.3.3 Synthesis of tetra(oxyethylene) a,wo-di-4-pentenyl ether (3sp4gl) (88)
Monomer 3sp4gl was prepare and purify using the same methodology as described
for 3sp2gl, yielded 10.5 g (72% yield) of colorless liquid. The following spectral
properties were observed: 1H NMR (CDC13): 6 = 1.64 (m, br, H4), 2.07 (m, br, H4), 3.43
(m, br, H4), 3.62 (m, br, H16), 4.93 (m, br, H4), 5.74 (m, br, H2). 13C NMR (CDC13): 8 =
28.78, 30.22, 70.10, 70.61, 70.69, 114.65, 138.28. FT-IR: (cm-1) 3076, 2866, 1640,
1449, 1416, 1350, 1323, 1296, 1248, 1118, 993, 912, 871. MS (70 eV): m/z= 330.2395
(M)+. (C18H3405) (330.44): Calcd. C 65.42, H 10.37; Found C 65.12, H 10.39.
3.6.3.4 Synthesis of tri(oxyethylene) a,to-di-7-octenyl ether (6sp3gl) (89)
Monomer 6sp3gl was prepare and purify using the same methodology as described
for 3sp2gl, yielded 9.2 g (55% yield) of colorless liquid. The following spectral
properties were observed: 1H NMR (CDC13): 6 = 1.30 (m, br, H12), 1.51 (m, br, H4), 1.98
(m, br, H4), 3.40 (m, br, H4), 3.53 (m, br, H12), 4.88 (m, br, H4), 5.70 (m, br, H2). 13C
NMR (CDC13): 6 = 25.87, 28.77, 28.86, 29.51, 33.63, 70.00, 70.55, 71.37, 114.10,
138.96. FT-IR: (cm-1) 3076, 2929, 2858, 1640, 1458, 1350, 1323, 1297, 1248, 1120,
995, 909, 726. MS (70 eV): m/z = 371.3168 (M+H)+. (C22H4204) (370.56): Calcd. C
71.31, H 11.42; Found C 71.33, H 11.53.
3.6.3.5 Synthesis of tetra(oxyethylene) a,o-di-7-octenyl ether (6sp4gl) (90)
Monomer 6sp4gl was prepare and purify using the same methodology as described
for 3sp2gl, yielded 6.7 g (58% yield) of colorless liquid. The following spectral
properties were observed: 1H NMR (CDC13): 6 = 1.31 (m, br, H12), 1.54 (m, br, H4), 1.98
