Surface grafting of polymers via living radical polymerization techniques; polymeric supports for combinatorial chemistry
By Nikolas Anton Amadeus Zwaneveld
A thesis submitted in fulfillment of the requirements for the degree of Doctor of Philosophy
School of Chemical Engineering and Industrial Chemistry University of New South Wales Sydney, Australia
March, 2006 ii iii
Abstract
The use of living radical polymerization methods has shown significant potential to control grafting of polymers from inert polymeric substrates. The objective of this thesis is to create advanced substrates for use in combinatorial chemistry applications through the use of γ-radiation as a radical source, and the use of RAFT, ATRP and RATRP living radical techniques to control grafting polymerization. The substrates grafted were polypropylene SynPhase lanterns from Mimotopes and are intended to be used as supports for combinatorial chemistry.
ATRP was used to graft polymers to SynPhase lanterns using a technique where the lantern was functionalized by exposing the lanterns to gamma-radiation from a 60Co radiation source in the presence of carbon tetra-bromide, producing short chain polystyrene tethered bromine atoms, and also with CBr4 directly functionalizing the surface. Styrene was then grafted off these lanterns using ATRP.
MMA was graft to the surface of SynPhase lanterns, using γ-radiation initiated RATRP at room temperature. It was found that the addition of the thermal initiator, AIBN, successfully increased the concentration of radicals to a level where we could achieve proper control of the polymerization.
RAFT was used to successfully control the grafting of styrene, acrylic acid and N,N’- dimethylacrylamide to polypropylene SynPhase Lanterns via a γ-initiated RAFT agent mediated free radical polymerization process using cumyl phenyldithioacetate and cumyl dithiobenzoate RAFT agents.
Amphiphilic brush copolymers were produced with a novel combined RAFT and ATRP system. Polystyrene-co-poly(vinylbenzyl chloride) created using gamma-radiation and controlled with the RAFT agent PEPDA was used as a backbone. The VBC moieties were then used as initiator sites for the ATRP grafting of t-BA to give a P(t-BA) brush that was then hydrolyzed to produce a PAA brush polymer. iv
FMOC loading tests were conducted on all these lanterns to assess their effectiveness as combinatorial chemistry supports. It was found that the loading could be controlled by adjusting the graft ratio of the lanterns and had a comparable loading to those commercially produced by Mimotopes. v
Declaration
I hereby declare that this submission is my own work and to the best of my knowledge it contains no material previously published or written by another person, nor material which to a substantial extent has been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis.
I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged.
______
Nikolas Zwaneveld vi
Acknowledgements
It has been an epic, oh such an epic… and there are many people who I owe so much to.
Firstly I must thank Julia and my family; Solara, Ansel, Lisa and Andrew, because without their love and support I would never have made it this far.
Next I must thank all those who have helped me academically to do this thesis; I need to thank Prof. Tom Davis, Christopher Barner-Kowollik and Hans Heuts who have been my supervisors throughout this thesis. Beyond my supervisors I have had the help of three great people who have poured hours if not weeks or months of their lives into helping me with my work; Tony Granville, Leonie Barner, Francoise Isaure. Without your help there is absolutely no way I would ever have finished this. I owe you all an infinite debt. My thanks goes especially out to Tony who not only edited the entirety of my thesis but also solved so many theoretical problems, helped with so much of my experiments and showed me the way into the light at the end of the tunnel. I would also like to thank Martina Stenzel, who always showed an amazing amount of interest in my work and kept me inspired.
I truly need to thank Mimotopes for their sponsorship of this project and the people there who have helped me out so much, Senake Perera, Francesca Ercole, Nick Ede.
I need to also thank all those people from the School of Chemical Engineering and Industrial Chemistry who not only would seem to make the impossible happen for me on short notice but also have become fantastic friends of mine; A/Prof. Mike Brungs, Prof. Rob Burford, Phil McAuley, John Starling, Phil and Paul from the workshop and many others. Especially I’d like to thank Istvan (Steve) who has become a great friend of mine and has helped me so many times throughout my thesis that I can’t even begin to count. Not only can he cook and brew, but the raw meat that he serves up is second to none. vii
In my many years at CAMD I need to thank all my friends who have come and gone. Especially I’d like to thank the Frenchies; Seb, Julien, Arnaud “bonjour salaud” Favier, Francoise who have always been an unending source of good humor. You have shown me that life should be a proper mix of work, soccer, beer and late nights. The Swedish; Niklas, Ingmar, Camila and Daniel thanks for teaching me about licorice with ammonia, Christmas crayfish, schnapps, and snus. The absent Dutchmen, Hans Heuts and Almar Postma who have been a great help and better friends despite being 1000s of kilometers away for most of my thesis. Mikey Whittaker who not only helped me with my work but who extended a hand in friendship that was always available. Thanks to Antonio and Nathan all those who would help me analyze HLUG’s over a cup of coffee. Thanks to the old-school boys Lachlan Yee, Dr Dave, Mr Dave, Evan, Henry, Joan and Simon who were around when it all began. Maribel our little Mexican girl, who I don’t think has ever not had a smile on her face the whole of her life. Nicole and Philipp Vana with whom I have developed a friendship that will last forever. Luca who showed me that Italian largely consists of random swear words strung together into a sentence.
To all my friends who concertedly tried to distract me from ever finishing, specifically; Gareth ‘just one more…’ Milton, Romily Webster, Daniel ‘in a cave’ Luong Van, Nathan Allan, Shawn ‘the wookie’ Sijnstra, Clare Hayward, Iain Findlay, Mike ‘I’ve got the full set’ Wheatley, Helen Vafiads, Angella Barrett, Shane Cox, Stefan Kaufman, all the doomies, Domus Luni, Lyon La Foret, the cricket boys and all those I’ve forgotten to mention.
A large thank you goes to those who corrected my engrish in my thesis, it wood be much worze if you not me correct. Thanks to Tom Murtagh, Rom, Dan, Shawn, Solara and of coarse Tony for fixing it all up goodly like and removing those words I just made up.
Everyone else who has helped me out during my time, I thank you all. It’s finished, it’s done…and now I can’t brain anymore. viii
Table of Contents
Abstract...... iii Declaration...... v Acknowledgements...... vi Table of Contents ...... viii Abbreviations ...... xiv 1. Introduction...... 1 1.1. Aims of the investigations...... 1 1.2. Outline of the thesis ...... 1 1. Introduction...... 1 2. Theoretical background ...... 1 3. Analysis techniques and synthesis of reagents...... 1 4. Graft polymerization from SynPhase lanterns using ATRP ...... 2 5. Gamma-radiation initiated reverse ATRP grafting polymerization ...... 2 6. Reversible Addition-Fragmentation Chain Transfer Polymerization of styrene...... 3 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers...... 3 8. Conclusions...... 4 9. Appendix 1...... 4 10. Appendix 2...... 4 1.3. Publications...... 5 2. Theoretical Background ...... 6 2.1. Solid phase organic synthesis and combinatorial chemistry...... 6 2.1.1. Factors affecting solid phase organic synthesis...... 9 2.1.2. The SynPhase lantern production process ...... 10 2.2. Graft polymerization ...... 10 2.2.1. Chemical attachment: “Grafting to” technique...... 11 2.2.2. Gamma radiation ...... 11 2.2.3. Oxygen functionalization...... 12 2.2.4. UV irradiation...... 13 2.2.5. Electron beam ...... 14 2.2.6. Plasma...... 15 2.3. Free radical polymerization...... 15 2.3.1. The mechanism of free radical polymerization ...... 16 2.3.2. Kinetics of Free Radical Polymerization...... 18 2.4. The evolution of living/controlled polymerization techniques ...... 21 ix
2.5. Methods in living radical polymerization ...... 24 2.5.1. Inferter...... 25 2.5.2. Nitroxide Mediated Polymerization (NMP) ...... 27 2.5.3. Atom Transfer Radical Polymerization (ATRP)...... 28 ATRP and the persistent radical effect...... 30 2.5.4. Reversible addition fragmentation chain transfer polymerization (RAFT)...... 32 The RAFT Mechanism ...... 32 2.6. The kinetics of living polymerization ...... 34 2.7. Radiation polymerization...... 36 2.7.1. Mechanism of radiation radical initiation...... 37 2.7.2. Chemical effects on radiation polymerization...... 39 2.7.3. Side reactions in radiation polymerization ...... 40 2.7.4. Living radical polymerization under γ-radiation ...... 41 2.8. Control of graft polymerization through the use of living radical polymerization...... 42 2.9. Conclusion ...... 45 3. Analysis techniques and synthesis of reagents ...... 47 3.1. 60Co γ-radiation facilities ...... 47 3.1.1. Introduction to 60Co radiation source at UNSW ...... 47 3.1.2. Calibration of γ-radiation source...... 49 Materials ...... 50 Experimental method...... 50 Results...... 51 3.1.3. Elimination of Radiation Bias in 60Co radiation source...... 53 3.2. Analytical techniques and Instrumentation...... 56 3.2.1. Gel permeation Chromatography (GPC) / Size Exclusion Chromatography (SEC)...... 56 Conventional GPC with primary calibration...... 57 Conventional GPC with universal calibration...... 57 Molecular weight sensitive detectors...... 57 GPC systems used as part of this thesis ...... 58 3.2.2. Nuclear Magnetic Resonance spectroscopy (NMR) ...... 58 Fourier transform NMR ...... 59 Effects on NMR spectra...... 59 3.2.3. Infrared spectroscopy...... 60 Attenuated Total Reflectance – Fourier Transform Infrared spectroscopy (ATR-FTIR) ...... 60 3.2.4. FMOC loading test...... 61 FMOC loading test of polystyrene based polymers...... 61 FMOC loading test of poly(acrylic acid) based polymers ...... 63 x
3.2.5. Gravimetric calculations and grafting ratio...... 64 3.3. Synthesis of RAFT agents...... 65 3.3.1. Synthesis of 1-phenylethyldithiobenzoate (PEDB) ...... 65 3.3.2. Synthesis of cumyl dithiobenzoate (CDB) ...... 66 3.3.3. Synthesis of 2-(2-cyanopropyl) dithiobenzoate (CPDB) ...... 67 3.3.4. Synthesis of cumyl phenyldithioacetate (CPDA) ...... 68 3.4. Synthesis and purification of ATRP related compounds ...... 69 3.4.1. Purification of CuBr...... 69
3.4.2. Synthesis of Me6TREN...... 69 4. ATRP graft polymerization from SynPhase lanterns...... 71 4.1. Aim...... 71 4.2. Introduction and theory...... 71 4.2.1. Surface functionalization through γ-radiation and carbon tetrabromide chain transfer ... 73 Mechanism of ATRP grafting from functionalized surfaces ...... 76 4.3. Solution ATRP...... 78 4.3.1. Experimental method...... 78 4.3.2. Solution ATRP results ...... 79 Effect of catalyst concentration of ATRP polymerization of styrene ...... 80 Effect of initiator concentration of ATRP of styrene...... 82 Effect of temperature on ATRP polymerization of styrene ...... 85 GPC analysis of ATRP polymers...... 88 4.3.3. Solution polymerization of other monomers...... 90 4.3.4. Conclusions ...... 91 4.4. Reinitiation of solution ATRP polymers ...... 92 4.5. ATRP controlled grafting off functionalized SynPhase lanterns...... 95 4.5.1. Determination of Cs for CBr4 at room temperature under constant γ-irradiation ...... 95 4.5.2. Functionalization and grafting of tethered Br functionalized lanterns ...... 97 The effect of CBr4 concentration on the functionalization of SynPhase lanterns...... 98 Effect of functionalization time on lantern grafting...... 103 4.5.3. Direct functionalization of lanterns in the absence of monomers ...... 106 4.6. FMOC-β-Alanine loading determination of ATRP grafted lanterns ...... 111 4.7. Conclusion ...... 113 5. γ-radiation initiated reverse ATRP graft polymerization ...... 115 5.1. Mechanism of RATRP polymerization using γ-radiation...... 117 5.2. Experimental method...... 121 5.3. RATRP of styrene using a thermal initiation...... 122 5.4. γ-radiation initiated RATRP grafting of styrene...... 123 xi
5.4.1. Mechanistic implications of the addition of AIBN to RATRP controlled systems ...... 125 5.5. γ-radiation initiated RATRP grafting of MMA...... 128 5.5.1. Molecular weight analysis of γ-radiation initiated RATRP polymers ...... 132 5.5.2. Effect of catalyst complex concentration on RATRP grafting of MMA ...... 133 5.5.3. Effect of dose rate on RATRP grafting of MMA...... 136 5.5.4. Effect of solution radical concentration on RATRP grafting of MMA ...... 140 5.6. Conclusion ...... 143 6. γ-radiation initiated RAFT grafting of styrene ...... 144 6.1. Introduction...... 144 6.1.1. Retardation...... 144 6.1.2. Molecular Architectures using RAFT...... 147 6.1.3. The mechanism of radiation grafting using RAFT polymerization ...... 150 6.2. Experimental method...... 152 Materials ...... 152 Grafting Experiments...... 152 Analysis ...... 153 6.3. Results and discussions...... 153 6.3.1. γ-initiated RAFT mediated graft polymerization of styrene onto SynPhase lanterns...... 153 6.3.2. Effect of RAFT agent concentration on γ-radiation grafting...... 156 6.3.3. Effect of dose rate on radiation grafting of styrene...... 157 6.4. Analysis of molecular weight distributions...... 158 6.5. FMOC-loading of grafted Lanterns ...... 160 6.6. Conclusions...... 165 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers initiated by γ-radiation...... 166 7.1. Effects of polymerization rates on RAFT mediated graft polymerization....168 7.1.1. Kinetics of RAFT controlled grafting ...... 171 7.2. Thermally initiated RAFT polymerization of N,N’-dimethylacrylamide.....172 7.2.1. Experimental Method ...... 172 Materials ...... 172 Polymerization procedure ...... 172 7.2.2. Thermal polymerization of DMA using CDB ...... 173 Determination for optimum conditions for thermal polymerization of DMA...... 175 Effect of RAFT agent concentration of RAFT polymerization of DMA...... 176 Effect of monomer concentration on thermally initiated RAFT polymerization of DMA...... 178 Effect of temperature on the RAFT mediated polymerization of DMA ...... 180 DMA grafting onto SynPhase lanterns using γ-radiation...... 183 xii
7.2.3. Experimental Method ...... 183 7.2.4. Results and discussions ...... 184 Effect of RAFT agent concentration on γ-radiation grafting of DMA...... 186 Effect of dose rate on radiation grafting of DMA onto SynPhase Lanterns...... 189 7.2.5. ATR-FTIR analysis of PDMA grafted surfaces ...... 191 7.2.6. GPC analysis of homopolymers from grafting polymerization reactions ...... 193 7.2.7. Conclusions ...... 195 7.3. RAFT mediated polymerization of acrylic and methacrylic acid ...... 196 7.3.1. Experimental method...... 196 Materials ...... 196 7.4. Thermally initiated polymerization of MAA and AA...... 197 Experimental method...... 197 7.4.1. Thermally initiated, RAFT mediated polymerization of MAA ...... 197 Selection of RAFT agent for AA polymerization ...... 199 7.5. RAFT controlled radiation grafting of AA onto SynPhase lanterns...... 201 7.5.1. Experimental method...... 201 7.5.2. γ-radiation initiated grafting of AA ...... 202 Effect of RAFT agent concentration on γ-radiation grafting of acrylic acid...... 206 Effect of γ-radiation dose rate on the grafting of acrylic acid...... 212 7.6. ATR-FTIR analysis of grafted PAA polymers ...... 214 7.7. FMOC-β-Alanine loading determination of PAA grafted lanterns ...... 217 7.8. Living behavior of γ-radiation produced RAFT polymers...... 219 7.8.1. Reinitiation of RAFT lanterns and creation of grafted block co-polymer substrates...... 220 7.9. Novel amphiphilic brush copolymers produced through combined RAFT and ATRP polymerization ...... 223 7.9.1. Solution polymerization of tert-butyl acrylate using ATRP...... 225 7.9.2. Thermally initiated solution amphiphilic brush copolymers ...... 230 Polymerization of solution brush backbone...... 230 Polymerization of side chains on solution brush...... 232 Hydrolysis of t-BA side chains on solution brush polymers...... 235 7.9.3. Surface-grafted amphiphilic brushes through combined RAFT and ATRP polymerization 238 Experimental Method...... 239 Results and discussions...... 240 7.10. Conclusions...... 247 8. Conclusions...... 249 8.1. Further work...... 252 9. Appendix 1...... 253 xiii
10. Appendix 2...... 257 10.1. Polymerization of methyl acrylate ...... 257 10.2. Polymerization of MMA using Me6TREN...... 259 10.3. Polymerization of MMA using PMDETA...... 261 10.4. Polymerization of tert-butyl acrylate ...... 263 11. References ...... 265 xiv
Abbreviations
AIBN 2,2’-Azobis(isobutyronitrile) AA Acrylic acid ATR Attenuated total reflectance ATRA Atom transfer radical addition ATRP Atom transfer radical polymerization BHT 2,6-di-tert-butyl-4-methylphenol bipy bipyridine BPO Benzyl peroxide CCT Catalytic chain transfer CDB Cumyl dithiobenzoate CMS Chloromethylstyrene CPDB Cyanopropyl dithiobenzoate CROP Cationic ring opening polymerization DCM Dichloromethane DMA N,N'-Dimethyl acrylamide DMAc N,N'-Dimethyl acetamide DMF N,N’-Dimethyl formamide DMSO Dimethylsulfoxide DPn Degree of polymerization DSC Differential scanning calorimetry DVB Divinyl benzene E-2-BIB Ethyl-2-bromoisobutyrate ESR Electron spin resonance FID Free induction decay FMOC 9-fluorenylmethoxycarbonyl FTIR Fourier transform infrared (spectroscopy) GMA Glycidyl methacrylate GPC Gel permeation chromatography HCl Hydrochloric acid xv
HEMA 2-Hydroxy ethyl methacrylate Iniferter initiator transfer terminator IR Infra red LDPE Low density polyethylene MA Methyl acrylate MAA Methacrylic acid MADIX Macromolecular design via interchange of xanthate MALDI- Matrix assisted laser desorption ionisation time-of-flight mass TOF-MS spectrometry
Me6TREN tris[2-(dimethylamino)-ethyl]amine MeOPEGMA Poly ethylene glycol methyl ether methacrylate MMA Methyl methacrylate NMP Nitroxide mediated polymerization NMR Nuclear magnetic resonance P(t-BA) poly(tert-butyl acrylate) PAA poly(acrylic acid) PDI Poly dispersity index PDMA Poly(N,N'-dimethylacrylamide) PDMS Polydimethylsiloxane PE polyethylene PEG poly(ethylene glycol) PEGMA Polyethyleneglycol methacrylate PEPDA Phenylethylphenyl dithioacetate PEPDB Phenylethylphenyl dithiobenzoate PGMA Poly(glycidyl methacrylate) PMAA Poly(methacrylic acid) PMDETA (N-[2-(Dimethylamino)ethyl]N,N',N'-trimethyl-1.2-ethanediamine) PMMA poly(methyl methacrylate) PP polypropylene PRE Persistent radical effect PS Polystyrene PVC Polyvinylchloride xvi
RAFT Reversible addition fragmentation chain transfer RATRP Reverse atom transfer radical polymerization ROP Ring opening polymerization RT Room temperature SEC Size exclusion chromatography SPOS Solid phase organic synthesis t-BA tert-butyl acrylate TEA triethyl amine TEMPO 2,2,6,6-tetramethylpiperidinyl-1-oxyl TFA Trifluoroacetic acid THF Tetrahydrofuran UV Ultra violet VBC Vinyl benzyl chloride 1. Introduction 1
1. Introduction
1.1. Aims of the investigations
The aim of this project is to use advanced living free radical polymerization techniques to control radiation-initiated graft-polymerization to create of novel and advanced substrates for the use in Solid Phase Organic Synthesis (SPOS) and combinatorial chemistry.
This thesis concerns several different living radical polymerization techniques that were implemented to control the radiation induced grafting of various monomers to a polypropylene substrate in the form of Mimotopes SynPhase lanterns. The living radical techniques that were used included reversible addition-fragmentation transfer polymerization (RAFT), atom transfer radical polymerization (ATRP), and reverse atom transfer radical polymerization (RATRP). These were used to graft a number of monomers of significance and use in SPOS.
1.2. Outline of the thesis
1. Introduction
Outlines the objectives of the thesis and outlines the layout of the thesis.
2. Theoretical background
This chapter gives a brief introduction to the fundamentals and concepts of free radical polymerization systems, living radical polymerization, radiation grafting, and combinatorial chemistry.
3. Analysis techniques and synthesis of reagents
Theoretical details are described for the experimental techniques and analytical equipment used throughout the research for this thesis. Most of the experimental 1. Introduction 2
explanations are presented in the relevant experimental chapters (see Chapters 3 to 6). This chapter contains details of equipment, experimental techniques, and reagent synthesis that are applicable to multiple chapters in this thesis.
4. Graft polymerization from SynPhase lanterns using ATRP
This work presents the use of ATRP to graft from specially functionalized SynPhase lanterns. SynPhase lanterns were functionalized by exposure to γ-radiation in the presence of CBr4. This led to a radical transfer reaction that left the polymeric surface functionalized with bromine atoms. These bromine atoms were then used as initiation sites for ATRP graft polymerization. The Cs of CBr4 at room temperature under γ- radiation was determined and this was used as a basis to functionalize a series of lanterns using styrene, methanol and toluene as solvents for the functionalization. It was found that using styrene as a solvent for the functionalization resulted in short chain, bromine-terminated polymer grafts that were effective as an initiator for ATRP graft polymerization. The lanterns that were directly functionalized in methanol and toluene were also effective as ATRP initiators. However it was found that toluene had about 50% fewer accessible initiation points than those lanterns produced in styrene and methanol. The ATRP system that was used consisted of CuBr, tris(2- dimethylaminoethyl)amine (or Me6TREN) and styrene as a monomer and was polymerized at 90oC under nitrogen. The effect of temperature, catalyst concentration, and initiator concentration were established for the polymerization system by using a solution initiator; once this was found the effect of CBr4 concentration on the functionalization reaction and the effect that functionalization time had on later ATRP polymerization reactions were determined. The lanterns were also assessed for their applicability as combinatorial chemistry supports by conducting loading tests using FMOC-β-alanine.
5. Gamma-radiation initiated reverse ATRP grafting polymerization
The γ-radiation initiated polymerization and γ-radiation induced graft polymerization were controlled using RATRP. It was found that styrene did not propagate fast enough for proper control of the polymerization but MMA proved to polymerize at a faster rate thus allowing full analysis of this system. It was found that the addition of a thermal 1. Introduction 3
initiator AIBN increased the rate of polymerization by augmenting the radical flux of the system. The system was successful in conducting RATRP of MMA under γ- irradiation with the addition of AIBN as an accelerator. The effect of dose rate, AIBN concentration, catalyst concentration, and progression of the polymerization over time were determined for both solution polymerization and graft polymerization. The system comprises of CuBr2, Me6TREN as a ligand, AIBN was added to increase the radical flux, and DMF was used as a solvent. A correlation was observed between the solution homopolymer and the grafted polymer, and the lanterns were assessed for their applicability as combinatorial chemistry supports by conducting loading tests using FMOC-β-alanine.
6. Reversible Addition-Fragmentation Chain Transfer Polymerization of styrene
The RAFT-controlled γ-radiation induced grafting of styrene from SynPhase lanterns was conducted using the room temperature RAFT agent cumyl phenyldithioacetate. The grafting kinetics appeared to show two different kinetic regimes. It was determined that this was caused by firstly grafting directly to the PP lantern and then later off the backbone of the PS grafts. The effect of RAFT agent concentration and dose rate were determined, and the graft polymerization was compared to uncontrolled γ-radiation induced styrene grafting. A correlation between the solution homopolymer and the grafted polymer, and the lanterns were assessed for their applicability as combinatorial chemistry supports by conducting loading tests using FMOC-β-alanine.
7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers
The RAFT controlled γ-radiation induced grafting of acrylic acid and dimethyl acrylamide from SynPhase lanterns was conducted using cumyl dithiobenzoate as a RAFT agent. Cumyl dithiobenzoate (CDB) was chosen as a RAFT agent as it is a slow- fragmenting RAFT agent at room temperature and thus would slow the polymerization of acrylic acid (AA) and N-N’-dimethylacrylamide (DMA) to an extent that cross- linking of the polymers wouldn’t occur. The effect of RAFT agent concentration and 1. Introduction 4
dose rate were determined and the graft polymerization was compared to uncontrolled γ-radiation induced grafting. A correlation between the solution homopolymer and the grafted polymer and the lanterns were assessed for their applicability as combinatorial chemistry supports by conducting loading tests using FMOC-β-alanine.
8. Conclusions
This chapter summarizes the major conclusions and summaries of the experiments conducted in this thesis.
9. Appendix 1
This appendix contains supplementary data for the calibration and modification of the 60Co radiation facilities.
10. Appendix 2
This appendix contains graphs for experiments that were conducted into ATRP controlled polymerization of acrylic monomers for Chapter 4. 1. Introduction 5
1.3. Publications
The following publications, arising from the work of this thesis have been published or appeared at conferences:
"Reversible Addition Fragmentation Chain Transfer (RAFT) Graft Polymerization of Styrene: Solid Phases for Organic and Peptide Synthesis" Barner, L.; Zwaneveld, N.; Perera, S.; Pham, Y.; Davis, T. P. J. Pol. Sci. Part A: Pol. Chem., 2002, 40, 4180
"Gamma Radiation Induced Grafting onto Solid Surfaces using the RAFT Process" Barner, L.; Zwaneveld, N.; Barner-Kowollik, C.; Davis, T. P., 25th APS, Armidale, NSW, February, 2002
"Controlled Radical Graft Polymerization Methods and Its Application to Solid Phase Combinatorial Chemistry" Zwaneveld, N.; Barner, L.; Heuts, H.; Perera, S.; Davis, T. P., 25th APS, Armidale, NSW, 2002
“Novel grafted surfaces via ATRP for combinatorial chemistry applications” Zwaneveld, N, Barner, L, Heuts, H, Perera, S, Davis, T. P., 26th APS, Noosa, 2003 2. Theoretical Background 6
2. Theoretical Background
This chapter contains the fundamentals needed to understand free radical polymerization and subsequently follow the kinetics and experiments in this thesis. Each experimental chapter contains the necessary theoretical background to understand the discussions.
2.1. Solid phase organic synthesis and combinatorial chemistry
The field of solid-phase organic synthesis (SPOS) was introduced in 1963 when Merrifield published this paper1 outlining the use of cross-linked polystyrene for peptide synthesis. Merrifield’s resins are polystyrene (PS) lightly cross-linked with divinyl benzene with the phenyl group on the styrene molecule chloromethylated to act as an anchor for peptide synthesis. This first discovery then allowed for the development of combinatorial chemistry and parallel synthetic methods which now essentially allows for the concurrent synthesis of hundreds of organic or biochemical molecules.2,3 The first instance of solid phase organic synthesis was in 1984 when Geysen et al.4,5 used a “multi-pin method” to demonstrate the synthesis of large sets of peptides.
Solid phase organic synthesis (SPOS) is the technique of synthesizing chemical compounds from a solid substrate. These compounds are then cleaved from the surface to yield a pure product. These methodologies are the basis of combinatorial chemistry and allow for the production of vast libraries of related compounds and large scale assay techniques for pharmaceutical development.
While there have been some advancements in the field, a significant amount of solid- phase synthesis today still uses these “Merrifield resins” and the similar “Wang resins”.6 Wang resins consist of a cross-linked polystyrene resin that is modified with 4- hydroxylbenzyl alcohol to increase hydrophilicity. There are several advanced products on the market that offer significant improvements and advantages over the resins. One 2. Theoretical Background 7
of these products is the SynPhase lantern product produced by Mimotopes.7,8 This product is essentially a solid polypropylene (PP) substrate onto which polystyrene has been grafted.
Figure 2.1: Polypropylene SynPhase lanterns used for the grafting experiments
The SynPhase lanterns are a significant advancement in supports for combinatorial chemistry and have had wide industrial uptake for the development of chemical peptide and pharmaceutical libraries. The SynPhase lanterns come in several forms depending on their anticipated usage. The lanterns come in 3 sizes with the standard ‘L Series’ lantern having a 15 μmol loading, while the ‘D series’ or double-lanterns are twice the size and have a loading of about 35 μmol, and the largest ‘A-series’ lanterns have a loading of about 75 μmol. These commercially available lanterns can come with either a hydrophobic polystyrene graft or a polyamide graft for hydrophilic applications, and a wide selection of linkers depending on the end use. The lanterns are specifically shaped to maximize available surface area for synthesis and also to easily allow for robotic automation in the development stage.
The advantages of the lanterns over traditional resins are listed below in Figure 2.2. 2. Theoretical Background 8
SynPhase lantern property Advantages over resin Surface polymer structure only • faster transfer of reagents to requires solvent wetting reaction sites • vigorous agitation is not required Low gelling tendency of the surface • easy removal of unreacted polymer reagents and by-products Modular design with defined loading • doesn’t require weighing Macroscopic, modular support • easy manual handling • no filtration • simple washing techniques • no resin shattering • no fines remaining in vessels • easily accommodates tags (colored or radio frequency transponders) • allows rapid reaction optimization Each lantern series has same loading • chemistry translates between L-, per unit surface area and per unit D- and A- series lanterns without working volume re-optimization Closely controlled manufacturing • minimal batch to batch or within process batch variation Optimized surface and shape • loading per unit working volume similar to resin (0.15 mmol/mL). • uniform reaction rates • minimizes working volumes • rapid drainage of reagents • compatible with automated synthesizers Figure 2.2: Summary of SynPhase Lanterns over traditional resin technologies
However, while the SynPhase lanterns offer significant advantages over resins there are still improvements that can be made. Advanced polymeric systems offer improved solvent compatibility through block copolymers, improved kinetics through polymeric control and the ability to control loading though easy adjustment of the surface grafted polymer. The problems we aim to overcome with this system include the inherent variability of radiation grafting and the need to frequently recalibrate to compensate for deterioration of the radiation source. There has also been a move towards more advanced polymeric systems to help alleviate these requirements and advance the lanterns to the next generation of supports. 2. Theoretical Background 9
2.1.1. Factors affecting solid phase organic synthesis
While SPOS has many advantages over traditional solution techniques such as improved sample handling, purer product, ability to drive reactions to completion with excess reagents, reduced side reactions and a significantly increased synthesis rate of compounds per chemist there are still some factors that have to be taken into account when conducting SPOS.
The dominant factor in SPOS efficiency is the choice of solvent. The compatibility of solvent, reagents and the support must be balanced. The solvent support interactions can be easily characterized by the degree of swelling that is achieved in the solvent used. Assuming a “surface depth” of 0.1μm then approximately 99% of active sites are inside the substrate and thus it makes sense that SPOS reactions are dependent on how well reagents can get to the reaction sites.9 It has long been known that the size of the resin particle makes a difference in SPOS kinetics10 and additionally it has been shown that the degree of cross-linking decreases the rate of reaction for SPOS by reducing the degree of swelling.11 One factor that is often forgotten is the reagent-polymer interaction. Li and Yan9 showed that there is also a strong effect of reagent/polymer compatibility has a strong effect on the reaction kinetics. As a general rule, a small degree of cross-linking results in faster kinetics, although a notable exception is when a poor solvent is used and negative swelling occurs. In that case a highly cross-linked polymer that maintains its structural integrity performs better.12
Clearly the interaction of the polymeric support and solvent is very important. Even with this knowledge the most common solid phase support in industrial use is still PS cross-linked with divinyl benzene (DVB)13, which is quite hydrophobic. A number of more hydrophilic supports have been developed. One of the more popular polymers is a PS resin cross-linked with polyethylene glycol (PEG); this has been developed as TentaGel14 and has shown improved kinetics in several solvents. Mimotopes have also developed a high performance polyamide based lantern for hydrophilic environments.15 2. Theoretical Background 10
2.1.2. The SynPhase lantern production process
Mimotopes produce their SynPhase lanterns through a radiation grafting process.
Blank PP lanterns are placed in the monomer solution and polymerized in the γ-source for the required time. Once the polymerization is complete the lanterns are washed in dichloromethane (DCM) to remove homopolymer. These washed lanterns are then functionalized with linkers and other groups depending on the future application.
2.2. Graft polymerization
There are many methods of grafting polymers to a substrate. As mentioned earlier, the SynPhase lanterns are produced by γ-radiation grafting of polymers and this thesis concentrates on this γ-radiation grafting onto polymeric substrates. However, there are many other methods available to graft polymers to a substrate.
Graft polymerization is the process of producing block copolymers, where the B block of the polymer originates off the polymeric backbone of the A block, such as shown here:
B B B B B B AAA A A A A A A B B B B
The term ‘graft polymer’ can refer to solution polymers of this structure, surface grafted polymers and modification of insoluble or cross-linked polymers. For the purposes of this thesis we will concentrate on grafting from insoluble substrates.
There are basically two ways of creating polymer grafts; the “Grafting to” technique and the “Grafting from” technique. The “Grafting to” technique involves the attachment of preformed polymers to a backbone polymer and the “Grafting from” technique involves growing the polymer chain off the backbone.
2. Theoretical Background 11
Kato et al.16 have written an extensive review into polymeric grafting which covers these techniques in greater detail. An outline of the most relevant grafting techniques is given in this section.
2.2.1. Chemical attachment: “Grafting to” technique
The simplest method of grafting involves the chemical attachment of preformed polymers to the surface and is a common method for the creation of brushes17, self assembled monolayer18, and similar graft systems. This technique is based on the reaction of an appropriate polymer end group with an appropriate substrate such as carbon black, graphite, gold and silicon or reactive substrates as well as some polymeric substrates.
One example of grafting to polymeric substrates is from Minko et al.17 who produced brushes from a polytetrafluoroethylene substrate, where the grafts were achieved by modifying the substrate with oxygen and ammonia plasmas to yield hydroxy and amino functionalities, which were in turn used to attach polymers from solution.
This technique has the disadvantage that it is limited to polymers and substrates that have compatible functionalities. The scope of polymeric grafts that can be produced though this technique is only limited to the attachment chemistry, and it is this chemistry that limits its application to combinatorial chemistry. Production of a stable graft with a chemical attachment that won’t be reacted by the aggressive SPOS cleaving conditions (usually very strong acid or base) is difficult. Additionally, interference from unreacted substrate sites can cause problems with chemical synthesis.
2.2.2. Gamma radiation
γ-radiation grafting is one of the most common commercial methods for grafting and modifying polymers. It has been shown that γ-polymerization produces a mixture of ions and radicals and that these undergo polymerization under different conditions. As 2. Theoretical Background 12
far back as the 1930s interest in polymers had started to develop and γ-radiation was found to induce polymerization in some simple monomers.19 However it was only in the 1950s that the first major works and patents relating to radiation grafting started to appear.20
The advantages of radiation grafting include: • the universality of the process which applies to many monomers and can graft off polymers of any nature (films, powders, fibers etc) • wide temperature range • active initiation sites are generated throughout the bulk material • there is no need for chemical additives and catalysts to initiate the graft
2.2.3. Oxygen functionalization
When a solid is put in a high energy radical source such as plasma, γ-radiation or UV radiation in the presence of oxygen a number of long-lived radicals, peroxides and hyperoxides are formed. These can be used to initiate polymerization for some time after the exposure. The solid is then immersed in monomer solution and the polymerization is allowed to occur. Most notably this is a common technique for plasma polymerization where it is also possible to use either oxygen or nitrogen plasmas to produce either stable radicals or groups that will readily degrade to radicals that can 21 initiate polymerization.
The mechanism for decay of the peroxide and hyperoxide radicals is shown in Scheme 2.1
heat peroxide O O 2 O
heat hyperoxide O OH O O OH
Scheme 2.1: Mechanism of radical formation from peroxide and hyperoxides from solid surfaces21 2. Theoretical Background 13
This mechanism proceeds through radicals that are trapped in the substrate, although this is dependent on the ability of the substrate to form stable long-lived radicals and 22 ions. Radicals can also be generated through thermal degradation of peroxides and hyperoxides that are generated by the reaction of the substrate with oxygen. These peroxide and hyperoxide radicals require heating to decay to polymerizing radicals.23
In general, graft yields are too low to find application in SPOS and this technique mainly finds application in surface modification. Another problem with the use of stable radicals and oxygen derived radicals is that significant degradation of the substrate polymer can occur, and this negatively affect further application of the materials.24
2.2.4. UV irradiation
High energy UV radiation has been found to generate surface radicals on a variety of solid substrates and has been widely studied for polymeric grafting.16 Polymerization usually is conducted in the presence of a photo sensitizer such as benzophenone16 however some monomers with an appropriate functionality, such as the phenyl group in styrene, can initiate without a sensitizer.25
Living UV grafting polymerization has been conducted by several groups. Yang et al.26 developed a novel system using benzophenone to graft acrylic acid and methacrylic acid off low density polyethylene (LDPE), as shown Scheme 2.2.
Lee et al.27 also grafted a variety of monomers to a modified glass surface using benzyl N,N-diethyldithiocarbamate as an inferter. The basis of this was chloromethyl styrene (CMS) to grafted glass plates. From this graft they converted the chlorine atoms to another inferter groups and used these to create brushes of the following monomers: N,N’-dimethylacrylamide (DMA) polyethylene glycol methacrylate (PEGMA), methacrylic acid (MAA), and sodium methacrylate. 2. Theoretical Background 14
UV H OH OH
O
UV UV OH monomer OH monomer
Scheme 2.2: Novel UV-initiated living grafting system controlled by benzophenone
RAFT polymerization has also been adapted for use in UV initiated polymerization. Quinn et al.28 used phenylethylphenyl dithioacetate (PEPDA), phenylethylphenyl dithiobenzoate (PEPDB) to polymerize styrene and MMA with a polydispersity of around 1.1-1.2. While the polymerization was successful, some degradation of the RAFT agents was observed.
Triblock polymers were have also been created using this methodology by You et al. using a bifunctional RAFT agent initiated by ultra-violet (UV).29
2.2.5. Electron beam
Many reactions can be initiated by electrical discharges but the situation is very complex and the products depend very much upon the nature of the discharge. In electron beam polymerization electrons are accelerated by an electric field and impact with molecules to excite atoms to form ions and free radicals. As the electric field is not usually uniform the energies transferred to the system in various parts of the beam differ and the efficiency of a reaction can vary according to its location in the discharge.21 2. Theoretical Background 15
2.2.6. Plasma
Plasma polymerization is a relatively new technique which involves reacting a monomer plasma with a polymer surface. The plasma can be produced in a number of ways, but the most common is through microwave or radio frequency excitation. In general, polymers produced through plasma techniques are highly cross-linked and likely to have a small degree of swelling, and thus offer poor performance as a SPOS support.
Due to the complexity of plasmas, the mechanism of plasma polymerization is not fully understood and a lot of work is being done in this area.25,30 Meng et al.30 conducted a comprehensive study using in situ optical and mass spectroscopy. This study showed that the initial excited species are formed in the vapor phase and the actual polymerization occurs on the substrate surface. Lee et al.31 conducted an in depth study of the different chemical and environmental parameters and how they effect the resulting polymer film.
Largely the work that has been published in relation to plasma polymerization concerns surface modification of polymers rather than the production of grafted polymers. As an example, Chang-Do et al.32 produced styrene grafted poly(vinylidene fluoride) films through plasma polymerization for the pervaporation of water-ethanol mixtures.
One way that plasmas are used to graft polymers rather than surface modification is through the use of plasma treated plastics as reactive substrates for chemical attachment as in oxygen functionalization as discussed above.
2.3. Free radical polymerization
There are many techniques for producing polymers, but free radical polymerization is currently the most widely used technique for the production of bulk polymers and other specialty polymers.33 The wide usage of this technique is primarily because of its wide- ranging applicability and versatility. Firstly, it can polymerize a very wide range of 2. Theoretical Background 16
monomers and conditions compared to other chain growth polymerization techniques like anionic or cationic polymerization. Secondly, it does not require expensive catalysts and operates at relatively mild conditions (-20 to 200oC) and because of the lack of highly active catalysts in the system it is generally less sensitive to impurities and contaminants.
Another reason for the large uptake of radical polymerization techniques is the wide range of industrial production methods that can be used to suit applications. These include standard solution and bulk polymerization as well as the environmentally- friendly suspension and emulsion techniques. This has made radical polymerization the process of choice in large portions of the polymer industry.33
2.3.1. The mechanism of free radical polymerization
Free radical polymerization is a step or chain growth polymerization technique that allows for the production of high molecular weight polymerization through the consecutive addition of vinyl functionalized monomers through radical attack on the double bond of the vinyl group. The polymerization process can be roughly broken down into 4 steps: 34 • Initiation: Radicals are generated. • Propagation: The generated radicals react with the double bonds in the vinyl monomer effectively growing the polymer chains. • Termination: The radical species are destroyed through various means. • Transfer: The radical species is moved from one molecule to another without terminating the radical. In this way the active radical concentration remains constant.
Initiation is usually, but not exclusively, achieved through the homolytic cleavage of species such as azo- or peroxy- compounds to yield a pair of radicals. The initiator species I can be degraded thermally, through high energy sources such as ultra violet 2. Theoretical Background 17
radiation, reaction with redox species or other techniques to produce radicals R• as shown in the below equation:
• I ⎯⎯→kd 2R Eq. 2.1
Alternatively, the initiating species may be formed from the solvent and monomer using thermal, ultraviolet radiation, γ-radiation, or via a redox process.35,36
These radicals then add to monomer species, M, creating a propagating polymer chain:
• • R + M ⎯⎯→ki RM Eq. 2.2
Once the polymerization has been initiated further monomer units are added by reaction of the radical with the unsaturated double bond in the monomer to create a polymer chain. This addition of monomer units is called propagation.
• k p • Eq. 2.3 RM + M ⎯⎯→ RM n+1
This propagation will continue adding monomer units until a termination event occurs. A termination even destroys the active radical and produces “dead polymer” i.e. polymer that can no longer continue polymerization. Termination can occur through either of two bimolecular methods: combination and disproportination. These are represented by Eq. 2.4 and Eq. 2.5 respectively:
• • + ⎯⎯→ktc Eq. 2.4 Rm Rn Pn+m
• • + ⎯⎯→ktd + Eq. 2.5 Rm Rn Pn Pm
In termination via combination, two radicals join to form a polymer with the length of the two combined polymer chains. In disproportination one of the radicals will abstract a hydrogen atom from the other polymer chain resulting in a dead alkyl group on one polymer end group and a double bond on the second polymer end group.
Monomers almost always terminate exclusively by one of these mechanisms, determined by steric and electronic effects. Thus some monomers terminate almost 2. Theoretical Background 18
exclusively by disproportination and others by combination. For example, styrene terminates predominantly by combination and methyl methacrylate predominantly by disproportination.37,38 A current topic of research is into chain length dependent termination reactions, where the rate and mode of termination is affected by the degree of conversion and the polymer chain length. There is a great deal of literature on this topic.39-41
The final radical reaction that can occur is chain transfer reactions, where the active radical is transferred from the growing polymer chain to another compound. Thus the radical concentration of the system is constant but the molecular weight and polydispersity of the system is affected. Chain transfer can occur through many mechanisms including transfer to monomer species (initiating a new polymer chain), transfer to solvent, transfer to initiator, transfer to polymer (which can result in side chain branching) or transfer to specially added chain transfer agents (such as mercaptans42, cobalt complexes43, halides or innumerable other compounds). The transfer reaction is represented below where S is the transfer agent:
• + ⎯⎯→ktr + • Eq. 2.6 Pn S Pn S or
• + ⎯⎯→ktr + • Eq. 2.7 Pn XA Pn X A
In this reaction, the species X is transferred to the macro-radical, combining with it to form a dead polymer chain. The new radical formed by the transfer reaction, depending on the nature, may go on to react with another monomer, continuing the polymerization process. The species transferred is often a hydrogen atom, but other species such as a halide, can also be transferred.44
2.3.2. Kinetics of Free Radical Polymerization
The free radical polymerization process is well understood and can be easily modeled into a kinetic scheme.45 From this kinetic scheme we can develop a rate equation and 2. Theoretical Background 19
readily predict the polymerization process such that complete control of the polymerization can be achieved. The initiation reaction is the rate-limiting step in most radical polymerization processes. The modeled initiation includes both the degradation of the initiator species and the initial addition of the first monomer unit and thus can be represented as:
d[RM • ] R = = 2 fk [I] Eq. 2.8 i dt d
where Ri is the rate of initiation and the initiator efficiency f is a measure of the number of radicals generated by the initiator that go on to actually start polymer growth.
Since the monomer is consumed by both the initiation and the propagation steps of the reaction the consumption rate can be represented by the sum of both these rates. That is:
d[M ] = R + R Eq. 2.9 dt i p
If we simplify this by assuming that all the polymer chains are long and thus there is negligible influence of Ri on d[M]/dt then we can simplify this representation to:
d[M ] • R = − = k [M ][RM ] Eq. 2.10 p dt p
● where [RM ] is the concentration of growing radicals. This is known as the long-chain assumption.36 However, as the name suggest this assumption has been shown to be 46-48 inaccurate for small radicals. Evidence has been presented to suggest that kp is in fact chain length dependent.48,49 Bearing this in mind the chain length of the growing polymer chain generally increases fast enough that the rate coefficient rapidly approaches a constant value and once this occurs one can again use the long-chain assumption and a constant kp value. In this way the propagation rate coefficient is often referred to as “constant” despite this not being strictly true.
Termination reactions through combination or disproportination are bimolecular processes and their rate Rt can be expressed by: 2. Theoretical Background 20
= • 2 Eq. 2.11 Rt 2kt [RM ]
where kt is the sum of both combination and disproportination i.e.:
= + Eq. 2.12 kt ktc ktd
As mentioned earlier the termination rate coefficient has been shown to be dependent upon the chain length of the macro-radical involved in the termination event, however under most circumstances it can be assumed that termination is constant across majority of the polymerization.50,51
Making a steady state assumption that the concentration of radicals has reached a stable state and is constant over time (a generally valid assumption, except for very beginning and end of the polymerization reaction) then the rate of initiation is equal to the rate of termination and the following overall equation emerges to represent the rate of polymerization Rp:
= = • 2 = Eq. 2.13 Ri 2fkd [I] 2kt [RM ] Rt
Then
• fk [I] [RM ] = d Eq. 2.14 kt
Substituting into Eq. 2.10 gives:
fk [I] = d Eq. 2.15 Rp k p [M ] kt
Assuming that the initiator concentration is constant over the period of the analysis, we can simplify this equation by setting a new constant term k which is a combination of all the constants in Eq. 2.15 then: 2. Theoretical Background 21
d[M ] − = k[M ] Eq. 2.16 dt
Integrating each side of the equation between time = 0 and time = t,
[M ] d[M ] t − ∫ = ∫ kdt Eq. 2.17 [M ] [M ]0 0
()()− = Eq. 2.18 ln [M ]0 ln [M ] kt
⎛[M ] ⎞ ln⎜ 0 ⎟ = kt ⎝ [M ] ⎠
= − Eq. 2.19 And since [M ] (1 x)[M ]0
⎛ 1 ⎞ ln⎜ ⎟ = kt Eq. 2.20 ⎝1− x ⎠
Adherence to this equation is proved by plotting ln(1/1-x) vs. t to produce a straight line with a slope k. This is often used as a proof that the polymerization is following the expected model.
The consequence of the ability of this kinetic model are that polymerization reactions can be effectively modeled on how they will proceed under all conditions, if certain reaction constants are know, this allows more detailed kinetic investigations into the polymerization.52,53
2.4. The evolution of living/controlled polymerization techniques
It was in the 1920s that the first proposition into the existence of long chain, large macromolecules was brought forward by Staudinger.54 From this proposition it took 2. Theoretical Background 22
thirty years before a basic mechanism was brought forward to produce these macromolecules through a chain polymerization mechanism. Originally the polymers produced were completely uncontrolled and of limited use and since then the driving force behind polymer production has been the ability to control these polymerization reactions and the resulting polymers.
The big limiting factor to the control of all chain polymerization techniques is that of bimolecular termination. Once this can be eliminated then further control of the polymeric system is possible. The first controlled/living polymeric system to be developed was in the late 1950s with the advent of anionic polymerization.55,56 The original polymerization system was used to produce a rubber from styrene and 1,3- butadiene that is still used commercially today.57,58
Bu Bu Bu Bu Li Li Li Li
n n n n n Initiating species Polystyrene Polystyrene-co-butadiene
Scheme 2.3: Mechanism of styrene-butadiene copolymer or S-Buna for commercial production using a butyl-lithium anionic process58
Nowadays living polymerization techniques are so well studied that not only do they allow for the production of very well-defined polymers but also the production of complex polymeric architectures and novel molecules.59-63
Living/controlled polymerization comprises of polymerization techniques that are used to control polymerization and, primarily, the termination reactions involved. Webster64 defined living polymerization behavior as: a. The polymerization proceeds to complete conversion, and further monomer addition leads to continued polymerization. b. The number average molecular weight is dependent on conversion. c. The number of polymer chains in the system remains constant during the polymerization. d. The molecular weight can be controlled through the reaction stoichiometry. 2. Theoretical Background 23
e. The polydispersity of the molecular weight distribution is low. f. Polymers with chain end functionality can be obtained quantitatively.
It is also important to look at the difference between living and controlled polymerization, two terms that are often used interchangeably. For a polymerization to be living, then the termination reactions must be negligible compared with the propagation and activation/deactivation steps. However, a living polymerization system may still have poor molecular weight control and polydispersity.65-67 Most recently, Darling et al.68 proposed that the term “living polymerization” be relaxed and defined as a process that yields living polymers. On the other hand, the term “controlled’ refers to whether the process can synthesize well-defined architectures, introduce chain-end functionalities (such as in catalytic chain transfer or CCT69,70), control the incorporation of copolymers, and have good targeted molecular weight based on the ratio between monomer and initiator.
The first living polymerization system was anionic polymerization developed by Szwarc et al.71 The active polymerization species in this case is a negatively-charged ion that propagates through vinyl bonds in monomer units. Control is achieved in this case by the electronic repulsion of the negatively charged chain ends that prevent two active polymer units terminating together. Thus, once steady state is achieved the concentration of active species remains constant and the chain length will grow linearly across the whole range of conversion. Szwarc named this system living polymerization because of this lack of termination. However while it was a great advance the system suffered from broad molecular weight distributions.
Since the first anionic system was developed by Szwarc many advances have been made and anionic polymerization is now an excellent method for producing low polydispersity controlled structures and polymers.72-76 However, problems that haven’t been overcome include the relatively limited range of monomers that can be used, the extreme polymerization conditions that are required (often around -30oC or even lower), and the susceptibility to catalyst poisoning.38 2. Theoretical Background 24
Following anionic polymerization the next big discovery was that of cationic ring opening polymerization of tetrahydrofuran.77-79 Being an ionic method, cationic polymerization generally suffers from the same weaknesses that anionic suffers from; restriction of monomers (must contain electron releasing substituents), extreme conditions and catalyst poisoning).80,81 One primary result from this work was the discovery of a new method of controlling the termination of the polymer chains. With cationic ring opening polymerization the control derives from an equilibrium between dormant and active polymerization species which reduces the concentration of active species and hence termination reactions. This new method of controlling the termination finally allowed for a way of controlling the polymerization without the need for repulsive groups as in anionic polymerization. This concept led to group transfer living polymerization82 which then led to new methods of anionic and cationic polymerization83-87 and coordination polymerization.88,89
Despite improvements, the inherent problems with ionic polymerization systems where still largely uneconomic to implement.71 Thus a large amount of effort has been put into developing living techniques using this same dormant-active equilibrium that was used with ring opening cationic polymerization. Radical polymerization compared to ionic method is open to a wider range of monomers, and does not require the extreme temperature conditions, and is less susceptible to impurities and contaminants.
2.5. Methods in living radical polymerization
After the 1980s several new systems were developed that lead to new living radical techniques that offered access to a wide range of monomers and better control of molecular weights, functionality and architecture. These included the now highly studied atom transfer radical polymerization (ATRP), reversible addition fragmentation transfer polymerization (RAFT) and nitroxide mediated polymerization (NMP).90-93
The concept of controlled or living free radical polymerization was introduced in 1982 by Otsu.65,94 Since then there have been a number of systems have had varied levels of success in developing controlled free radical polymerization. 2. Theoretical Background 25
Essentially all the living radical polymerization techniques (except for RAFT polymerization and other methods that aren’t controlled by reversible termination) follow the mechanism of Scheme 2.4. This mechanism shows the reversible and rapid transfer between the dormant and active species (radical) that can then undergo polymerization. The essence of this is that any radical species are rapidly deactivated; keeping the radical concentration low and thus limiting the bimolecular termination that greatly affects normal free radical polymerizations.
ka RX R X kd
kp
Scheme 2.4: Living free radical polymerization mechanism controlled by reversible termination94
The main modern living free radical polymerization techniques are outlined in the following section.
2.5.1. Inferter
The term inferter comes from a combination of the words initiator, transfer and terminator and is basically a compound that can do all of these actions. The result is that these compounds can produce reasonably well-controlled polymers through a reversible termination reaction.
The first instance of inferter usage was in 1982 by Otsu65 when he polymerized styrene in the presence of dithiocarbamates. Also in this paper he first introduced the term “living radical polymerization”. The living character of the polymerization is achieved through the rapid reversible termination reaction that occurs.
The mechanism for inferter action is based around the homolytic cleavage of a weakened covalent bond in the inferter. This bond is broken by the addition of energy to the system and produces two radicals; one radical goes on to polymerize the monomers 2. Theoretical Background 26
while the other is significantly more stable and doesn’t initiate polymerization. However, this stable radical can recombine with the active radical, reversibly terminating the polymerization through a bimolecular combination reaction:
H2 C H 2 5 Energy H2 C2H5 C C S C N SCN C H S 2 5 S C2H5
n n
M kp
Scheme 2.5: Mechanism for inferter polymerization of styrene65-67
Many compounds have been found to act as inferter but it has been found that organosulphur compounds are the most effective.66,67 In general though, the cleavage reaction is not overly selective, and bond breaking in other parts of the molecule can result in unstable radicals. These radicals can be slow to initiate and result in poor molecular weight control and polydispersity.95
Still, some inferter systems have been developed that can reliably produce polymers of low polydispersity ~1.3.96
C H 2 5 C2H5 C2H5 H2 S S C S C N N S S N C H S 2 5 C2H5 S S C2H5
(1) (2) (3)
CH3
S OC2H5 N N H2C O S
(4) (5)
Figure 2.3: Some known inferters: (1) Diphenyldisulphide97, (2) benzyl N,N- diethyldithiocarbamate98, (3) tetraethylthiuram disulphide99, (4) phenyltriphenylazomethane100, (5) methacryloyl-O-ethyl xanthate96 2. Theoretical Background 27
2.5.2. Nitroxide Mediated Polymerization (NMP)
In 1985, Solomon and Rizzardo patented a method for controlling radical polymerization through the use of stable nitroxide radicals.101 This was termed nitroxide mediated polymerization or NMP. The advantage of the NMP system over an inferter system is that initiation can come through any normal radical initiation method. This is usually an azo- or peroxide initiator, but the system can also be initiated through an alkoxylamine fragment or other methods. Similar to inferter controlled polymerization, NMP uses a stable radical to reversibly terminate the polymerization, thus limiting the radical concentration and the chance of non-reversible termination reactions.
This reversible termination reaction allows the polymer chains to grow incrementally with conversion and producing a linear relationship between molecular weight and conversion as well as quite low polydispersity indexes (~1.5). NMP has been shown to work well with styrenics101,102, acrylates and acrylamides.103,104
H3C CH3 H3C CH3 O N O N
H3C CH3 H3C CH3 TEMPO
M kp
Figure 2.4: Nitroxide mediated radical polymerization of styrene using TEMPO as presented by George et al.102 2. Theoretical Background 28
2.5.3. Atom Transfer Radical Polymerization (ATRP)
ATRP was first published by Sawamoto105 in Japan and by Matyjaszewski106 in the USA almost simultaneously. Sawamoto termed the process simply “metal catalyzed living radical polymerization” where Matyjaszewski107 called the process “atom transfer radical polymerization”, which is the name that is commonly used to describe this process now and will also be the terminology used in this thesis.
ATRP is based on the organic chemistry synthesis for the metal catalyzed radical addition to alkenes called the Kharasch reaction or Atom Transfer Radical Addition (ATRA).106 In ATRP, instead of the reaction being a one step process, the radical is allowed to propagate and polymerize a vinyl monomer.
Essentially the process involves the metal complex (M – metal, X – halide, L – ligand) breaking the alkyl R-X bond and removing the halide. The halide is then incorporated into the metal complex thus changing the oxidation state of the metal and leaving a radical on the R- group which can propagate. This process is reversible and lies fairly heavily on the dormant side thus reducing the concentration of radicals and minimizing bimolecular termination.
n n+1 RX M XnLm R M Xn+1Lm
R1 H2CC R2
R1 R H H 1 2 n 2 n+1 R C X M XnLm R C M Xn+1Lm
R2 R2
Figure 2.5: Kharasch or ATRA reaction scheme 2. Theoretical Background 29
n n+1 RX M XnLm R M Xn+1Lm
R1 H2CC R2
R 1 R1 H H 2 n 2 n+1 R C X M XnLm R C M Xn+1Lm
R2 R2
R1 H2CC R2
R1 R H 1 n 2 n+1 R C X M XnLm R C M Xn+1Lm H2 R2 R2
Figure 2.6: ATRP mechanism as proposed by Sawamoto108
The ATRP system consists of a number of components: • The initiator is one of many alkyl halides that have been developed. Many of the initiators have been designed to either mimic existing monomers, thus matching the reactivity of an ATRP macromonomer or compounds designed to add functionality to the polymer. • The second component is a metal complex. There have been several systems developed based on Ru, Ni, Pl, Cu, Mo, Rh and Fe.109,110 Of these copper is by far the most commonly used although ruthenium and nickel have received quite a bit of attention. With this various ligands are used to complex the metal increasing solubility and tune the electronic state of the metal to allow the reduction of the alkyl halide. • The last part of the system is the monomer and solvent. One of the great advantages of ATRP is that it is applicable to many monomers including (meth)acrylates111,112, styrenics113, acrylamides114, vinyl pyridines115, acrylonitrile116, (meth)acrylic acids.117 In fact, the only real restriction is on 2. Theoretical Background 30
monomers and solvents that can, themselves act as ligands, thus destroying the metal complex.
Two reviews have been published by Sawamoto108 and Matyjaszewski118 and further information in regards to the available variety of ATRP systems can be found there.
One great advantage of ATRP is in its applicability for production of advanced and novel polymeric structures. As the dormant species is a stable compound, it is a simple process to stop the polymerization and reinitiate it to produce block copolymers and other more complicated polymers such as stars119-121, combs122 and graft polymers.123-125 Also, as mentioned earlier, because the initiator can be any of a variety of alkyl halides this allows for the addition of functionality to the start of the polymer chain. Alternatively the halide atom can be used as a point for post-functionalization of the polymer.126,127
Compared to NMP, ATRP is slower, requires special initiators and the resulting polymers contain metal complexes which restrict its use in food and medical applications.
A comment should be made on the useful temperature range for ATRP. In general because of the higher activation energy for the radical propagation than for the radical termination reactions (i.e. the kp/kd ratio) better control is achieved at higher temperature. However, at increased temperature chain transfer and other side reactions also become more pronounced. Catalyst solubility is also increased at higher temperatures, but decomposition can occur if the temperature is pushed too high. Thus we find that optimal temperature depends on the monomer, catalyst and targeted molecular weight.118
ATRP and the persistent radical effect
The mechanism that allows ATRP to provide such excellent control and selectivity in polymerization reactions can be explained by the persistent radical effect (PRE). Fischer et al.128,129 was the first to elucidate the PRE which uses a persistent radical species to control termination reactions with an equilibrium between a dormant adducts 2. Theoretical Background 31
and their corresponding transient and persistent radical species. This inhibits the rate or bimolecular radical coupling and disproportination which for most organic free radicals is near the diffusion-controlled limit, thus these radical destroying reactions tend to dominate radical reactions.
The PRE occurs when concentrations of transient and persistent radicals are formed at equal rates in a single step. Because the transient radicals can undergo fast termination reactions via coupling or disproportination, their concentration decreases and the concentration of the persistent radical builds up. Eventually the concentration of persistent species is sufficiently large that the rate at which the propagating radicals react with the persistent radicals in a deactivation (or reversible termination) step is much faster than the rate at which the propagating radicals react with each other in an irreversible termination step. Thus, addition chemistry can be performed involving free- radical intermediates that are highly selective for addition over radical coupling and disproportination. In ATRP the persistent species is the copper(II) species and the transient species is the radically active polymer chain P•.
There are a couple of results from this. Firstly, bimodal termination and other radical side reactions are limited because of the low active radical concentration in the system. Secondly, there is a tendency for all the radical species to form into dormant halide- capped species until the concentration of Cu(II) deactivator is reduced to a level where equilibrium allows the formation of sufficient radical species that polymerization can occur. This manifests in an induction period while the radical concentration builds enough to reduce the Cu(II) concentration. While these radicals are being reacted into dormant halide compounds, there is a higher concentration of radicals in the initial stages. This increased radical concentration decreases reasonably quickly, but allows for the formation of bimolecular termination products in the initial stages of the polymerization and appears as low molecular weight tailing.
This contrasts with RAFT in which termination products tents to appear at the later stages of polymerization where the concentration of monomer is decreased and thus the termination steps become more likely, resulting in a high molecular weight shoulder. 2. Theoretical Background 32
2.5.4. Reversible addition fragmentation chain transfer polymerization (RAFT)
Reversible Addition Fragmentation Chain Transfer Polymerization or RAFT is the most recent living radical polymerization technique developed. It was developed in Australia by Ezio Rizzardo’s team at CSIRO in 1998.130 RAFT has been found to work with a very large range of monomers and to be highly versatile.131,132 The versatility of RAFT lies in its similarity to normal radical polymerization techniques. The polymerization can be conducted in exactly the same manner as normal radical polymerization with a RAFT agent added. The resulting polymerization can form low molecular weight polymers with active end groups and low polydispersity131,133 and thus RAFT meets most of the requirements of living radical polymerization.134,135
Macromolecular design via interchange of xanthate or MADIX is a similar system to RAFT that uses xanthates as transfer agents but follows the same mechanism.136,137
The RAFT Mechanism
RAFT uses thiocarbonyl compounds such as dithioesters130, dithiocarbamates138, trithiocarbamates139, and xanthates140,141 to control the radical polymerization of a wide range of monomers. The original mechanism put forward by the CSIRO team has remained unchanged and is shown below in Scheme 2.6.
The initiating radical is generated from degradation of an initiator species (I). This radical then initiates a propagating chain that then adds to the thiocarbonyl group in the dithio compound (II), creating the intermediate (2). It has been previously shown142,143 that the rate of this addition governs the effectiveness of the particular dithio compound as a RAFT agent. The R-group of the dithio compound is then released as a radical and this species may then initiate another polymer chain (III). The dithio moiety of the polymer chain formed during this process (3) may then continue to act as a transfer agent, establishing the equilibrium (IV). In this way, all the polymer chains grow at approximately the same rate leading to a linear evolution of the molecular weight with conversion and a low polydispersity index. Further, the terminal dithio moiety 2. Theoretical Background 33
(i) Initiation
monomer I P1 ki
(ii) Chain Transfer
P S S S S R Pm S S R m R Pm Z Z Z (1) (2) (3)
(iii) Reinitiation
monomer monomer Pn Pn+1 R P1 kp kp1
(iv) Chain Equalibrium
kβ k-β P S S S S Pm Pn S S Pm n Pm Pn k Z k-β Z β Z (1) (2) (3)
(v) Termination
Scheme 2.6: CSIRO’s RAFT mechanism as represented by Barner-Kwollik144 in species (3) leads to the possibility of either reinitiating the polymer with some different monomer to synthesize block copolymers, or functionalizing the end-group of the polymer. Some dead polymer chains will be formed via termination events in the reaction. This may occur through either a combination or disproportination event depending on the monomer. This will lead to a slight broadening of the molecular weight distribution, and at high conversion a high molecular weight shoulder sometimes forms from combination reactions.
There are a wide range of RAFT agents available and their selection and effectiveness largely depends on the transfer constant of the RAFT agent. The Z-group is the functional group that remains statically attached to the RAFT agent and acts to stabilize the radical intermediates. The R-group or leaving group dispatches from the RAFT agent during the fragmentation step and goes on to initiate polymer chains. The transfer constant is effectively determined by the selection of both R- and Z- groups. For an 2. Theoretical Background 34
effective RAFT agent the R-group must be a good radical leaving group and then the activity of the reagent is modified by changing the nature of the Z-group.145
S S R S S R
Z O Z
Dithioester Xanthate
S S R S S R
N S Z Z' Z Dithiocarbamate Trithiocarbonate
Figure 2.7: Structure of different classes of reagents currently used as RAFT/MADIX transfer agents
As RAFT is a free radical polymerization mechanism, some amount of dead polymer must be produced.145 This is effectively determined by the number of chains initiated by initiator radicals.146 In general, if the polymerization is tuned such that in comparison to the uncontrolled polymerization the molecular weight is tenfold lower, then less than 10% dead chains can be expected.145
2.6. The kinetics of living polymerization
In living free radical polymerization, unlike conventional free radical polymerization, the rate of initiation must be greater than the rate of propagation for all the polymer chains to grow in parallel. Due to the limited termination and chain-transfer events we can greatly simplify our reaction model and thus the rate of propagation can be given as91:
d[M ] R = − = k [P ][M ] Eq. 2.21 p dt p n
where Pn is the concentration of active growing polymer chains. This is approximately equal to the initial initiator concentration in an ideal case. 2. Theoretical Background 35
Integrating Eq. 2.21 then we get
⎛[M ] ⎞ ln⎜ 0 ⎟ = k [P ]t Eq. 2.22 ⎝ [M ] ⎠ p n
where [M]0 is the initial concentration of monomer at time = 0 and [M] is the [M] concentration of monomer at time = t. The ratio 0 is linked to the monomer [M] conversion x as shown in Eq. 2.23:
[M ] [M ] 0 = 0 = 1 = 1 = 1 [M ] [M ] − [P] [M ] − [P] [P] 1 − x Eq. 2.23 0 0 1 − [M ]0 [M ]0 where [P] is concentration of polymer at time t. Thus, if we plot ln(1/(1-x)) vs. time to give straight line we have evidence of constant concentration of active species Pn and this also gives the apparent rate of polymerization kp[Pn].
Since the degree of polymerization (DPn) at any point should be the same for each polymer chain then
[M ] −[M ] = 0 DPn Eq. 2.24 [I]0
where [I]0 is the initial concentration of initiator.
Similarly, the number average molecular weight can be given by the number of units of monomer converted to polymer divided by the number of polymer chains and is given by Eq. 2.25
[M ] × x × Mn Mn = 0 monomer Eq. 2.25 [I]0
where x is monomer conversion and Mnmonomer is the molecular weight of the monomer. 2. Theoretical Background 36
Thus theoretical Mn at 100% conversion is given by
[M ] × Mn = 0 monomer Mntheoretical Eq. 2.26 [I]0
Overall the linear kinetic plots, parallel growth of each chain, narrow PDI and the good agreement between theoretical and experimental results define a living radical technique.
The one slight variation on this theory is for RAFT mediated living polymerization. The difference in RAFT polymerization is that we assume that all the polymer chains will be initiated by the liberated R-groups off the RAFT agent rather than the initiator fragment. This is valid because both macroRAFT agents and RAFT agent molecules can react with radicals in the polymerization. Thus in theory, every RAFT agent will eventually react and each R-group will form the start of a polymer chain. The initiator in this situation determines the concentration of active radicals in solution not the number of polymer chains, and is there to ‘kick start’ the RAFT process. We need to modify Equations 25-27 to take this into consideration.
[M ] −[M ] DP = 0 Eq. 2.27 n [RAFT ]
[M ] × x × Mn Mn = 0 monomer Eq. 2.28 [RAFT ]
[M ] × Mn Mn = 0 monomer Eq. 2.29 theoretical [RAFT]
2.7. Radiation polymerization
As mentioned previously, radiation grafting will be the dominant grafting method considered in this thesis, thus more detailed discussion will be given to it over other 2. Theoretical Background 37
grafting methods. The use of high-energy radiation in polymerization has been used for many years since it was first observed in 1924 by Lind et al.147 The process uses radiation-derived radicals as a source of initiation for the polymerization, instead of the more common method of using thermal initiators.
These sources include photochemical (most commonly ultraviolet radiation148), electrical discharge (electron beam), nuclear radiation (α and γ-radiation) and plasma polymerization (indirectly using microwave or radio frequency to generate plasmas).
2.7.1. Mechanism of radiation radical initiation
Under the right conditions most long chain addition polymers can be produced from high-energy irradiation. The radiation simply acts as an initiator producing free radicals (usually hydrogen atoms) and ions into the system.149 Although the radiation acts as an initiation source the actual route to radical generation is somewhat more complicated. Radiation primarily produces ions, and excited molecules and radicals are indirectly produced by these excited molecules.19
For irradiation at room temperature it is generally valid to assume that radical polymerization is the only polymerization mechanism occurring.150,151 Thus, the initiation step is the production of free radicals which may be represented as such:
A ⎯Rad⎯→⎯ 2R• Eq. 2.30 where A is any molecule present in the system whether it is monomer, solvent, or any • other compound. R represents a primary radical which can then go on to react with unsaturated monomer molecules to form a larger radical:
R CC R CC Eq. 2.31 or in a general form; 2. Theoretical Background 38
R• + M → RM • Eq. 2.32 where M is a monomer unit.
If we put this initiation step into the normal free radical mechanism we get the following rate equation:
d[M ] • rate = − = k [RM ][M ] Eq. 2.33 dt p for steady state conditions, the rate of radical generation is equal to the rate of disappearance of radicals (by bimolecular termination). Thus the velocity of each of these reactions can be written as
= • 2 Eq. 2.34 vi 2kt [RM ]
or
1/ 2 • ⎛v ⎞ [RM ] = ⎜ i ⎟ Eq. 2.35 ⎝ 2kt ⎠
Substituting into Eq. 2.35 into Eq. 2.33,
⎛ ⎞1/ 2 = vi Eq. 2.36 rate k p ⎜ ⎟ [M ] ⎝ kt ⎠
The rate of formation of radicals will be proportional to the rate at which energy is absorbed by the system or the dose rate. Thus,
rate ∝ ()dose 1/ 2 Eq. 2.37 2. Theoretical Background 39
combining with Eq. 2.33 yield:
d[M ] • (dose)1/ 2 ∝ − = k [RM ][M ] Eq. 2.38 dt p
Deviations from this relationship can exist through competition reactions of the primary radicals. The primary radicals can do three things: they can initiate polymerization, they can react with a growing polymer chain to terminate it, and they can react with another primary radical to recombine themselves. At high monomer concentrations then the initiation reaction will dominate, however deviations from the second two reactions can become apparent at low monomer concentrations or high dose rates.
2.7.2. Chemical effects on radiation polymerization
There are some additional effects that are associated with γ-polymerization that should be noted. Most importantly, the rate at which radicals are generated is dependent on both the dose and the radiative susceptibility of the compounds being irradiated. This is called the G-value or also known as “radiative yield” of the material. The G-value is defined as the number of molecules changed per 100eV adsorbed.
Due to the difference in G-value between chemicals, it is common to use solvents with a high G-value to increase the yield of a low G-value monomer. An example of this is that styrene is more efficiently polymerized by irradiation in an aqueous emulsion because the radiolytic yield of radicals from water is greater than that of styrene.21
There are also other methods of increasing the rate of polymerization through addition of various compounds to the polymerization solution. These work on either the principle that they absorb more radiation than the monomer/solvent mix and thus generate more radicals, or they stabilize the formed radicals thus preventing termination. Some additives that have been used include mineral acids, Mohr’s salt, photo initiators, thermal initiators, inclusion compounds like urea, and polyfunctional monomers (especially acrylates and methacrylates).152,153 2. Theoretical Background 40
2.7.3. Side reactions in radiation polymerization
The chemical changes produced by radiation don’t specifically produce radicals through hydrogen combination with a monomer.
Firstly, radiation produces a mixture of ions (mostly cations) and radicals. Thus, there is competition between ionic polymerization and radical polymerization in every system. This can often result in bimodal distributions through the contribution of both radical and cationic polymerization mechanisms. Matsuzaki et al.154 did a comprehensive study into the nature of gamma and UV initiated polymerizations. He found that at room temperature and in basic solvents, polymerization only takes place via a radical mechanism. In contrast he showed that at low temperature (-78oC) and with the addition of radical scavengers, there was negligible radical polymerization and cationic polymerization took place.
Secondly, the reacting radical species A (see Eq. 2.30) can be derived from a monomer, solvent, polymer or any other compound in the solution. This can result in variations in radical reactivity, differing polymeric end groups, or side chain grafting of the polymer if the radical forms on the side chain. One notable example of this is the strong effect of oxygen on radiation polymerization systems. Oxygen is a highly effective radical inhibitor that traps the radical through the formation of oxygen containing compounds such as hydroxides, peroxides and hyperoxides (see 2.2.3. Oxygen functionalization).
Polymers can also be affected by the radiation in different ways. Depending on the chemical constituency of the polymer chain and various other factors such as the polymeric concentration, polymers can undergo changes such as back-bone radical production, chain-scission and cross-linking. These changes are dependent on many factors, and can create many changes to the characteristics of the produced polymers, and have been extensively studied.19,21,155 2. Theoretical Background 41
2.7.4. Living radical polymerization under γ- radiation
Quinn et al.156,157 was the first to write a series of papers looking at room temperature RAFT polymerization initiated by γ-radiation, using a newly developed RAFT agent called 1-phenylethyldithioacetate (PEPDA). This new RAFT agent used the addition of a methylene spacer between the Z-phenyl group to destabilize the intermediate RAFT radical and allow for effective low temperature usage of this RAFT agent. Gamma- initiated RAFT polymerization has also been conducted using xanthate transfer agents158 and trithiocarbamates.159
Pan et al. also conducted γ-initiated polymerization RAFT polymerization of acrylic acid (AA)160, methyl methacrylate (MMA), methyl acrylate (MA), styrene161,162, using several RAFT agents. Bai originally proposed a reversible termination reaction160-162 that can be seen in Figure 2.8 for the γ-radiation initiated polymerization. However, in a later paper dealing with the γ-initiated polymerization using benzyl 9H-carbizole-9- carbothioate they conceded that the theory for reversible termination lacked evidence and that the RAFT mechanism is the more likely explanation.163 2. Theoretical Background 42
S S R γ S S (I) R Z Z
monomer (II) R Pn
S S S S Pn (III) Pn Z Z
γ S S Pn (IV) S S Pn Z Z M
Figure 2.8: Proposed mechanism for reversible termination in γ-initiated living polymerization161
You et al. used a trithiocarbonate RAFT agent to make a PMA-PS-PMA triblock copolymer using γ-initiated RAFT polymerization.
Angot also demonstrated that these polymerizations show characteristics of living polymerization, including the ability to carry out reinitiation to block/gradient copolymers.120 Miwa et al.164,165 used the pre-irradiation technique induced by γ-irradiation in air as an initiation source for nitroxide mediated grafting of styrene onto a polypropylene substrate. The molecular weight of the grafted polystyrene was controlled and had a narrow polydispersity.
2.8. Control of graft polymerization through the use of living radical polymerization
Grafted polymers created through the above grafting methods generally consist of a broad molecular weight distribution and high polydispersity. It is therefore desirable to establish methods for controlled grafting onto polymeric surfaces, so that new structures 2. Theoretical Background 43
can be designed.92,93,108,118,166 However, there has been little written about grafting from insoluble substrates.
In recent years, there have been several investigations into the control of grafting polymerization including CCT167, NMP102,168,169, RAFT138,145,170,171 and ATRP.106,172,173
Strehling et al.174 reported on the synthesis of PP-graft-PS by a combination of metallocene and nitroxide mediated radical polymerization. The grafting was performed by a metallocene/borate-catalysed polymerization of alkoxyamine substituted alkenes. The result was poly(olefins) containing initiating groups for living free radical polymerization. These initiating groups were then polymerized with styrene to give well defined graft copolymers. The grafts were analyzed by cleaving them from the poly(olefin) backbone using trimethylsilyl iodide. The molecular weight adhered to theoretical values reasonably closely, indicating that the graft polymerizations had living characteristics.
Controlled ATRP surface graft polymerization has been well documented in grafting from a variety of surfaces including polymeric micro-spheres175, off flat polymeric surfaces123, inside capillaries176, for use in proton conducting membranes177, and on silicon178 and gold.179 Schellekens et al.180 conducted ATRP polymerization of methylmethacrylate off a substrate of hydroxy end-functionalized poly(ethylene-co-butylene). The hydroxy end group on the polyolefin was functionalized by 2-bromoisobutyryle bromide which provided the bromine functionality that was used as the initiation point for the ATRP polymerization. The resultant copolymers were well controlled, having a polydispersity index (PDI) of around 1.4 with little terminated polymer.
Surface-initiated living free-radical polymerization has also been conducted off silica surfaces. Blomberg et al.124 attached trichlorosilyl-substituted alkoxyamine initiating groups to the surface silanol groups of silica nanoparticles using them to grow functionalized linear chains of polystyrene homo- and copolymers. After cross-linking of the polymer shell and decomposition of the core with hydrofluoric acid, a robust and hollow polymeric nanocapsule was obtained. 2. Theoretical Background 44
The use of RAFT to control grafting reactions has also been reasonably well studied.181 Some of the controlled RAFT grafting systems that have been used include grafting from cadmium selenide nanoparticles182, DVB micro-spheres183 and a fluorinated polyimide using peroxides generated by pre-treatment with ozone as an initiator.184
Hawker et al. have also conducted a great deal of work using NMP to control surface graft polymerizations.93 This was achieved this through covalently attaching functionalized alkoxyamines to various substrates for use as an initiator. NMP offered good control over the grafting density, degree of polymerization, thickness of graft and functionality of the graft. NMP has also allowed for the development of block copolymers and other methods for adding functionality to polymeric substrates.
In addition to solution grafting and grafting from inorganic substrates, several groups have been conducting grafts from poly(olefin) substrates using controlled polymerization methods.
Angot et al.120 and Ayres et al.185 used cross-linked polystyrene resin beads as initiators from ATRP polymerization. Wang resins were used as a substrate for this experiment and the hydroxide functionalities on the Wang resin were functionalized as shown in Figure 2.9.
P OH O
Wang resin
O CH3 THF/TEA Br room temp Br CH3
O P O O Br
Figure 2.9: Method used by Angot for functionalizing Wang resins to form ATRP initiators120 2. Theoretical Background 45
These initiators were then used to perform copper-mediated ATRP polymerization of methacrylates and dimethylacrylamide.
Grafting of poly(styrene-alt-maleic anhydride) was conducted by two groups.186,187 De Brouwer used RAFT to control the grafting off a commercial polyolefin Krayton L- 1203 using an attached cyanopropyl dithiobenzoate RAFT agent. De Brouwer also demonstrated using UV radiation to degrade the RAFT agent after polymerization as a means of deactivating the polymer. Wu et al. used γ-radiation to graft polystyrene and maleic anhydride using benzyl dithiobenzoate.
Pan et al.188 used RAFT to graft water soluble and temperature responsive polymers onto multi-walled carbon nano-tubes. N-isopropylacrylamide was grafted from the nanotube by a dithiobenzoate type RAFT agent attached through the R-group. Pan also used cationic polymerization to graft polystyrene to an expanded graphite sheet using + - 189 CO ClO4 as a catalyst.
Matyjaszewski et al.125 used ATRP to graft off an insoluble polyolefin substrate. The substrate used was a commercial PE-PGMA copolymer; the glycidyl groups were initially functionalized to give a bromo-isobutyrate group that was used as the initiator for the polymerization. The polymers were cleaved from the substrate and the molecular weight was found to increase linearly, and the polydispersity index eventually reached around 1.4.
2.9. Conclusion
As can be seen some work has been done on controlling polymeric grafting using living radical polymerization methods. Methods such as RATRP, ATRP and RAFT have vast possibility for controlling these reactions and also open the possibility of producing advanced block and polymeric structures. 2. Theoretical Background 46
This thesis will cover the use of these polymers as novel substrates for combinatorial applications. These products should provide significant advances over current SPOS supports and advance the field significantly. 3. Analysis techniques and synthesis of reagents 47
3. Analysis techniques and synthesis of reagents
This chapter will outline some general experimental and analytical setups that are used throughout this thesis including: details of the setup and calibration of the 60Co γ- radiation source, analytical techniques and equipment that were used in multiple chapters throughout this thesis, and synthesis of RAFT agents and ATRP ligands.
3.1. 60Co γ-radiation facilities
3.1.1. Introduction to 60Co radiation source at UNSW
The University of New South Wales 60Co γ-radiation source is the type of γ-source that rises out of a water pool rather than the more modern γ-sources that lower the sample into an irradiating area. The source is housed in a shielded room when in use, and immersed in a large water tank when dormant. A winch mechanism is used to move the source from the water to the sample table. The following figure shows the layout of the room.
The sample table has been modified over the last 2 years to eliminate a lack of symmetry in the radiation source, which will be discussed in detail later.
Samples to be exposed to the γ-radiation are placed in a sample table and the γ-source is winched out of a water pool that blocks the radiation. The table consists of two circular trays, each with 20mm holes drilled in concentric rings centered on the γ-sources raising wire, as depicted below in Figure 3.2. 3. Analysis techniques and synthesis of reagents 48
Figure 3.1: γ-radiation source layout and setup190
Figure 3.2: Rotating sample table and γ-radiation source 3. Analysis techniques and synthesis of reagents 49
Radiation obeys an inverse-square relationship for level of intensity in relation to the distance from the source. Therefore, placing your sample in the various holes allows for control of the dose rate.
Radiation Row Number of holes Distance (cm) intensity (%) a 17 10.95 100% b 25 15.05 53% c 33 18.95 33% d 39 22.40 24% e 45 25.90 18%
Table 3.1: Sample table layout showing the number of holes per row, the distance from the centre of the sample hole to the γ-radiation source and the relative intensity of dose rate as you increase your distance from the source
Due to the wide number of units and systems that are around for the classification of radiation intensities, it may be useful to refer to Table 3.2 to help compare various radiation units.
Radiation Units S.I. Units Imperial Units Conversion Activity Becquerel Bq Curie Ci 1 Ci = 3.7*1010 Bq Exposure Coulomb/kg C kg-1 Roentgen R 1 C kg-1 = 3876 R Absorbed dose Gray Gy Rad Rad 1 Gy = 100 Rad Equivalent dose Sievert Sv Rem Rem 1 Sv = 100 Rem Effective dose Sievert Sv Rem Rem 1 Sv = 100 Rem
Table 3.2: Commonly used S.I. and imperial radiation units
3.1.2. Calibration of γ-radiation source
The radiation calibration was conducted on installation using the Fricke Method outlined by Draganić et al.1 This method uses a prepared dosimeter solution that determines the adsorbed dose by using the oxidation of ferrous ions in an aqueous sulphuric acid solution followed by UV-vis spectroscopy. 3. Analysis techniques and synthesis of reagents 50
Fe2+ ⎯γ⎯→−radiation⎯⎯ Fe3+ + e−
+ − + + ⎯⎯→ O2 2e 2H H 2O
2+ + + + ⎯⎯→ 3+ + 2Fe O2 2H 2Fe H 2O
Scheme 3.1: Reaction scheme for the redox reaction in Fricke dosimetery
Materials
Ferric sulphate (99% minimum) was used as received from Ajax Chemicals Ltd. Concentrated sulfuric acid (98%w/w) and sodium chloride (99.9%) was obtained from APS and no further purification was required. The School of Chemical Engineering and Industrial Chemistry distillation apparatus produced distilled water used in this experiment.
Experimental method
The dosimeter solution was prepared by mixing ferrous sulfate (0.14g, 0.5x10-3 moles), NaCl (0.03g, 0.5x10-3 moles) and 250 ml of distilled water were mixed in a 500 ml volumetric flask. 11 ml of concentrated sulfuric acid was then added slowly with shaking (note that the addition of sulfuric acid generates large amounts of heat and if inadequate stirring occurs then the solution may boil). Once the dosimeter solution had cooled to room temperature the volumetric flask was topped up with distilled water to 500 ml and allowed to sit for 24 hours, allowing the solution to reach ionic equilibrium.
4 ml of solution was then pipetted into 20 ml sample vials. These were then sealed using septa to prevent spillage and exposed in the γ-source for exactly 1hr.
Once the vials had been removed from the γ-source they were placed in a 1 ml quartz cuvette. The UV-Vis absorption of the solution was measured using a Varian Cary 300 Scan: Uv-Visible spectrophotometer by scanning between 250 and 350 nm using the unirradiated sample as a background. 3. Analysis techniques and synthesis of reagents 51
Samples were analyzed in random order to remove any systematic error that could be generated through ageing of the samples prior to exposure and after exposure.
The adsorbed dose is then calculated from the concentration of ferric ions in the solution after irradiation and accuracy is better than 3%.191
Results
The range of dosages that can be measured using this system is normally 2x103 – 4x104 Rads. However the range may be extended to 2x105 Rads by increasing the concentration of ferrous ions to 0.02 M, and saturating the system with oxygen prior to exposure.
Dose rate is calculated from the following equation as derived by Draganić at al.191: D = 2.74x10 4.ΔOD Rads where: D = Dose in Rads OD = Optical density at 305nm (25°C) Where this equation has been derived from the following equation N.OD.100 D = eV / g k.Q.103.G N = Avogadro’s number ( = 6.02*1023 molecules per mole) Q = The density of the dosimeter solution ( = 1.03 g/ml) K = Molar extinction coefficient ( = 2187 l.mol-1cm-1) G = the G-value, which is the number of molecules converted per 100eV of absorbed energy ( = 15.6 molecules/100eV)
The dose rate was measured with changes in the angle of rotation from an arbitrary point and also measuring the variation of dose with distance from the centre of the source. Figure 3.3 shows how the measured dose rate varies with the distance from the centre of the source and also how it varies in an angular fashion around the source. 3. Analysis techniques and synthesis of reagents 52
20000
18000
16000
14000
12000
10000
8000 Dose rate (Rad/h) 6000
4000
2000
0 8 10121416182022242628303234 Distance from source (cm)
90 120 60 30000
20000 150 30
10000
0 180 0
10000 Dose rate (rad/h)
20000 210 330
30000 240 300 270
Figure 3.3: Variation of γ-radiation dose rate with distance from the source and variation with angle. Angle is measured from an arbitrary axis and the dose rate was measured with a Fricke Dosimeter at a distance of 10.95cm from the centre of the source 3. Analysis techniques and synthesis of reagents 53
As can be seen from Figure 3.3 the radiation was asymmetric and the dose rate varied around the source. This can be explained through asymmetric degradation of the 60Co pellets, producing metals that attenuate the radiation dosage at different angles. Normally this isn’t a problem, but given the age of our γ-source and the degree to which the cobalt has degraded, this variation has become significant. The result is that there is a significant bias in the γ-source and that this must be remedied prior to any scientific work being conducted.
The dose rate was found to decrease exponentially with distance from the centre of the source as expected from theory. However, the scatter in the results caused by the radiation bias caused too much noise for any useful analysis of the results.
3.1.3. Elimination of Radiation Bias in 60Co radiation source
To alleviate the problem of the bias in the radiation, the radiation source was modified with a rotating table that allowed each sample to be exposed to the same amount of radiation.
The modification consisted essentially of a rotating platform driven by a 10A motor. This was achieved through the use of a lower table that held a set of castor wheels which in turn allowed an upper table to freely rotate when driven by the motor. The upper and lower tables had a hole in the centre cut out for the draw-rod, and a slot cut to allow the tables to be fitted properly. In addition there was a steel support brace that locked around the base plate of the γ-source which allowed the whole structure to be more solidly located and also allowed the lower plate to be held properly. This lower table was situated on 5 brackets and the top table was located to the support brace through the use of 5 locator wheels that allowed the top table to be centered without affecting the ease of rotation.
The drive system consists of a 3A / 24V DC motor, with a rubber wheel attached. The motor was driven from a 20V DC transformer and the speed is controlled through a 3. Analysis techniques and synthesis of reagents 54
specially produced Pulse-Width Modulation motor controller. Schematics and diagrams of the modified γ-radiation source are available in Appendix 1.
Once the sample table was modified, the γ-source was recalibrated to establish whether the exposed dose rate was now homogenized.
As can be seen from Figure 3.4 there is a significant decrease in the scatter of the results (standard deviation was reduced from 6,593 Gy/h to 506 Gy/h) and there is no longer any variation of the dose rate around the γ-source. This shows that we have achieved our objective of reducing the variation in the dose rate. The dose rate was found to be 0.1052 kGy/h in the first row of the sample holder and thus the dose rate for each of the sample holders can be seen in Table 3.3 and the full set of calibration data is available in Appendix 1.
Dose rate Row Distance (cm) (kGy/h) a 10.95 0.105 b 15.05 0.056 c 18.95 0.035 d 22.40 0.025 e 25.90 0.019
Table 3.3: Calibrated radiation dose rate calculated for each row of the sample holder 3. Analysis techniques and synthesis of reagents 55
20000 Without Rotation 18000 With Rotation 16000
14000
12000
10000
8000 Dose rate(Rad/h) 6000
4000
2000
0 8 10121416182022242628303234 Distance from source (cm)
90 120 60 30000
20000 150 30
10000
0 180 0
10000 Dose rate (rad/h)
20000 210 330
30000 240 300 270
Figure 3.4: Variation of γ-radiation dose rate with distance from the centre of the source and variation around the source. Results show dose rate with and without rotation of the sample table, measured using Fricke Dosimeter. 3. Analysis techniques and synthesis of reagents 56
3.2. Analytical techniques and Instrumentation
This section will outline the analytical equipment used such as GPC systems, FTIR and NMR systems. Additionally common chemical analysis techniques and calculation methods that are used throughout the thesis will also be outlined here.
3.2.1. Gel permeation Chromatography (GPC) / Size Exclusion Chromatography (SEC)
Gel permeation chromatography or size exclusion chromatography is the most common technique used to analyze molecular weight and molecular weight distributions of polymers. It is a high pressure liquid chromatography method, using the differential retention of dissolved polymers through a column packed with cross-linked polymeric beads with pores of varying size. As the hydrodynamic volume of the polymer particle increases (increased molecular weight) then the polymer is retained less by the column, thus has a faster elution time.38
This technique typically uses concentration sensitive detectors which directly measure the elution concentration of the sample. The elution concentration is the converted to a hydrodynamic volume distribution and then to a molecular weight distribution.192 The hydrodynamic volume is a function of both its molecular weight and its intrinsic viscosity. The intrinsic viscosity is a function of the chemical structure of the polymer, the degree it swells in the eluent, the temperature, as well as the molecular weight of the polymer.193
Common concentration detectors include: UV-Vis detectors, differential refractive index, and thermal conductivity detectors. 3. Analysis techniques and synthesis of reagents 57
Conventional GPC with primary calibration
This technique uses a constructed calibration curve made from narrow molecular weight standards relating the elution volume (or time) to the molecular weight of the polymer standard. Although this is extremely accurate this method is limited to polymers where narrow molecular weight standards are available.
Conventional GPC with universal calibration
For many polymers where narrow standards are not available, a standard calibration such as MMA or polystyrene can be used. It is then assumed that the calibration curve is universal and thus different polymers can be analyzed by using the intrinsic viscosity/molecular weight relationship referred to as the Mark-Houwink-Sakurada equation:
η = α Eq. 3.1 [] KM v
η = + α Eq. 3.2 log[ ] log K log M v
This equation relates the intrinsic viscosity [η] to the molecular weight of the analyzed polymer through two constants; K andα. The two constants are specific to a polymer- solvent system. Although this technique is widely used, it has been criticized for errors induced in the fact that the Mark-Houwink-Sakurada relationship is not strictly linear over a wide molecular weight range.194
Molecular weight sensitive detectors
The optimum method for analyzing polymers for which there is no primary standard is through the use of molecular weight sensitive detectors. This also allows for the analysis of novel monomers, branched structures, block copolymers and other polymeric structures. The two existing molecular weight sensitive detectors are differential viscometry and light scattering detectors.
The technique is based on matching the signals of the molecular weight detector with a standard concentration detector for each time period. While the second detector allows 3. Analysis techniques and synthesis of reagents 58
us to analyze different polymer samples it also increases the complexity of the system can introduce inaccuracies through inter-detector delay.195
GPC systems used as part of this thesis
Two systems were used during the experiments conducted as part of this thesis. The first uses THF as an eluent for hydrophobic polymers, and the second uses DMAc for more hydrophilic polymers. The equipment is outlined below.
THF GPC system Molecular weight distributions for polymers soluble in THF were measured via gel permeation chromatography (GPC) on a Shimadzu modular system, comprising an auto injector, a Polymer Laboratories 5.0 μm bead-size guard column (50 × 7.5mm) followed by three linear columns (105, 104, and 103 Å), and a differential refractive-index detector. The eluent was tetrahydrofuran (THF, 1 mL/min) and the system was calibrated with polystyrene standards ranging in molecular weight from 106 to 200 g mol-1.
DMA GPC system Molecular weight distributions were measured via gel permeation chromatography (GPC) on a Shimadzu modular system, comprising an auto injector, a Phenomenex 5.0 μm bead-size guard column (50 × 7.5mm) followed by three 5.0 μm Phenomenex phenogel-OH linear columns (105, 104, and 103 Å), and a differential refractive-index detector. The eluent was N,N’-dimethylacetamide (DCM, 1 mL/min) with 0.05% BHT added as a radical inhibitor and 0.05% LiBr added to reduce aggregation of the polymers. The system was calibrated with polystyrene standards and PMMA ranging in molecular weight from 106 to 200 g mol-1.
3.2.2. Nuclear Magnetic Resonance spectroscopy (NMR)
NMR is a spectroscopic method that is widely used for the analysis of many chemical compounds. It is based on the absorption of electromagnetic radiation in the radio- 3. Analysis techniques and synthesis of reagents 59
frequency range (4-600MHz) by atomic nuclei. However, unlike UV-visible, and infrared spectroscopy it is based on the absorption by the nucleus of atoms rather than the vibration of functional groups.
Fourier transform NMR
In Fourier transform NMR, the nuclei are exposed to brief pulses of intense radiation instead of scanning the radio-frequency range as is done in standard NMR. In general the pulse time τ=10 μs and the time between pulses are in the order of several seconds. During this period a time-domain radio frequency signal called a Free Induction Decay (FID) is emitted by the nuclei as it returns to its ground state.
The FID can be detected by a radio receiver perpendicular to static magnetic field and can then be digitized and processed by a computer. Normally this time-domain result is then averaged with repeat measurements to reduce noise, and is converted to a frequency domain through the use of a mathematical Fourier transform.
Effects on NMR spectra
The radio frequency radiation that is adsorbed by the nucleus is affected by several factors. The first is that chemical shifts occur from either shielding or deshielding of the proton by nearby groups resulting in a chemical shift. The second effect is spin-spin coupling caused by the interaction between the proton and protons that are bonded one carbon away. Spin-spin coupling causes the proton peak to be split into several smaller peaks.196
Both chemical shifts and coupling patterns are used in identification and structural analysis. Experimentally, peak separations resulting from chemical shift are proportional to the field strength or frequency of the applied field.197
NMR spectroscopy was carried out in this thesis using a Bruker 300 MHz spectrometer. 3. Analysis techniques and synthesis of reagents 60
3.2.3. Infrared spectroscopy
Infrared or IR spectroscopy is a spectroscopic method that involves the excitation and vibration of functional groups through exposure to the electromagnetic region between the visible and microwave regions (4000-400 cm-1).
Theoretically the IR spectrum is characteristic of the entire molecules and a peak by peak correlation will give excellent evidence for the identity of the compound. However, there are many slight variations and overlapping peaks that prevent this from occurring. In practice, unless enantiomers are being analyzed, it is unlikely that you will get the exact same spectra. It is true however that many functional groups give absorptions in key bands and these are generally used as part of a tool for identification.
The IR spectrum can be broken up into several parts. Less than 100cm-1 the energy is converted into molecular rotation and the peaks appear as discrete lines. Between 100 and 10,000 cm-1, the energy is converted into vibrations in the organic molecule, however, because some of the energy is also converted into molecular rotation the spectra appears as somewhat broadened peaks. The stretching (symmetric and asymmetric), bending or scissoring (symmetric or asymmetric), out of plane bending or wagging molecular motions are most commonly used in analysis.
The equipment used to carry out FTIR analysis as part of this thesis was using a Bruker IFS 66/S system.
Attenuated Total Reflectance – Fourier Transform Infrared spectroscopy (ATR-FTIR)
ATR-FTIR is a method for analyzing solid samples using infrared spectroscopy. The basis of this is that you reflect an infrared beam off the sample and measure the absorption from the reflected beam. Unfortunately the reflected radiation is of a very low intensity. To get around this the infrared beam is bounced through a crystal (usually Sapphire) at an angle such that the beam is totally internally reflected. The solid sample sits on the crystal, and the result is that the beam will be reflected off the sample 3. Analysis techniques and synthesis of reagents 61
multiple times. This effectively amplifies the absorption signal, as illustrated in Figure 3.5197
SAMPLE
Source Detector
Figure 3.5: Representation of ATR-FTIR operation.
3.2.4. FMOC loading test
One of the factors governing synthetic performance is the number of ‘active’ groups on the surface; that is, the number of functional groups that can easily be reached by the reactants, and can be quantified by a FMOC loading test. 9-fluorenylmethoxycarbonyl (FMOC)-β-alanine is coupled to the functional groups of the polymer on the surface. After thoroughly washing, the FMOC groups are cleaved from the lanterns and the amount of free FMOC can be determined by UV-spectroscopy, thus providing a measure for the loading capacity of the lanterns.198
O OH O N H O
Figure 3.6: FMOC-β-alanine
FMOC loading test of polystyrene based polymers
The procedure for aminomethylation199 of polystyrene polymers followed by FMOC coupling and load determination was conducted as follows: 3. Analysis techniques and synthesis of reagents 62
Hydroxymethylphthalimide (0.734 g) was dissolved in 49 mL of a dry mixture of 20% trifluoroacetic acid (TFA) in DCM that had been dried overnight with a spatula of P2O5. Methanesulfonic acid (2.46 mL) was then added slowly with stirring. This solution was placed over the lanterns and they were slowly shaken at room temperature for 24 hours. It’s important to release the pressure from the container at regular intervals during the first hour to prevent pressure build up. The lanterns were drained and washed with 50mL of 20 % TFA/DCM solution (for 10 min), twice with 50 mL of DCM (15 min and 45 min) and finally with 50 mL of methanol (15 min). The lanterns were then refluxed overnight with 75 mL of 5% hydrazine hydrate in methanol. The lanterns were then drained and refluxed lanterns four times in methanol for 25 min. The lanterns were finally washed by soaking for 15min in DCM, followed by 1% TFA/DCM and DCM again. Dry in vacuo overnight.
After drying the lanterns were washed in the following washing steps: 1. 100 mL of 5% triethylamine (TEA) solution (in 50% DMF/DCM) for 10 min. 2. Repeat 3. 100 mL of 50% DMF/DCM, 5 min 4. 100 mL DCM, 5 min 5. 100 mL 50% DMF/DCM, 5 min 6. Final wash of 100 mL DCM, 5 min. 7. Dry in vacuo overnight.
React lanterns overnight at room temperature with a solution of FMOC-β-alanine (3.72 g), N-hydroxybenzotriazole.H2O (2.2 g), N,N'-diisopropyl carbodiimide (1.88 ml) in 100 mL of 20% DMF/DCM.
To cleave the FMOC from the lanterns a single lantern was reacted with 20mL of 20% piperidine in DMF for 45 minutes. To determine the loading, 1 mL of this was solution was diluted with 10mL of 20% piperidine in DMF solution. Absorbance was then measured at 301 nm against a blank 20% piperidine solution in N,N’- dimethylformamide (DMF). 3. Analysis techniques and synthesis of reagents 63
FMOC loading in nmole per Lantern is calculated via Eq. 3.3. derived from the Beer- Lambert law 200 absorbance× 220 loading (nmoles / lantern) = Eq. 3.3 0.0078 (the factor 220 is a dilution factor, and the factor 0.0078 is the extinction coefficient.)
1. hydroxymethylphthalimide Methanesulfonic acid * 20% TFA/DCM *
2. hydrazine hydrate
n n
H2N
FMOC-β-Alanine N-hydroxybenzotriazole.H2O N,N' diisopropyl carbodiimide H N
* *
n piperidine n
HN HN O O
NH NH HO O O O
Scheme 3.2: FMOC loading test. Aminomethylation followed by FMOC coupling and decoupling with piperidine201
FMOC loading test of poly(acrylic acid) based polymers
FMOC loading tests can also be conducted using an acid functionality such as on PAA as the anchor group. This is conducted in the same manner as the PS loading with a reaction coupling BOC protected hexamethylene diamine to the acid group instead of the aminomethylation of the phenyl group as is used in the PS loading test. 3. Analysis techniques and synthesis of reagents 64
O NH N 2 O H O O O H N BOC-ethylene diamine OH O NH O
HCl Dioxane
O
NH2 O NH
O OH O N H O FMOC-β-alanine
O O O H O NH N O HN O
Scheme 3.3: Mechanism for functionalization and loading of FMOC-β-alanine onto PAA lanterns
3.2.5. Gravimetric calculations and grafting ratio
The grafting ratio of a grafted lantern is the percentage weight increase of the surface, and is a common method for expressing the mass increase of a grafted material. The graft ratio of each grafted lantern was calculated by the following equation
weight − weight initial = graft PP × Eq. 3.4 Grafting ratio (wt%) initial 100% weightPP
Where weightgraft is the weight of each lantern after grafting and thorough washing with initial DCM and weightPP is the initial weight of each lantern. 3. Analysis techniques and synthesis of reagents 65
3.3. Synthesis of RAFT agents
Various RAFT agents were used during this thesis and many of them were used in different chapters. The synthesis of these RAFT agents is outlined in the following section.
3.3.1. Synthesis of 1-phenylethyldithiobenzoate (PEDB)
1-PEDB was synthesized as described by Perrier et al.202
Cl
S SH S S CH2
NaMeO Sulphur
dithio benzoic acid 1-Phenyl Ethyl dithiobenzoate
Scheme 3.4: Synthesis of 1-phenylethyldithiobenzoate
Benzyl chloride (12.6 g, 0.1 moles) was added drop wise over one hour to a round bottomed flask containing elemental sulphur (6.4 g, 0.2 moles), 25% sodium methoxide solution in methanol (36 g) and methanol (30 g). The resulting brown solution was then heated and allowed to reflux at 80°C overnight. After cooling to room temperature, the mixture was filtered to remove the white solid (sodium chloride), and then the methanol was removed via rotary evaporation. The resulting brown solid was then re-dissolved in distilled water (100 ml), and washed three times with diethyl ether (3x50 ml). A final layer of ether (50 ml) was added to the solution and the two phase mixture was then acidified with 32% aqueous HCl until the aqueous layer lost its characteristic brown color and the top layer was deep purple. The etherous layer was extracted (dithiobenzoic acid) dried over anhydrous calcium chloride, then the ether was removed by rotary evaporation.
This product, dithiobenzoic acid, was dissolved in 10ml of hexane and reacted with a 20% excess of styrene overnight with a small amount of para-toluene sulfonic acid 3. Analysis techniques and synthesis of reagents 66
(1%). The product was isolated by column chromatography through a silica packing using n-hexane as an eluent.
3.3.2. Synthesis of cumyl dithiobenzoate (CDB)
Cl S SH S S CH2 NaMeO Sulphur
dithio benzoic acid Cumyl dithiobenzoate
Scheme 3.5: Synthesis of cumyl dithiobenzoate
Benzyl chloride (12.6 g, 0.1 moles) was added drop wise over one hour to a round bottomed flask containing elemental sulphur (6.4 g, 0.2 moles), 25% sodium methoxide solution in methanol (36 g) and methanol (43 g). The resulting brown solution was then heated and allowed to reflux at 80°C overnight. After cooling to room temperature, the mixture was filtered to remove the white solid (sodium chloride) and then the methanol removed via rotary evaporation. The resulting brown solid was then re-dissolved in distilled water (100 ml), and washed three times with diethyl ether (3x50 ml). A final layer of ether (50 ml) was added to the solution and the two phase mixture was then acidified with 32% aqueous HCl until the aqueous layer lost its characteristic brown color and the top layer was deep purple. The etherous layer was extracted (dithiobenzoic acid) dried over anhydrous calcium chloride before the ether was removed by rotary evaporation.
The product was dissolved in n-hexane (10ml) and reacted with α-methyl styrene (20% excess) in the presence of 1% para-toluene sulfonic acid. The product was then isolated by column chromatography using silica gel and n-hexane as the eluent. 3. Analysis techniques and synthesis of reagents 67
3.3.3. Synthesis of 2-(2-cyanopropyl) dithiobenzoate (CPDB)
Cl S S CN S SH S S Na CH2 1. K Fe CN) NaMeO NaOH 3 ( 6 Sulphur 2. AIBN
dithio benzoic acid sodium dithiobenzoate Cyanopropyl dithiobenzoate
Scheme 3.6: Synthesis of 2-(2-cyanopropyl) dithiobenzoate
2-(2-cyanopropyl) dithiobenzoate was synthesized using the following method. Benzyl chloride (12.6 g) was added drop wise over one hour to a round bottomed flask containing elemental sulphur (6.4 g), 25% sodium methoxide solution in methanol (40 g) and methanol (40 g). The resulting brown solution was then heated and allowed to reflux at 80°C overnight. After cooling to room temperature, the mixture was filtered to remove the white solid (sodium chloride) and then the methanol removed via rotary evaporation. The resulting brown solid was then re-dissolved in distilled water (100 ml), and washed three times with diethyl ether (3x50 ml). A final layer of ether (50 ml) was added to the solution and the two phase mixture was then acidified with 32% aqueous HCl until the aqueous layer lost its characteristic brown color and the top layer was deep purple. The etherous layer was extracted (dithiobenzoic acid) and deionised water (120 ml) and 1.0 N NaOH (240 ml) were added to extract Sodium Dithiobenzoate to the aqueous layer. This washing process was repeated two more times to finally led to a solution of sodium dithiobenzoate.
Potassium ferricyanide (13.17g, 0.04mol) was dissolved in deionised water (200 ml). sodium dithiobenzoate solution (140 ml) was transferred to a conical flask equipped with a magnetic stir bar. Potassium ferricyanide solution was added drop wise to the sodium dithiobenzoate via an addition funnel over a period of 1 hour under vigorous stirring. The red precipitate was filtered and washed with deionised water until the washings became colorless. The solid was dried in vacuo at room temperature overnight. The product was recrystallised from ethanol. 3. Analysis techniques and synthesis of reagents 68
A solution of Azobisisobutyronitrile (AIBN) (5.84g, 0.021mol) and bis(thiobenzoyl) disulfide (4.25g, 0.014 moles) in ethyl acetate (80 ml) was heated at reflux for 18 h. After removal of the volatiles in vacuo the crude product was subjected to column chromatography with ethyl acetate: n-hexane (2:3) as eluent to afford 2-cyanoprop-2-yl dithiobenzoate as red liquid. After freezing overnight at -20oC the product crystallized to form a red solid.
3.3.4. Synthesis of cumyl phenyldithioacetate (CPDA)
S SH Cl C MgCl S S CH 2 CH2 CH2
1.CS2 Mg 2. H2O
phenyldithioaceticacid cumyl phenyldithioacetate
Scheme 3.7: Synthesis mechanism for cumyl phenyldithioacetate
CPDA was prepared via the following method.203 Benzyl Chloride (20g, 0.158 mol) was added drop wise to a mixture of magnesium turnings (3.75 g, 0.155 mol) in dry diethyl ether (100 ml). Following the vigorous initial reaction, the mixture was then chilled and carbon disulfide (12.0 g, 0.158 mol) was added drop wise over 30min. The mixture was then stirred at 0oC for 2 hrs. The mixture was then poured onto ice water (300 ml) and the aqueous portion was collected following three washes with diethyl ether. A final layer of diethyl ether was added, and the mixture was acidified using 30% aqueous HCl. The phenyldithioacetic acid (approx 7 g, 42 mmol, 27%) was collected via rotary evaporation of the ether. The acid was then reacted with α-methyl styrene (9.0 g, 76 mmol) with a small amount of acid catalyst (p-toluenesulfonic acid, 0.1g, 0.5 mmol) in
CCl4 (10.0 g). The product was then precipitated in cold methanol and recrystallized from methanol as large orange crystals (4.2g, 15mmol, 36% yield). The purity determined by NMR was better than 99%. 3. Analysis techniques and synthesis of reagents 69
3.4. Synthesis and purification of ATRP related compounds
3.4.1. Purification of CuBr
5g of CuBr was made into a slurry and ground using 1N sulphonic acid. Another 500ml of 1N sulphonic acid was added and the solution was stirred for 1hr. The solid CuBr was then filtered off and washed with the following: 2x100ml of glacial acetic acid, 3x30ml of absolute ethanol and 5x100ml of dry ether. The purified CuBr was then tried under vacuum at room temperature and stored in a dry atmosphere until required.
3.4.2. Synthesis of Me6TREN
204 The synthesis of Me6TREN was based on the methods outlined by Ciampolini and Queffelec.205 It should be noted that Queffelec’s synthesis method suggests using dicloromethane (DCM) as a solvent for purification by solvent extraction. Our experiments using DCM resulted in a white, water-soluble solid forming during evaporation of the DCM. It was found through NMR that this was the quaternary amine salt formed by the following reaction:
CH2Cl2 RN RNCH2Cl Cl NaOH
Scheme 3.8: Mechanism of quaternary amine formation.206
The reaction normally occurs with an alkyl halide such as CH3Cl, CH3Br or CH3I but in this case, it occurred with a compound containing two halide groups. The tertiary amine extracts one of the chlorine groups from the DCM to form the quaternary amine salt. This reaction was avoided by the use of diethyl ether as the extraction solvent.
The reaction for the formation of the ligand is given below in Scheme 3.9. 3. Analysis techniques and synthesis of reagents 70
N N 1. Formic acid Water Formalderhyde H3C H2N N NH2 N CH3 2. Sat NaOH H3C H2N N H3C CH3 CH3
Scheme 3.9: Mechanism of formation of Me6TREN
Me6TREN was synthesized by the following method:
A 250ml round bottom flask fitted with a condenser was purged with nitrogen for at least 15 minutes, placed in an ice bath and charged with a solution of formaldehyde (29.4g of 35 wt% solution in water, 0.34 moles) and formic acid (31.4g of 99% solution in water, 0.34 moles) in 40 ml of water. TREN (10g, 0.068 moles) in 60ml of water was added drop-wise over a period of 1hr. The mixture was then warmed to room temperature and then refluxed at 95oC overnight. The reaction was then allowed to cool and volatiles were removed by rotovaporation. The brown residue was then treated with saturated NaOH solution until pH > 12. The oily layer was then extracted into diethyl ether with 3x50ml and the organic phase was dried with anhydrous magnesium sulfate and rotary evaporated to give a yellow oil. This yellow oil was then vacuum distilled with the product coming off at 69-74oC.
Yield was 73% and the purity was >95% by NMR; Signals from the starting TREN 1 were below 1% with respect to Me6TREN. H NMR (CDCl3) δ: 2.44 (dd, 6H), 2.16 (dd, 6H), 2.00 (s, 18H). It is worth noting that Queflec incorrectly labels the two dd signals as 12 equivalent hydrogens each, where they should only represent 6 hydrogens. 4. ATRP graft polymerization from SynPhase lanterns 71
4. ATRP graft polymerization from SynPhase lanterns
4.1. Aim
The objectives of this chapter are to conduct ATRP controlled grafting off functionalizes substrates. First, it was necessary to conduct a full set of solution polymerization so that a set of polymerization conditions could be found that produced controlled polymers. Next, lanterns were functionalized with carbon tetrabromide in styrene to produce short-chain tethered functionalized lanterns and also directly functionalized with carbon tetrabromide in methanol. These lanterns were then used as initiators to conduct ATRP grafting reactions. The applicability of these lanterns as SPOS supports was also assessed through a FMOC loading tests.
4.2. Introduction and theory
Atom transfer radical polymerization (ATRP) is a living polymerization method based around the Kharasch chemical synthetic method. ATRP was simultaneously discovered by Sawamoto and Matyjaszewski in the 1980’s and has proven to be a superior method for the production of narrow molecular weight polymers and advanced block copolymers.84,106 A more comprehensive outline of the ATRP mechanism and is available in Chapter 2.
Several polymeric grafting systems have been produced that use ATRP to graft off a polymeric backbone, however all of these are from dissolved polymeric backbones. The only way that ATRP polymerization has been conducted off insoluble substrates is from silicon, silica or gold surfaces. These will be are discussed together with discussions on grafting off dissolved polymeric substrates below.
There are a substantial range of examples where ATRP has been used to graft off dissolved polymeric backbones. A selection of the more interesting examples is discussed below. Firstly, Paik et al.207 conducted graft polymerization of styrene and meth(acrylates) off a functionalized polyvinyl chloride (PVC) polymer substrate 4. ATRP graft polymerization from SynPhase lanterns 72
functionalized with 1 mol% chloroacetate groups. He showed that the polymer molecular weight increased in each case by grafting onto the PVC backbone and the polydispersities of the polymers were effectively unchanged. Differential scanning calorimetry (DSC) was used to prove that there was a reduction in the crystallinity of the PVC. It is interesting to see that it was necessary to functionalize the PVC to make it an effective initiator for ATRP, as the chloride groups that are part of the PVC are bonded too strongly to the backbone, and thus are not able to act as initiators for the polymerization reaction.
Another example of ATRP from functionalized polymeric materials, is the grafting of styrene, isobornyl acrylate, and MMA from EXXPRO which is a commercially available poly(isobutylene-co-p-methylstyrene-co-p-bromomethylstyrene).208-210 In a similar way ATRP was used to graft well-defined polymers from polyethylene using a polyethylene-poly(glycidyl methacrylate) with the epoxide groups transformed into α- bromoesters.125
Another method of functionalizing a backbone polymer for use as an ATRP initiator is through the bromination of chlorination of the phenyl groups in polystyrene. This has been achieved using polyethylene-polystyrene copolymers211 and also syndiotactic polystyrene.212
Literature on ATRP grafting from surfaces is, as previously mentioned, somewhat more restricted. However there are several examples of ATRP being used to graft from insoluble substrates.
In a novel piece of work that has found commercial application, ATRP was used to graft polyacrylamide from modified silica particles and silicon wafers.213,214 The former was successfully then used in packed HPLC columns to separate basic proteins.
Block copolymers have also been grafted from silica surfaces; one example of this was when polystyrene-co-poly(tert-butyl acrylate) was grafted onto silica wafers by Matyjaszewski et al.215 This paper also demonstrates that the addition of a small amount of copper (II) halide at the beginning of the polymerization provides enough deactivator to make the use of a sacrificial initiator unnecessary. 4. ATRP graft polymerization from SynPhase lanterns 73
As an example of the range of systems that ATRP can be used to control, is shown by the ambient temperature ATRP of MMA using 2-bromoisobutyrates attached to a gold surface with Cu-Br and tris[2-(dimethylamino)-ethyl]amine as the catalyst. This led to densely chemically bound PMMA brushes on the gold surface.216,217
As can be seen, ATRP grafting from both surfaces and polymeric backbones is not a new concept. The ease of creating functional surfaces from common grafting substrates such as silica, silicon and metal surfaces such as gold has made it an obvious route for creating graft polymers. However, to my knowledge, there has not yet been work that has produced a grafted polymer from an insoluble polymeric surface.
This work will centre on the use of ATRP to control grafting off specially functionalized Mimotopes SynPhase lanterns. The first step was to create a functionalized surface with a halide group that will allow us to initiate polymerization. This was done by exploiting the chain-transfer properties of carbon tetrabromide and is discussed in the following section
4.2.1. Surface functionalization through γ- radiation and carbon tetrabromide chain transfer
The method that we have developed for the functionalization of our polymeric surfaces to allow us to conduct ATRP grafting, uses the chain transfer properties of carbon tetrabromide coupled with the ability of γ-radiation to induce surface bound radicals in normally chemically-inert polymers such as polypropylene (PP) and polyethylene (PE). The concept is that the radiation generates surface radicals that then polymerize a monomer in solution. When a significant amount of carbon tetrabromide is added to this and it results in chain transfer reactions where the polymer radical is transferred to the • CBr4 moiety, leaving a bromine atom capping the polymer chain and a CBr3 radical.
Carbon tetrabromide is a well known chain transfer agent218 that essentially reacts with a growing radical to leave a bromine group and expel a carbon-tribromide radical as per the mechanism shown below 4. ATRP graft polymerization from SynPhase lanterns 74
R1 R1
CBr4 Br CBr3
R2 R2
Scheme 4.1: Mechanism for carbon tetrabromide radical transfer reaction
The main effect of a chain transfer agent is that it produces a competition between the chain transfer reaction and the polymeric addition reaction. This reduces the resulting molecular weight from what would be expected without the addition of the chain transfer agent that can be quantified by the Mayo equation. The Mayo equation relates the number average degree of polymerization xn to the number average degree of polymerization for the system without a chain transfer agent present, to the concentrations of chain transfer agent, monomer and the chain transfer constant Cs.
⎛ ⎞ 1 = ⎜ 1 ⎟ + [S] Mayo equation: ⎜ ⎟ Cs Eq. 4.1 x x [M ] n ⎝ n ⎠0
This equation states that the effect of chain-transfer agents being added to a polymerization is that the molecular weight of the polymer will be reduced by competition between the chain transfer reactions and the polymeric propagation reactions.
A table of Cs values from commonly used chain transfer agents is shown below in Table 4.1. 4. ATRP graft polymerization from SynPhase lanterns 75
Cs (for polymerization) Styrene Methyl Methacrylate 60oC 80oC 60oC 80oC Benzene 0.28 x 10-5 1.5 x 10-5 0.4 x 10-5 2.4 x 10-5 Carbon tetrachloride 0.0092 0.013 0.0005 0.0024 Carbon tetrabromide 1.78 2.3 0.27 0.33 1-Octanethiol 19.0 (50oC) - - - 1-Dodecanethiol 14.8 - - -
Table 4.1: Reference of chain transfer constants for common chain transfer agents219
The SynPhase lanterns were functionalized by placing in a solution of styrene monomer with a very high concentration of CBr4 and exposing it to γ-radiation for 48 hours. As such, short-chains bromine-terminated chains of polystyrene are grafted to the surface through the chain transfer reactions of CBr4. This is termed the “functionalization stage” and is shown below:
Functionalization
gamma radiation Initiation SS SS
SS SS T Propagation
CBr SS T CBr4 SS T Br 3 Chain-transfer
SS = solid support T = Tether polymer
Figure 4.1: Mechanism for surface functionalization of SynPhase lanterns using carbon tetrabromide chain transfer
The result is that a bromine atom will be bonded to the surface of the lantern by a tethering polymer chain T. The next step is to conduct ATRP polymerization from these tethered bromine atoms, which will be discussed in greater detail in the next section. 4. ATRP graft polymerization from SynPhase lanterns 76
Mechanism of ATRP grafting from functionalized surfaces
The mechanism for grafting off these tethered surface bromine atoms can be derived by modifying the standard mechanism for a solution polymerization off a standard ATRP initiator such as ethyl-2-bromoisobutyrate or 1-bromoethylbenzene (represented as R-Br in the below equation)
RBr CuBr-Ligand R P CuBr2-Ligand
kp M
Scheme 4.2: ATRP kinetic equilibrium and polymerization reaction
If we substitute the functionalized substrate, SS-T-Br, for R-Br in the solution polymerization, then we get the below mechanism:
SS T Br CuBr-Ligand SS T P CuBr2-Ligand
kp M T = Tether polymer P = ATRP polymer
Scheme 4.3: Mechanism for ATRP controlled graft polymerization from functionalized SynPhase lanterns
This shows the reversible reductive extraction of the tethered bromine atom to yield a
CuBr2 complex and a polymeric radical that can then undergo propagation. From this mechanism we can derive a set of equations that can kinetically describe the polymerization.
The general kinetic equation118 that gives the rate of polymerization in a radical polymerization system has be given by Eq. 4.2 below:
= • Eq. 4.2 Rp k p[M ][P ] 4. ATRP graft polymerization from SynPhase lanterns 77
• Where Rp is the rate of polymerization, [M] is the monomer concentration and [P ] is the radical concentration.
Matyjaszewski118 modified this equation to take into consideration the ATRP mechanism, resulting in the equation:
= − d[M ] = × [CuBr] Rp k p Keq[M ][I]0 Eq. 4.3 dt [CuBr2 ]
Where Keq is the equilibrium constant for the ATRP equilibrium and [I]0 is the initial initiator concentration.
In the case of our grafting experiments we can say that the initial initiator concentration
[I]0 is equal to the sum of the number of moles of surface-bound bromine groups
( msuface−Br )divided by the total volume of the reaction (V).
(m − ) [I] = suface Br Eq. 4.4 0 V
If there is no termination in an ATRP polymerization system it follows that the theoretical molecular weight will be given by the number of moles of monomer converted to polymer, divided by the number of polymer chains. Thus the degree of polymerization will be given by118:
Δ = [M ] DPn Eq. 4.5 [I]0
This can then be modified to give the theoretical molecular weight for any given conversion (x) as
[M ] .x.M monomer = 0 n Eq. 4.6 Mnth [I]0
The following section will describe firstly a set of ATRP reactions using standard solution initiators that will allow for the optimization of the polymerization conditions. Following this, graft polymerization will be conducted off a number of different functionalized polymeric substrates using these conditions. 4. ATRP graft polymerization from SynPhase lanterns 78
4.3. Solution ATRP
As part of developing a system for using ATRP to graft from insoluble surfaces, it is advantageous to investigate the ATRP system being used with a standard solution initiator. This will allow us to develop an understanding of the effects that various parameters will have on the polymerization kinetics and the resulting polymers.
The system that was decided to be most appropriate to our given application uses copper bromide as the metal halide, tris(2-dimethylaminoethyl)amine (Me6TREN) as the ligand and ethyl-2-bromoisobutryate (E-2-BIB) as the initiator.
N
O CuBr H C 3 N N CH3 H3C N O H3C CH3 CH3 Br
Copper (I) bromide tris(2-dimethylaminoethyl)amine ethyl-2-bromoisobutyrate
Figure 4.2: Components of ATRP system being used
This system has been used in a number of publications and has been found to be an efficient and robust system for the polymerization of styrenics, (meth)acrylates, and for a variety of other monomers.205,220,221
4.3.1. Experimental method
These polymerizations were conducted using Me6TREN as a ligand and CuBr as the transition metal catalyst. The initiator in each polymerization was ethyl-2- bromoisobutyrate, which was used because of its application in a wide variety of ATRP systems.108
The Me6TREN and copper bromide were prepared as described in Chapter 3. The solvents and ethyl-2-bromisobutyrate were purchased from Aldrich and used as received. 4. ATRP graft polymerization from SynPhase lanterns 79
Monomer, initiator, CuBr and solvent were weighed into a Schenck and were degassed by sparging with nitrogen for 30 min. The ligand was injected from a degassed syringe and the solution stirred to aid in the complex formation. The Schenck vial was then immersed in an oil bath at the required temperature. Samples were taken with a degassed syringe and placed in a pre-weighed aluminium plan and dried overnight in the fume cupboard.
Polymers were then dried for 24hrs in vacuo and dissolved in tetrahydrofuran (THF) for GPC analysis using the THF GPC system outlined in Chapter 3.
4.3.2. Solution ATRP results
Due to the multi-component nature of ATRP polymerization, there is a multitude of variables that have an effect on the polymerization. These variables include standard polymerization conditions such as temperature, initiator concentration, and solvent type and content. Beyond this, there are changes in the metal complex, initiator structure, and ratio between initiator, metal and ligand. However, since the objectives of this section are to develop a set of conditions that will effectively control solution polymerization and allow expansion to graft polymerization, it not necessary to investigate all these variables. The various effects on ATRP systems have been extensively investigated elsewhere and two comprehensive reviews are recommended as further reading on this topic.108,118
CuBr and Me6TREN form a 1:1 coordination complex and as such the ratio between these will be held constant for all these experiments and the effect of the complex concentration, initiator concentration and temperature will be investigated.
The effect of ligand to Cu(I) ratio has been investigated by Huang et al.222 who found that at a ligand:CuBr ratio of < 1 there was inefficient initiation, and higher concentrations of ligand transfer to ligand reactions were contributing, that resulted in a flattening out of the molecular weight at higher conversion. It should be noted that often 4. ATRP graft polymerization from SynPhase lanterns 80
in ATRP concentrations are referred to in molar ratios between various components rather than absolute concentrations. This convention will be used through this thesis.
For comparisons sake it is often useful to calculate the apparent rate constant kapp of the polymerization experiment. This is defined as the slope of the first order kinetic plot and allows for quantitative comparison of different polymerizations and is a compounded
× [CuBr] rate constant defined as k p Keq[I]0 . [CuBr2 ]
Effect of catalyst concentration of ATRP polymerization of styrene
The established ATRP theory states that the actual amount of catalyst should not have any effect on the rate of polymerization or on the molecular weight of the polymer produced. The only requirement is that the catalyst must be available in sufficient quantities to adequately control the polymerization. However, it should not overly effect the polymerization otherwise.
The effect of the catalyst concentration on the ATRP of styrene was determined by polymerizing styrene in bulk with E-2-BIB as an initiator; CuBr as the copper catalyst, and Me6TREN as the ligand in a ratio of 1000:1 monomer:initiator. The ratio between ligand and metal was kept constant at 1:1 ratio. The polymerization was conducted at 90oC in an oil bath, with samples taken regularly using a degassed syringe.
Figure 4.3 shows the first-order kinetic plot of three polymerizations that were conducted while varying the catalyst concentration. There were some small changes in the slope of these graphs and kapp was found to be 0.0455, 0.0386 and 0.0699 for the 2000:1, 1000:1 and 500:1 monomer to catalyst ratios respectively. While there are changes in kapp there doesn’t seem to be a significant catalyst concentration effect on the polymerization. In a normal solution polymerization, increasing the catalyst concentration will shift the equilibrium towards the dormant side, slowing the polymerization. However, in a system where all the catalyst is not dissolved (i.e. a heterogeneous polymerization) then the solution catalyst concentration is determined by 4. ATRP graft polymerization from SynPhase lanterns 81
1.0
0.8
0.6 ln(1/1-x) 0.4
0.2
0.0 0 5 10 15 20 25 Time (hr)
Monomer:Catalyst ratio 2000:1 1000:1 500:1
1.6 1.4
1.2 PDI 100000 1.0 0.0 0.1 0.2 0.3 0.4 80000
60000
Mn 40000
20000
0 0.0 0.1 0.2 0.3 0.4 Conversion Monomer:Catalyst ratio 2000:1 1000:1 500:1
Figure 4.3: Effect of catalyst ratio on ATRP polymerization of styrene at 90oC and a styrene:E-2-BIB ratio of 1000:1 and varying the styrene:CuBr:Me6TREN ratio. Theoretical molecular weight shown by the solid line 4. ATRP graft polymerization from SynPhase lanterns 82
the solubility of the catalyst and would not affect the rate of polymerization, as seen here.
The data also shows that the molecular weights of the polymers are not affected by the catalyst concentration. The molecular weights for the 2000:1 and 1000:1 concentrations are exactly the same and the 500:1 catalyst ratio polymerization which showed a slight variation from this consistent molecular weight. This is as the initiator concentration in each polymerization is constant, and thus the molecular weights produced should be constant as predicted by Eq. 4.6. It will be noticed that the molecular weight measured varied from the theoretical molecular weight, and this is possibly caused by inefficiency in the initiation reaction.
It can also be seen that at the start of the polymerization, the polydispersity is around 1.6 and then as the polymerization proceeds, this index decreases to around 1.1. This is because in the initial stages of the polymerization we find that there is an artificially high concentration of radicals than later, due to the PRE, and thus there is an increased amount of bimolecular termination. Once the ATRP equilibrium has stabilized, there is enough CuBr2 deactivator present that termination decreases to an almost non-existent level and the polydispersity of the polymers decrease as a typical PDI evolution for an ATRP.
It can be thus concluded that, as far as control of the polymerization is concerned, there is no significant difference between the three catalyst concentrations and the concentration of the other species in the system will have a greater effect on the polymerization.
Effect of initiator concentration of ATRP of styrene
Theoretically, if we inspect Eq. 4.3 and Eq. 4.6 we can see that the molecular weight of the polymer is determined by the initiator concentration. The rate of polymerization will also be effected by the initiator concentration, however the relationship is somewhat complicated by the ATRP equilibrium and that the ratio between CuBr and CuBr2 will also be effected by the initiator concentration. 4. ATRP graft polymerization from SynPhase lanterns 83
= − d[M ] = × [CuBr] R p k p K eq [M ][I]0 Eq. 4.3 dt [CuBr2 ]
[M ] .x.M monomer = 0 n Eq. 4.6 Mnth [I]0
To test this, an experiment varying the concentration of E-2-BIB without varying the other experimental parameters was conducted. The polymerization was prepared in the same manner as described at the beginning of this chapter using a 1000:1:1 ratio o between styrene:CuBr:Me6TREN at a polymerization temperature of 90 C. The ratio of styrene to E-2-BIB was varied between 2000:1, 1000:1, 500:1.
It can be seen in Figure 4.4 that there is little change in the rate of polymerization, and the molecular weight decreases with increased initiator concentration. This is quantified by looking at kapp which is calculated to be 0.0409, 0.0473 and 0.571 with increasing monomer initiator ratio from 2000:1, 1000:1 and 500:1 respectively. These experimental molecular weights correlate very well with those determined by theory. PDI of all polymers showed that the polymerizations were well controlled and of a living nature.
This effectively shows that it is possible to adjust the target molecular weight of the polymerization system simply by adjusting the initiator concentration with good control over the range of initiator contents that were analyzed. 4. ATRP graft polymerization from SynPhase lanterns 84
0.5
0.4
0.3
0.2 ln(1/1-x)
0.1
0.0 0123456 Time (h) Monomer:E-2-BIB ratio 2000:1 1000:1 500:1
1.4
1.2 PDI 60000 1.0 0.0 0.1 0.2 0.3 0.4
40000 Mn 20000
0 0.0 0.1 0.2 0.3 0.4 Conversion Monomer:E-2-BIB ratio th 2000:1 1000:1 500:1 Mn
Figure 4.4: Effect of initiator concentration on the ATRP polymerization of o styrene. Styrene was polymerized at 90 C with a styrene:CuBr:Me6TREN ratio of 1000:1:1 and varying E-2-BIB concentration. 4. ATRP graft polymerization from SynPhase lanterns 85
Effect of temperature on ATRP polymerization of styrene
Temperature has significant effects on ATRP polymerization systems beyond the effects normally associated with polymerization systems. Increasing the temperature in a polymerization system will normally increase the kp and kt of the monomer. This will increase the rate of polymerization and the amount of termination, thus broadening the molecular weight. In ATRP, temperature will also increase the solubility of the catalyst and can also have an effect on the equilibrium constant of ATRP reaction. A range of optimum temperatures should exist for this system below which the rate of polymerization is too slow to effectively produce polymers, and above which the polymerization starts to loose control.
The polymerization was conducted in the same manner as outlined at the beginning of the chapter, using a ratio of 1000:1:1:1 between styrene:CuBr:Me6TREN:E-2-BIB at temperatures of 60oC, 90oC, 110oC and 130oC.
The calculated kapp was found to be 0.0154, 0.0487, 0.1793 and 0.4999 for the increasing temperatures. The molecular weight conversely is constant over the temperature range; however there was still a variation from the theoretical molecular weight. Additionally it should be noted that the molecular weight control declines at higher temperatures which can be attributed to self-initiation of styrene that will occur at this temperature.
Additional information in relation to the activation energy of the reaction is available from the Arrhenius equation (Eq. 4.7)223:
⎛ E ⎞ ⎜ − ⎟ Eq. 4.7 k = Ae⎝ RT ⎠
Which can be rearranged to give:
E ln(k) = − + ln(A) Eq. 4.8 RT 4. ATRP graft polymerization from SynPhase lanterns 86
1.4
1.2
1.0
0.8
0.6 ln(1/1-x)
0.4
0.2
0.0 0123456 Time (h) Polymerization temperature 60oC 90oC 110oC 130oC
1.4
1.2 PDI
1.0 1200000.0 0.2 0.4 0.6
100000
80000
60000 Mn 40000
20000
0 0.0 0.2 0.4 0.6 Conversion Polymerization temperature 60oC 90oC 110oC 130oC
Figure 4.5: ATRP of styrene; Effect of polymerization temperature on polymerization. Polymerization was conducted at varying reaction temperature with a styrene:CuBr:Me6TREN:E-2-BIB ratio of 1000:1:1:1. 4. ATRP graft polymerization from SynPhase lanterns 87
Where k is the rate constant for the polymerization, E is activation energy of the reaction R is the universal gas constant and A is termed the pre-exponential factor or often known as frequency factor.
If we substitute Eq. 4.2 into this equation then we get the following equation:
⎛ R ⎞ ⎜ p ⎟ = − E + ln⎜ • ⎟ ln(A) Eq. 4.9 ⎝[M ][P ]⎠ RT
Which if we assume that [M] and [P•] are constant then the equation rearranges to
E ln()R = − + C Eq. 4.10 p RT where C a combination of all the constants and is given by:
C = ln(A[M ][P• ])
If we assume that the density of the solution is relatively constant over the conversion range, which should be true for a bulk polymerization, the assumption that [M] is constant is valid. Since the polymerization is assumed to be at a steady state for the majority of the polymerization, then [P•] should be constant for most of the polymerization, excluding the very beginning before the equilibrium is established and at the very end where monomer concentration is significantly decreased.
Thus, if we plot ln(Rp) vs. 1/T then we should get a straight line with a slope of –E/R. This can be seen in Figure 4.6.
The resulting line has a slope of -7606.53 and an intercept of 18.14 which gives us an activation energy of 63.24 kJ mol-1. This is very similar to the value obtained by Cheng et al.224 who found the activation energy for the polymerization of styrene using bis- (1,10-phenanthroline) copper bromide to be 75 kj/mol. 4. ATRP graft polymerization from SynPhase lanterns 88
0
-1
-2 ) p
ln(R -3
-4
-5 0.0024 0.0025 0.0026 0.0027 0.0028 0.0029 0.0030 1/T (K)
Figure 4.6: Arrhenius plot of ATRP of styrene
GPC analysis of ATRP polymers
The actual molecular weight distributions of the generated polymers can provide significant information to us about what is occurring mechanistically in the system. A set of sample GPC spectra are shown below and will be discussed in terms of the mechanistic features that they represent. 4. ATRP graft polymerization from SynPhase lanterns 89
Increased polymerization time Normalized Response Normalized
3.5 4.0 4.5 5.0 log(M ) w
Figure 4.7: Molecular weight distribution showing the evolution of molecular weight with time for the ATRP controlled polymerization of styrene
o o o T=60oC T=90 C T=110 C T=130 C Normalized Response Normalized
3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 5.8 log (M ) w
Figure 4.8: Effect of temperature on the molecular weight distributions of ATRP polystyrene 4. ATRP graft polymerization from SynPhase lanterns 90
In Figure 4.7 it can be seen that as the polymerization proceeds and the conversion increases, the molecular weight of the polymers increase. However it should also be noted, that the molecular weight distributions are also narrowing as the polymerization progresses. This is a result of the persistent radical effect that exists as part of ATRP polymerization, which stipulates that in the early stages of the polymerization there will be an increased rate of termination while there is a high concentration of radicals in the 129 system with insufficient CuBr2 concentration to deactivate it. As the polymerization progresses, the polydispersity gets narrower and a slight shoulder starts to form on the low molecular weight side of the trace. This is from early termination reactions that came from this PRE that was discussed earlier.
Figure 4.8 shows the molecular weight distributions of polymers produced at different temperatures and it can be seen here that under optimum conditions the molecular weight distributions are narrow and symmetrical. However at 60oC it can be seen that that molecular weight is significantly broader that 90oC or 110oC and at 130oC there is a significant shoulder on the low molecular weight side that is caused by the increased termination of polymer chains during the polymerization process.
4.3.3. Solution polymerization of other monomers
While styrene is being primarily used as a monomer in this chapter because of its applicability in combinatorial chemistry applications and ease of polymerization, this ATRP system is applicable to polymerizing a wide range of monomers. In order to look at some examples of monomers that can be polymerized by ATRP beyond styrene, several other monomers were polymerized, including MA, MMA and t-BA.
All the polymerizations were conducted as described in the above procedures with a monomer:CuBr:ligand:initiator ratio of 1000:1:1:1. Temperatures varied depending on the monomer and the results can be seen in Table 4.2.
MMA and t-BA were additionally polymerized using a different ligand that had shown promise polymerizing acrylic monomers. The major advantage of this monomer is that 4. ATRP graft polymerization from SynPhase lanterns 91
it is commercially available thus, does not require extensive synthesis procedures like
Me6TREN does. This ligand is (N-[2-(Dimethylamino)ethyl]N,N',N'-trimethyl-1.2- ethanediamine) or PMDETA and is the results for this polymerization are shown below.
Detailed graphs showing molecular weight evolution and the first-order kinetic plots for these polymerizations can be seen in Appendix 1.
Th Monomer Ligand Temperature Time Conversion Mn Mn PDI o MA Me6TREN 90 C 4 h 51% 13,49644,011 1.12 o MMA Me6TREN 60 C 4 h 31% 41,04031,456 1.24 MMA PMDETA 60oC 4 h 32% 43,17032,472 1.05 t-BA PMDETA 60oC 5 h 13% 14,181 9,766 1.21 t-BA PMDETA 90oC 6 h 40% 34,73730,454 1.12
Table 4.2: Summary of results for solution polymerization of acrylic monomers
There are some variations from the theoretical molecular weight in the tables given above, however it must be remembered that the molecular weight given for the polymer is from GPC analysis vs. MMA standards and as such does not offer an absolute measure of the true molecular weight.
It has thus been shown that this polymerization is applicable to a range of acrylic monomers as well as styrene.
4.3.4. Conclusions
These experiments showed that the Me6TREN/CuBr system is quite robust and generally produces well controlled polymers for a wide variety of conditions. It was found that the optimum conditions for control of the polymerization were an initiator ratio between 500:1 and 2000:1, catalyst concentration of 1000:1 or 2000:1 and a temperature between 90oC and 110oC. The optimal conditions that will be used for ATRP grafting later were decided to be 1000:1:1 ratio of Monomer:ligand:CuBr and 90oC temperature. 4. ATRP graft polymerization from SynPhase lanterns 92
4.4. Reinitiation of solution ATRP polymers
When producing living radical polymers it becomes necessary to show that the polymerization is truly living rather than just controlled. This is done by reinitiating a polymer that was created in a previous polymerization. This reinitiation is also considered the first step in producing more advanced polymers such as block copolymers and stars.
The first step is to produce an ATRP macro initiator with a target molecular weight of 5000 g/mol at 50% conversion, which was prepared as follows: Styrene (0.19 moles, 20g), CuBr (1.9x10-4 moles, 0.027g) and E-2-BIB (1.9x10-3 moles, 0.371g) were weighed into a Schenck vial and degassed by -4 sparging with nitrogen for 30 minutes. Me6TREN (1.9x10 moles, 0.043g) was injected via a degassed syringe and the polymerization was started by immersion into an oil bath at 90oC for 12 hours.
The resulting polymer was passes through a basic alumina column and then precipitated into methanol to remove the copper catalyst and purify the polymer.
th The resulting polymer had a Mn of 4775 (Mn = 5190), PDI of 1.17 at 46% conversion.
This polymer was then used as a macroinitiator to polymerize styrene in the following manner: Styrene (0.19 moles, 20g), CuBr (1.9x10-4 moles, 0.027g) and macroinitiator (1.9x10-3 moles, 0.9809g) were weighed into a Schenck vial and degassed by -4 sparging with nitrogen for 30 minutes. Me6TREN (1.9x10 moles, 0.043g) was injected via a degassed syringe and the polymerization was started by immersion into an oil bath at 90oC for 6 hours taking regular samples via a degassed syringe.
The results can be seen below in Figure 4.9. 4. ATRP graft polymerization from SynPhase lanterns 93
1.0
0.8
0.6
0.4 ln(1/1-x)
0.2
0.0 01234567 Time (h)
1.6 1.4
1.2 PDI 100000 1.0 0.00.10.20.30.40.50.60.7 80000
60000
Mn 40000
20000
0 0.00.10.20.30.40.50.60.7 Conversion ATRP reinitiation macroinit re-initiated Mth n
Figure 4.9: ATRP reinitiation of a styrene macroinitiator using ATRP. Polymer was reinitiated from ATRP styrene Mn=4775. Polymerization was conducted at o 90 C with a styrene:CuBr:Me6TREN:macro-initiator ratio of 1000:1:1:10. 4. ATRP graft polymerization from SynPhase lanterns 94
The reinitiation of dormant ATRP polymer shows very similar kinetics to a normal ATRP polymerization, indicating that the polymerization still adheres to the ATRP mechanism and kinetics. The first order kinetic plot shows a linear increase which in turn, implies that the radical concentration in the system is constant, therefore there is little termination. Another indication of living behavior in a polymerization system is a linear increase in molecular weight with conversion, as can be see in Figure 4.9.
The polydispersity of the reinitiated polymer is around 1.2, which is close to the PDI of the macroinitiator. If we inspect the individual molecular weight distributions, as can be seen from Figure 4.10, we can see that there is no macroinitiator remaining in the final polymers which shows that all the macroinitiator has been effectively reinitiated. It will be noticed that the molecular weight of the polymers produced is slightly lower than predicted by theory. A possible explanation for this is that there is a slight discrepancy between the measured molecular weight and the true molecular weight of the macroinitiator. This shows that the polymerization system we are using produces living polymers that can be easily reinitiated simply by adding a new monomer and catalyst solution.
Increased polymerization time Normalized Response
3.0 3.5 4.0 4.5 5.0 log(M ) w
Figure 4.10: Molecular weight distributions of reinitiation of styrene from a dormant macroinitiator. Dashed line indicates macroinitiator 4. ATRP graft polymerization from SynPhase lanterns 95
4.5. ATRP controlled grafting off functionalized SynPhase lanterns
Now that it has been shown that Me6TREN and PMDETA can be used to polymerize a variety of monomers off solution initiators, it should be possible to take this system and apply it to grafting of polymers from a solid surface with bound halide groups.
As mentioned at the beginning of this chapter, this will be conducted by first functionalizing the surface using CBr4 and γ-radiation and then taking these functionalized lanterns and using them to initiate ATRP polymerization.
4.5.1. Determination of Cs for CBr4 at room temperature under constant γ-irradiation
The concentration of carbon tetrabromide used in the initial functionalization reaction will reduce the length of the initial polymer macro-initiator. This relationship is quantified by the Mayo equation that related the degree of polymerization with a chain transfer agent to the degree of polymerization without a chain transfer agent added, and the concentration of the chain transfer agent.
⎛ ⎞ 1 = ⎜ 1 ⎟ + [S] Mayo equation: ⎜ ⎟ Cs Eq. 4.11 x x [M ] n ⎝ n ⎠0
If we can determine the chain transfer constant Cs for the same conditions that are used in the functionalization reaction then it can be assumed that the molecular weight of the tethering polymer chain is approximately the same as that calculated.
The chain transfer constant can be determined by conducting a series of polymerizations and plotting the inverse of the degree of polymerization vs. the concentration of transfer agent. 4. ATRP graft polymerization from SynPhase lanterns 96
The method for this was as follows: Carbon tetrabromide was weighed into glass vials along with 5ml of styrene. These vials were sealed with septa and degassed by sparging with nitrogen for 30min. The vials were then placed into the γ-source for 48hr at a dose rate of 0.105 kGy/H. Once the polymerization was complete the vials were removed and poured into pre-weighed aluminum pans and dried over-night.
The polymers were then washed for 15min with ethanol to remove CBr4 and the wash solution was decanted off. Polymers were then dissolved in THF and the molecular weight was measured using GPC.
The concentrations of CBr4 that were used were: 0 M, 0.01 M, 0.025 M, 0.05 M, 0.075 M, 0.1 M, 0.15 M, 0.2 M, 0.5 M.
The results from this experiment showed that unfortunately concentrations of CBr4 above 0.1 M produced polymers that were outside the exclusion limit of the GPC and as such weren’t analyzable. This however doesn’t affect results, as enough data points were available to accurately determine the Cs for CBr4 at room temperature.
The Mayo plot is show below in Figure 4.11 and the result of the linear regression on this plot gives the equation
1 = 0.1770×[CBr4] + 0.0021. x n
From this regression the Cs value can be calculated and was found to be 1.6645. This can be compared to the literature value 1.78 at 60oC initiated by AIBN.219 4. ATRP graft polymerization from SynPhase lanterns 97
0.015
0.010 n 1/X
0.005
0.000 0.00 0.01 0.02 0.03 0.04 0.05 0.06 [CBr ] 4
Figure 4.11: Mayo plot for determining the Cs value for CBr4 at room temperature for γ-radiation initiated polymerization of styrene
4.5.2. Functionalization and grafting of tethered Br functionalized lanterns
The first method of ATRP grafting of polymers to lanterns that was undertaken was using lanterns that had a Br atom tethered to the substrate with a short-chain polystyrene chain.
These were produced by functionalizing the lantern in the presence of styrene monomer and a high concentration of CBr4 as shown in the mechanism below: 4. ATRP graft polymerization from SynPhase lanterns 98
Functionalization
gamma radiation Initiation SS SS
SS SS T Propagation
CBr SS T CBr4 SS T Br 3 Chain-transfer
SS = solid support T = Tether polymer
Graft polymerization
SS T Br CuBr-Ligand SS T P CuBr2-Ligand
kp M P = ATRP polymer Solution polymerization
RBr CuBr-Ligand R P CuBr2-Ligand
kp M
Figure 4.12: Mechanism for ATRP controlled grafting of styrene from functionalized SynPhase lanterns
The effect of CBr4 concentration on the functionalization of SynPhase lanterns
The theoretical molecular weight of the tethering polymer chains can be determined from the Mayo equation and the Cs value that was found earlier. It should be the case that since the number of tethered CBr4 units is only dependent on the number of radicals generated on the surface of the lantern, and then the actual number of Br sites should be constant across the range of CBr4 concentrations. The increased CBr4 concentration 4. ATRP graft polymerization from SynPhase lanterns 99
should linearly decrease the molecular weight of the tethering polymer chains, and consequentially the graft ratio of the functionalization reaction.
The functionalization experiment was conducted as follows: SynPhase lanterns were washed in DCM for 24 hours and then dried in
vacuo to remove any contaminants. CBr4 was weighed into a glass vial with 10ml of monomer and 5 lanterns. The vial was then sealed with septa and degassed by sparging with nitrogen for 30min. The vial was then exposed to γ-radiation for 48hours at a dose rate of 0.105 kG/h. After this, to remove ungrafted homopolymer, the lanterns were washed in DCM for 2 weeks while changing the wash solution every second day.
The results for this functionalization experiment are shown below. 4. ATRP graft polymerization from SynPhase lanterns 100
10
8
6
4 graft ratio (wt%)
2
0 0.00 0.05 0.10 0.15 0.20 0.25 0.30 [CBr ] 4
Figure 4.13: Functionalization of lanterns; effect of CBr4 concentration of the functionalization graft ratio. Results can be compared to a sample with no CBr4 that yielded a mass change of 32.0%. ATRP polymerization was conducted at 90oC with styrene:CuBr:Me6TREN ratio of 1000:1:1 and 5 functionalized lanterns.
These results show that there is a change in mass of the lantern indicating that there has been some grafting onto the surface. It’s possible to then convert the CBr4 concentration to a molecular weight that is derived from the Mayo equation and the parameters determined earlier. The results seen in Figure 4.14, show that there is a linear correlation between the graft ratio and the molecular weight determined from the Mayo equation. Since there is a linear correlation between the graft ratio and the molecular weight, the number of bromine atoms tethered to the substrate can be determined to be constant and the changes in lantern mass are caused by changes in the chain length of the tethering polymer chain.
= × Eq. 4.12 graft _mass moles M n 4. ATRP graft polymerization from SynPhase lanterns 101
35
30
25
20
15
graft ratiograft (wt%) 10
5
0 0 10000 20000 30000 40000 50000 M (Mayo) n
Figure 4.14: Functionalization graft ratio changes for the functionalization of
SynPhase lanterns using CBr4. Graft ratio plotted against the molecular weight of grafted polymer predicted by the Mayo equation
Thus the assertion that changing the CBr4 concentration does not alter the number of tethered bromine atoms, but merely changes the length of the tethering polymer chain is correct.
These lanterns were then used as a surface to conduct ATRP controlled grafting by the following procedure:
Styrene (10.4g, 0.1 mol) and CuBr (0.0143g, 1x10-4 mol) were placed in a glass vial. To this the functionalized lanterns were added along with a blank unfunctionalised PP lantern. These vials were septa sealed and degassed by -4 sparging with nitrogen for 30min. Me6TREN (0.0230g, 1x10 mol) was then injected via a degassed syringe. The polymerization was started by immersing the samples in an oil bath at 90oC. 4. ATRP graft polymerization from SynPhase lanterns 102
Once the polymerization was complete the solution polymers were dried and analyzed by GPC. The lanterns were washed in DCM for 2 weeks while changing the wash solution every 2 days to remove ungrafted homopolymer.
The theory says that all the lanterns should have the same initiator density, no matter what the CBr4 concentration that they were grafted in, and the CBr4 should only change the length of the tethered chain. As can be seen in Figure 4.15, there is very little difference in the amount of polymer that is grafted by the ATRP process.
The lanterns functionalized with 0.13 M and 0.16M CBr4 concentration grafted up to a graft ratio of 0.55 while the lanterns functionalized with 0.04 M CBr4 showed very little
ATRP grafted polymer. It is possible that the lower CBr4 concentration during the functionalization stage resulted using up all the CBr4 and thus there thus there is a limited number of initiation sites available. However, the more likely explanation is that the experiment with 0.004 M CBr4 had some oxygen ingress during polymerization that poisoned the polymerization. 4. ATRP graft polymerization from SynPhase lanterns 103
60
50
40
30
20 Graft ratio ratio Graft (wt%)
10
0 0 2 4 6 8 101214161820222426 Time (h) Functionalization [CBr ] 0.04 M 0.13 M 0.16 M blank 4
Figure 4.15: Graft polymerization of styrene off lanterns functionalized with o varying [CBr4]. ATRP polymerization was conducted at 90 C with styrene:CuBr:Me6TREN ratio of 1000:1:1 and 5 functionalized lanterns.
Effect of functionalization time on lantern grafting
The amount of Br tethered to the surface of the polymer should be a function of the number of radicals generated on the lantern only, which is in turn, a function of the length of time the lanterns are exposed to the γ-radiation. It thus follows that the degree of functionalization of the lanterns can be controlled by adjusting the length of time that the lanterns are exposed to the γ-source in the functionalization phase.
To test this, lanterns were functionalized using the following method: 5 blank PP lanterns were washed in DCM for 24hr and dried in vacuo. These
lanterns were then placed in a glass vial with styrene (10g) and CBr4 (0.331g, 1x10-3 moles) to make up a 0.1M solution. The vial was then sealed with septa and degassed for 30min by sparging with nitrogen and then placed in the γ-source at a 0.105 kGh-1 dose rate for 48hrs. 4. ATRP graft polymerization from SynPhase lanterns 104
Lanterns were then weighed and washed for 1 week in DCM, changing
solution every second day to remove unreacted CBr4 and styrene homopolymer. The lanterns were then dried in vacuo overnight.
As was expected from theory, it was found that the mass of the lanterns increased as the functionalization time increased. This represents the increase in the number of chains polymerizing from the surface as the exposed radiation dose increases. If the lanterns produced at 4hrs are ignored then there was a strong linear correlation between the functionalization time and the mass gained during the functionalization step caused by an increase in the number of tethered Br chains attached to the surface.
25
20
15
10 Graft ratioGraft (wt%)
5
0 0 1020304050 time (h)
Figure 4.16: Effect of functionalization time on the mass gain of lanterns functionalized in 0.1 M CBr4 solution in styrene. ATRP polymerization was o conducted at 90 C with styrene:CuBr:Me6TREN ratio of 1000:1:1 and 5 functionalized lanterns. 4. ATRP graft polymerization from SynPhase lanterns 105
These lanterns were then used as a surface to conduct ATRP grafting from the surface using Me6TREN/CuBr as a catalyst system.
The polymerization was conducted using the following method: One lantern from each functionalization set was weighed and placed in a glass vial with the polymerization solution comprising of styrene (10.4g, 0.1 mole) and CuBr (0.143g, 1x10-4 moles) and sealed with septa. The vial was then degassed by sparging with nitrogen for 30min. The polymerization was -4 then initiated by injecting Me6TREN (0.023g, 1x10 moles) and placing the vial in an oil bath at 90oC.
Four reactions were made up in this manner and sampling was conducted by removing one of these reactors and pouring the contents into a pre-weighed aluminum pan.
Post reaction, the lanterns were washed in DCM for 2 weeks replacing the DCM every 48hr and then dried in vacuo. Solution homopolymers were dissolved in THF for GPC analysis.
The results of this polymerization can be seen below in Figure 4.17.
It can be seen that as the functionalization time increases, the number of tethered bromine chains effectively increase the initiator density on the substrate. From this, if we conduct ATRP grafting off these surfaces, it would be expected that the longer the functionalization time, the faster that the graft ratio will increase, and the more polymer will be grafted to the lantern. This is exactly what occurred and as can be seen above the graft polymerization follows a linear trend increasing with time. 4. ATRP graft polymerization from SynPhase lanterns 106
45
40
35
30
25
20
15 Graft ratio (wt%) ratio Graft
10
5
0 02468 time (h) Functionalization time: 4hr 24hr 36hr 48hr
Figure 4.17: ATRP graft polymerization off functionalized lanterns; effect of functionalization time on ATRP graft polymerization. ATRP polymerization was o conducted at 90 C with styrene:CuBr:Me6TREN ratio of 1000:1:1 and 5 functionalized lanterns.
4.5.3. Direct functionalization of lanterns in the absence of monomers
It is possible to apply the radical reaction between CBr4 and the polymeric alkyl radical to a system without monomer present. In that case the “chain-transfer” reaction would occur with the alkyl radical directly generated on the surface of the PP lantern surface. This reaction would be better described as radical transfer rather than chain transfer, and theoretically allows for the direct functionalization of the lanterns’ surface without the need for a tethering polymer. 4. ATRP graft polymerization from SynPhase lanterns 107
gamma CH3 CH3 Br CH radiation 3 [CBr4] CH2 CH CH2 C CH2 C
n n n
Scheme 4.4: Mechanism for direct radical bromination of PP using radiation and
CBr4 in the absence of monomer
The full mechanism for the functionalization and consequential ATRP polymerization is shown Figure 4.4 below.
Functionalization
gamma radiation SS SS Initiation
SS CBr4 SS Br CBr3 Chain-transfer
SS = solid support
Graft polymerization
SS Br CuBr-Ligand SS P CuBr2-Ligand
kp M P = ATRP polymer
Scheme 4.5: Mechanism for direct functionalization of SynPhase lanterns and subsequent ATRP grafting from the surface
The functionalization was conducted using methanol and toluene as a solvent using the following method: 5 blank PP lanterns were washed in DCM for 24hr and dried in vacuo. These
lanterns were then placed in a glass vial with the solvent (10g) and CBr4 (0.331g, 1x10-3 moles) to make up a 0.1M solution. The vial was then sealed 4. ATRP graft polymerization from SynPhase lanterns 108
with septa and degassed for 30min by sparging with nitrogen and then placed in the γ-source at a 0.105 kGh-1 dose rate for 48hrs.
Lanterns were then weighed and washed for 1 week in DCM, changing
solution every second day to remove unreacted CBr4 and styrene homopolymer. The lanterns were then dried in vacuo over-night.
There was an observable change in the mass of the lanterns after functionalization indicating that there was some degree of functionalization of the surface as shown in Table 4.3
Functionalization Reaction Solvent graft ratio
Functionalization Methanol 0.061 in methanol
Functionalization Toluene 0.042 in toluene
Functionalization Styrene 0.220 in styrene
Table 4.3: Results for the direct functionalization of lanterns in the absence of monomers
These lanterns were then used as surfaces to conduct ATRP grafting from the surfaces in the exact same way as was conducted previously off the lanterns functionalized with tethered Br atoms. The method for ATRP grafting from the directly functionalized substrates is given below:
A lantern from each functionalization set were weighed and placed in a glass vial with the polymerization solution comprising styrene (10.4g, 0.1 mole) and CuBr (0.143g, 1x10-4 moles) and sealed with a septa. This vial was then degassed by sparging with nitrogen for 30min. The polymerization was then -4 initiated by injecting Me6TREN (0.023g, 1x10 moles) and placing the vial in an oil bath at 90oC. 4. ATRP graft polymerization from SynPhase lanterns 109
Four reactions were made up in this manner and sampling was conducted by removing one of these reactors and pouring the contents into a pre-weighed aluminum pan.
Post reaction, the lanterns were washed in DCM for 2 weeks replacing the DCM every 48hr and then dried in vacuo. Solution homopolymers were dissolved in THF for GPC analysis.
The results of this polymerization can be seen below in Figure 4.18.
50
40
30
20 Graft ratio (wt%) ratio Graft
10
0 02468 time (h) Functionalization solvent: Styrene Toluene Methanol blank
Figure 4.18: Comparison of ATRP grafting from tethered and directly functionalized lanterns. ATRP polymerization was conducted at 90oC with styrene:CuBr:Me6TREN ratio of 1000:1:1 and 5 functionalized lanterns
It can be seen that the polymerization was effective in grafting polymers to the substrates, and consequentially that the direct functionalization method in the absence of monomer was effective in producing a brominated surface that could be used to initiate 4. ATRP graft polymerization from SynPhase lanterns 110
ATRP controlled graft polymerization. Interestingly, there was an appreciable difference between the lanterns functionalized in methanol and in toluene. A possible explanation for this is that the PP lanterns will swell in toluene but not in methanol, thus that the functionalization is taking place both on the surface of the lantern as well as internally. This internal functionalization will not be as accessible to the ATRP catalyst and monomer and thus may be inert as an initiator for the ATRP grafting reaction. Additionally, since methanol has a higher G-value than toluene it would result in a higher radical yield and increased surface grafting.
While radiation generates a wide variety of different radicals across the material it is anticipated that since the most stable radical that can be formed is the tertiary radical from hydrogen abstraction, a significant number of the bromine functionalizations being formed will be tertiary type functionalization. This mechanism can be seen in Figure 4.19. Matyjaszewski118 states that most alkyl halides can act as ARTP initiators, and quality of the initiator is generally tertiary > secondary > primary. Thus, these tertiary formed bromine functionalities can be expected to work reasonably efficiently as ATRP initiators.
CH2 CH2 CH2
CH3 CH3
HC CH3 C C Br Radiation CBr4 CH2 CH2 CH2
HC CH3 HC CH3 HC CH3
CH2 CH2 CH2
Polypropylene Surface radical Bromine functionalised substrate surface
Figure 4.19: Proposed mechanism for direct bromine functionalization of PP lanterns.
It can be thus concluded that direct functionalization by CBr4 dissolved in methanol is comparative in its effectiveness to that of lanterns functionalized with a styrene tether on the bromine atoms. 4. ATRP graft polymerization from SynPhase lanterns 111
4.6. FMOC-β-Alanine loading determination of ATRP grafted lanterns
The functionalization of a grafted surface using FMOC-β-alanine and the subsequent chemical cleavage is a standard industry method for quantifying the effectiveness of a support as a combinatorial chemical scaffold.
The procedure is to aminomethylate the styrene and then couples this with a FMOC-β- alanine moiety. The FMOC-β-alanine is then chemically cleaved, and then the amount of FMOC cleaved from the surface is determined using UV-Vis spectroscopy. The exact procedure for this is outlined in Chapter 3.
Loading tests were conducted on the lanterns produced that were grafted with styrene with varying functionalization times. This effectively shows how the loading of the lanterns changes with increasing graft density.
20
18
16
14
12
10 mol/lantern) μ 8
6 Loading ( 4
2
0 0 5 10 15 20 25 30 35 40 45 Graft Ratio (wt%) Functionalization time: 4hr 24hr 36hr 48hr
Figure 4.20: FMOC-β-alanine loading of lanterns grafted with styrene by ATRP from lanterns functionalized with CBr4 in styrene. 4. ATRP graft polymerization from SynPhase lanterns 112
It can be seen here that the loading of the lanterns increases linearly with an increasing graft ratio and isn’t dependent on the functionalization time at all. This implies that the loading is not affected by the graft density and is alone dependent on the actual mass of grafted polymer. It has been thus shown that the loading on the lantern can be controlled by adjusting the amount of polymer grafted through ATRP, either by adjusting the chain length of the grafted polymer by increasing polymerization time or by increasing the number of polymer chains by increasing the functionalization time.
As with the tethered Br functionalized lanterns, loading tests were also conducted on these lanterns and as can be seen in Figure 4.21, there is an excellent correlation between the lanterns grafted by ATRP from styrene functionalized lanterns and those directly functionalized in methanol.
20
15 mol/lantern)
μ 10
Loading ( Loading 5
0 0 1020304050 Graft Ratio (wt%) Functionalization solvent: Styrene Toluene Methanol Blank
Figure 4.21: Comparison of the FMOC-β-alanine loading of grafted polymers on lanterns directly functionalized by CBr4 in methanol and those functionalized in styrene monomer 4. ATRP graft polymerization from SynPhase lanterns 113
The loading of the lanterns was again found to be only a function of the graft ratio of the polymer and not a function of the functionalization method, whether the lanterns were functionalized directly or with a polystyrene tether.
The loading of these lanterns can be compared to that of a commercially produced lantern from Mimotopes that was tested at the same time and yielded a loading of 12.4 μmol/lantern.
4.7. Conclusion
This chapter has shown that it is viable to use the chain transfer properties of carbon tetra-bromide to functionalize polymeric surfaces such that they may then be used as initiators for ATRP grafting. Lanterns were functionalized using short chain polystyrene tethered bromine atoms and also with CBr4 directly functionalizing the surface.
The styrene-Br functionalized lanterns were produced through exposing the lanterns to
γ-radiation in a mixture of styrene and CBr4. These were then used as initiators for ATRP grafting of styrene off the surfaces, up to a graft ratio of 0.45. It was found that changing the CBr4 concentration in the functionalization stage would affect the length of the tethering polymer chain, but did not effect the concentration of bromine on the surface, and consequentially did not affect the later ATRP grafting experiments. The amount of accessible initiator sites was also found to be a function of the time the lanterns were exposed during the functionalization stage. This was from the increased dose of radiation generating increased surface radicals. This effect carried on to the ATRP grafting reactions where it was found that the amount of polymer grafted to the lantern increased as the functionalization time was increased.
SynPhase lanterns were also directly functionalized by exposing the lanterns to γ- radiation with CBr4 in both methanol and toluene. These were then also used as initiators from ATRP graft polymerization and the lanterns were grafted up to a graft ratio of 0.45. It was found that the lanterns functionalized with methanol as a solvent, contained a similar amount of accessible initiator sites as the styrene functionalized 4. ATRP graft polymerization from SynPhase lanterns 114
lanterns. The lanterns functionalized in toluene were found to have significantly less accessible initiator groups.
FMOC loading tests were conducted on these lanterns, and it was found that the loading increased linearly with increasing graft ratio of the lanterns, and was independent on the functionalization method and CBr4 concentration. The lanterns produced through this method were found to have a comparative loading to those commercially produced by Mimotopes. 5. γ-radiation initiated reverse ATRP graft polymerization 115
5. γ-radiation initiated reverse ATRP graft polymerization
One of the features of ATRP polymerization is that the catalytic equilibrium is reversible and thus it was possible to develop what is termed reverse ATRP where a radical initiator and a Cu(II) complex are used to control the polymerization as opposed to conventional ATRP where a halide compound is used as initiator and a Cu(I) complex acts as the catalyst. This system traditionally comprises a thermal radical initiator such as AIBN to initiate the polymerization, which then proceeds along the standard ATRP mechanism shown in Scheme 5.1:
ka RBr CuBr R CuBr2 kd
kp
Scheme 5.1: Generic ATRP/RATRP polymerization mechanism
While thermal initiators are the normal radical source for these polymerizations, γ- radiation can also be used as the radical source to initiate this polymerization. This opens the possibility of controlling graft polymerization through an ATRP/RATRP mechanism without the need for pre-functionalization of a surface, as the γ-radiation generates surface radicals without the need for chemical modification.
There has been only one paper published on room-temperature RATRP polymerization, written by Wang et al.225 This system polymerizes MMA initiated by benzyl peroxide (BPO) and uses Cu(I)-bipy as the complex. The interest here is that both the radical species and the Cu(II) compound for controlling the RATRP are generated from a redox reaction between the BPO and CuBr complex as shown Scheme 5.2. 5. γ-radiation initiated reverse ATRP graft polymerization 116
O O O O II O O CuIBr O Cu Br/ O
Scheme 5.2: BPO redox initiation mechanism used to initiate room temperature RATRP by Wang et al.
As such, this system can be thought of as a redox-initiated conventional ATRP system rather than a RATRP system. The polymer produced as a result reached about 70% conversion after 96 hours, and had a molecular weight of 71300 and a PDI of 1.45. Polymers produced at earlier stages of the polymerization had a lower polydispersity down to 1.2.
The polymerization presented by Wang et al. uses a modified RATRP mechanism, and because of this won’t present some of the properties often seen in RATRP polymerization. As the initial copper compound is not Cu(II) but a Cu(I) compound, there will be little initial Cu(II) deactivator in the system where there would be a very large concentration in a traditional RATRP system. Thus in the induction period effects that are noticed in a RATRP system while the Cu(II) deactivator concentration is reduced, will be significantly less pronounced and polymerization should start almost instantly.
As mentioned earlier it could be argued that this system is a redox initiated conventional ATRP system rather than a RATRP, but this depends on whether RATRP systems are defined by the fact that they are radically initiated or by the fact that they are initially controlled by a Cu(II) complex rather than a Cu(I) complex. It has also been noted that BPO and peroxide initiators have problems in ATRP and RATRP systems. Matyjaszewski226 conducted a comparative study of azo- and peroxide initiators and found that the same BPO initiation system as was used by Wang et al. was unable to control polymerization of styrene unless a sufficient excess of CuBr was added. This was ascribed to electron transfer from Cu(I) to BPO and coordination of benzoate anions to copper. Such an induced decomposition reoxidizes Cu(I) to Cu(II) species and disables the catalytic reaction. 5. γ-radiation initiated reverse ATRP graft polymerization 117
γ-radiation initiation offers the opportunity to bypass the problems associated with peroxide initiators by directly producing radicals without the need for an initiator.
Some of the problems that may occur with our γ-radiation initiated RATRP polymerization system include:
1. Solubility problems with the catalytic complex. In general the CuBr2-ligand complex is only sparingly soluble in many solvents and this may be accentuated by the room temperature polymerization conditions. 2. Loss of control caused by increases in the ATRP reaction equilibrium constant
Keq, created by temperature dependencies in the activation and deactivation rate constants. 3. Slow rates of polymer propagation at room temperature. 4. Problems associated with γ-radiation as a constant initiation source.
5. Amount of bimolecular termination being too high. At low temperature the kp/kt
ratio tends to increase because of a decrease in kp.
Aside from the problems with using peroxide initiators in RATRP, Wang et al.225 does however show that the RATRP mechanism is valid at room temperature. Specifically, he shows the reaction equilibrium is such to allow for the production of controlled polymers and MMA can polymerize fast enough at room temperature to produce quantitative polymer yields. Thus, γ-initiated, room-temperature polymerization should be viable and offers a novel method of effectively producing polymers at room temperature.
5.1. Mechanism of RATRP polymerization using γ- radiation
The mechanistic scheme for RATRP polymerization using γ-radiation can be represented by a modification of Matyjaszewski’s118 mechanism that includes the γ- initiation. 5. γ-radiation initiated reverse ATRP graft polymerization 118
(I) γ Initiation
gamma monomer M M P1 krad-soln krad = krad-soln + krad-ss gamma monomer SS SS SS-P1 (SS is Solid Support) krad-ss
(II) Equalibrium and propagation
ka CuBr2 Pn CuBr Br Pn kd
kp
ka CuBr2 SS-Pn CuBr Br SS-Pn kd
kp
(III) Termination
Pn Pm Pn+m
SS-Pn Pm SS-Pn+m
Scheme 5.3: RATRP polymerization and grafting mechanism
The kinetic equation for the rate of polymerization in a radical polymerization system is given by Eq. 4.2: 118
= • Eq. 4.2: Rp k p[M ][P ]
• where Rp is the rate of polymerization, [M] is the monomer concentration and [P ] is the radical concentration.
This can then be modified as described by Matyjaszewski118 to give an kinetic equation for ATRP polymerization (Eq. 5.1): 5. γ-radiation initiated reverse ATRP graft polymerization 119
= − d[M ] = × [CuBr] Rp k p Keq[M ][I]0 Eq. 5.1 dt [CuBr2 ]
Where Keq is the equilibrium constant for the ATRP equilibrium and [I]0 is the initial initiator concentration.
To modify this for γ-initiation, define the radical concentration in terms of radiation dose rate rather than initiator concentration. Since the radical concentration is proportional to the dose exposed,
[P• ] ∝ Dose simplify the mechanism of radical generation from γ-radiation to:
DR.frad M M then the amount of radicals produced from radiation can be given by Eq. 5.2:
• = × [CuBr] [P ]rad Keq frad DR.t Eq. 5.2 [CuBr2 ]
where DR is the radiation dose rate and frad.is an efficiency factor that describes the efficiency of generation of active, polymerizable radicals.
Substituting Eq. 5.2 this into the general polymer kinetic equation (Eq. 4.2) gives:
= − d[M ] = × [CuBr] Rp Keqk p frad [M ]DR.t Eq. 5.3 dt [CuBr2 ]
Integrating Eq. 5.3 and rearranging in terms of conversion x, gives Eq. 5.4:
⎛ 1 ⎞ [CuBr] ln⎜ ⎟ = 1 K k f [M ]DR.t 2 × Eq. 5.4 − 2 eq p rad ⎝1 x ⎠ [CuBr2 ]
The major conclusion from this is that given the rate of radical generation and frad is constant over time, the resulting rate of monomer consumption is 2nd order with respect 5. γ-radiation initiated reverse ATRP graft polymerization 120
to time. Eq. 5.4 can then be used to predict the resulting rates of polymerization for a given system.
In addition to the system being 2nd order with respect to time, it can be seen from Eq.
5.3 that the rate is of a negative order with respect to CuBr2 concentration. If the concentrations of CuBr2 are too high then very slow rates of polymerization will be seen. It becomes important to balance the initial concentration of CuBr2 with the radical concentration. However, since the radical concentration is dynamic the rate of polymerization will change throughout the polymerization and the CuBr2 concentration needs to be sufficiently high to leave enough CuBr2 to control the polymerization at later stages, but low enough not to significantly inhibit polymerization early on. This will be discussed in greater detail in the next section.
It is also possible to derive equations describing the molecular weight and polydispersities in a living radical polymerization in addition to the kinetic equations. For a living polymerization system without termination the degree of polymerization is given by the equation118
Δ = [M ] DPn Eq. 5.5 [I]0 which can then be modified to give the theoretical molecular weight for any given conversion (x) as:
[M ] .x.M monomer = 0 n Eq. 5.6 Mnth [I]0 Modifying Eq. 5.6 to the radiation induced radical concentration: [M ] .x.M monomer = 0 n Eq. 5.7 Mnth frad DR.t For γ-radiation initiated RATRP; the persistent radical effect will significantly affect the polymerization kinetics in several ways. Firstly, the radical concentration needs to be sufficiently high to reduce the CuBr2 concentration and allow polymerization to occur.
This needs to be balanced with a sufficient concentration of CuBr2 to control the polymerization and limit the bimolecular termination reactions. The result of this is that there is likely to be a large induction period while these equilibriums are established. 5. γ-radiation initiated reverse ATRP graft polymerization 121
5.2. Experimental method
All materials were prepared as follows: Monomers (styrene and MMA) were purchased from Sigma-Aldrich and
were deinhibited by percolating through a basic alumina column. Me6TREN
was synthesized and CuBr2 was purified as per the methods described in Chapter 3. DMF was dried over molecular sieves. AIBN was acquired from Qenos and purified twice by recrystalization from hot methanol.
All remaining reagents were analytical grade and acquired from Sigma- Aldrich and used as received without further purification.
The polymerization was conducted as follows:
Solid CuBr2 was placed in a vial with all the remaining components except the ligand and degassed by sparging with nitrogen for 60 minutes. This solution was then transferred using a degassed syringe to individually degassed reactor vials. These vials were then degassed for another 15 minutes by sparging with nitrogen and individual aliquots of ligand were injected into each vial to form the complex.
The solutions were then placed in the γ-source at room temperature for the required time.
The thermal polymerization was conducted similarly; however the vials were placed in an oil bath for the polymerization time rather than the γ-source.
Post polymerization, the solution was decanted into pre-weighed aluminium pans and dried in vacuo. Lanterns were washed for 2 weeks in DCM changing the solution every second day to remove homopolymer. The solution polymer was dissolved in THF for GPC analysis. 5. γ-radiation initiated reverse ATRP graft polymerization 122
5.3. RATRP of styrene using a thermal initiation
To provide a comparison to the more complicated radiation initiated polymerization system a standard thermally initiated RATRP polymerization of styrene was conducted.
CuBr2 NC NC Br Me6TREN AIBN Fragment
n
Scheme 5.4: Mechanism of RATRP controlled polymerization of styrene
The polymerization was prepared as described earlier in the chapter. The polymerization system used styrene as a monomer, AIBN as the initiator and Me6TREN as the ligand.
These were used at a 1000:1:1:1 ratio of monomer:CuBr2:ligand:AIBN. The polymerization was conducted at 110oC in an oil bath.
The resulting polymerization can be seen in Figure 5.1 and showed a linear first-order kinetic plot, with a slow initiation in the early stages of the polymerization. The resulting polymer was well controlled (PDI 1.0-1.1) and molecular weight was linear with respect to conversion.
This experiment showed that RATRP polymerization of styrene was possible using these reagents and would be used as a basis for the room temperature radiation induced polymerization of styrene outlined in the next section. 5. γ-radiation initiated reverse ATRP graft polymerization 123
0.8
0.6
0.4 ln(1/(1-x))
0.2
0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (h)
Figure 5.1: Pseudo first order kinetic plot for the thermally initiated RATRP of o styrene conducted at 110 C and a styrene:CuBr2:Me6TREN:AIBN ratio of 1000:1:1:1
5.4. γ-radiation initiated RATRP grafting of styrene
Since it has been shown that the current system polymerizes styrene well when initiated using thermal initiators at high temperatures, the objective is now to develop a system that can be used for γ-radiation initiated polymerization via RATRP at room temperature.
Initial experiments were conducted using styrene polymerized using CuBr2 and
Me6TREN as the ligand. 10% DMF was added to the system to aid in the dissolution of 227 the CuBr2 catalyst as it has been used previously as a solvent for ATRP complexes.
Both 1000:1:1 and 500:1:1 monomer to CuBr2 to ligand ratios were used. Once the polymerization solutions were prepared as described at the beginning of the chapter 5. γ-radiation initiated reverse ATRP graft polymerization 124
polymerization was initiated by placing the reaction vessels in the γ-radiation source at a dose rate of 0.056 kGy/h.
It was found that the polymerization did not reach more than 1-2% conversion and the resulting polymers were poorly controlled. The grafting that did occur was uncontrolled and unpredictable as can be seen in Figure 5.2. The grafting was significantly greater than what the conversion of linear polymer would imply. Thus it was concluded that the probable cause of this was that the CuBr2 complex was inhibiting solution polymerization, but the grafting was completely unaffected and probably proceeded by an uncontrolled free radical mechanism.
35
30
25
20
15 Graft ratio(wt%) 10
5
0 0 20406080100120140160180 time (h) Monomer to catalyst ratio: 1000:1; 500:1
Figure 5.2: γ-initiated RATRP polymerization of styrene under γ-irradiation at a dose rate of 0.056 kGy/h. Styrene:CuBr2:Me6TREN ratios were conducted at both 1000:1:1 and 500:1:1: concentrations and 10% DMF was added as a solvent.
It was concluded that the styrene system had number of problems and that the source of the problems was slow initiation rate and that there was insufficient radical 5. γ-radiation initiated reverse ATRP graft polymerization 125
concentration to reduce the CuBr2 concentration to a level where polymerization could properly proceed.
5.4.1. Mechanistic implications of the addition of AIBN to RATRP controlled systems
There are two ways of overcoming the problems with the styrene system. Firstly it is possible to reduce the CuBr2 concentration so that fewer radicals are required to reduce the concentration of CuBr2 deactivator to a level where polymerization can proceed. However, while a reduction in the amount of catalyst complex in the system will reduce the required amount of generated radicals, a reduction beyond 1000:1 monomer to complex ratio would not provide adequate catalyst to control the polymerization. The other method was to increase the radical concentration in the system either through increasing the dose rate or better, through the addition of radical sensitizers that increase the radical yield in a system without affecting any of the reactions.
Chaplin et al.153 successfully described the addition of AIBN and other thermal initiators to accelerate radiation polymerization. The addition of AIBN would offer a good possibility of increasing the radical concentration and alleviating the problems thus associated without compromising control of the polymerization.
To increase the rate of polymerization it was also decided that a faster polymerizing monomer would be used, thus MMA was chosen. The change to MMA should have two effects: the higher kp of MMA should improve the yield of polymer in this system and MMA does not undergo self initiation as styrene does. The ligand was also changed to PMDETA as it is commercially available ligand that is effective for MMA polymerization.228
It has been shown that the addition of free radical initiators to NMP and other PRE controlled polymerization systems can increase the rate of polymerization up to threefold without impacting the molecular weights or polydispersities of the system.129,229,230 The addition of free radical initiators has also been noted to increase rates of polymerization γ-radiation polymerization.153 However, Fukuda has shown that 5. γ-radiation initiated reverse ATRP graft polymerization 126
the addition of a free radical initiator (in this case AIBN) has further mechanistic effects beyond the simple increase in the radical flux that was discussed earlier.229,230
If we make the simplification that an AIBN initiated fragment behaves in the same manner mechanistically as a species derived from the action of radiation upon a monomer molecule then we can greatly simplify the kinetic equations. The modified mechanism is shown in Scheme 5.5:
(I) γ Initiation
gamma monomer M M P1 krad-soln krad = krad-soln + krad-ss gamma monomer SS SS SS-P1 krad-ss
gamma monomer NC N N CN 2 NC P1 kaibn
(II) Equalibrium and propagation
ka CuBr2 Pn CuBr Br Pn kd
kp
ka CuBr2 SS-Pn CuBr Br SS-Pn kd
kp
(III) Termination
Pn Pm Pn+m
SS-Pn Pm SS-Pn+m
Scheme 5.5: RATRP grafting mechanism with modifications to include radicals formed from AIBN 5. γ-radiation initiated reverse ATRP graft polymerization 127
The kinetic equations that were derived at the beginning of this chapter can likewise be modified to take the addition of the AIBN into consideration.
Given that the amount of radical derived from AIBN is proportional to the initial concentration of AIBN (assuming a burst initiation from the AIBN) then
• ∝ [P ]aibn [AIBN] then the radicals produced from the AIBN can be expressed as:
• = × [CuBr] [P ]aibn Keq faibn[AIBN]0 Eq. 5.8 [CuBr2 ]
where fAIBN is an efficiency factor that describes the number of active radicals that are generated from the initial AIBN concentration (it should be noted that this factor incorporates the fact that AIBN degrades to form two identical radicals). Thus total radical concentration from both radiation and AIBN sources will be:
• = ()+ × [CuBr] [P ] faibn[AIBN]0 frad DR.t Keq Eq. 5.9 [CuBr2 ]
Substituting into the general radical polymerization kinetic equation (Eq. 4.3) from Chapter 4, give Eq. 5.10:
d[M ] [CuBr] R = = ()f [AIBN] + f DR.t ×[M ]k K p aibn 0 rad p eq Eq. 5.10 dt [CuBr2 ] and integrating both sides
⎛ 1 ⎞ [CuBr] ln⎜ ⎟ = ()f [AIBN] t + 1 f DR.t 2 × k K Eq. 5.11 − aibn 0 2 rad p eq ⎝1 x ⎠ [CuBr2 ] thus we find that the polymerization will display characteristics of being 1st or 2nd order with respect to time depending on the degree of dominance of the two radical sources. This equation can then be used to predict the resulting rates of polymerization for a given system. If the AIBN is the dominant source of radicals in the system then a plot of 5. γ-radiation initiated reverse ATRP graft polymerization 128
ln(1/1-x) vs. t should give a straight line, however without dominance of AIBN radicals then we would expect a linear plot of ln(1/1-x) vs. t2.
As with other living polymerization systems the degree of polymerization is given by:108
Δ = [M ] DPn Eq. 5.12 [I]0
This can then be modified to give the theoretical molecular weight as
[M ] .x.M monomer = 0 n Eq. 5.13 Mnth [I]0 Modifying this equation to include both the radicals derived from AIBN and the radiation induced radical concentration:
[M ] .x.M monomer Mn = 0 n Eq. 5.14 th + f AIBN [AIBN]0 frad DR.t These equations can now be used to predict the molecular weight and conversion of RATRP polymerization conducted under γ-radiation with the addition of AIBN.
5.5. γ-radiation initiated RATRP grafting of MMA
As mentioned previously, it was desired to move to a MMA based polymerization system and change the ligand to PMDETA to try to accelerate the polymerization. The polymer solution was prepared as described at the beginning of the chapter and used two lanterns, MMA as a monomer, PMDETA as the ligand and CuBr2 as the copper compound, with a 1000:1:1 ratio of monomer to CuBr2 to ligand. 20 vol% of DMF was added to aid in the dissolution of CuBr2 and AIBN was added in a ratio of 0.5:1 relative to CuBr2 to increase the radical concentration in the solution polymerization. Once the solutions were prepared and degassed by sparging they were initiated by placing in the 60Co radiation source at a dose rate of 0.056 kGy/h.
The polymerization results from this experiment are shown in Figure 5.3. 5. γ-radiation initiated reverse ATRP graft polymerization 129
0.25
0.20
0.15
0.10 ln(1/(1-x))
0.05
0.00 0 10000 20000 30000 40000 50000 Time2 (h2)
Figure 5.3: γ-initiated RATRP polymerization of MMA at a 1000:1:1 ratio with the addition of 20% DMF to aid dissolution and a 0.5:1 ratio of AIBN to CuBr2 to increase the radical flux
The conversion reached 20% after 220 hours and, as such, the problem of low conversion of the polymerization reaction was alleviated. The molecular weight reached 64000 and showed an increasing molecular weight over time and a polydispersity ranged between 1.2 and 1.6. It can be seen in Figure 5.3 that the polymerization order shows a linear correlation with the square of time. This indicates that the polymerization exhibits a second order dependence on time, fitting with our hypothesis that the radical concentration increases linearly with time. The molecular weight of the polymer was also seen to increase over time as expected for a living polymerization system; however, there was a significant amount of scatter in the GPC results and it was not possible to determine any linear trends. Likewise the graft ratio of the polymers increased with time, but the degree of scatter in the results made analysis extremely difficult.
The probable cause for the lack of consistent molecular weight control is that the catalyst ratio was too low to control this polymerization, and thus the monomer to 5. γ-radiation initiated reverse ATRP graft polymerization 130
catalyst ratio was increased to 500:1. The result was successful polymerization and grafting as can be seen below in Figure 5.4, Figure 5.5 and Figure 5.6.
1.5
1.0 ln(1/(1-x)) 0.5
0.0 0 5000 10000 15000 20000 Time2 (h2) Original; Repeat
Figure 5.4: γ-initiated RATRP polymerization of MMA at a 500:1:1 ratio with the addition of 20% DMF to aid dissolution and a 0.5:1 ratio of AIBN to CuBr2 to increase the radical flux
The increased catalyst concentration greatly improved control over the system. As with the previous experiment the first order kinetic plot is linear with respect to t2, indicating there is a linear increase in radical concentration. However, unlike the 1000:1 polymerization, the molecular weight analysis showed a good linear dependence on conversion and the polydispersity is quite small, around 1.2. Thus, successful RATRP polymerization was conducted showing improved control with the increased catalyst concentration. 5. γ-radiation initiated reverse ATRP graft polymerization 131
1.50
1.25 pdi 400000 1.00 02040 350000 300000 250000 200000 Mn 150000 100000 50000 0 02040 Conversion (%) Original; Repeat
Figure 5.5: Molecular weight data for γ-initiated RATRP of MMA with an
M:CuBr2:L:AIBN ratio of 500:1:1:0.5
With the increased catalytic complex concentration it was also noticed that the graft polymerization was now controlled, with the graft ratio increasing linearly with time.
40
35
30
25
20
15 Graft ratio (wt%) 10
5
0 0 20406080100120140 Time (h) Origional; Repeat
Figure 5.6: Grafting of MMA to SynPhase substrates using RATRP with an
M:CuBr2:L:AIBN ratio of 500:1:1:0.5 5. γ-radiation initiated reverse ATRP graft polymerization 132
5.5.1. Molecular weight analysis of γ-radiation initiated RATRP polymers
The homopolymer formed from the radiation-initiated RAFT polymerization was found to have a PDI between 1.2 and 1.5.
Figure 5.7 shows the molecular weight distributions from the experiments from section 5.5 and it can be seen that the reason for the higher PDI is that there is significant tailing at the low molecular weight region of the trace. This tailing is a mechanistic feature of γ-radiation initiated RATRP. Rather than being from termination reactions throughout the polymerization, as the tailing due to continual initiation during the reaction time. The AIBN in the solution should degrade fairly quickly under γ-radiation which will give an initial burst of radicals such that the majority of polymer chains are initiated early in the polymerization process. Thus, the dominating radical source is the AIBN degradation.
Increased polymerization time Normalized Response
4.0 4.5 5.0 5.5 6.0 log(M ) w
Figure 5.7: Molecular weight distributions for γ-radiation initiated RATRP polymerization of MMA 5. γ-radiation initiated reverse ATRP graft polymerization 133
Thus it can be concluded that while the polymers produced don’t have a PDI as low as thermally-initiated RATRP polymers, the polymers produced are still of a controlled molecular weight.
5.5.2. Effect of catalyst complex concentration on RATRP grafting of MMA
By keeping the AIBN to copper complex ratio constant in this system, the concentration of active and dormant polymer chains is changed without changing the equilibrium concentrations. Because of this equilibrium it is necessary to also alter the AIBN concentration to look at the effects of adjusting the catalyst concentration on the polymerization. The CuBr2 to PMDETA to AIBN ratio was kept constant at 1:2:0.5. The effect of increasing AIBN concentration is effective in increasing the radical concentration and thus an increase in the rate of polymerization, and a decrease in the molecular weight. The change of complex concentration should have little effect on the rates of polymerization and molecular weight apart from the persistent radical effects and the possible increase in induction period. Catalyst concentration is increased so that control can be maintained over the increased radical concentration.
The polymerization was prepared as described at the beginning of the chapter with
MMA as the monomer, PMDETA as the ligand and CuBr2. The polymerizations were prepared by serial dilution and the ratios of the ligand AIBN and CuBr2 were varied relative to the monomer concentration as shown below:
Monomer:CuBr2 Monomer:Ligand Monomer:AIBN 250 : 1 250 : 2 250 : 0.5 375 : 1 375 : 2 375 : 0.5 562 : 1 562 : 2 562 : 0.5 843 : 1 843 : 2 843 : 0.5 1265 : 1 1265 : 2 1265 : 0.5 1898 : 1 1898 : 2 1898 : 0.5
Table 5.1: Reagent ratios used to investigate the effect of complex concentration on γ-radiation RATRP polymerization of MMA 5. γ-radiation initiated reverse ATRP graft polymerization 134
Again DMF was used at 20 vol% to increase solubility of the CuBr2 complex. Once the solutions were degassed they were placed in the γ-source at a dose rate of 0.056 kGy/h for 27 hours.
The results from this experiment can be predicted by Eq. 5.11 that was derived earlier in the chapter.
⎛ 1 ⎞ [CuBr] ln⎜ ⎟ = ()f [AIBN]t + 1 f DR.t 2 × k K Eq. 5.11 − aibn 2 rad p eq ⎝1 x ⎠ [CuBr2 ]
Since the reaction should reach an equilibrium concentration of CuBr and CuBr2, this experiment should be dominated by the increased AIBN concentration, but without showing molecular weight broadening effects that are associated with increasing the initiator concentration significantly beyond the catalytic complex concentration. The results of this experiment indicated that both the conversion and graft ratio increased linearly with complex/initiator concentration as seen in Figure 5.8.
The molecular weight evolution of this system can be predicted as derived earlier by Eq. 5.14
[M ] .x.M monomer Mn = 0 n Eq. 5.14 th + f AIBN [AIBN]0 frad DR.t
As expected from theory, the molecular weight decreased linearly with an increase in the catalyst ratio. The conclusion here is that the increase in relative concentration of initiator to monomer ratio causes an increase in radical concentration, and thus an increase in rate of polymerization and a decrease in the corresponding molecular weight. The resulting polydispersity index of the molecular weight was again between 1.2 and 1.5. 5. γ-radiation initiated reverse ATRP graft polymerization 135
0.05
0.04
0.03
0.02 ln(1/1-x)
0.01
0.00 0.00 0.01 0.02 0.03 Complex conc. (mol/L)
180
160
140
120
100
80
60 Graft Ratio Ratio (wt%) Graft 40
20
0 0.000 0.005 0.010 0.015 0.020 0.025 Complex conc. (mol/L)
Figure 5.8: Effect of initial CuBr2 concentration on RATRP polymerization of MMA. AIBN and PMDETA were maintained at a 0.5:1 and 2:1 ratio with respect to CuBr2. Polymerized at 0.056 kGy/h for 27 h at the concentrations specified above. 5. γ-radiation initiated reverse ATRP graft polymerization 136
5.5.3. Effect of dose rate on RATRP grafting of MMA
It’s possible to alter the radiation dose rate in the polymerization system by varying the distance that the samples are placed from the radiation source, to exploit the fact that radiation decays with an inverse-square relationship with distance.
Since the radiation is the source of radicals in this system, adjustment of the dose rate will effectively adjust the amount of radicals in the system. The mechanistic implications of this will be discussed below.
As the γ-source is a constant initiation source the result is that the concentration of radicals in the system increases. However, because of the activation and deactivation equilibrium:
ka RBr CuBr R CuBr2 kd
kp
The situation is more complicated than a linear increase in the radical concentration. The equilibrium constant for the above equation is thus:
[R• ][CuBr ] K = 2 eq [R − Br][CuBr]
This means that as the radiation produces radicals the number of radicals in the system will increase but not at the same rate as they are being created because of a shift in this equilibrium. Looking at the kinetic equation (Eq. 5.10) derived earlier we can see that the rate of polymerization and should increase linearly with the dose rate as the increasing dose rate effectively increases the radical concentration in the system.
d[M ] [CuBr] R = − = ()f [AIBN] + f DR.t ×[M ]k K p aibn rad p eq Eq. 5.15 dt [CuBr2 ]
5. γ-radiation initiated reverse ATRP graft polymerization 137
The effect of dose rate on the RATRP of MMA was determined by conducting an experiment as follows. The polymerization was prepared as described at the beginning of the chapter with MMA used as the monomer, PMDETA as the ligand, and CuBr2 in a ratio of 500:2:1. 10 vol% DMF was added to aid in the dissolution of the CuBr2 complex. AIBN was added in a 0.5:1 ratio with the CuBr2. The polymerization solution was then placed in the γ-radiation source at varying dose rates of 0.019, 0.025, 0.035, 0.056 and 0.105 kGy/h for 83 hours.
40
35
30
25
20
15 Conversion (%)
10
5
0 0.000 0.025 0.050 0.075 0.100 Dose Rate (kGy/H)
Figure 5.9: Effect of dose rate on homopolymer produced in RATRP grafting of
MMA. MMA:PMDETA:CuBr2:AIBN ratio was 500:2:1:0.5 and a 10% volume of DMF was added as a solvent
Eq. 5.14 shows the theoretical molecular weight equation derived at the beginning of this chapter.
[M ] .x.M monomer Mn = 0 n Eq. 5.14 th + f AIBN [AIBN]0 frad DR.t 5. γ-radiation initiated reverse ATRP graft polymerization 138
From Eq. 5.14, the expected molecular weight for a system is proportional to the inverse of the dose rate. However, the molecular weight analysis is slightly more difficult for this system since the samples were taken after a constant time and the conversion varied between each sample. To properly interpret the molecular weight data it is necessary to normalize the molecular weight with respect to conversion, such that the individual molecular weights can be predicted. There should be a linear relationship between the normalized molecular weight and the inverse of the dose rate, as can be seen in Figure 5.10.
2500000
2000000
1500000
1000000 Mn/conversion
500000
0 0 102030405060 1/Dose rate (H/kGy)
Figure 5.10: RATRP grafting of MMA under constant γ-radiation; Effect of dose rate on the normalized molecular weight
Rearanging Eq. 5.14 and conducting a linear regression on the above graph yields the following realationship M [M ] ..M monomer [M ] ..M monomer n = 0 n + 0 n
x frad DR.t f AIBN [AIBN]0
Y = 42280x + 122769 5. γ-radiation initiated reverse ATRP graft polymerization 139
-4 Thus it is calculated that frad = 3.05x10 and fAIBN = 2.37. As such this tells us two things, firstly that all the AIBN is used up in the initial stages of the polymerization as predicted and the expected maximum value for fAIBN should be 2, but this can be explained by the high amount of error in the graph. Secondly, this tells us that AIBN derived radicals contribute a tenfold greater number of radicals than the radiation for these polymerization conditions. However, there is still a significant contribution by radiation derived radicals evident.
The polymeric graft ratio is a combination of the number of surface polymer chains and the molecular weight of these polymer chains. Figure 5.11 shows that the graft ratio is linear with respect to inverse dose rate, following the same trend as the molecular weight. Thus, the changes in graft ratio are predominantly driven by the changes in the molecular weight of the grafted and solution polymers.
60
50
40
30
20 Graft ratio (wt%)
10
0 0 5 10 15 20 25 30 35 40 45 50 55 1/(dose rate) (kGy/h)-1
Figure 5.11: RATRP grafting of MMA under constant γ-radiation; Graft ratio vs. inverse dose rate with a MMA:PMDETA:CuBr2:AIBN ratio of 500:2:1:0.5 and a 10% volume of DMF was added as a solvent 5. γ-radiation initiated reverse ATRP graft polymerization 140
5.5.4. Effect of solution radical concentration on RATRP grafting of MMA
Due to the nature of the equilibrium in ATRP and reverse ATRP polymerizations, there is a significant build up of radicals into dormant species before polymerization can take place. Once the radicals are formed there is a minimum concentration of radicals required before controlled polymerization can take place and the polymerization will start. Radiation initiation compared to conventional thermal polymerization generates fewer radicals, so it is advantageous to increase the radical flux to maintain the active- dormant equilibrium at an optimum level. The addition of thermal initiators, in particular AIBN, to the system has been noted in the past to accelerate radiation polymerization by increasing the radical flux.153
The addition of AIBN to a RATRP system thus should result in an overall drop in
CuBr2 concentration in the equilibrium. This should in turn accelerate the rate of polymerization and likewise reduce the molecular weight of the resulting polymer for a given conversion.
The polymerization was undertaken to determine the effect of AIBN concentration on the polymerization of MMA. The solutions were prepared as described at the beginning of the chapter. MMA as the monomer, PMDETA as the ligand, and CuBr2 were used in a 500:1:1 ratio. 10 vol% DMF was added to aid in the dissolution of the CuBr2 complex and the AIBN concentration was changed between 7.5, 15, 30, 45 and 75 mmol/L. Polymerization was initiated by placing the solution vials in the γ-radiation source at a dose rate of 0.056 kGy/h and polymerized for 96 hours.
Figure 5.12 shows that increasing the concentration of AIBN results in an increase in the conversion. Likewise there is a decrease in the molecular weight as the AIBN concentration increases. This is because of competition between an increase in the number of polymer chains which would reduce molecular weight and an increased rate of polymerization which would increase the expected molecular weight. Figure 5.12 5. γ-radiation initiated reverse ATRP graft polymerization 141
60 55 50 45 40 35 30 25 20 Conversion (%) 15 10 5 0 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 AIBN conc (mol/L)
1.5 1.4 1.3
1.2 PDI 1.1 1000000 1.0 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
800000
600000
400000 Mn/Conversion 200000
0 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 AIBN concentration (mol/L)
Figure 5.12: Effect of AIBN concentration on conversion of γ-initiated RATRP polymerization of MMA with a MMA:PMDETA:CuBr2 ratio of 500:2:1, the AIBN concentration was varied between 7.5, 15, 30, 45 and 75 mmol/L and a 10% volume of DMF was added as a solvent 5. γ-radiation initiated reverse ATRP graft polymerization 142
shows the effect of AIBN concentration on the molecular weight, normalized against conversion. This graph shows the decrease in molecular weight expected from an increase in radical concentration.
The changes in AIBN concentration across this experiment result in a 62% reduction in molecular weight of the solution polymer, which in turn is expressed by a 55% reduction in the graft ratio. The AIBN effectively reduces the polymer molecular weight while the number of grafted polymer chains remains constant. This is because the surface grafted polymer chains can only be initiated by the effect of the γ-radiation on the substrate and the addition of the solution initiator AIBN doesn’t effect the number of grafted polymer chains.
35
30
25
20
15
Graft Ratio (wt%) Graft 10
5
0 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 AIBN conc. (mol/L)
Figure 5.13: Effect of AIBN concentration on the grafting of MMA controlled by RATRP 5. γ-radiation initiated reverse ATRP graft polymerization 143
5.6. Conclusion
γ-radiation initiated polymerization of MMA was successfully controlled using RATRP. Increasing the radical flux of the polymerization through addition of the thermal initiator AIBN increased the concentration of radicals to a level where control of the polymerization could be achieved. The resulting polymers had a linear first-order kinetic plot and molecular weight increased linearly with conversion. The polymers had a PDI of between 1.2 and 1.5 and this broadening due to γ-radiation causing constant initiation of short chain polymers throughout the polymerization. It was found that frad = -4 3.05x10 and fAIBN = 2.37 and as such there was a significant contribution of AIBN radicals to the solution polymerization mechanism and that all the AIBN is consumed during the polymerization.
The effect of AIBN concentration, dose-rate and the evolution over time were analyzed. This is the first time that radiation initiated polymerization has been controlled using RATRP and is a significant discovery and provides a new method for grafting of polymers using living radical techniques.
This is a significant result as it not only opens a new technique for grafting polymers using γ-radiation but also is the first instance that RATRP has been properly conducted from a radical source at room temperature.
There is further scope for work on this topic, which could potentially provide greater insight into the kinetics and the exact mechanism of RATRP and ATRP polymerization. 6.γ-radiation initiated RAFT grafting of styrene 144
6. γ-radiation initiated RAFT grafting of styrene
6.1. Introduction
A general introduction into RAFT polymerization can be found in Chapter 2; however there are several topics intrinsic to RAFT polymerization that will be discussed here. Primarily this is the tendency of RAFT polymerization to proceed at a slower rate than conventional polymerization. This is termed ‘retardation’ and has been the basis of significant research into its causes and the implications that this has on the mechanism controlling RAFT polymerization.
Further detail will be provided here on the production of novel architecture molecules using RAFT polymerization.
6.1.1. Retardation
Most RAFT polymerizations ideally proceed with rates of polymerization within 20% of free radical rates of polymerization. However, in some circumstances significant retardation of the polymerization can occur. This is especially noted with the largely popular cumyl dithiobenzoate (CDB), which significant retardation has been observed at high RAFT agent concentrations and during the initial stages of emulsion polymerization.145 Both retardation and inhibition have been associated with macroRAFT radical characteristics146
Moad145 et al. suggests that retardation is caused by one or more of the below effects: 1. Slow fragmentation of the RAFT intermediate (2) formed from the initial RAFT agent (1) 2. Slow fragmentation of RAFT intermediate (2) formed from the polymeric RAFT agent 3. Slow reinitiation by the expelled radical R 6. γ-radiation initiated RAFT grafting of styrene 145
4. Specificity for the expelled radical to add to the RAFT agent rather than monomer (i.e. transfer constant is too high)
Retardation occurs exclusively when an agent with a radical stabilizing Z-group is used. If the Z-group is replaced with a less stabilizing group then retardation is suppressed or even disappears.146
This retardation has been the subject of much contention over the years. The formation of the intermediate (2) has been confirmed by electron spin resonance (ESR) from CSIRO, and the radical concentration was found to be approximately 8x10-7 M for styrene polymerization controlled by CDB.231 The accuracy of this was questioned given that in other systems there is evidence for radical dissociation onto the sulphur.142,146,232,233 This dissociation likely to introduce other factors into the determination and ESR of sulphur centered radicals is quite complex.234
Barner-Kwollik et al.235 suggested the intermediate radical has significant lifetime because radicals being supplied through initiation were not producing the predicted concentration of polymer. Vana et al.146 calculated intermediate macroRAFT radical concentration at 5x10-4 M through simulation of the same conditions as the CSIRO paper.
Monteiro and co workers236,237 suggest a different scenario, that fragmentation of intermediate radical is very fast (10-5 s-1) and retardation is caused by macroRAFT radical reacting with free polymeric radicals, kt,intermediate, and they believe the CSIRO mechanism needs amending with an irreversible termination step that causes radical loss.237 This radical concentration is in agreement with CSIRO’s ESR measurements.231 In a later paper Vana et al.146 did point out some drawbacks with this conclusion. Firstly intermediate termination hasn’t been identified by techniques such as MALDI-TOF- MS. He does concede that RAFT polymers show a significant amount of end-group fragmentation during ionization, and this may prevent identification through mass- spectroscopy techniques. Secondly, kt,intermediate has to be very high to induce observed retardation. 6.γ-radiation initiated RAFT grafting of styrene 146
Fakuda et al.238 also proposed termination of the macroRAFT radical to form a 3-armed star. To do this he synthesized a star product through non-polymeric means. However, the synthesis of this compound does not necessarily relate it to RAFT polymerization.146 Using electron spray mass spectroscopy techniques, Vana et al. managed to identify only the dithioester functionalized chain ends. The mechanism for Fakuda’s proposed termination is show below.
R
Pn S S Pm Z
Radical
k−β S S Pm Pn S S Pm Pn S S Pn kβ Pm kβ Z k−β Z Z (1) (2) (3)
kreactivate
Pn S S Pn
Pm S S Pm
possible reversible termination product
Figure 6.1: Possible mechanism of star type macroRAFT termination
Vana and co-workers146 designed an experiment to detect the presence of long lived and stable intermediates via γ-irradiation initiation of styrene using CDB and the conditions outlined by Barner-Kowollik et al.239 Their work suggested that the RAFT process involves a stable radical sink, removing radicals from the reaction process. This sink could be either a radical or non-radical reservoir. As part of this work an experiment was designed to enhance cross termination, however no cross-termination product could be identified. 6. γ-radiation initiated RAFT grafting of styrene 147
Further modeling has been conducted by different groups which have supported various models. From Wang and Zhu’s240 modeling of the RAFT process it was found that 4 -1 238 kβ =10 s , which agrees with Kwak but differs by 6 orders of magnitude from 235 -2 -1 Barner-Kwollic’s calculations that showed k-β = 10 s .
Coote and Radom241 used high level ab initio orbital calculations and showed the disulphur intermediate is relatively stable (for systems where R= CH3, CH2COOCH3, -1 C(CH3)2CN and Z=CH3, phenyl, benzyl). They found k-β = 9.8 s which is several orders of magnitude different to Kwak.238 These calculations unequivocally support the evidence that the intermediate RAFT radical is by no means short lived.
Ah-Toy et al.242 conducted a detailed study using GPC and ESI-MS techniques in an attempt to find evidence of terminated RAFT intermediates. This detailed study failed to find any evidence of three or four arm terminated stars and thus concluded that this is not likely to be part of the RAFT mechanism.
Additionally, retardation is noted to be highly dependent on the RAFT agent used. In fact there are RAFT agents that exhibit no noticeable retardation. If irreversible termination of the macroRAFT radical was the dominant reason for the retardation, then we would expect to see significant retardation for most RAFT systems and this is not the case.146
The origins of retardation in the RAFT system are well beyond the scope of this thesis, however it does highlight the fact that the RAFT mechanism is not completely understood yet and there is a whole field of lively discussion still on the RAFT mechanism.
6.1.2. Molecular Architectures using RAFT
The RAFT mechanism and the chemical versatility of the RAFT process means that RAFT polymerization has been rapidly adapted to the production of polymers with novel architectures such as blocks, brushes and stars.144 There are two possible methods from producing more complicated RAFT architectures; either by attachment through the 6.γ-radiation initiated RAFT grafting of styrene 148
R- or the Z- group. If the attachment point is the R- group effectively you have a pure polymer and the RAFT agent is only incorporated to control the polymerization. If the attachment point is the Z- group, then the RAFT agent forms an integral part of the polymer and the structure is effectively produces from the reaction of a polymer “arm” with the RAFT agent. These methods also go by a number of other names including “arm first” or “grafting to” for the Z- group method, and “core first” or “grafting from” for the R- group method.
Firstly to produce more complicated polymers using RAFT, you have to look at what factors affect RAFT polymerization. The effectiveness of a particular RAFT agent is dependent on 3 main factors145
1. The reactivity of the thiocarbonyl to radical addition 2. The leaving group ability of R compared to that of the propagating polymer radical. 3. The ability of the leaving group R to reinitiate polymerization.
Thus to produce more novel polymers you need to tune the RAFT agent by tailoring these parameters to the specific application.
S S S S
Core
Core S S S S S S
S S
(a) (b)
Figure 6.2: Representations of (a) R group approach to novel RAFT polymers and (b) Z group approach for the formation of stars 6. γ-radiation initiated RAFT grafting of styrene 149
One of the first novel polymers to be published was from the CSIRO group who produced a variety of AB and ABA block copolymers using dithioesters.133 The more interesting ABA polymers where produced using a bifunctional RAFT agent in a two step process. The B block was grown first and then after isolation and purification the two A blocks were grown out from the centre B block. This method has allowed for a variety of triblocks to be produced.
Using these multifunctional RAFT agents many novel structures have been created including stars133,243, graft polymers132,244 and dendrimers.245,246 Some examples include Shinoda and Matyjaszewski133 who made a water soluble graft of MMA and polydimethylsiloxane (PDMS) methacrylate. The result was well a defined block copolymer without a prevalence of unreacted side chains.
S S CH2
O C4H9 Si O Si O O n
Figure 6.3: Shinoda and Matyjaszewski’s method for producing MMA-PDMS brush copolymers
Matyjaszewski and co-workers133 also made the first alternating RAFT block copolymer. Proof of the structure of this polymer came through NMR analysis to get the structure and GPC to analyze the polymer chains. The polymer was found to have a narrow molecular weight and a well ordered tacticity and molecular weight distribution. This shows that through appropriate use of RAFT agents we can create and control all facets of the polymerization process.
There have also been a number of more complicated block copolymers formed through a combination of RAFT and cationic ring opening polymerizations (CROP). He et al.247 polymerized 5-6-benzo-2-methylene-1,3-dioxepane controlled by a mixture of RAFT and CROP. The same group also extended this work to produced a number of block copolymers and star copolymers from this combination of RAFT and CROP.248-251 6.γ-radiation initiated RAFT grafting of styrene 150
6.1.3. The mechanism of radiation grafting using RAFT polymerization
In a paper from our group, Quinn et al.252 reported on the synthesis of RAFT agents that can be used for low temperature polymerization. This included polymerization under UV radiation28 and 60Co γ-radiation. These novel RAFT agents carry a benzyl moiety as the Z-group as opposed to the more normal phenyl group to destabilize the intermediate macroRAFT radical (species (2) and (4) in Figure 6.4) and reduce retardation.253 A typical ambient temperature RAFT agent, cumyl phenyldithioacetate (CPDA), is shown in Figure 6.5.
This chapter extends this work and applies it to 60Co γ-radiation grafting polymerization off polymeric surfaces.
The proposed reaction mechanism for reversible addition fragmentation chain transfer polymerization under a constant source of γ-radiation is shown in Figure 6.4. The basic steps of the RAFT mechanism are extended by reaction steps that effect grafting onto surfaces solid surfaces such as polypropylene.
Radiation initiation reactions are discussed in greater detail in Chapter 2. For this work we can assume the initiation step for this mechanism involves the action of γ-radiation on the various organic compounds in the solution (I). We can assume that the only active species produced are those that are derived from monomer units and radicals formed on the substrate.198 These styrene radicals and surface radicals can then initiate propagating chains which subsequently react with the RAFT agent (II) to for a RAFT radical intermediate (2).
These reaction steps, summarized under (II), have been termed the pre-equilibrium. In the pre-equilibrium the leaving group radicals R are released from the initial RAFT agent and subsequently re-initiate macromolecular growth in the reaction step (IIIb). The core of the RAFT process is the main equilibrium given by the reaction steps (IVa) and (IVb). 6. γ-radiation initiated RAFT grafting of styrene 151
(I)
gamma Monomer M M Pm
gamma Monomer SS SS SS-Pm
(II) P S S S S R Pm S S R m R Pm Z Z Z (1) (2) (3)
monomer monomer (IIIa) (IIIb) SS SS P1 R P1
(IIIc) monomer Pn Pn+1
(IVa) S S P Pm S S Pm S S Pn n Pn Pn Z Z Z (1) (2) (3)
S S P SS (IVb) Pm S S Pm S S Pn SS n Pn SS Pn Z Z Z (1) (2) (3)
(V) P P n Pm n+m
SS-Pn Pm SS-Pn+m
Figure 6.4: Mechanism for graft polymerization controlled by RAFT
Within this core process, free macro radicals that are either attached to the PP surface or are free to move in the surrounding reaction mixture, are adding to polymeric RAFT agent to from macroRAFT intermediate radicals (4), which in turn may fragment by releasing the starting materials or the formerly attached polymeric entity. Reaction step (V) gives the bimolecular termination reactions between free or surface attached macro radicals. In the case of styrene, these termination reactions are believed to proceed almost exclusively via combination. 6.γ-radiation initiated RAFT grafting of styrene 152
S S R-group
Z-group
Figure 6.5: Cumylphenyl dithioacetate low temperature RAFT agent
One great advantage for the use of RAFT in controlling graft polymerization is that the growing tethered chains generated on the surface are in a dynamic equilibrium with un- tethered chains in solution. This allows for a ‘post-mortem’ analysis of the linear chains by NMR or GPC that should closely reflect that of the surface bound chains.
6.2. Experimental method
Materials
Styrene (Aldrich, 99 %) was deinhibited by passing through a column of activated basic alumina. SynPhase Lanterns made of polypropylene were provided by Mimotopes Pty. Ltd. and were used as polymeric surfaces for the experiments. The lanterns were washed with dichloromethane for 24 hours and dried in vacuo prior to use. Each Lantern has a surface area of approx. 3.6 cm2 and a weight of approx. 0.07 g. Each Lantern was weighed prior to grafting.
Grafting Experiments
Solutions of cumyl phenyldithioacetate (CPDA) were prepared gravimetrically, ranging from [CPDA] = 1x10-2 to 2x10-3 M. Bulk polymerizations were conducted in 5 mL aliquots of solution per lantern in glass sample vials. The vials were capped with rubber septa, and deoxygenated by purging with nitrogen gas for 15 minutes each. The samples were placed in an insulated room with a 60Co source at room temperature, each series in duplicate and at dose rates of 0.18, 0.08, 0.07, 0.05 and 0.03 kGy h-1. Samples were taken after different time intervals. The conversion of polymer in solution of each sample was determined gravimetrically after drying initially in a fume hood, and then in a vacuum oven at 30 °C for 24 hours. The Lanterns were rinsed with dichloromethane 6. γ-radiation initiated RAFT grafting of styrene 153
(DCM), dried in a fume hood, and then in a vacuum oven. The mass gain of the grafted Lanterns was determined via weighing. In order to remove the homo-polystyrene, which can be enclosed in the grafted surface, the lanterns were washed extensively with DCM. Each lantern was then placed in a bottle and 20 mL of DCM were added. The bottles were put on a shaker and shook for up to one week. The solvent was changed every few days. This process was repeated until no homo-polystyrene could be identified in the solution via precipitation with methanol. After removing all homo-polystyrene, the lanterns were dried in a fume hood, and then in a vacuum oven at 30°C for 24 hours. The mass gain of the Lanterns due to the grafted polystyrene was determined gravimetrically. In order to investigate differences between controlled and conventional free radical grafting of polystyrene onto solid surfaces, styrene was also grafted onto lanterns without using RAFT agents or other chain transfer agents.
Analysis
The homopolymer produced in solution will be analyzed gravimetrically to give conversion and by GPC analysis give molecular weight details.
The grafted polymer was analyzed gravimetrically to give the graft ratio. FMOC load testing was conducted to give information about the load potential of the grafts, and the grafted polymer was further analyzed through the use of ATR-FTIR.
The molecular weight of the solution homopolymers was measured using the THF based GPC system outlined in Chapter 3.
6.3. Results and discussions
6.3.1. γ-initiated RAFT mediated graft polymerization of styrene onto SynPhase lanterns
The use of radiation to initiate grafting of polymers onto polymeric substrates is a quite well known technique.254-256 This direct grafting method includes irradiation of both the monomer and the polymer. The concentration of radicals generated by the radiation is characterized by the free-radical radiation chemical yield (GR) of the substance. This is 6.γ-radiation initiated RAFT grafting of styrene 154
also sometimes known as the G-Value of the substance. The G-value of the substance is essentially the number of radicals (or other active species) that is generated per gram of material per 100eV of exposed. For grafting of polystyrene onto a polypropylene backbone, the grafting reaction is significantly favored over homopolymerization of the polystyrene. Also grafting onto a polystyrene backbone is not favored due to the low G- value of polystyrene. This can be seen by the GR values below
149 GR(styrene) is 0.69 radicals per 100 eV 257 GR(polypropylene) at 77 K is 4.8 radicals per 100 eV 257 GR(polystyrene) at 77 K vary between 0.0009 and 1.8 radicals per 100 eV
149 Chapiro estimates GR(polystyrene) from 1.5 to 3.0 radicals per 100 eV at room temperature. As free radical yields of polypropylene and polystyrene are higher than that for styrene, grafting of styrene onto polypropylene and polystyrene is favored over homo polymerization of styrene.
Figure 6.6 shows a typical graft reaction showing the graft ratio vs. time. This experiment was for the grafting of polystyrene to polypropylene lanterns at a dose rate of 0.18 kGy h-1 at room temperature and using a CPDA concentration of 6x10-3 M. The interesting thing to note here is the two distinct regimes that can be seen. Regime 1 follows a linear relationship between graft ratio and time up to a certain point, at which the rate slows down (however it remains linear). This property is seen in both the conventional and RAFT mediated polymerization, however the changes between the two regimes occur at different points. Regime 1 ends for the conventional process at about 70 wt%, while the RAFT mediated polymerization changes at about 50 wt% graft ratio. Overall the RAFT polymerization proceeds at a slower rate compared with the conventional free radical polymerization. The crossover between the two regimes seems to occur between 25 and 30 h reaction time, independent of the polymerization method. 6. γ-radiation initiated RAFT grafting of styrene 155
140 no CPDA CPDA 120
100
/ wt% / 80 Regime 2
60
Grafting Ratio Grafting 40
20 Regime 1
0 0 102030405060708090 time / h
Figure 6.6: Grafting ratio vs. time plot of a RAFT agent mediated and a conventional free radical styrene polymerization onto PP Lanterns, dose rate = -1 -3 0.18 kGy h , [CPDA] 0 = 6x10 M, T ≈ 20ºC. It should be noted that the lines in Figure 2 and 4 only indicate regimes 1 and 2 and do not correspond to any kinetic model
Minko et al.258 stipulated that through the analysis of radical graft polymerization of surface bound initiators, the polymeric surface goes through several phases depending on the degree of grafting. It was found that after a certain time the surface is completely covered by graft polymer coils, and at this point grafting of new polymer chains requires stretching of the coils until the maximum possible surface graft is achieved. Minko assumed at after the surface is covered by grafted chains two processes may occur; a chain transfer reactions to bulk reactants, or a ‘force out’ regime, where new chains are forced out into the bulk solution. For both cases, grafting can either stop or result in multilayer covering of the substrate. 6.γ-radiation initiated RAFT grafting of styrene 156
While Minko et al.’s work was based on a surface bound initiator, it can be extrapolated to a γ-radiation initiated system, in this case essentially radicals are created in the bulk solution, and on the surface of the substrate. The surface radicals can initiate new polymer chains without hindrance until the surface is completely covered by graft polymer. At this stage the surface radicals continue to graft, but at a slower rate as they are shielded by the grafted polymer layer that restricts access to the monomer. In addition to this, the possibility of radicals forming on the grafted polystyrene chains exists. As discussed earlier, the degree off grafting off the polymeric styrene will be significantly less than off the polypropylene however it will still occur. This implies that grafting is still favored, but the grafting ratio decreases as GR(polystyrene) is smaller than GR(polypropylene). Thus, the number of chains per time that can be grafted onto the polystyrene layer is smaller than the number that can be grafted per time onto the polypropylene surface, resulting in a lower polymer graft density. Thus we can hypothesize that Region 2, during the grafting process originates in the complexity of a heterogeneous grafting mechanism.
6.3.2. Effect of RAFT agent concentration on γ- radiation grafting
The effect of the initial RAFT agent concentration was analyzed by irradiating RAFT agent solutions with between 1x10-2 and 2x10-3 M PEPDA for 48 hrs at 0.08kGy h-1. As expected, it was found that while increasing the initial RAFT agent concentration decreased the graft ratio. With conventional bench-top RAFT polymerizations increasing the concentration of RAFT agent decreases the polymer chain length. Thus if the number of grafted chains is constant (i.e. GR(polypropylene) is independent of RAFT agent concentration). From this it is clear that the graft ratio must decrease. This result can be seen below in Figure 6.7. 6. γ-radiation initiated RAFT grafting of styrene 157
80
70
60 / wt% 50
40 Grafting Ratio Grafting
30
20 0.002 0.004 0.006 0.008 0.010 Initial RAFT-agent Concentration / mol L-1
Figure 6.7: Graft ratio vs. initial RAFT agent concentration plot, reaction time 47hr, 0.08 kGy h-1 and T≈ 20oC
6.3.3. Effect of dose rate on radiation grafting of styrene
The effect of dose rate on the radiation grafting of styrene was also analyzed. The polymerizations were performed at dose rates of 0.18 and 0.08 kGy h-1, and the results are shown in Figure 6.9. As expected the graft ratio increased with increasing dose rates. With the decreased dose rate, fewer radicals are produced on the surface and in solution, thus slowing the rate of polymerization. Inspection of Figure 6.9 shows the same trends as Figure 6.8, with the two grafting regimes and again a crossover between the regimes at about 25 to 30 hrs. This suggests that the number of radicals generated on the surface and hence the number of grafted polymer chains depends only on the dose rate of the γ-source and on GR(polypropylene), not on whether the polymerization process is free radical or RAFT mediated. Essentially the molecular weight, polydispersity and polymer properties are controlled by the addition of the RAFT agent. However, this doesn’t affect the actual mechanism of grafting. The physical effects of 6.γ-radiation initiated RAFT grafting of styrene 158
the grafting of styrene to the surface are independent of the presence of the RAFT agent or not. Thus, the crossover between the two regimes should show a dependency on the dose rate. Unfortunately in this work, the difference between the dose rates (0.18 and 0.08 kGy h-1) does not allow us to see any significant shift in the crossover time.
140 0.18 kGy h-1 0.08 kGy h-1 120
100
/ wt% 80
60
Grafting Ratio Grafting 40
20
0 0 1020304050607080 time / h
-1 -3 Figure 6.9: Grafting ratio vs. time plot for 0.18 and 0.08 kGy h , [CPDA]0 = 6x10 mol L-1 and T ≈ 20ºC
6.4. Analysis of molecular weight distributions
Analysis of the free polymer produced in solution during graft polymerization is a standard method for analyzing graft polymerization reactions. It is possible to draw a correlation between the solution polymer and the grafted polymer. For RAFT mediated living polymerization, theory stipulates that the grafted polymer should be exactly the same as the solution polymer. However, Stenzel et al.259 showed through cleaving of the arms off RAFT stars from cyclodextrin, sucrose and glucose that there were some variations from theoretical in the growth of the arms. Stenzel hypothesized that since these get larger with conversion they are due to inactivity of some of the RAFT groups 6. γ-radiation initiated RAFT grafting of styrene 159
on the star core due to steric hindrance, and inter- and intra- molecular termination reactions. For radiation grafting reactions, the density of initiation sites that are currently active is significantly smaller than that of a star core, suggesting that we can ignore this interference and assume strong correlation between the solution polymer molecular weight, and that of the grafted polymer.
Strehling et al.174 and Miwa et al.165 investigated the cleavage of grafted polystyrene from its point of attachment with trimethylsilyl iodide and triflouroacetic acid respectively. Both these groups used grafting methods that link the grafted polymer to the backbone with an ester linkage that can be easily cleaved. However, with this work, the polystyrene is linked to the backbone through a carbon-carbon bond, which gives no option of selectively cleaving the grafts from the substrate. Thus, we used the free solution polystyrene (PSfree) to analyze the graft polymers, and to gain an indication of the number average molecular weight (Mn) and the polydispersity (ρ). These can be seen below in Figure 6.10.
14000 M 2.4 n M theo n 12000 PDI 2.2
10000 2.0
-1 8000 1.8 / g mol / g
n 6000 1.6 M
4000 1.4
2000 1.2
0 1.0 0.00 0.02 0.04 0.06 0.08 0.10 conversion
Figure 6.10: Evolution of number average molecular weight, Mn, and polydispersity index with monomer conversion for polystyrene formed in solution, -3 -1 PSfree: [CPDA] 0 = 6x10 mol L and T ≈ 20ºC 6.γ-radiation initiated RAFT grafting of styrene 160
The data clearly shows that the number average molecular weight PSfree is proportional to the conversion, consistent with a living polymerization. The polydispersity index remains low (<1.2) throughout the entire polymerization. The theoretical molecular weight is slightly higher than the experimental molecular weight.
Strehling et al.174 and Miwa et al.165 reported similar results to those found here. 165 Moreover, Miwa and co-workers found that Mn of PSgraft corresponds closely to
PSfree. However, with increasing conversion the Mn of PSgraft is slightly higher (ca. 10%) 260 than the Mn of PSfree. Luzinov et al. have shown theoretically for a conventional free radical polymerization, the molecular weights of grafted chains and the chains in the solution are equal if the propagation and termination coefficients in the bulk and for the surface grafting are equal. Thus, assuming little termination in graft polymerization and similar propagation rates between grafting and solution polymerization, our analysis here is valid.
6.5. FMOC-loading of grafted Lanterns
In practice, one of the most important characteristics of a solid phase is the loading capacity (defined by the number of accessible reaction sites on the surface). One measurement of the loading capacity is an FMOC-test. The exact details of how this loading test is conducted can be seen earlier in Chapter 3. Effectively this test measures the number of accessible active sites that are available for use as an anchor point for SPOS.
Grafted Lanterns were produced using dose rates of 0.18, 0.08, 0.07, 0.05 and 0.03 kGy h-1 and initial RAFT agent concentrations between 1x10-2 and 2x10-3 M were used. The loading vs. the molecular weight of the free polystyrene PSfree and loading vs. graft ratio for both conventional free radical and RAFT-mediated free-radical polymerization can be seen below in Figure 6.11 and Figure 6.12. 6. γ-radiation initiated RAFT grafting of styrene 161
200
160
/ wt% / 120
80
Grafting Ratio Grafting 40
0
500000
400000 -1
300000 / g mol g / free 200000 of PS w
M 100000
0 0 10203040 Loading / μmol
Figure 6.11: Grafting ratio and weight average molecular weight of PSfree vs. loading for a conventional free radical styrene polymerization onto PP Lanterns, T ≈ 20ºC. For the specific grafting conditions see text. 6.γ-radiation initiated RAFT grafting of styrene 162
100
80 / wt% 60
40 Grafting Ratio Grafting 20
0
10000 -1 8000
/ g mol / g 6000 free
of PS of 4000 w M 2000
0 0 5 10 15 20 25 Loading / μmol
Figure 6.12: Grafting ratio and weight average molecular weight of PSfree vs. loading for a RAFT agent mediated free radical styrene polymerization onto PP Lanterns, T ≈ 20ºC. For the specific grafting conditions see text. 6. γ-radiation initiated RAFT grafting of styrene 163
In the same way that the graft ratio shows two distinct regions, we find that the loading also has two distinct regions. The crossover between these two regimes occurs at a graft ratio of approximately 50 wt% for the RAFT mediated polymerization, and at about 70 wt% for the conventional polymerization. It is also interesting to note that there is a good linear correlation between the Mw (PSfree) and loading with the RAFT mediated polymerization, which also shows a distinction between the two regimes and crosses -1 over between Mn (PSfree) of between 4000 and 5000 g mol . However, the conventional radical polymerization shows no correlation at all between molecular weight and loading. This can be explained by the differences between living radical and conventional radical polymerization. In conventional radical polymerization the polymeric material that is initially formed is of a high molecular weight and this distribution does not change until very high conversion is reached. In addition the polymers produced through conventional radical techniques are of a high polydispersity. Conversely for living free radical polymerization, the polymers start form from a low molecular and increase linearly with conversion. This means that the loading of a grafted Lantern can only show a linear relationship with respect to the molecular weight of the polymer in the solution, if the grafted polymer is also produced via a living polymerization. From this we can conclude that the loading can be directly related to the molecular weight of the free polymer Mn (PSfree) and as such this provides good evidence that the RAFT mediated grafting of styrene onto polypropylene is in fact a living radical process.
Further analysis of Figure 6.12 shows that in regime 1, for an increase of Mw of about 1160 g Mol-1 gives us a loading increase of about 1.0 μmol but in region two the same increase in Mw gives a loading increase of 3.6 μmol.
If we plot the loading per gram of PS instead of the total loading, then the change in regime is even more clearly evident, as can be seen in Figure 6.13. Thus for Regime 1 -1 the loading capacity increases at about ~ 75 μmol per 1 g of PSgraft per 4000 g mol of
PSfree. In regime two the loading capacity generated is approximately 210 μmol per 1 g -1 of PSgraft per 4000 g mol of PSfree.
As the loading test measures the number of ‘active’ groups on the lantern, these results led us to the conclusion that the polymer grafted in regime two provides a substrate 6.γ-radiation initiated RAFT grafting of styrene 164
more suited to solid phase chemistry than in regime 1, and that the polymer in regime 2 is more accessible by reagents.
400 -1
300 mol g μ / graft 200
100 Loading per 1 g of PS of g 1 per Loading 0
0 2000 4000 6000 8000 10000 12000 M of PS / g mol-1 W free
Figure 6.13: Loading per PSgraft vs. weight average molecular weight of PSfree, T ≈ 20ºC. Experimental data used for this Figure are the same as in Figure 7. For the specific grafting conditions see text. 6. γ-radiation initiated RAFT grafting of styrene 165
6.6. Conclusions
Styrene was successfully grafted to polypropylene SynPhase Lanterns via a γ-initiated RAFT-mediated free-radical polymerization process using cumyl phenyldithioacetate. This grafting was contrasted with conventional free radical polymerization.
Both the RAFT mediated polymerization and the conventional free radical polymerization showed two distinct regimes in relation to the rate of grafting. These two regimes showed linear dependence of graft ratio and conversion and are hypothesized to correspond to conditions where only part of the substrate is covered by graft polymer and regime two where the whole substrate is covered by a layer of graft polymer. It is hypothesized that transfer of radicals from the substrate and direct radical generation on the grafted polystyrene result in continued grafting of polystyrene and generation of multi-layers of grafted polymer.
It was found that the graft ratio could be effectively controlled by the concentration of RAFT agent and varying the dose rate of the γ-source. The molecular weight of the free polystyrene was analyzed and found to show a linear correlation vs. conversion and a low polydispersity index and thus is a good indication of living free radical behavior.
Loading tests were conducted on the lanterns and it was found that the loading showed the same two regime trend as the graft ratio, and also showed a linear dependence on the molecular weight of free polystyrene Mw (PSfree).
In this work, we have demonstrated that γ-initiated RAFT polymerization can be used to graft polymers with low polydispersity onto polymeric surfaces. This method opens the door to designing new surfaces with block copolymer, branched and star structures. γ- initiated RAFT polymerization is a one-step method that leads to a stable and robust bonding between the surface and the grafted polymer. The surface does not have to be functionalized prior to polymerization. During polymerization a stable carbon-carbon- bond between the surface and the grafted polymer is established. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 166
7. RAFT mediated grafting of fast polymerizing- hydrophilic monomers initiated by γ-radiation
While a large part of the field of combinatorial chemistry is centered on the use of polystyrene based systems such as the SynPhase lanterns, Wang and Merrifield resins, there is still a large scope for other polymers to be used as solid phase supports. Of particular interest are hydrophilic polymers that are used in biological fields of combinatorial chemistry such as peptide synthesis.261,262
One major problem with the use of many of the hydrophilic polymers is that they polymerize very fast and have a tendency to cross-link. The use of RAFT techniques to control this polymerization could possibly open up new polymeric supports for the field of combinatorial chemistry and SPOS.263
The polymers of N,N’-dimethylacrylamide, methacrylic acid and acrylic acid were selected as being of interest for grafting onto supports. Extending this work, novel block copolymers and brushes were also grafted to demonstrate the wide range of advanced polymeric structures that living polymerization opens up.
The polymerization of high kp monomers creates some significant hurdles that need to be overcome to control their polymerization. RAFT polymerization, which generally proceeds at the same rate or 20% slower than the comparable uncontrolled free radical 145 polymerization. Unfortunately the polymerization of some high kp monomers, this is too fast. Polymerizing at this speed still makes control difficult and risks cross-linking and other side reactions. It was chosen to use CDB as a RAFT agent for these polymerizations. CDB at low temperature fragments very slowly, and this will increase the amount of time that radicals spend in the intermediate macro-radical state. This has two effects: firstly it slows the polymerization to a manageable rate and secondly, because of the reduced active radical concentration it reduces side reactions. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 167
Poly(dimethylacrylamide) is an important polymer in production of solid phase supports because of its hydrophilicity and the tertiary amine functionality that is used as a linker point or as part of other chemical reactions.185,264 However, one of the problems with DMA is that it polymerizes very fast (29,198 l/mol/s at 30oC compared to 51.9 l/mol/s for styrene219) and is difficult to control. The use of RAFT to control the grafting of DMA onto substrates can provide distinct advantages.
Poly(acrylic acid) and poly(methacrylic acid) are extensively used for biomedical applications due to the polymers biocompatibility, hydrophilicity and functionality.265 In particular PAA has been noted for being able to produce polymers that respond to stimuli such as temperature and pH.266 However, AA and MAA are also noted for easily cross-linking and forming gels during polymerization.267
Beyond controlling the polymerization, RAFT provides a mechanism for producing more advanced polymeric structures such as block copolymers and brushes. These polymeric structures open the possibility of producing supports that are useful in a wide range of hydrophilic and hydrophobic solvents.
The common techniques for modifying the solvent compatibility of a combinatorial chemistry surface involve changing the graft polymer, modifying polymer end-groups or modifying cross-linker agents. However, these techniques all still require the backbone polymer to be significantly solvated by the reaction solution. Brush copolymers offer a method for completely changing this restriction. In a brush copolymer the solvating properties come from the backbone polymer and the linking properties will come entirely from the side chains.
Substrate
Side-chains
Backbone
Figure 7.1: Representation of grafted brush copolymer 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 168
A PDMA-co-PS block copolymer was grafted as a demonstration piece for the possibilities that block copolymer options that RAFT can provide. The styrene block is an excellent anchor point for linkers; however it has the disadvantage of being quite hydrophobic. Thus coupling PDMA to the styrene polymer should provide a greater solvent compatibility range over pure grafted DMA or polystyrene.
In addition to the block copolymer, a novel brush copolymer was produced in the aim of producing an entirely novel SPOS surface. The brush produced was a polystyrene-graft- poly(acrylic acid) and was produced through a novel system using both RAFT and ATRP polymerizations. RAFT was used to graft a copolymer of polystyrene-co-poly(4- vinylbenzyl chloride). The 4-vinylbenzyl chloride is essentially a chloromethylated styrene molecule and can be used as an ATRP initiator. ATRP was then used to graft tert-butyl acrylate (t-BA), which was subsequently hydrolyzed to give a grafted poly(acrylic acid). The resulting amphiphilic graft polymer brush was then tested for its use as a support in SPOS.
Amphiphilic graft copolymers produced in a similar manner using VBC have been previously created with a practical application in personal care by polymerizing various meth(acrylates), methacrylic acid and vinyl-benzylchloride (named as p- chloromethylstyrene in the paper) by conventional free radical means. The VBC groups were then used to initiate ATRP of methacrylic acid, t-BA or HEMA-TMS giving amphiphilic graft polymers.268,269
Thus, the use of RAFT to control the polymerization of hydrophilic monomers and produce novel block and brush copolymers was demonstrated.
7.1. Effects of polymerization rates on RAFT mediated graft polymerization
Using RAFT to control grafting reactions adds some level of complexity to the polymerization kinetics and mechanism. The effects on the reactivity and polymerization rates are worth commenting on in detail, as they have significant mechanistic implications. These factors become more apparent in this system where we 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 169
are using fast polymerizing monomers at room temperature. Several factors that need to be considered when looking at the rates of polymerization with RAFT grafting systems are outlined below.
S S R kβ P S S R k-β Pm S S m R Pm Z k-β Z kβ Z (1) (2) (3)
Scheme 7.1: RAFT transfer mechanism
Firstly, it is necessary for the kp of the monomer to be matched to the reactivity stable of the RAFT agent. If kp >> k-β then the RAFT agent will be ineffective as the rate of radical transfer will be too low to provide a uniform chain length for each polymer chain.
Secondly, the RAFT agent and polymer must be compatible. If the polymer-RAFT agent bond is too strong then there will be very slow fragmentation of the intermediate (2). This will again result in poor transfer between the active polymer chains and broadening of the molecular weight distributions.
Also for grafting polymerization under γ-radiation, the ratio of kp compared with the rate of initiation of radicals on the solid substrate becomes important. In normal solution
RAFT polymerization it is found that the rate of polymerization (Rp) is proportional to the concentration of radicals in the system and the molecular weight is proportional to the RAFT agent concentration, as can be seen from Eq. 7.1 and Eq. 7.2 below 92:
d[M ] • R = − = k [P ][M ] Eq. 7.1 p dt p n
[M ] × x × M monomer M = 0 n Eq. 7.2 n [RAFT ]
monomer where Mn is the expected molecular weight, Mn is the molecular weight of the monomer and x is the conversion of the polymerization at time = t. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 170
This theory states that in solution RAFT polymerization, the majority of the polymer chains will be initiated by a RAFT R-group and the number of active radicals (not number of polymer chains) is dictated by the number of initiator radicals in the system. This means that the molecular weight of the polymer will be a function of RAFT agent concentration, and the rate of polymerization will be a function of the initiator concentration. However, in γ-radiation grafting polymerization there are two simultaneous initiation reactions taking place. The solution polymerization occurs through a conventional RAFT process with all the chains initiated near simultaneously by R-groups. The grafting process however cannot be initiated by R-group radicals and the graft chains are formed from the interaction of γ-radiation with the solid substrate, leading to surface-bound radicals. The solution rate of polymerization will theoretically increase with time because of the effects of the γ-radiation acting as a constant initiation source; however, this can be taken into account. The difficulty arises with surface grafting, where the number of surface chains will be proportional to the time it has been exposed to the γ-source.
The mass of grafted polymer can be defined by Eq. 7.1 where GR is the radiation susceptibility of the substrate, DR is the dose rate and t is the exposure time.219
= × × Eq. 7.3 (#.surface.radicals) GR DR t
This graft polymer will still be controlled as it still participates in the transfer reactions with the solution radicals and RAFT agent. The result of this is that the kp must be slow enough such that enough time is given to allow the formation of surface radicals. If kp>>ki(surface) then the result will be that the conversion of the solution polymer will be reaching completion with little surface grafting.
Finally when polymerizing high kp monomers such as DMA, AA and MAA, the side reactions, cross-linking, gel and Tromsdorff effects become more dominant. Thus, it is frequently advisable to slow these polymerizations to limit these reactions.
As has just been demonstrated, it is quite desirable for the rate of polymerization to be slowed for some instances and a slow fragmenting RAFT agent. While at 60-800C CDB is a efficient RAFT agent, at room temperatures its quite slow fragmenting156 and can 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 171
then be used to achieve these polymerization rate reductions. This slows the polymerization by slowing the kβ fragmentation step which effectively slows Rp by taking radicals temporarily out of the polymerization system.
7.1.1.Kinetics of RAFT controlled grafting
If we say that the mass of grafted polymer (mgraft) is equal to the number of polymer chains multiplied by the molecular weight of the polymer chains then we can derive this equation: (#.graft.chains) × M = n mgraft Eq. 7.4 N A where NA is Avogadro’s number.
Since not all the radicals formed on the surface of the polymer will go on to form polymer chains, the number of grafted chains is given by the amount of radicals produced multiplied by an efficiency factor f as shown:
(#.graft.chains) = f ×(#.surface.radicals) Eq. 7.5
Combining Eq. 7.5 and Eq. 7.4:
= × × × Eq. 7.6 (#.graft.chains) f GR dose.rate t
where GR is the radiation susceptibility of the substrate.
The molecular weight of the polymer can be given by Eq. 7.7;91
[M ] × x× Mn Mn = 0 monomer Eq. 7.7 [RAFT]
We can combine Eq. 7.4, Eq. 7.6 and Eq. 7.7 we get the following: f × G × DR × t ×[M ] × x × Mn m = R 0 monomer Eq. 7.8 graft × N A [RAFT ] This equation can now be used to predict the grafting ratio and graft masses during RAFT mediated γ-initiated polymerization. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 172
7.2. Thermally initiated RAFT polymerization of N,N’-dimethylacrylamide
Prior to investigating the γ-initiated, RAFT-controlled polymerization of DMA, it was necessary to examine the thermally initiated polymerization system. This will show what factors affect the polymerization of DMA and allows a system to be developed that can then be moved to room-temperature γ-initiated polymerization with minimal modification.
60oC AIBN O NC S S N CDB O N
n
Scheme 7.2: Mechanism for thermally initiated polymerization of DMA
7.2.1. Experimental Method
Materials
RAFT agents were synthesized as per the methods described in Chapter 3.
N,N’-dimethylacrylamide (Aldrich, 99%) was deinhibited by passing through a column of activated basic alumina, AIBN was acquired from Aldrich and recrystallised twice from methanol to remove impurities. Analytical grade DMF, benzene, toluene and ethanol were acquired from Aldrich and used as received.
Polymerization procedure
The general procedure for the bench top solution polymerization of DMA was as follows: RAFT agent, AIBN, solvent and monomer were weighed into a Schenck vial. The vial was sealed with a septa and copper wire. The polymerization solution was degassed by sparging with nitrogen for 30 minutes and polymerization was initiated by immersion of the Schenck vial into an oil bath at 600C. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 173
Samples were taken at regular intervals by use of a degassed syringe, and injected into a pre-weighed aluminium pan. Polymers were dried in the fume-cupboard overnight and then for 24hrs in vacuo. Conversion was calculated through gravimetry and then samples were dissolved in DMAc for GPC molecular weight analysis.
7.2.2. Thermal polymerization of DMA using CDB
One of the major advantages of living polymerization is that the molecular weight increases linearly with conversion. This is opposed to conventional free radical polymerization, in which each polymer chain will polymerize as soon as the radical is formed and will continue to polymerize until it is terminated. The result is that for conventional free radical polymerization the molecular weight of the polymer is formed depending on the kp and kt of the polymer only, and is apparently constant throughout the polymerization. This “apparent” constant molecular weight is from subsequent radical formation, rapid molecular weight gain and termination events. This makes control of both the molecular weight and conversion very difficult.
The polymerization was conducted as per the procedure outlined above using 50 wt % ethanol as a solvent and [AIBN] = 4x10-3 M. For the RAFT mediated [CDB] = 8x10-3 M and polymerizations were conducted in an oil bath at 600C for up to 8hrs.
As can be seen from Figure 7.2, the free radical polymerization rapidly produces polymers of a high molecular weight and polydispersity where as the CDB controlled polymerization the molecular weight evolves linearly with increasing conversion. Additionally the polymerization has reached a stable endpoint where polymerization no longer continues after about 1.5hrs (60-70% conversion). 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 174
100
80
60
40 Conversion (%)Conversion
20
0 0123456789 Time (h)
[CDB]=8*10-3 M ; Conventional FR
3.0 2.5 2.0 PDI 1.5 100000 1.0 0 20406080100 90000 80000 70000 60000 50000 Mn 40000 30000 20000 10000 0 0 20406080100 Conversion (%)
Conventional FR; [CDB]=8*10-3
Figure 7.2: Comparison of free radical and RAFT mediated polymerization of DMA in 50 wt % ethanol [AIBN] = 4x10-3 M [CDB] = 8x10-3 M at 600C for up to 8hrs. The straight line in the molecular weight data represents the theoretical molecular weight 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 175
It can be readily seen that the addition of CDB slows the polymerization of DMA to a level where accurate control of polymerization is possible. Control of a conventional polymerization of DMA at this temperature is very difficult as the reaction reaches 100% conversion in less than 2 hours.
Determination for optimum conditions for thermal polymerization of DMA
Before moving on to graft polymerization it was important to determine a set of optimum conditions for the polymerization. The concentration of RAFT agent and monomer has a significant impact on the application of the grafting process since conversion rarely reaches more than 30 or 40% in γ-radiation grafting there is significant waste. Thus, the aim is to minimize the required amounts of the expensive RAFT agents and monomers while still allowing for good controlled polymerization and grafting.
The original literature demonstrating the RAFT mediated polymerization of DMA is from McCormick et al.170,270,271 DMA was polymerized in both benzene using CDB and in water using a novel water soluble RAFT agent. McCormick found the system reached o 96% conversion after 156 hours at 60 C and GPC analysis showed a Mn of 53100 and a
PDI of 1.24 where the theoretical Mn was 38400. This shows that it is possible to polymerize DMA using CDB and that there are possibilities for the use of CDB to control the polymerization of DMA under γ-radiation at room-temperature.
Initially polymerization using the same system as McCormick was trialed by preparing a polymerization using the same method as outlined above. 50 vol% benzene was used as a solvent and the initiator concentration [AIBN] = 4.05x10-3 M, with [CDB] = 8.1x10-3 M. The polymerization was initiated after sparging with nitrogen for 30min by immersion in an oil bath at 600C.
The CDB/benzene system was found to produce well-defined living polymers. However, a 2 hour induction period was noted and the toxicity of benzene is a deterrent to its use. Ethanol was chosen as a solvent because it has been used as a solvent in 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 176
RAFT polymerization previously272 and it is compatible with PDMA and CDB. In fact PDMA is more soluble in ethanol than benzene and improved kinetics would be expected.
In an effort to alleviate the induction period that was observed, an alternative RAFT agent was used (using the same procedure as CDB polymerization in benzene). 1-PEDB is noted to be less susceptible to retardation and inhibition than CDB and was found to produce well defined and controlled polymers. However, the polymerization still showed a 2 hour induction period. Thus it was concluded that the induction period was caused by either oxygen ingress or RAFT agent impurity rather than from the properties of CDB as a RAFT agent. Additionally an induction period is not overly of concern because the γ-radiation induced polymerization would be expected to show strong inhibition and retardation due to the nature of using CDB at room temperature.
Thus the system that was chosen to investigate was a CDB ethanol system.
Effect of RAFT agent concentration of RAFT polymerization of DMA
The concentrations of RAFT agent and monomer concentration also have a significant effect on the produced polymers and rates of polymerization. The simplest way to adjust the molecular weight of the polymer is to adjust the RAFT agent concentration. The kinetic equations derived in Chapter 2 demonstrate that the rate of polymerization (Rp) is a function of the initiator concentration and the monomer concentration. Likewise the molecular weight of the final polymer is a function of the initial monomer concentration, percent conversion, and RAFT agent concentration.
fk [I] = d Eq. 2.15 Rp k p[M ] kt
[M ] .x.Mn Mn = 0 monomer Eq. 2.28 [RAFT]
A series of polymerization runs were conducted to determine the effect of RAFT agent concentration on the thermally initiated polymerization of DMA. The polymerization 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 177
setup was the same as outlined above with the solvent being 50 vol% ethanol and the -3 concentration of [AIBN]0 = 8 x10 M and the concentration of CDB was variously 16 -3 -3 -3 x10 M, 8x10 M and 4x10 M.
Figure 7.3 shows the effect of the CDB concentration on the thermal polymerization of DMA. As expected the molecular weight decreases with increasing CDB concentration as the increased CDB concentration effectively allows for more polymer chains to be initiated. The resulting polymers adhered to a linear first-order kinetic plot indicating that the concentration of radicals in the system is constant and this is in turn an indication of the living behavior of the system. The 16x10-3 M and 8x10-3 M concentration of CDB produced well-controlled polymers with a polydispersity of around 1.1 – 1.15. The lowest concentration of CDB still produced controlled polymer, however, the polydispersity was slightly higher and increased slightly towards the end of the polymerization due to bimolecular termination reactions.
2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 ln(1/1-x) 0.8 0.6 0.4 0.2 0.0 0123456789 Time (h)
[CDB] 16*10-3 M; 8*10-3 M 4*10-3 M
Figure 7.3: Effect of RAFT agent concentration on the thermally initiated polymerization of DMA 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 178
In Chapter 2, it was shown that the rate of polymerization for RAFT mediated polymerization can be represented by the following equation:
fk [I] = d Rp k p [M ] kt
This stipulates that the rate of polymerization is a function of the initiator concentration and efficiency and should be independent of the RAFT agent concentration. However, as can be seen in Figure 7.3 the rate of polymerization changes significantly with the changes in RAFT agent concentration. Analysis of the data shows that the rate of polymerization decreases linearly with an increase in the RAFT agent concentration. This effect can be explained by a slow fragmentation of the RAFT intermediate. This would significantly slow the rate of polymerization, at a rate dependent on the concentration of RAFT agent.
Overall it seems that 8x10-3 M CDB concentration is a good concentration, balancing enough RAFT agent to achieve control with a reduced RAFT concentration minimizing costs and reagent consumption.
Effect of monomer concentration on thermally initiated RAFT polymerization of DMA
By changing the initial monomer concentration the molecular weight of the resulting polymer (for a given conversion) can be adjusted in accordance with the equations derived in Chapter 2. Decreasing the monomer concentration will also slow the rate of polymerization. Additionally, decreasing the concentration of monomer will decrease the chance of monomer-based side reactions such as transfer to monomer and cross- linking reactions.267
The polymerization was conducted using the method outlined above and varying the ethanol solvent content from 0 vol%, 25 vol%, 50 vol% and 75 vol% at 60oC. The RAFT agent was CDB and its concentration was 8x10-3 M, while the initiator -3 concentration [AIBN]0 = 4.05 x10 M. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 179
Figure 7.4 shows the effect of monomer concentration on the thermally initiated polymerization of DMA. As can be seen the polymerization still coincides with the theoretical molecular weight evolution. The resulting polymers are well defined with polydispersities between 1.1 and 1.3. The polymerizations again follow a linear first order kinetic plot and all the monomer concentrations have a similar rate of polymerization. The one exception is that the bulk DMA polymerization shows a significantly lower rate of polymerization to the polymerizations that contain a solvent. This can be explained by a solvent interaction with the RAFT agent that could destabilize the RAFT intermediate and led to an accelerated polymerization compared to the bulk polymerization. It can also be seen from the graph that Rp is clearly proportional to the concentration of monomer, as can be expected by examining the above equation.
1.4 1.3 1.2 PDI 1.1 100000 1.0 0204060 90000 80000 70000 60000 50000 Mn 40000 30000 20000 10000 0 0204060 Conversion (%)
DMA:Ethanol ratio Bulk; 7.5:2.5; 5:5 ; 2.5:7.5
Figure 7.4: Effect of monomer concentration on the thermally initiated polymerization of DMA (solid lines represent theoretical molecular weight)
The optimum monomer content was found to be 50 vol% monomer as the bulk polymerization and very low monomer concentrations showed either a broadening of the molecular weight or slow polymerization rates. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 180
Effect of temperature on the RAFT mediated polymerization of DMA
The temperature dependence of the polymerization of DMA was also investigated. This is very important as it demonstrates that the polymerization system can be applied to lower temperature initiation sources, such as γ-radiation induced polymerization.273
The polymerization setup was the same as described at the beginning of this section with 50 vol% ethanol used as a solvent and [CDB] = 8x10-3 M and [AIBN] = 8x10-3 M. The solutions were sealed into vials, degassed and initiated by immersion into oil baths at various temperatures (room temperature, 50oC, 60oC and 70oC)
The experiments showed that as temperature decreased the rate of polymerization slowed significantly due to reduced rates of initiation and polymerization because of reductions in kd and kp. However, even at low temperature the polymer produced was still controlled and of low polydispersity indicating that the CDB controlled RAFT process will probably control the polymerization at room temperature.
It was found that 60oC was the optimum temperature for the thermal polymerization of DMA. At higher temperature, the polydispersity index tended to increase at higher conversion and the polymerization started to deviate from a first-order kinetic. This is possibly caused by the increase in radical concentration because of increased kd, or perhaps degradation of the RAFT end groups.274 It was found that at 50oC polymerization proceeded at an exceedingly slow rate and at room temperature and no noticeable polymer was produced. While the polymerization at 50oC was very slow (2.5% conversion after 8 hours), it was still controlled and produced polymers with polydispersity between 1.1 and 1.2. This indicates that CDB is still active as a RAFT agent at lower temperatures, and that the slowed polymerization is due to significantly decreased rate of initiation from the AIBN and from the reduced kp of DMA at this temperature. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 181
3.0 2.8 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 ln(1/1-x) 1.0 0.8 0.6 0.4 0.2 0.0 0123456789 Time (h) Polymerization temperature Room temp 50oC 60oC 70oC
1.4 1.3 1.2 pdi 1.1 100000 1.0 0 20406080100 90000 80000 70000 60000 50000 Mn 40000 30000 20000 10000 0 0 20406080100 Conversion (%) Polymerization temperature Room temp; 50oC 60oC 70oC
Figure 7.5: Temperature dependence of thermally initiated, RAFT mediated polymerization of DMA in 50 vol% ethanol [CDB]=8x10-3 M and [AIBN]=8x10-3 M 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 182
In conclusion it was found that CDB effectively controls the polymerization of DMA at elevated temperatures with optimum conditions found to be 50 vol% ethanol as a solvent and [CDB] = 8 x10-3 M, [AIBN] = 4 x10-3 M. The temperature dependence experiments suggested that this system may be applied to room temperature polymerization without significant modification. This comprised the next step of the investigation. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 183
DMA grafting onto SynPhase lanterns using γ-radiation
gamma radiation room temp. S S DMA O CDB N
Figure 7.6: Mechanism of γ-initiated grafting of DMA controlled through the use of CDB
From the work on thermally initiated solution polymerization, the viability of CDB- controlled polymerization of DMA was established and a CDB/ethanol system was found to work effectively. Thus this system was used to control the γ-initiated grafting polymerization of DMA onto SynPhase Lanterns.
7.2.3. Experimental Method
All reagents were prepared as for the solution polymerization of DMA.
SynPhase Lanterns made of polypropylene were provided by Mimotopes Pty. Ltd. and used as polymeric surfaces for the experiments. Each Lantern has a surface area of approx. 3.6 cm2 and a weight of approx. 0.07 g. The SynPhase lanterns were prepared by washing for 24hrs in dichloromethane to remove contaminants, dried in vacuo overnight prior to use, and weighed prior to grafting.
RAFT agent, monomer and solvent were combined and stirred for 5 minutes to dissolve the RAFT agent. Two SynPhase lanterns were weighed and placed in a glass vial. 5ml of monomer solution per lantern was added to the vials, and they were then sealed with septa and copper wire. The solution was then degassed by sparging with nitrogen for 30 minutes. The vials were then placed in the γ-source for the required time. Once the polymerization was complete the solution was dried first in the fume cupboard and then for 24 hours in vacuo. The resulting solution polymer was weighed to give the conversion, and samples were dissolved in DMAc for GPC molecular weight analysis. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 184
To remove ungrafted homopolymer, the lanterns were rinsed in DMF and then washed for 3 weeks in DMF with stirring and changing the wash solution every two days. The washing was deemed complete when no visible polymer would precipitate out of the wash solution. Gravimetry was performed to determine the graft and polymer weight increases.
7.2.4. Results and discussions
Polymerizations were conducted as per the above described procedure using 50 vol% -3 ethanol as a solvent and [CDB] = 8x10 M. The polymerization was conducted over 92 hours and exposed at a dose rate of 0.056 kGy h-1.
DMA was successfully grafted to the SynPhase lanterns, as determined by gravimetry. The addition of CDB slowed the polymerization to a level where control could be gained over the polymerization. Without the addition of CDB the free radical polymerization reached completion and formed a gel that couldn’t be analyzed within 8 hours.
Analysis of the solution polymer shows that the molecular weight increases linearly with time and that there was excellent control of the polymerization, producing polymers with a PDI around 1.10. The analysis of the solution polymers can be seen in Figure 7.7.
The graft ratio of the polymer was also found to increase linearly with time, indicating that the grafting reactions parallel the solution polymerization. Thus the assumption that the solution polymer can provide an indication of what is happening on the surface is valid. Figure 7.8 shows the linear increase of the graft ratio with respect to time.
It should be noted that, in Figure 7.8, there is a 24 hour induction period before polymerization starts. This is probably caused by the slow formation of polymerization species from CDB under γ-radiation at room temperatures. This is expected as CDB tends to have long induction periods and large retardation at lower temperatures.157 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 185
1.4 1.3 1.2 PDI 1.1 6000 1.0 02468 5000
4000
3000 Mn 2000
1000
0 012345678 Conversion (%)
Figure 7.7: Radiation induced grafting of DMA to SynPhase lanterns 50 vol% ethanol and [CDB] = 8x10-3 M. Solid line represents theoretical molecular weight
20
15
10 Graft ratio (wt%)
5
0 0 20406080100 Time (h)
Figure 7.8: Evolution of graft ratio for γ-radiation induced grafting of DMA onto SynPhase lanterns50 vol% ethanol and [CDB] = 8x10-3 M 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 186
Now that a working polymerization system has been established the various effects on that polymerization need to be analyzed. The graft polymerization should be effected by the radiation dose rate, temperature, concentration of RAFT agent as well as the solvent type and percentage. Since our γ-source is not equipped with heating facilities we looked at the effect of these other factors on the graft polymerization.
Effect of RAFT agent concentration on γ-radiation grafting of DMA
The RAFT polymerization theory predicts that the rate of polymerization will be a function of the radical concentration or dose rate and that the molecular weight will be a function of the concentration of RAFT agent as given by Eq. 2.26 from Chapter 2.
[M ] ×conversion× Mn Mn = 0 monomer Eq. 2.26 theoretical [RAFT]
Thus, in theory, the rate of polymerization should be constant with respect to RAFT agent concentration. This theory is based on the assumption that the R-group of the RAFT agent is released and initiates a polymer chain. Therefore the number of polymer chains in solution is equal to the number of R-groups released and thus related to the concentration of RAFT agent, assuming that the concentration of RAFT agent is significantly greater than the concentration of initiator. It is obvious then that increasing the concentration of RAFT agent increases the number of polymer chains in solution and thus decreases the molecular weight proportionately for a given conversion.
While this is the case for solution polymerization it is not true for the grafted polymer. The grafted polymer can only be initiated by radicals produced on the surface by γ- radiation and not from RAFT R-groups. As such the number of graft polymers is dependent only on the exposure time and dose rate of the radiation and is then a linear function of time. Unlike the solution polymerization, the R-groups cannot initiate a surface-bound polymer (if “transfer to solid” reactions are ignored). However, the grafted molecular weight will still be affected by the RAFT agent concentration in the same way as the solution polymerization. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 187
The slow fragmenting RAFT agent is being used to essentially reduce the rate of polymerization such that adequate control of the polymerization is possible and it was found that the rate of polymerization was significantly slower than the uncontrolled free radical polymerization. This effect makes the rate of polymerization dependent on the RAFT agent concentration as increasing the concentration of RAFT agent drives the reaction back towards the dormant unfragmented state.
There are two effects involved with increasing the concentration of RAFT agent. The first is the reduction in the rate of polymerization due to the slowed fragmentation of RAFT species and the second is the reduction in the molecular weight of the solution polymer due to the initiation of the R-groups from the RAFT agent. To test this, a polymerization was conducted as described at the beginning of this section using 50 vol% ethanol as a solvent, 0.056 kGy h-1 dose rate and varying the concentration of CDB between 16.2x10-3 M, 8x10-3 M, 4.02x10-3 M. The polymerizations were conducted over a 72 hour period with regular sampling.
The molecular weight dependence on the RAFT agent concentration is shown in Figure 7.9 and from this it can be seen that the molecular weight decreases with RAFT agent concentration and well defined polymers are produced for RAFT agent concentrations of 16x10-3 M and 8x10-3 M concentrations. The lower 4x10-3 M concentration showed some broadening of the molecular weight distributions at higher conversion, which was probably caused by the RAFT concentration being too low to adequately control the polymerization. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 188
1.4
1.2 pdi 10000 1.0 0.0 2.5 5.0 7.5 10.0
8000
6000
Mn 4000
2000
0 0.0 2.5 5.0 7.5 10.0 Conversion (%)
[CDB] 16*10-3M; 8*10-3 M 4*10-3M
Figure 7.9: Effect of changing CDB concentration on γ-induced polymerization and grafting of DMA onto SynPhase lanterns with 50 vol% ethanol as a solvent, 0.056 kGy h-1 dose rate and varying the concentration of CDB between 16.2x10-3 M, 8x10-3 M, 4.02x10-3 M. Solid lines represent theoretical molecular weights (higher conversion data points for 4x10-3 M concentration of CDB were omitted for clarity)
Looking at Figure 7.10 it can be seen that there is a clear decrease in the amount of polymer grafted to the lantern as the concentration of CDB was increased. This can be explained by both the reduction in polymerization rate due to the slow-fragmenting CDB species and by the decrease in molecular weight of the grafted polymer with increased RAFT agent concentration (for a fixed number of grafted polymer chains). f × G × DR × t ×[M ] × x × Mn m = R 0 monomer Eq. 7.8 graft × N A [RAFT ] Looking at Eq. 7.8 that was derived at the beginning of the chapter it can be seen that if the graft polymerization is following a RAFT mechanism then we would expect that the graft mass or the graft ratio would be linear with respect to the inverse of the raft agent concentration. This was exactly what was seen in Figure 7.10. This implies that the 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 189
polymerization follows the derived RAFT polymerization kinetic equation, and thus can be taken as evidence that the grafting is RAFT controlled.
180
160
140
120
100
80
60 Graft ratio (wt/%)
40
20
0 0 250 500 750 1000 1250 1/[CDB]
Figure 7.10: γ-radiation initiated grafting of DMA onto SynPhase lanterns. Effect of [CDB] on grafting ratio
Effect of dose rate on radiation grafting of DMA onto SynPhase Lanterns
The radiation dose rate in γ-initiated polymerization is essentially a constant radical source and is equivalent to the initiator concentration in a normal thermally initiated polymerization. The polymerization was conducted using the above procedure outlined with 50 vol% ethanol used as the solvent, [CDB] =8x10-3 M. Dose rate was varied between 0.105, 0.056, 0.035, 0.025 and 0.019 kGy h-1. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 190
9
8
7
6
5
4
3 Graft ratio (wt%)
2
1
0 0.000 0.025 0.050 0.075 0.100 Dose Rate (kGy/h)
Figure 7.11: γ-radiation initiated RAFT grafting of DMA; Effect of dose rate with 50 vol% ethanol used as the solvent, [CDB] =8x10-3 M. Dose rate was varied between 0.105, 0.056, 0.035, 0.025 and 0.019 kGy h-1
Figure 7.11 shows a linear increase of graft ration with dose rate which is as expected, as a higher dose rate would generate a higher concentration of radicals. This would in turn lead to an increase in the rate of polymerization, because the rate of monomer consumption is dependent on both the concentration of monomer and radical species. Experiments were undertaken to determine the order of reaction with respect to dose rate. Assuming the general formula:
d[M ] − = R = k[M ]a [DR]b dt p where DR is the dose rate and –d[M]/dt is the rate of polymerization, k is a pseudo-first- order rate constant and [M] is the monomer concentration, a is the order with respect to monomer concentration, and b is the order with respect to dose rate. Taking the logarithm of both sides of the equation yields: 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 191
⎛ d[M ]⎞ log(R ) = log⎜ ⎟ = logk + a log[M ]+ blog[DR] P ⎝ dt ⎠
By only taking values for the rate of polymerization at low conversion the first two terms can be considered constant. Therefore a plot of log(d[M]/dt) vs. log(dose rate) should give a straight line with a slope corresponding to b. The rate of polymerization
Rp was determined by finding the rate of change of monomer concentration over time, and the results can be seen in Figure 7.12.
-5.0
-5.2
-5.4
-5.6 ) P -5.8
log (R log -6.0
-6.2
-6.4 y=0.6529x-3.9845 R2=0.9672 -6.6
-4.2 -4.0 -3.8 -3.6 -3.4 -3.2 -3.0 -2.8 -2.6 -2.4 -2.2 log (dose rate)
Figure 7.12: Determination of reaction order in radiation grafting of DMA with respect to dose rate
The reaction order was found to be 0.65 ± 0.08 with respect to dose rate. Quinn et al.156 reported an order of 0.5 for the RAFT mediated polymerization of styrene under a constant source of γ-radiation. Chapiro et al.275 also reported a similar value for their study of free radical polymerization of styrene using radiation initiation.
7.2.5. ATR-FTIR analysis of PDMA grafted surfaces
ATR-FTIR spectra were acquired to analyze the grafted surfaces and the tethered polymers. FTIR allows us to look at the various chemical functional groups on the 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 192
surface and this in turn allows us to see that the polymers are actually grafted to the surface rather than swollen inside the lantern.
The full FTIR spectra can be seen in Figure 7.13, and it shows the chemical changes in the material as the graft ratio of PDMA increases.
1.0 998 973 3450 2951 2917 2868 2837 1621 1500 1454 1404 1376 1256 1149 1099 1057
0.8
0.6
0.4
Absorbance 0.2
0.0
-0.2
4000 3500 3000 2500 2000 1500 1000 Frequency (cm-1)
Figure 7.13: ATR-FTIR spectra of DMA grafted lantern increasing in PDMA graft mass compared to ungrafted PP lanterns (ungrafted PP lantern is shown offset)
The large broad peak at ~3500 cm-1 corresponds to a region that the only absorbances known are caused by strong intermolecular hydrogen bonding from O-H or N-H groups in alcohols, primary or secondary amine/amide compounds. This peak intensifies with increased PDMA graft ratio and thus probably caused by absorption of water onto the surface by the hydrophilic PDMA polymer, explaining why it intensifies with increasing graft ratio. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 193
Peaks between 3000 and 2800 cm-1 are caused by miscellaneous alkyl C-H stretching -1 -1 -1 -1 bonds: 2951 cm νas CH3, 2917 cm νas CH2, 2868 cm νs CH3, 2837 cm νs CH2. The overbearing intensity in this area comes from the PP substrate.
-1 -1 The peaks 1454 cm δas CH2 and 1376 cm δas CH3 are from CH symmetric and asymmetric bending in the polymeric backbone and on the PP substrate.
The other important absorption spectrum is the carbonyl C=O stretching absorption that can be seen at 1621cm-1. This is universally known as the Amide I band. It can be seen that the intensity of this band increases with the concentration of the PDMA on the surface as the graft ratio increases.
0.20 0.4
0.15 Increase Increase [PDMA] 0.10 [PDMA] 0.2 Absorbance 0.05 Absorbance
0.00 0.0
-0.05 3500 3000 1700 1650 1600 1550 Frequency (cm-1) Frequency (cm-1)
Figure 7.14: Key absorption peaks for ATR-FTIR analysis of grafted PDMA
As can be seen in Figure 7.14, the PP substrate is being gradually covered by the PDMA chains until such a point that no more PP appears in the FTIR spectra.
7.2.6. GPC analysis of homopolymers from grafting polymerization reactions
Analysis of the molecular weight distribution can provide large amounts of information beyond the actual numerical molecular weight values. A sample of the molecular weight distributions is shown below and discussed in relation to mechanistic effects. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 194
D CB A 1.0
Increasing 0.8 [RAFT]
0.6
0.4
Normalized Response 0.2
0.0
3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 log(M ) w
Figure 7.15: Molecular weight distributions showing effect of [CDB] on RAFT controlled, γ-initiated polymerization of DMA (including replicates). A: [CDB] = 8x10-4 M, B: [CDB] = 2x10-3 M, C: [CDB] = 4x10-3 M, D: [CDB] = 8x10-3 M
Several features of the polymerization are demonstrated in these molecular weight distributions. Overall, as the concentration of RAFT agent increases the molecular weight distributions are shifted to a lower molecular weight as predicted by theory. Trace A shows a very low concentration of CDB, where there is an inadequate amount of RAFT agent to control the polymerization, thus there is a large high molecular weight shoulder from bimolecular termination and another shoulder on the low molecular weight side.
Traces B and C thus contain good amounts of CDB and thus show symmetrical low polydispersity molecular weight distributions. Trace B shows a slight high molecular weight shoulder from bimolecular termination and Trace C effectively eliminates this with the increased RAFT agent concentration. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 195
Trace D contains enough RAFT agent to effectively reduce the molecular weight to oligomer levels. Multiple peaks are detected representing both solvent contributions to the GPC and individual contributions from different sized oligomers.
Thus it can be seen that good control of the polymerization can be achieved through the selection of appropriate concentrations of RAFT agent.
7.2.7. Conclusions
Studies of thermally initiated polymerization of DMA, radiation initiated room temperature polymerization and radiation initiated grafting of DMA were conducted. CDB was found to work effectively in controlling the polymerization at both room temperature and elevated temperature while ethanol was found to be a good solvent for both the polymerization system and the resulting PDMA. The thermal initiated polymerization produced polymers with Mn up to 40,000 g/mol and had a PDI between
1.1 and 1.3. The radiation grafted polymers had Mn up to 8,000 g/mol and the resulting PDI was between 1.1 and 1.3. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 196
7.3. RAFT mediated polymerization of acrylic and methacrylic acid
X X = H, or CH3 60oC X AIBN O NC S S HO Anisol O CDB HO
Scheme 7.3: Mechanism for thermally initiated polymerization of acrylic acid and methacrylic acid
Acrylic and methacrylic acid polymers are widely used in the pharmaceutical and biochemical industries because of their water solubility, biocompatibility and functionality.276 One of the difficulties with the polymerization of methacrylic and acrylic acids are that the monomers polymerize at an extremely fast pace and have a tendency to cross-link at high monomer concentrations.267 One of the aims of this work is to polymerize methacrylic and acrylic acids with a reduced risk of cross-linking and to control the polymerization such that advanced structures can be made from these polymers.
In the same way that a slow fragmenting RAFT agent was used for the grafting of PDMA, we find that a slow fragmenting RAFT agent allows for good control over AA and MAA polymerization. The reason for this is conventional free-radical polymerization rates can result in cross-linking and poor control of the polymerization of DMA.
7.3.1. Experimental method
Materials
Methacrylic acid and acrylic acid (Aldrich, 99%) were vacuum distilled with the addition of BHT to inhibit polymerization during the distillation process. RAFT agents were synthesized as described above, and AIBN was acquired from Aldrich and 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 197
recrystallised twice from methanol to remove impurities. Analytical grade DMF, Dimethyl sulfoxide (DMSO), benzene, toluene and ethanol was acquired from Aldrich were used as received. PEPDA was synthesized as described in chapter 3.
7.4. Thermally initiated polymerization of MAA and AA
Initial thermal polymerization experiments were conducted to help find a system that could then be used to control both the thermal polymerization and γ-radiation initiated polymerization and grafting of MAA and AA at room temperature.
Experimental method
The general procedure for the bench top solution polymerization of MAA or AA is as follows. RAFT agent, AIBN, solvent and monomer were weighed into a Schenck vial. The vial was sealed with a septa and copper wire. The polymerization solution was degassed by sparging with nitrogen for 30 minutes. The polymerization was conducted by immersing the samples in an oil bath at the appropriate temperature. Samples were taken at regular intervals by use of a degassed syringe and injected into a pre-weighed aluminium pan. The resulting polymers were dried in the fume-cupboard overnight and then for 24 hours in vacuo. Conversion was calculated through gravimetry and then samples were dissolved in DMAc for molecular weight analysis.
7.4.1. Thermally initiated, RAFT mediated polymerization of MAA
Initial experiments were conducted into the thermally initiated polymerization of MAA. The polymerization was conducted as described above with a 4x10-3 M concentration of AIBN, CPDB was used as the RAFT agent at a concentration of 4x10-3 M, and DMF was used as the solvent at 50 wt %. The initial experiments showed that a linear pseudo- first-order kinetic plot was generated (Figure 7.16), molecular weights between 10,000 and 60,000 were produced, and polydispersity was well controlled, varying between 1.2 and 1.5. The slight deviation from linearity after 7 hours is most likely caused by a gel 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 198
effect in the polymerization as at this stage the polymerization has reached 92% conversion.
3.0 2.8 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 ln(1/1-x) 1.0 0.8 0.6 0.4 0.2 0.0 0123456789 Time (h)
Figure 7.16: Thermally initiated polymerization of MAA controlled by 4x10-3 M CPDB, [AIBN] = 4x10-3 M, and 50 wt %.DMF at 60oC
While the polymerization showed good control and produced good polymers it was found there were significant difficulties in analyzing MAA. Samples prepared for GPC analysis took up to a week to dissolve in dimethylacetamide (DMAc) to allow analysis and polymerization solutions could easily gel and cross-link.
Attempts were made to control this cross-linking by changing the RAFT agent between CPDB and CDB, while comparing it to conventional uncontrolled polymerization. It was found that there was little difference between the two RAFT agents and that they polymerized with similar rates of polymerization and results. For both CPDB and CDB the molecular weights were between 20,000 and 80,000 molecular weight and polydispersity was between 1.3 and 1.5. The uncontrolled polymerization of MAA 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 199
rapidly produced high molecular weight polymers (265,000) of a broad polydispersity index PDI = 2.0 - 3.0.
It was concluded that AA would be a better choice of monomer as it should be more soluble in DMAc and less prone to cross-linking, thus making the polymerization and analysis far simpler.
Selection of RAFT agent for AA polymerization
Conventional free radical AA polymerization was found to reach completion within 30 minutes and formed a solid gel. The produced polymer had a molecular weight of 110,000 and a PDI of 2.73 which is indicative of the side reactions such as cross- linking, transfer to monomer and polymer. This demonstrates the major difficulty in polymerizing AA.
Trials for two different RAFT agents were performed to control the polymerization of acrylic acid: CPDB and the more common CDB. CPDB is an interesting RAFT agent as the leaving group (R-group) is equivalent to an AIBN fragment and thus kinetically the initiation and reaction of AIBN and the fragmentation of the CPDB should be the same.
The polymerizations were conducted using [AIBN] = 4x10-3 M, [RAFT agent] = 4x10-3 M and DMF at 50 wt% as the solvent. The polymerization was initiated by immersion in an oil bath at 600C.
The comparison between CDB and CPDB can be seen in Figure 7.17. As can be seen there is no significant difference between the polymerization rates using CPDB or CDB. This would suggest that the fragmentation rate is dictated by AA rather than the R- group, the Z-group should have a greater effect on the polymerization rate by affecting the stability of the binary intermediate. The pseudo-first-order kinetic plot clearly shows a straight line and the molecular weights are controlled with a PDI that is around 1.1 in the early stages of the polymerization. Unfortunately, as can be seen in Figure 7.17, the polydispersity increases to a value of around 1.45 as the reaction proceeds, which while not optimal, is a significant improvement on the uncontrolled radical polymerization. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 200
0.6
0.5
0.4
0.3 ln(1/1-x) 0.2
0.1
0.0 012345 Time (h) RAFT agent: CDB; CPDB
1.4
1.2 PDI 100000 1.0 02040
80000
60000
Mn 40000
20000
0 02040 Conversion (%) RAFT agent: CDB; CPDB 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 201
Figure 7.17: Effect of RAFT agent type on thermally initiated polymerization of AA with [RAFT agent] = 4x10-3 M [AIBN] = 4x10-3 M, and DMF at 50 wt% as the solvent at 60oC (theoretical molecular weight is shown by unbroken line)
Looking at Figure 7.17 it can be seen that while the molecular weight increases linearly with conversion, as expected in a living process, it does not correlate with the theoretical Mn. The reason that the molecular weight data does not match the theoretical molecular weight is that the experimental molecular weight is determined by GPC analysis vs. polystyrene standards. This means that the molecular weight data is internally consistent but doesn’t give us an absolute molecular weight determination. There is no way of acquiring the true molecular weight without the use of PAA standards or an absolute molecular weight detector such as a light scattering detector or determining Mark-Houwink parameters for PAA in DMAc.
From these experiments we can conclude that there is no significant advantage in using CPDB over CDB. Due to the increased difficulty in producing and purifying CDPB it was decided that CDB would be used for further polymerizations.
7.5. RAFT controlled radiation grafting of AA onto SynPhase lanterns
7.5.1. Experimental method
All reagents were prepared as in the solution polymerization of AA.
SynPhase Lanterns made of polypropylene were provided by Mimotopes Pty. Ltd. and were used as polymeric surfaces for the experiments. Each Lantern has a surface area of approx. 3.6 cm2 and a weight of approx. 0.07 g. The SynPhase lanterns were prepared by washing for 24 hours in dichloromethane to remove contaminants and dried in vacuo overnight prior to use and were weighed prior to grafting.
RAFT agent, monomer and DMF 50 vol% solvent were combined and stirred for 5 minutes to dissolve the RAFT agent. Two SynPhase lanterns were weighed and placed 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 202
in a glass vial; 5ml of monomer solution per lantern was added to the vials and they were then sealed with septa and copper wire. The solution was then degassed by sparging with nitrogen for 30 minutes. The vials were then placed in the γ-source for the required time. Once the polymerization was complete the solution was dried first in the fume cupboard and then for 24hrs in vacuo. The resulting solution polymer was weighed to give the conversion and samples were dissolved in DMAc for GPC molecular weight analysis.
The lanterns were rinsed with DMF and then washed for 3 weeks in DMF while stirring and changing the wash solution every two days. The washing was deemed complete when no visible polymer would precipitate out of the wash solution. Gravimetry was performed to determine the graft and polymer weight increases.
7.5.2. γ-radiation initiated grafting of AA
gamma radiation room temp. S S AA O CDB HO
Scheme 7.4: Mechanism of γ-radiation initiated grafting of AA controlled using RAFT
The RAFT mediated grafting of acrylic acid onto SynPhase lanterns was investigated using CDB. A comparison between conventional free radical and CDB mediated polymerization can be seen in Figure 7.18. The polymerization was conducted with [CDB] = 4x10-3 M and a dose rate of 0.056 kGy h-1.
It was found that without addition of RAFT agent to the polymerization solution, acrylic acid will gel within 48hrs of γ-irradiation at 0.056 kGyh-1. Additionally it was found that the RAFT agent PEPDA, which was specially developed for room temperature polymerization of styrene, caused the polymerization to reach completion within 4 hours, which is a similar polymerization rate to conventional free radical polymerization. This RAFT agent effectively eliminates the retardation in the RAFT polymerization and brings the Rp value close to that of conventional free radical 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 203
polymerization. Unfortunately, at conventional polymerization rates the PAA easily cross-links and the result of this was very quick gelling of the polymer and an inability to control the polymerization.
Thus, CDB was used to control the grafting reactions of AA to SynPhase lanterns. As can be seen from Figure 7.18 the addition of CDB allows for control of the grafting polymerization. Without CDB the polymerization proceeds to completion and gels within 4 hours, while with the addition of CDB the polymerization proceeds at a more manageable pace reacting 50% within 70 hrs. The polymers produced through the CDB mediated polymerization of AA are well controlled and follow a linear progression of conversion with respect to time.
0.8
0.6
0.4 ln(1/1-x)
0.2
0.0 0 20406080100 Time (h)
Conventional FR; RAFT mediated
Figure 7.18: First-order kinetic plot comparing the grafting of AA onto SynPhase lanterns by conventional free radical and RAFT mediated polymerization The polymerization was conducted with [CDB] = 4x10-3 M and a dose rate of 0.056 kGy h-1. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 204
Analysis of the grafted lanterns showed good control of the polymerization and grafting. The graft ratio was proved to increase linearly with conversion and the molecular weight of the solution polymer was also shown to increase linearly with conversion. The molecular weight increased linearly with conversion, but deviated from the expected theoretical molecular weight because the molecular weight from GPC was against polystyrene standards and thus can only be used for internal comparison.
30
25
20
15
Graft ratio (wt%) Graft 10
5
0 0 102030405060708090100 Time (h) Conventional FR; RAFT mediated
Figure 7.19: Graft ratio for AA grafted to lanterns from the polymerization shown in Figure 7.18
It should be noted that the surface bound radicals that produce grafted polymer result only from the interaction between the γ-radiation and the polypropylene substrate. As such the number of growing graft chains will increase with time. Since the uncontrolled grafting of acrylic acid proceeds very fast and reaches full conversion within 4 hours, there is little time for surface radicals to be generated, thus resulting in very little grafted polymer. This can be clearly seen in Figure 7.19 where at completion of the uncontrolled polymerization (within 24 hours) there is less than 0.05 graft ratio increase 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 205
whereas it is possible to reach graft ratios of up to 0.3 or even greater through the use of CDB.
GPC analysis was conducted and the molecular weight distributions were further analyzed to give greater insight into the polymerizations.
increasing conversion Normalised Response Normalised
3.5 4.0 4.5 5.0 5.5 log(M ) w
Figure 7.20: Molecular weight distributions showing evolution of molecular weight distribution from CDB controlled graft polymerization of AA in DMF
From these molecular weight distributions it can be seen that the molecular weight increases over time. The final two Molecular weight distributions show a high molecular weight shoulder caused by bimolecular coupling at the later stages of the polymerization, when the monomer concentration is decreasing and thus termination reactions become more apparent.
The other interesting feature is the low molecular weight tailing that exists on all the traces. This is an effect of γ-radiation as a constant source of initiation. For conventional RAFT polymerization it is assumed that all the radicals are generated in the very early stages of polymerization and thus all the polymer chains start at the same time (usually 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 206
assumed to start from RAFT agent R-groups). However, in γ-radiation initiated polymerization the radiation source is constantly producing radicals and thus while a large percentage of the polymer chains are initiated from RAFT agent R-groups in the initial stages of the polymerization; there is some degree of polymer chains being initiated as the polymerization advances. The result is that there is constant addition of new polymer chains and that a low molecular weight tail is produced. The degree of tailing will be dependent on the length of exposure to the radiation source and should be relatively small, depending on the ratio of polymer chains initiated from RAFT agent R- groups and the number of polymer chains initiated from radicals derived from the γ- radiation.
Effect of RAFT agent concentration on γ-radiation grafting of acrylic acid
As mentioned earlier the addition of CDB to the system allows for both the control of the molecular weight and the slowing of the polymerization to a useful rate. Two experiments were conducted, the first varying the CDB concentration between 16.2x10-3 M, 8.1 x10-3 M, 4.05 x10-3 M and 8x10-4 M and with a dose rate of 0.056 kGyh-1 for a fixed time of 95 hours. A second polymerization was also undertaken using CDB concentrations of 16.2 x10-3 M, 8.1 x10-3 M, 4.05 x10-3 M with regular removal of sample vials over time to determine the time dependence with respect to RAFT agent concentration.
The results shown in Figure 7.21 clearly show the effect of using CDB to control the molecular weight of the PAA polymerization. The molecular weight evolution is linear with respect to conversion, indicating that there is little termination and the radical concentration is constant and the PDI is well controlled at a value of around 1.2 until the later stages of the polymerization where bimolecular termination becomes more predominant. It can also be seen that the addition of CDB is significantly slowing the polymerization. For a normal RAFT polymerization we expect the rate of polymerization to be roughly the same as that of a conventional polymerization. However, in this case, the slow fragmenting RAFT agent slows the polymerization significantly. The result of this is that the rate of polymerization is dependent on the RAFT agent concentration. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 207
This slowing of the polymerization effectively increases the graft ratio that can be achieved through radiation grafting. Since the surface bound radicals can only be produced through the interaction of radiation with the substrate, the number of grafted chains depends on the time that it is exposed to the γ-radiation. This means that the longer the amount of time the lanterns are exposed to the γ-radiation, the greater the number of grafted polymer chains. If the polymerization is polymerized at the same rate as conventional free radical acrylic acid polymerization then the polymerization would reach completion within 4 hours and the amount of grafted polymer would be significantly lower than for a longer polymerization. It is thus advantageous to slow the rate of polymerization to increase the number of grafted chains. The use of CDB at room temperature results in a significant retardation and slowing of the rate of polymerization of the acrylic acid allowing significant amounts of acrylic acid to be grafted. Despite this retardation, the polymer produced is still controlled. Also, the solution polymer still shows controlled characteristics with a low polydispersity and linearly increasing molecular weight with respect to conversion.
The equation that was derived at the beginning of this chapter to describe radiation polymerization can be used to interpret these results in greater detail.
f × G × DR × t ×[M ] × x × Mn m = R 0 monomer Eq. 7.8 graft × N A [RAFT ]
This equation shows that there should be a linear relationship between the graft ratio and the inverse of the CDB concentration. This linear dependence of the grafting ratio can be seen in Figure 7.22. These results show the graft ratio after 95 hours of exposure in the γ-source. Since the lanterns have been in the γ-source for a fixed period of time the number of grafted chains should be the same and this graph represents the reduction in the chain length of these grafted chains. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 208
4.0
3.5
3.0
2.5
2.0
ln(1/1-x) 1.5
1.0
0.5
0.0 04080120160 Time (h) [CDB] = 4.05x10-3 M ; [CDB] = 8.1x10-3 M ; [CDB] = 16.2x10-3 M
2.0
1.5 PDI 200000 1.0 0 20406080100
150000
100000 Mn
50000
0 0 20406080100 Conversion (%) [CDB] = 4.05x10-3 M ; [CDB] = 8.1x10-3 M ; [CDB] = 16.2x10-3 M
Figure 7.21: Effect of [CDB] on γ-radiation induced grafting of AA onto SynPhase lantern. [CDB] = 16.2x10-3 M, 8.1 x10-3 M, 4.05 x10-3 M and 8x10-4 M and DR = 0.056 kGyh-1 for 95 hours. In molecular weight data the solid lines represent theoretical molecular weight 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 209
220
200
180
160
140
120
100
80
Graft ratioGraft (wt%) 60
40
20
0 0 100 200 300 400 500 1/[CDB]
Figure 7.22: Effect of [CDB] on γ-radiation induced grafting of AA onto SynPhase lanterns after 95 hours. [CDB] = 16.2x10-3 M, 8.1 x10-3 M, 4.05 x10-3 M and 8x10-4 M and DR = 0.056 kGyh-1.
The fact that there is a linear relationship between the graft ratio and inverse RAFT agent concentration that the molecular weight of the grafted polymer is controlled and close to that of the solution polymer. This relationship is also demonstrated in Figure 7.23 which shows a linear relationship between the solution molecular weight and the graft ratio. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 210
180000
160000
140000
120000
100000 (g/mol)
n 80000 M 60000
40000
20000
0 0 20 40 60 80 100 120 140 160 180 200 220 Graft Ratio
Figure 7.23: Comparison of molecular weight of solution polymer and graft ratio of the grafted PAA lanterns
The effect that the concentration of RAFT agent has on the polymer molecular weight distribution can be seen in Figure 7.24.
Several features of the polymerization are demonstrated in these molecular weight distributions. Overall as you increase the concentration of RAFT agent the molecular weight distributions are shifted to a lower molecular weight as predicted by theory. Trace A shows a very low concentration of CDB which does not provide enough RAFT-agent to adequately control the polymerization; thus there is a large high molecular weight shoulder from bimolecular termination and another shoulder on the low molecular weight side. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 211
D CB A 1.0 Increase
[CDB]
0.8
0.6
0.4 Response
0.2
0.0
3.5 4.0 4.5 5.0 5.5 6.0 6.5 log (Mw)
Figure 7.24: Molecular weight distributions showing effect of [CDB] on RAFT controlled, γ-initiated polymerization of AA (including replicates). A: [CDB] = 2x10-3 M, B: [CDB] = 4x10-3 M, C: [CDB] = 8x10-3 M, D: [CDB] = 1.6x10-2 M
Traces B and C contain good amounts of CDB and thus show good low polydispersity molecular weight distributions. Both traces show a slight high molecular weight shoulder from bimolecular termination with Trace C reducing this by the increased RAFT agent concentration.
Trace D contains more CDB and has effectively eliminated the bimolecular termination shoulder that was apparent to both Traces B and C.
Thus it can be seen that good control of the polymerization can be achieved through the selection of appropriate concentrations of RAFT agent. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 212
Effect of γ-radiation dose rate on the grafting of acrylic acid
As with the grafting of DMA to lanterns the dose rate should effectively reduce the rate of polymerization by reducing the radical concentration. In this experiment the same setup was used, with 50 vol% DMF as a solvent, RAFT agent concentration of 8x10-3M and the dose rate was varied between 0.105, 0.056, 0.035, 0.025 and 0.019 kGy h-1. The polymers were placed in the γ-source for 95 hrs and removed and analyzed as described at the beginning of this section.
As can be seen from Figure 7.25 there is a linear dependence of the graft ratio on the dose rate of the radiation it is exposed to. The molecular weight evolution is linear with respect to conversion and the polymers are well controlled with PDI of between 1.05 and 1.3. This can be seen in Figure 7.25. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 213
140
120
100
80
60 Graft ratio (wt%) 40
20
0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 Dose rate (kGy/h)
1.4
1.2 PDI 100000 1.0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
80000
60000
Mn 40000
20000
0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 Dose rate (kGy/h)
Figure 7.25: Effect of radiation dose rate on the CDB mediated grafting of acrylic acid in 50 vol% DMF as a solvent. [CDB] = 8x10-3M and the dose rate was varied between 0.105, 0.056, 0.035, 0.025 and 0.019 kGy h-1. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 214
The order of the polymerization with respect to dose rate was determined from the equation below derived by Quinn et al.156
⎛ d[M ]⎞ log⎜ ⎟ = log k + a log[M ] + b log[DR] ⎝ dt ⎠
By plotting log(Rp) vs. log(dose rate) in the same manner as with grafting PDMA. This should give a straight line with a slope corresponding to b. The results indicate an order of 0.73 compared to 0.65 for PDMA.
-2.6 -2.8 -3.0 -3.2 -3.4 -3.6 ) P -3.8
log (R -4.0 -4.2 -4.4 -4.6 y=0.730x-1.42279 R2=0.95336 -4.8 -5.0 -4.4 -4.2 -4.0 -3.8 -3.6 -3.4 -3.2 -3.0 -2.8 -2.6 -2.4 -2.2 -2.0 log (dose rate)
Figure 7.26: Determination of reaction order with respect to dose rate for CDB polymerization of AA.
7.6. ATR-FTIR analysis of grafted PAA polymers
To evaluate the grafting of PAA to polypropylene substrates FTIR spectra were obtained by using a Brunker FTIR using an ATR attachment and the method outlined in chapter 3. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 215
1.2
1.0 A
0.8 B C 0.6
0.4 D E Absorbance 0.2 F
PAA grafted 0.0
-0.2 Blank PP
4000 3500 3000 2500 2000 1500 1000 Frequency (cm-1)
Figure 7.27: FTIR spectra of PAA grafted polymers.
The hydroxyl groups in the acrylic acid rarely exist as free hydroxyls unless they are highly solvated in a non-protic solvent197, more likely is that the acid dimerizes through strong hydrogen bonding with another acid group (either on the same chain or on a different polymer chain), as shown below.
O O O O
H H H H
O O O O
Scheme 7.5: Mechanism for intermolecular hydrogen bonding in PAA as seen by FTIR. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 216
The acid dimer (in the region marked ‘A’ in Figure 7.27) displays a very broad, intense O-H stretching absorption across the region of 3300-2500 cm-1, centered near 3000 cm- 1.The free hydroxyl would adsorb around 3250 cm-1, however the dimer adsorbs at a lower frequency because of the resonance effects in this hydrogen bonding. In this spectrum multiple peaks can be discerned deriving from the O-H stretching. The full O- H stretching absorption is a combination of free O-H (3500 cm-1), dimeric O-H stretching (3300-3000 cm-1) and various resonances and overtones of there absorptions (2750-2500 cm-1).
The region marked ‘B’ in Figure 7.27 is caused by alkyl C-H stretching vibrations from both PAA backbone and the PP substrate. The multiple peaks are caused by symmetric -1 -1 and asymmetric stretching of CH2 and CH3 as follows: 2951 cm νas CH3, 2918 cm νas -1 -1 CH2, 2868 cm νs CH3, 2837 cm νs CH2.
The peak in region ‘C’ is from absorption by C=O stretching in the acid species 1702 cm-1. The carboxylic dimer has a centre of symmetry and only the asymmetric C=O stretching adsorbs in the IR range. Hydrogen bonding and resonance weakens the C=O bond, resulting in absorption at a lower frequency than the free acid. This is shown in Figure 7.27 by the asymmetry and tailing of the peak towards the lower frequency range.
The peaks in region ‘D’ are from CH symmetric and asymmetric bending in the -1 polymeric backbone and on the PP substrate and appear at 1452 cm δas CH2 and 1376 -1 cm δas CH3
Region ‘E’ shows two peaks and is caused by C-O stretching and O-H bending. The C- O stretching band at 1237 and 1165 cm-1 appears as a doublet because of the interaction of the C-O stretching and C-O-H in-plane bending. The C-O-H band appears at 1410 -1 cm and is half hidden by the CH2 asymmetric bending of the CH2 group adjacent to the carbonyl.
The final peak in region ‘F’ at 797 cm-1 is a characteristic peak of dimeric carboxylic acids and results from the out-of-plane bending of the bonded O-H. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 217
This shows the various characteristic peaks of the acrylic acid functional groups as they are grafted to the substrate.
7.7. FMOC-β-Alanine loading determination of PAA grafted lanterns
The functionalization a surface using FMOC-β-alanine and the subsequent chemical cleavage is a standard industry method for quantifying the effectiveness of a support as a combinatorial chemical scaffold.
The procedure effectively functionalizes the acrylic acid with a FMOC-β-alanine moiety which is then chemically cleaved and then the amount of FMOC cleaved from the surface is determined using UV-Vis spectroscopy. The exact procedure for this is outlined in Chapter 3.
Figure 7.28 shows the results from these loading tests and as can be seen there is a linear relation between the graft ratio and the loading of the lanterns. Additionally there is also a linear relation between the loading and the molecular weight of the solution polymer produced during the grafting process. This shows that the molecular weight of the grafted polymer can then be used to tailor the lantern to the required loading. It should be noted that the lower 3 points on the above molecular weight data were below the calibration range of the GPC and thus while there is grafted polymer on the surface, accurate molecular weight determinations is impossible.
For comparison, a commercial PDMA-PAA lantern produced by Mimotopes was also tested during the same test run and resulted in a loading of 16.4%, thus indicating that the RAFT controlled lanterns are seen to be a significant improvement on the uncontrolled grafted lanterns on the market. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 218
80000
70000
60000
50000
40000 (g/mol) n 30000 M 20000
10000
0 0 1020304050 50
40
30
20 Graft Ratio (wt%) Ratio Graft 10
0 0 1020304050 Loading(μmol)
Figure 7.28: FMOC-β-alanine loading determination of PAA grafted lanterns grafted using . 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 219
7.8. Living behavior of γ-radiation produced RAFT polymers
It is important to be able to establish that these RAFT polymerizations are actually living polymerization reactions. One of the important properties for establishing living behavior is reinitiation of the RAFT end-group in the polymer chains. Thus, to prove that the polymers produced during radiation grafting experiments are living, a solution homopolymer that was produced during γ-initiated polymerization of MAA was purified and reinitiated with DMA.
The PMMA macro-initiator was a CDB controlled PMAA polymer with a Mn = 13600 and a PDI of 1.24. The macro-initiator was thoroughly dried in vacuo and weighed into a Schenck vial. Then 5g of DMSO was used to dissolve the polymer and 5g of DMA was added with 4.87x10-6 moles of AIBN. The polymerization solution was degassed by sparging with nitrogen for 30 min and was started by heating to 60oC in an oil bath and samples were taken at regular intervals by use of a degassed syringe.
As can be seen from Figure 7.29, the RAFT groups were successfully reinitiated and the polymerization follows the expected living kinetics. The first order kinetic plot is linear with respect to time as expected from a system that is proceeding with a constant radical concentration. The molecular weight of the produced polymer increases linearly with conversion which consequently indicates that there are no termination reactions and that the end-groups remain active after the initial polymerization under γ-radiation. Thus, it has been shown that the RAFT polymers produced under γ-irradiation and controlled by CDB, retail their living ability and can be further reinitiated 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 220
0.50
0.45
0.40
0.35
0.30
0.25
0.20 ln(1/1-x)
0.15
0.10
0.05
0.00 012345 Time (h)
1.5 1.4
1.3 PDI 1.2
125000 0 10203040
100000
75000 Mn 50000
25000
0 0 10203040 Conversion (%)
Figure 7.29: Thermal reinitiation of γ-radiation PMMA polymer The PMMA -6 macro-initiator was CDB PMAA, Mn = 13600 and PDI = 1.24. [AIBN] =4.87x10 .
7.8.1. Reinitiation of RAFT lanterns and creation of grafted block co-polymer substrates
Following from the above proof that the solution RAFT homopolymers are able to be reinitiated, we wish to prove that the RAFT lanterns are also able to be reinitiated. This not only proves the living nature of the radiation grafting reaction but opens up one of 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 221
the major advantages of RAFT polymerization over conventional free radical polymerization, which is the ability to reinitiate polymer chains and create block copolymers. To get the maximum available number of chains available for reinitiation, lanterns were grafted with a high density of short chain-length PDMA. To do this the lanterns were grafted with a significantly higher CDB concentration than previous polymerizations (0.05 M as opposed to 8x10-3M).
The lanterns were produced by the following method: A solution of CDB (2.5x10-3 moles, 0.068g) in 25 ml of DMA and 25ml of DMF solution was made. Glass vials each containing five lanterns and 10ml of the above solution was prepared and sealed with septa. The samples were degassed for 30 minutes by sparging with nitrogen. The lanterns were polymerized for 211 hrs at a dose rate of 0.056 kGy h-1. Once polymerization was complete the lanterns were washed for 2 weeks in DMF, changing the solution every 2 days to remove homopolymer and solution polymers were dried overnight in vacuo. The dried polymer was weighed to give conversion and dissolved in DMAc and the molecular weight was analyzed by GPC.
Analysis of the lantern and solution polymer found that the polymer had a Mn = 44,533, PDI of 1.15 at 4.28% conversion and the lanterns had an average graft ratio of 0.37.
Once the lanterns had been dried they were used as a macro initiator to reinitiate lanterns and graft styrene, effectively producing a grafted DMA-block-PS lantern and proving that the grafted polymer chains were still ‘living’.
Reinitiating off a grafted surface provides some additional challenges over conventional solution RAFT reinitiation. While the grafted polymer has a high concentration of RAFT agent, in actual molar terms, there is inadequate RAFT agent to control the 5-10 ml of monomer that is polymerized both on the surface and in solution. Thus, additional RAFT agent must be added to increase the overall RAFT agent concentration to a normal level.
The method for reinitiating the lanterns was as follows: 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 222
Four of the DMA grafted lanterns were taken and sealed in a glass vial with a solution of CDB (2.04x10-4 moles, 0.055g), AIBN (1.02x10-4 moles, 0.016g) and styrene (0.192 moles, 20g). The vial was degassed by sparging with nitrogen for 1 h and then polymerization was initiated by immersing the vial in an oil bath at 80oC for 24hrs.
The results showed a 31% mass increase of the total lantern mass, which equates to a 109% increase in the grafted polymer or 0.0298g per lantern. Analysis of the solution polymer showed that conversion was 51% after 24hrs and molecular weight analysis showed Mn = 35,121, PDI = 1.18.
The importance of this is significant. This shows that not only can RAFT be used to control grafting polymerization onto polypropylene supports, but it also retains its living end-groups that can be used to graft block copolymers. The ability to produce grafted block copolymers effectively opens a whole multitude of new polymeric systems for combinatorial chemical surfaces. This can allow for the tuning of the polymeric supports to any solvent/reagent system. By altering the ratios and chemical content of the graft polymers this opens a whole world of possibilities including the grafting of novel monomers. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 223
7.9. Novel amphiphilic brush copolymers produced through combined RAFT and ATRP polymerization
Living radical systems apart from their ability to provide excellent control over the molecular weight and polydispersity of a polymer also provide a basis for the production of novel polymers that could provide distinct advantages over standard polymers in the production of polymeric supports. The use of well defined polymeric brushes as supports for SPOS could provide several distinct advantages. The first is that the density of accessible functional groups could be significantly increased from standard polymers through the high density grafts that occur in brushes. Secondly the formation of the rod-like brush structures could increase the rate of reaction and solvation of the surfaces because the polymers will not tend to collapse and entangle when in a poor solvent. Finally brushes will allow us to tailor the hydrophilic environment around the anchor point by varying the hydrophilicity of the backbone and side chain polymers.
Through the use of RAFT and ATRP combined into the one system we have the possibility of producing a novel new grafted brush polymer system that can offer a wide variety of possibilities for producing new supports. The basis of this system is the use of the monomer 4-vinylbenzyl chloride (VBC). VBC is an interesting monomer because the vinyl group polymerizes in a similar manner to styrene and the chloromethyl group can act as an ATRP initiator. Thus we can use a VBC based polymer as the backbone to produce brush copolymers using ATRP. This is a similar concept to Venkatesh et al.277 who used 2(2-bromoisobutyryl-oxy)ethyl methacrylate as the monomer that was used as the initiator group rather than VBC. Importantly Venkatesh did not notice any incompatibility issues between the RAFT and ATRP or poisoning between the two groups. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 224
Cl
4-vinylbenzyl chloride
Figure 7.30: 4-Vinylbenzyl chloride
It is of interest to produce an amphiphilic brush system that theoretically will have a wide ranging solvent compatibility. Since the backbone polymer is hydrophobic polystyrene based then it would be of interest to graft acrylic acid side chains to provide a hydrophilic nature to the grafted surface.
While RAFT works well to polymerize acrylic acid it is not possible to polymerize it directly through ATRP techniques. This is because the highly polar nature of the acid group interferes with the copper complex that controls ATRP polymerization. A common method for making poly(acrylic acid) polymers through means where direct polymerization is not possible is through the hydrolysis of tert-butyl acrylate (t-BA). t- BA can be hydrolyzed through either a base catalyzed or acid catalyzed hydrolysis mechanism, with acid catalyzed being the most effective mechanism.
n TFA n O H2O O HO DCM O O H
Scheme 7.6: Mechanism for acid catalyzed hydrolysis of the tert-butyl group in P(t- BA)
On the assumption that the chloromethyl functionality doesn’t react with the RAFT agent and that more importantly, the presence of sulphur compounds doesn’t poison the ATRP copper catalyst, then it is possible to combine both RAFT and ATRP to make these brushes. The first step will be to produce the backbone polymer of polystyrene-co- VBC which will be controlled by RAFT. This polymer will then be purified and used without further reaction as an ATRP initiator to produce side chains. These t-BA side 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 225
chains were then hydrolyzed using TFA to produce AA side-chains. The whole process is outlined in Scheme 7.7.
RAFT ATRP AIBN S S S S
S S PMDETA Cl CuBr t-BA m n m n
Cl O
O o Cl
Hydrolysis
TFA DCM
S S
m n
O
HO
o Cl Scheme 7.7: Mechanism for polymer brush production
7.9.1. Solution polymerization of tert-butyl acrylate using ATRP
Initially t-BA was homopolymerized to test the viability of polymerizing t-BA using ATRP. The polymerization was conducted as follows:
The polymerization was conducted at 60oC using a 500:1:1:2 ratio of monomer to initiator to copper bromide to ligand. The ligand was PMDETA, initiator 2-EBIB and anisole was used as a solvent (50 vol%). The solution was degassed by sparging with nitrogen for 1hr prior to polymerization. The polymerization was conducted for over a 24hr period with samples taken regularly with a purged syringe.
This polymer was controlled with a linear evolution of molecular weight with respect to time and a PDI of around 1.3. The resulting polymer after 24hrs had a molecular weight of 16,809 at 34% conversion (theoretical Mn = 22,555). This polymer was then purified 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 226
by dissolving in DCM and then percolating through a basic alumina column to remove the copper catalyst and precipitating into methanol.
The resulting polymer was then hydrolyzed using TFA to produce a linear AA polymer. From gravimetric analysis it can be concluded that there was a 92% conversion of t-BA to acrylic acid from the TFA hydrolysis, calculated by the following equation:
moles P(t−BA) × Mn AA conversion(AA) = ×100% Eq. 7.9 mass AA
An NMR spectrum was taken to confirm the conversion of P(t-BA) to PAA and the NMR spectra can be seen in Figure 7.31.
Peak A: 1.73 backbone methene group from P(t-BA) and PAA, Peak B: 1.43 methyne group from polymer backbone, Peak C: 2.19 methane peak from tert-butyl group. Peak D: 12.19 (singlet, broad.1H) from acidic hydrogen in acrylic acid.
Solvents: 7.25 residual CDCl3, 3.14 water in DMSO, 2.48 residual DMSO solvent peak.
The NMR clearly shows the evolution of the strong peak over 12 ppm that is characteristic of hydrogen on a carboxylic acid. This is confirming evidence that the t- BA has been effectively hydrolyzed. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 227
B A H HH 12.192 3.304 2.478 2.185 1.729 1.432
O OH O O
D H3C CH3
CH3 C
Poly(acrylic acid) Poly(tert-butyl acrylate)
C
D A B
Poly(AA)
C A B Poly(t-BA)
16.0 15.0 14.0 13.0 12.0 11.0 10.0 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 ppm (t1t1) 3.304 2.478 2.185 1.729 1.432
Poly(AA) 1.00 0.47 1.49
C A B
Poly(t-BA) 1.00 0.48 10.02
C A B
4.0 3.0 2.0 1.0 ppm (t1)
Figure 7.31: H1-NMR of PAA produced from hydrolysis of solution t-BA homopolymer taken in DMSO 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 228
1.0 Pre-hydrolysis Post-hydrolysis
0.5 Response
0.0
3.0 3.5 4.0 4.5 5.0 5.5 log (Mw)
Figure 7.32: GPC analysis of P(t-BA) polymer and the resulting PAA polymer after hydrolysis
The molecular weight distribution in Figure 7.32 also shows an clear increase in molecular weight as the polymer is hydrolyzed. This is because the hydrophilicity of the polymer is changing as the polymer is hydrolyzed, and thus the PAA will appear as a higher molecular weight (vs. polystyrene standards) than the P(t-BA) polymer. This is because there will be a change in the radius of gyration (Rg) of the polymer as it swells to a varying levels in the GPC solvent as the hydrodynamic behavior changes. The important thing to notice with the molecular weight distributions is that there is not a low molecular weight shoulder on the PAA post-hydrolysis trace that would indicate a degree of unhydrolysed polymer. This is consistent with the NMR and gravimetric results that show a very high degree of hydrolysis.
Finally, FTIR spectra were acquired before and after hydrolysis to look at the changes in the polymer functional groups that occurred through the hydrolysis process. The spectra were acquired using the FTIR-ATR equipment described in Chapter 3. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 229
0.8 D 0.6 B
0.4 A C 0.2 E
Absorbance 0.0 PAA
-0.2
-0.4 t-BA
-0.6 4000 3000 2000 1000 Frequency (cm-1)
Figure 7.33: FTIR-ATR spectra of P(t-BA) pre- and post-hydrolysis
There are a couple of key absorptions on this spectrum that should be noted.
The P (t-BA) spectrum shows absorptions at 2978 cm-1 and 2872 cm-1 from asymmetric and symmetric CH3 stretching vibrations. The characteristic carbonyl peak (Peak B) from C=O stretching appears at 1722 cm-1 and broadens significantly as the P(t-BA) is hydrolyzed. The hydrolyzed acrylic acid peaks tends to form as very broad peaks because of the strong hydrogen bonding between the acidic hydrogen and the carbonyl group in an adjacent acrylic acid. This can be seen by looking at the acidic hydrogen in Region A in the above spectra absorbing across the range of 3500-2500 cm-1 and the broadening of the 1722 cm-1 carbonyl absorption as the acidic group’s hydrogen bond to the carbonyl. Region C is from CH3 asymmetric and symmetric bending in the tert- butyl group and appears at 1451 cm-1 and 1366 cm-1, and as expected there disappear with the elimination of the tert-butyl group during the hydrolysis. The 2 peaks in Region D are from C-O stretching (1252 cm-1) which is shifted slightly downfield during the hydrolysis, and the C-(C=O)-O bending vibration at 1141 cm-1. These 2 peaks blend into a single broad peak as the hydrolysis takes place. The final peak in 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 230
Region E appears at 845 cm-1 and is caused by C-H bending vibrations in the tert-butyl group. This peak disappears as the tert-butyl group is eliminated during hydrolysis.
As can be seen there are several characteristic changes in the spectra confirming the hydrolysis of P(t-BA) to PAA.
7.9.2. Thermally initiated solution amphiphilic brush copolymers
Before grafting brushes it is important to understand the factors affecting the brush formation reactions. The objective here is to first produce the brushes in solution from a thermal initiator to evaluate the polymerization and brush production processes to assess the viability of the process.
Polymerization of solution brush backbone
The target polymer was a 5% VBC-styrene copolymer produced using CDB as the mediating RAFT agent and AIBN as the initiator. The polymerization was conducted as follows:
CDB (0.3228g, 1.19 x 10-3 mol), AIBN (0.0388g, 2.38 x 10-4 mol) were dissolved in a mixture of styrene (18.5g, 0.177 mol) and 4-vinylbenzyl chloride (1.36g, 0.0089 mol). The solution was transferred to a Schenck vial and sealed with rubber septa. The solution was degassed for one hour by sparging with nitrogen gas and then the polymerization was started by immersion of the vial in an 80oC oil bath. After 13 hours the vial was removed and the polymer was transferred to a weighed aluminium pan and allowed to dry overnight. The dried polymer was dissolved in DCM and precipitated into methanol to purify it. The polymer was then dried and precipitation was repeated another two times. The resulting polymer was analyzed by GPC. Yield = 40.5 %, Mn = 6734, PDI = 1.25, theoretical molecular weight was Mnth = 6740. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 231
The number of moles of chloromethyl initiator groups per mole of polymer is found below:
The degree of polymerization DPn (number of monomer units per unit of polymer) is
Mn polymer DPn = Mnmonomer Mn polymer DPn = (styreneratio)Mnstyrene + (1− styreneratio)MnVBC 6734 DPn = 0.95×104.1+ 0.05×152.6 DPn = 62.99
Thus the number of moles of chloromethyl groups per unit of polymer is equal to 5% of this or 3.15 chloromethyl groups per polymer chain.
The VBC content was confirmed through NMR analysis. While quantitative NMR analysis of polymeric samples is difficult because of the long relaxation times associated with the polymeric backbone, it was possible to get reasonably accurate results by reducing the time between scans to 10 seconds rather than the default 1 second.
The peak allocation is as follows, it is expected that the VBC and styrene will show indiscernible spectra except for the chloromethyl group. 7.08ppm (doublet, 3H) o- phenyl hydrogens, 6.57 ppm (triplet, 2H) m-phenyl and p-phenyl hydrogens, 4.58 ppm (singlet, 2H) from the chloromethyl hydrogens in the VBC polymer, 1.83 ppm (singlet, 1H) backbone methyne hydrogens, 1.42 ppm (singlet, 2H) backbone methylene hydrogens. Solvent peaks: 7.26 ppm residual CDCl3, 2.17 ppm acetone, 1.59 ppm water in CDCl3. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 232
Figure 7.34: NMR of polymeric PS-VBC brush backbone
From this if we base our calculations on the 1.83 ppm peak then the percentage VBC content is
∫ peak(4.46) ÷ 2 %VBC = = 6.71% ∫ peak(4.46) ÷ 2 + ∫ peak(1.83) which is slightly higher than the expected 5%. However, this discrepancy can be ascribed to the fact that accurate quantitative NMR is difficult due to the slow relaxation times of polymeric hydrogens. It is possible that VBC and styrene also incorporate at different rates into the polymer at different rates, however because of their similarity in structure it would seem unlikely that there is a significant difference in reactivity ratios.
Polymerization of side chains on solution brush
This back bone was then used as a macro initiator for the ATRP polymerization of t- BA, and additionally the ATRP polymerization of styrene off the chloromethyl groups on the VBC. The side chains were polymerized as follows: 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 233
Either t-BA or styrene (3.94 M) and anisole (50 wt%) were mixed with CuBr (7.97x10-3 M) and the brush-macroinitiator (5%VBC/styrene, 2.53x10-3 M, equivalent to 7.97x10-3 M initiator concentration). The solution was degassed by sparging with nitrogen for 1 hour. Once the sample was degassed the PMDETA ligand (15.9 x10-3 M) was injected via a degassed syringe with shaking, and the polymerization was initiated by immersion into an oil bath at 80oC. Samples were taken by degassed syringe and dried in vacuo before being dissolved for GPC analysis.
Once the backbone was changed to 5% VBC content the polymers produced were well controlled and easily analyzed by GPC. The results can be seen in Figure 7.35. As expected for a living process the pseudo first-order kinetic plot is linear and the molecular weight is tightly controlled and increases with a linear dependence on conversion.
The reasons that the backbone molecular weight does not correlate with the linear evolution of the brush polymers is because the process of grafting side chains to the brush will significantly alter the Rg of the polymer and thus correlation between absolute molecular weight and the measured molecular weight will change with the degree of side-chain grafting.
Firstly this shows that the chloromethyl group on VBC is a viable initiator to produce living ATRP polymers. It also shows that the polymerization of t-BA off the polymer backbone proceeds in a living manner producing a linear correlation between molecular weight and conversion as well as implying a constant radical concentration through the linear first order kinetic plot. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 234
0.150
0.125
0.100
0.075 ln(1/1-x) 0.050
0.025
0.000 0 5 10 15 20 25 Time (h) t-BA brush; PS brush
1.4
1.2 PDI 15000 1.0 024681012 12500
10000
7500 Mn 5000
2500
0 024681012 Conversion (%) Backbone; t-BA brush; PS brush
Figure 7.35: Chain reinitiation of solution brushes with t-BA and styrene using ATRP 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 235
Hydrolysis of t-BA side chains on solution brush polymers
These t-BA polymer brushers were then purified and hydrolyzed using TFA by the following method:
The polymer was dissolved in 20ml of DCM and percolated through a basic alumina column to remove ATRP catalyst and precipitated in methanol. The dried polymer was then dissolved in 20ml of DCM and hydrolyzed using a fivefold excess of TFA and shaking for 24 hours at room temperature. The remaining DCM and TFA were removed by blowing air across the sample before drying under vacuum for 24 hours prior to analysis.
The resulting polymers were analyzed by H1-NMR and FTIR to determine whether the polymers had been successfully hydrolyzed.
The H1-NMR spectra can be seen below in Figure 7.36. 3.988 3.986 3.983 3.980 3.940
Hydrolysed PAA
P(t-BA) grafted
PS-co-VBC backbone
15.0 10.0 5.0 0.0 ppm (t1)
Figure 7.36: H1-NMR of t-BA graft brush and polymeric backbone 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 236
This NMR shows the three stages of the solution brush creation process. All of the spectra show the vibration from the aromatic groups in styrene and VBC as two broad peaks around 7 ppm. The peaks at 1.85 ppm and 1.43 ppm are from the methyne and methylene groups on the polymeric backbone and appear through all three spectra. The PS-co-VBC backbone shows the characteristic absorption at around 4.3 ppm from the chloromethyl group on the VBC molecule. This peak subsequently disappears as the VBC group polymerizes. The tert-butyl groups appear as single peaks at 2.2 ppm and this subsequently reduces but does not disappear once the tert-butyl groups are hydrolyzed. This indicates that not all of the P(t-BA) has been successfully hydrolyzed. Finally a peak at 3.98 ppm appears in the hydrolyzed PAA polymer from the acidic hydrogen. It would be expected that this peak appears at a significantly higher field than it does (for example in straight PAA it appears at about 12 ppm). However because of the acidic nature of this hydrogen it has a strong tendency to undergo hydrogen bonding and thus it is possible that the hydrogen is shifted down to this area through the strong interaction between the hydrophilic and hydrophobic groups in the polymer. The solvent peaks are so strong in these spectra because of the complicated nature of the polymeric brushes and the hydrophilicity of the acrylic acid makes elimination of solvent contamination difficult.
In addition to the NMR spectra, FTIR spectra were taken to look at the functional groups on the polymer brushes. The FTIR spectra were taken using the ATR equipment and techniques described in Chapter 3 and the spectra of the PS-co-VBC backbone, the P(t-BA) grafted brush and the hydrolyzed PAA brush are shown in Figure 7.37. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 237
0.8 C 0.6 A B
0.4 D
PAA 0.2 Absorbance 0.0 P(t-BA)
Backbone -0.2
-0.4 4000 3500 3000 2500 2000 1500 1000 Frequency (cm-1)
Figure 7.37: FTIR-ATR spectra of solution polystyrene-poly(acrylic acid) brushes
In the above FTIR spectra, the P(t-BA) and PS-co-VBC spectra shows absorptions between 3100 and 2800 (Region A) from the various carbon-hydrogen stretching vibration on the polymeric backbone and the aromatic rings in the styrene and VBC molecules. The characteristic carbonyl peak (Peak B) from C=O stretching appears at 1724 cm-1 and broadens significantly as the P(t-BA) is hydrolyzed. The hydrolyzed acrylic acid peaks tend to form as very broad peaks because of the strong hydrogen bonding between the acidic hydrogen and the carbonyl group in an adjacent acrylic acid. This can be seen by looking at the acidic hydrogen in Region A in the above spectra absorbing across the range of 3500-2500 cm-1 and the broadening of the 1722 cm-1 carbonyl absorption as the acidic groups’ hydrogen bond to the carbonyl. The two peaks in Region C are from C-O stretching (1254 cm-1) which is shifted slightly downfield during the hydrolysis, and the C-(C=O)-O bending vibration at 1144 cm-1. These 2 peaks blend into a single broad peak as the hydrolysis takes place. The final peak in Region D appears at 845 cm-1 and is caused by C-H bending vibrations in the tert-butyl group. This peak disappears as the tert-butyl group is eliminated during hydrolysis. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 238
As can be seen there are several characteristic changes in the spectra confirming the hydrolysis of P(t-BA) to PAA.
7.9.3. Surface-grafted amphiphilic brushes through combined RAFT and ATRP polymerization
Once it had been shown possible to create polymer brushes through the combined use of RAFT and ATRP, it was desirable to take this system and apply it to producing advanced surfaces for SPOS by grafting these brushes to SynPhase substrates.
The principles for producing grafted brushes are exactly the same as for producing the solution polymers with slight modifications in the RAFT step. Since we want the backbone attached to a SynPhase Lantern, γ-radiation is used as an initiation source as opposed to AIBN in the solution brush production. Thus the backbone polymer is made by copolymerizing styrene and VBC in the γ-source in the presence of CPDA, the low temperature RAFT agent used in Chapter 5. This lantern is then washed and used as the initiator for bench top polymerization of the chloromethyl groups using ATRP. The mechanism is outlined in Scheme 7.8. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 239
RAFT
ATRP Cl S S S S
S S PMDETA CH CH2 CH2 2 CuBr t-BA m m n n Cl O
O
o Cl
Hydrolysis
TFA DCM
S S
CH2
m n
O
HO
o Cl
Scheme 7.8: Mechanism for grafting of polymeric brushes from SynPhase Lanterns
Experimental Method
The backbone was grafted to the lanterns with VBC feed ratio varied between 0, 0.05, 0.1 and 0.15. The styrene, VBC and CPDA were mixed and 10ml of this solution was placed over ten pre-weighed lanterns and sealed with septa. These solutions were then degassed by sparging for 1 hour and then exposed to the γ-source for 48 hours at a dose rate of 0.056 kGyH-1. Once the polymerization had completed the solution was decanted off into pre-weighed aluminium pans and the lanterns were washed with 20ml of DCM until no more homopolymer was removed (usually 2 weeks).
Once the lanterns were completely washed the side chains were polymerized by ATRP. This was done in the following manner: the lanterns were weighed and placed in a glass vial, septa sealed and degassed by sparging with nitrogen for 15 min; t-BA (65.25g, 0.373 moles), CuBr (0.1460g, 7.5x10-4 moles) , E-2-BIB (0.1995g, 7.5x10-4 moles) were mixed together and then degassed by sparging for 30 minutes; PMDETA (0.3545g) was then introduced through a degassed syringe and 10ml of this solution was 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 240
syringed over each set of lanterns, the polymerization was initiated by immersing the polymerization solution in an oil bath at 80oC.
Once the polymerization was complete the polymerization solution was poured into pre- weighted aluminium pans and dried in vacuo prior to gravimetric and GPC analysis. The polymerized lanterns were washed in DCM for 2 weeks, changing the DCM daily to remove unwanted homo-polymer from the lanterns.
A batch of 10% VBC lanterns was produced separately from these kinetic runs for the hydrolysis experiments. The lanterns for the hydrolysis experiments had a graft ratio of 0.69 and the solution polymer had the following properties: Mn = 22400, Mnth = 20765, PDI = 1.22 at 32.4% conversion which adheres very well to the predicted molecular weight.
Hydrolysis of these lanterns was conducted by placing 6 of the lanterns in 60ml of DCM with 5 mol equivalent of TFA (2.18g). The lanterns were then stirred overnight at room temperature and the DCM and TFA were removed by blowing compressed air over the solution. The lanterns were then washed for an hour with 50ml of DCM (twice), water, and then methanol. The polymer was dried overnight in vacuo.
Results and discussions
P(t-BA) was successfully grafted to the prepared polymeric sub-substrates to yield grafted brush copolymers. Analysis of the solution homopolymer yielded a straight line correlation between molecular weight and conversion and polydispersity index that reached 1.1-1.2 at the later stages of polymerization. The results from this analysis can be seen in Figure 7.38. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 241
1.4
1.2 PDI 25000 1.0 02040 22500 20000 17500 15000 12500 Mn 10000 7500 5000 2500 0 0 5 10 15 20 25 30 35 40 45 50 Conversion (%)
Figure 7.38: Analysis of solution polymer from t-BA brush chain extension of PP substrate
In Figure 7.39, it can be clearly seen that successful grafting of t-BA to the PP substrate was conducted and the graft ratio of P(t-BA) was found to increase with increasing VBC content in the original grafted macro-initiator. Sample blanks, consisting of both ungrafted and PS grafted (no VBC) PP lanterns, were also included and showed no increase in mass during the t-BA polymerization. This indicates that the grafting mass increase is from polymerization of the chloromethyl groups in the VBC rather than swelling of the lanterns with homopolymer.
This shows that not only has t-BA been successfully grafted to polymeric substrates with a combination of RAFT and ATRP, but also that this polymerization is controlled and that the polymer is grafted in a manner that correlates to the VBC content of the initial polymer. This in turn implies that the polymer is grafted from the chloromethyl groups on the substrate, and there is little ungrafted homopolymer that remains in the substrate. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 242
60
50
40
30
Graft ratio (wt%) 20
10
0 0.0 2.5 5.0 7.5 10.0 12.5 time (h) 5% VBC; 10% VBC; 15% VBC
Figure 7.39: Results for grafting of t-BA onto surface grafted PS-VBC copolymer using ATRP
In addition to these kinetic experiments a batch of t-BA brush lanterns were created and then hydrolyzed using TFA to yield a grafted amphiphilic block copolymer. It was shown that t-BA was successfully grafted up to 0.738g onto the lanterns in brush form. This t-BA polymer was then hydrolyzed with 30.9% of the t-BA polymer being converted into AA. This amount of hydrolysis is relatively low compared to the hydrolysis of t-BA homopolymer which reached 92% conversion. This can be explained by the fact that some of the t-BA polymer will be inaccessible in the polypropylene substrate bulk and thus not hydrolyzed. Alternatively, since PAA is insoluble in DCM it was found that as the t-BA is hydrolyzed it tended to precipitate out of the DCM solvent. This could result in the polymer chains expanding or collapsing, depending on solvent exposure. The collapsed chains may be inhibiting the reaction and this could somewhat protect the remaining t-BA chains from the hydrolysis reagents. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 243
The calculations for this hydrolysis are shown in Table 7.1: RAFT ATRP grafting Conversion of Reaction step backbone of t-BA side t-BA side grafting chains chains to PAA Mass before 0.0713 0.1061 0.1810 Mass After 0.1086 0.1800 0.1710 Mass change 0.0372 0.0738 -0.0100
Table 7.1: Summary of results from brush lantern synthesis.
From this data we can calculate on a per lantern basis that the number of moles of styrene grafted to the lantern
= × mass(RAFTgraft) = × 0.0373 moles(styrene) 0.90 effect. 0.90 Mnmonomer 106.9 moles(styrene) = 3.1×10−4 moles
effect. where Mnmonomer is the effective monomer molecular weight and is a weighted average of the styrene and VBC molecular weights. The moles of VBC can also be calculated
= × mass(RAFTgraft) = × 0.0373 moles(VBC) 0.10 effect. 0.10 Mnmonomer 106.9 moles(VBC) = 1.7×10−5 moles
Similarly the moles of t-BA grafted to the side chains would be
mass(ATRPgraft) 0.0738 moles(t − BA) = = Mnt−BA 126.2 moles(t − BA) = 5.8×10−4 moles and the t-BA grafts will then have a degree of polymerization (DPn) of
moles(t − BA) 5.8×10−4 DPn(t − BA) = = moles(VBC) 1.7×10−5 DPn(t − BA) = 33.05 Mn(t − BA) = 4235
Now in theory the mass loss for the conversion of t-BA to AA is 56.1 mass units, as can be seen from the below mechanism: 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 244
Loss of
O O C4H8 fragment O O H mass = 56.1 thus the percentage of t-BA polymer converted into AA polymer can be given as shown below:
massloss 0.0100g conversion = = − × × −4 × moles(t BA) Mn fragment 5.8 10 56.1 conversion = 30.9%
The conversion to acrylic acid should show quite significant changes in the hydrophilic behavior of the lanterns. This can be seen by looking at the swelling behavior of the lanterns under different solvents.
This behavior was tested by the following procedure: The lanterns were dried in vacuo and weighed prior to any solvent contact. They were then immersed in the desired solvent for 1 hour to swell the polymer. After this time, the lanterns were removed and placed on paper towel to dry for 5 minutes and then weighed.
The degree of swelling is thus calculated from the Eq. 7.10:
m − m D.O.S = swollen dry Eq. 7.10 mdry
where mswollen is the mass of the lantern swollen with solvent and mswollen is the mass of the dry lantern.
The results of this swelling study can be seen in Figure 7.40. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 245
70.00% PS-co-VBC substrate P(t-BA) grafted 60.00% PAA - Post hydrolysis
50.00%
40.00%
30.00% Degree of swelling 20.00%
10.00%
0.00% Water Ethanol Chloroform n-hexane Solvent
Figure 7.40: Degree of swelling in grafted lanterns
The results of this study show significant differences in solvent uptake between each of the stages of the lantern production. In ethanol, the hydrolyzed acrylic acid polymers swelled more than twice as much as the P(t-BA) and PS-co-VBC polymer lanterns. Likewise in chloroform and hexane the P(t-BA) lanterns swelled significantly better than both the PS-co-VBC and PAA polymers because of the more hydrophilic nature.
Unfortunately, the degree of swelling in water could not be measured by this method because of the low vapor pressure and high surface tension of water. This meant that the lantern was still covered by surface water that wasn’t swollen into the polymers, and as such it wasn’t possible to determine which lantern swelled more because the water swelling the polymers was overshadowed by the surface water absorbed onto the surface. For the other solvents, lower surface tension resulted in the surface solvent wicking onto the paper and the higher vapor pressure meant increased evaporation of the surface solvent such that after five minutes the lanterns were free of unabsorbed solvent. 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 246
The lanterns were then further analyzed by ATR-FTIR as described in Chapter 3. The results can be seen in Figure 7.41.
1.0 A
0.8
0.6
0.4 B C D PAA 0.2 Absorbance P(t-BA) 0.0 PS-co-VBC -0.2
-0.4
4000 3500 3000 2500 2000 1500 1000 Frequency (cm-1)
Figure 7.41: FTIR analysis lanterns grafted with PS-co-VBC, then sequential grafting of P(t-BA) to the lanterns and hydrolysis to PAA
This spectrum shows four key absorbances. The first is Region A which shows absorptions from various CH, CH2 and CH3 stretching vibrations on the polymeric backbone, and the aromatic rings in the styrene and VBC molecules. The characteristic carbonyl peak (Peak B) from C=O stretching appears at 1724 cm-1 when the P(t-BA) is grafted and broadens significantly as the P(t-BA) is hydrolyzed. This is the same as seen in the previous hydrolysis of the solution brushes and P(t-BA) homopolymer.
The broad peak from the acidic hydrogen stretching that usually appears between 3500 and 2500 cm-1 is conspicuously absent in this spectra. It is probably that the absorption is present but not visible because of several effects that are coming into play. The first is that the 30% conversion of P(t-BA) to PAA mean that the PAA peaks are relatively small compared to P(t-BA), PS and PP substrate peaks. Also as discussed previously, because of the strong interaction between the acidic hydrogen and various carbonyl 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 247
bonds in the polymers there is significant broadening of this acidic hydrogen absorption. This is possibly further accentuated by the fact that the polymers are now tethered to a substrate, possibly further restricting vibration of the PAA groups and further broadening the peak. This could result in an absorption that is so broad that it blends into the background of the spectra.
The two peaks in Region C are from C-O stretching (1254 cm-1) which is shifted slightly down field during the hydrolysis, and the C-(C=O)-O bending vibration at 1144 cm-1. These two peaks blend into a single broad peak as the hydrolysis takes place. The final peak in Region D appears at 845 cm-1 and is caused by C-H bending vibrations in the tert-butyl group. This peak is significantly reduced as the tert-butyl group is eliminated during hydrolysis.
These key spectra can be seen below in Figure 7.42.
0.2 0.10
P(t-BA) 0.08
0.06 P(t-BA) PS-co-VBC Absorbance Absorbance 0.04
PAA
0.02 PAA PS-co-VBC 0.0 0.00 1400 1350 1300 1250 1200 1150 1100 1050 880 860 840 820 800 Frequency (cm-1) Frequency (cm-1)
Figure 7.42: Key absorptions from PAA brush grafting process
From the gravimetric analysis, FTIR spectroscopy and swelling tests we can see that P(t-BA) was grafted to the PS-co-VBC substrate and was then successfully hydrolyzed to PAA grafts.
7.10. Conclusions
This chapter showed the successful production of well-defined homopolymers of DMA and AA from a thermally initiated system controlled by CDB. The results from this were used as a basis for developing a RAFT controlled, γ-initiated polymerization 7. RAFT mediated grafting of fast polymerizing-hydrophilic monomers 248
system. CDB was used as the RAFT agent to control this polymerization as it provided controlled polymers and molecular weight, and acted to slow the polymerization of DMA and AA to an acceptable level. It was found that CDB worked well and produced controlled polymers grafted to the surface, and the graft ratio of the grafted lanterns was found to be dependent upon monomer concentration, RAFT agent concentration and dose rate, as would be expected from a RAFT mediated polymerization process.
Further to this, amphiphilic brush copolymers were produced through a novel combined RAFT and ATRP system. Styrene and vinylbenzylchloride were polymerized through γ- radiation and controlled by the use of PEPDA. This polymer was then used as an initiator for ATRP polymerization of t-BA to produce a P(t-BA) brush. This brush was then hydrolyzed to produce a PAA brush polymer. It was found that the brush could be controlled through adjustment of the VBC content of the backbone and this in turn could be used to control the hydrophilic-hydrophobic nature of the polymer. It was also found that the RAFT agent remained intact throughout the polymerization and the RAFT end-group of the polymer chain could be reinitiated both before and after t-BA had been grafted to the VBC group. 8. Conclusions 249
8. Conclusions
The primary objective of this thesis has been to develop new methods for controlling grafting of polymers to inert polymeric substrates using living radical polymerization techniques. These will then be used to create advanced polymeric substrates for use in combinatorial chemistry applications. The substrates used were polypropylene SynPhase lanterns from Mimotopes that were grafted using RAFT, ATRP and RATRP living radical techniques.
ATRP was used to graft polymers to SynPhase lanterns by first functionalizing the lanterns such that a halide group was present on the surface that could be used as an initiator. This was achieved by exposing the lanterns to γ-radiation from a 60Co radiation source in the presence of carbon tetra-bromide, which acts as a radical transfer agent. Lanterns were functionalized using styrene, methanol and toluene as a solvent to produce either short-chain tethered Br atoms or direct functionalization of the surface.
The styrene-Br functionalized lanterns were then used as initiators for conventional ATRP grafting of styrene off the surfaces at 90oC in bulk, which produced lanterns with a graft ratio of up to 0.45. It was found that changing the CBr4 concentration in the functionalization stage would affect the length of the tethering polymer chain but did not effect the concentration of bromine on the surface, and consequentially did not affect the later ATRP grafting experiments. The amount of accessible initiator sites was also found to be a function of the time the lanterns were exposed during the functionalization stage. This was from the increased dose of radiation generating increased surface radicals. This effect carried on to the ATRP grafting reactions where it was found that the amount of polymer grafted to the lantern increased as the functionalization time was increased.
The SynPhase lanterns that were directly functionalized by exposing the lanterns to γ- radiation with CBr4 in methanol and toluene were also used as initiators from ATRP graft polymerization and the lanterns were grafted up to a graft ratio of 45%. It was 8. Conclusions 250
found that the lanterns functionalized with methanol as a solvent contained a similar amount of accessible initiator sites as the styrene functionalized lanterns. The lanterns functionalized in toluene were found to have significantly less accessible initiator groups.
FMOC loading tests were conducted on these lanterns to assess their effectiveness as combinatorial chemistry supports and it was found that the loading increased linearly with increasing graft ratio of the lanterns and was independent on the functionalization method and CBr4 concentration. The lanterns produced through this method were found to have a comparative loading to those commercially produced by Mimotopes.
RATRP was used to graft MMA to the surface of SynPhase lanterns using γ-radiation initiated RATRP at room temperature. The system that was used to control the polymerization consisted of either Me6TREN or PMDETA as a ligand, and CuBr2 as the transition metal and was initiated by constant irradiation from a 60Co γ-radiation source. It was found that increasing the radical flux of the polymerization through addition of the thermal initiator AIBN successfully increased the concentration of radicals to a level where proper control of the polymerization could be achieved. The resulting polymers had a linear first-order kinetic plot and molecular weight increased linearly with conversion. The polymers had a PDI of between 1.2 and 1.5; and this broadening was caused by the nature of γ-radiation causing a constant initiation of small chain polymers throughout the polymerization.
The effect of AIBN concentration, dose-rate, and the polymer evolution over time were analyzed. This is the first time that radiation initiated polymerization has been controlled using RATRP and is a significant discovery and provides a new method for grafting of polymers using living radical techniques.
Grafting of styrene to polypropylene SynPhase Lanterns via a γ-initiated RAFT agent mediated free radical polymerization process using cumyl phenyldithioacetate.
Both the RAFT-mediated polymerization and conventional free radical polymerization showed two distinct regimes in relation to the rate of grafting. These two regimes 8. Conclusions 251
showed linear dependence of graft ratio and conversion and are hypothesized to correspond to conditions where only part of the substrate is covered by graft polymer and where the whole substrate is covered by a layer of graft polymer. It is hypothesized that transfer of radicals from the substrate and direct radical generation on the grafted polystyrene result in continued grafting of polystyrene and generation of multi-layers of grafted polymer.
It was found that the graft ratio could be effectively controlled by the concentration of RAFT agent and also variation in the dose rate of the γ-source. The molecular weight of the free polystyrene was analyzed and found to show a linear correlation versus conversion and a low polydispersity index, and thus is a good indication of living free radical behavior.
Loading tests were conducted on the Lanterns and it was found that the loading showed the same two regime trend as the graft ratio and showed a linear dependence on the molecular weight of free polystyrene.
RAFT was also used to control the grafting of well-defined homopolymers of DMA and AA from a thermally initiated system controlled by CDB. The results from this were used as a basis for developing a RAFT controlled, γ-initiated polymerization system. CDB was used as the RAFT agent to control this polymerization as it provided controlled polymers and molecular weights and acted to slow the polymerization of DMA and AA to an acceptable level. It was found that CDB worked well and produced controlled polymers grafted to the surface. The graft ratio of the grafted lanterns was found to be dependent upon monomer concentration, RAFT agent concentration and dose rate, as would be expected from a RAFT mediated polymerization process.
Further to this amphiphilic brush copolymers were produced through a novel combined RAFT and ATRP system. Styrene and vinylbenzylchloride were polymerized through γ- radiation and controlled by the use of PEPDA. This polymer was then used as an initiator for ATRP polymerization of t-BA to produce a P(t-BA) brush. This brush was then hydrolyzed to produce a PAA brush polymer. It was found that the brush could be controlled through adjustment of the VBC content of the backbone and this in turn 8. Conclusions 252
could be used to control the hydrophilic-hydrophobic nature of the polymer. It was also found that the RAFT agent remained intact throughout the polymerization and the RAFT end-group of the polymer chain could be reinitiated both before and after t-BA had been grafted to the VBC group.
8.1. Further work
This thesis has delved into the intricacies and complicated mechanisms that dominate the field of grafting polymers from insoluble surfaces. This process has traditionally been highly uncontrolled and poorly understood but advances in our ability to control this grafting will allow for the development of many advanced products. In particular some work that is getting a lot of press at the moment is the development of honeycomb structures278, grafting off colloids279 and surfaces to be used in electronics and detection applications280.
The application of the methodologies in this thesis are already seeing application, however there is still much more work to be done in the development of controlled grafting from surfaces. In particular RATRP grafting from surfaces offers the advantage over other methods that no chemical modification of the surface is required to conduct this grafting. However, a significant amount of work still needs to be done to fully realize the potential of this method, both in the theory and its application.
The use of radiation grafting methods has fallen out of favor in the last couple of decades in scientific fields; however in industry radiation grafting methods still find significant application. If we are to continue to develop and support industry then the scientific community must continue to research advanced radiation methods such as these. 9. Appendix 1 253
9. Appendix 1
This appendix contains supplementary material regarding the setup and operation of the γ-radiation source.
The data in this appendix contains the complete results for the calibration of the γ- radiation source using a Fricke Dosimiter, as described in Chapter 3. This data compares the observed dose rate of the dosimeter with and without rotation of the sample table and demonstrates the effectiveness of the rotating table in homogenizing dose rate that sample is exposed to.
Figure 9.1: γ-source sample tray and rotation mechanism, top view. 9. Appendix 1 254
Figure 9.2: γ-source sample tray and rotation mechanism, side view 9. Appendix 1 255
Dose rate Sample Degrees OD (Rad/h) 1 21 0.4049 9,754 2 64 0.4769 11,725 3 106 0.6824 17,357 4 148 1.1376 29,828 5 191 0.3375 7,907 6 233 0.4561 11,155 7 275 0.6939 17,671 8 318 0.6266 15,826 9 360 0.7323 18,724 Average 15,550 Standard Deviation 6593
Table 9.1: Results from γ-source calibration with Fricke Dosimiter without table rotation 9. Appendix 1 256
Dose rate Sample Degrees OD (Rad/h) 1 21 0.3596 10,764 2 64 0.3437 10,292 3 85 0.3571 10,692 4 106 0.3210 9,612 5 127 0.3273 9,800 6 148 0.3342 10,007 7 169 0.3671 10,991 8 191 0.3739 11,193 9 233 0.3487 10,440 10 254 0.3584 10,729 11 275 0.3508 10,504 12 296 0.3853 11,534 13 318 0.3518 10,531 14 339 0.3416 10,228 15 360 0.3510 10,508 Average 10,522 Standard Deviation 506
Figure 9.3: Results from γ-source calibration with Fricke Dosimiter with rotating sample table 10. Appendix 2 257
10. Appendix 2
This appendix contains details of ATRP polymerization of the acrylate monomers MA, MMA and t-BA using ATRP polymerization. The experimental procedures are outlined below in each section.
10.1. Polymerization of methyl acrylate
Methyl acrylate was polymerized as follows: MA (0.0581 moles, 5g), anisol (5ml), CuBr (5.81x10-5 moles, 0.0083g) and E-2-BIB (5.81x10-5 moles, 0.0113g) were weighed into a Schenck vial and -4 degassed by sparging with nitrogen for 30 minutes. Me6TREN (8.715x10 moles, 0.0201g) was injected via a degassed syringe and the polymerization was started by immersion into an oil bath at 90oC for 12hrs. 10. Appendix 2 258
0.8
0.6
0.4 ln(1/1-x)
0.2
0.0 012345 Time (h)
1.4
1.2 pdi 60000 1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 50000
40000
30000 Mn 20000
10000
0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Conversion
Figure 10.1: ATRP polymerization of MA at 60oC. A ratio of Monomer:CuBr:lig:E-2-BIB was 1000:1:1.5:1 at 90oC was used. Straight line represents theoretical molecular weight 10. Appendix 2 259
10.2. Polymerization of MMA using Me6TREN
Methyl methacrylate was polymerized in the following manner: MMA (0.0497 moles, 5g), toluene (5g), CuBr (4.97x10-5 moles, 0.0049g) and E-2-BIB (4.97x10-5 moles, 0.0097g) were weighed into a Schenck vial - and degassed by sparging with nitrogen for 30 minutes. Me6TREN (7.46x10 4 moles, 0.01719g) was injected via a degassed syringe and the polymerization was started by immersion into an oil bath at 60oC. 10. Appendix 2 260
0.6
0.4 ln(1/1-x)
0.2
0.0 0 5 Time (h)
1.4
1.2 pdi 80000 1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6
60000
40000 Mn
20000
0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Conversion
Figure 10.2: Results from the ATRP polymerization of MMA using Me6TREN at 60oC. A ratio of Monomer:CuBr:lig:E-2-BIB was 1000:1:1.5:1 was used. Straight line represents theoretical molecular weight 10. Appendix 2 261
10.3. Polymerization of MMA using PMDETA
Methyl methacrylate was polymerized in the following manner: MMA (0.0497 moles, 5g), toluene (5g), CuBr (4.97x10-5 moles, 0.0049g) and E-2-BIB (4.97x10-5 moles, 0.0097g) were weighed into a Schenck vial and degassed by sparging with nitrogen for 30 minutes. PMDETA (7.46 x10- 4 moles, 0.01719g) was injected via a degassed syringe and the polymerization was started by immersion into an oil bath at 60oC. 10. Appendix 2 262
0.30
0.25
0.20
0.15 ln(1/1-x) 0.10
0.05
0.00 012345 Time (h)
1.3 1.2
1.1 pdi 60000 1.0 0.00 0.05 0.10 0.15 0.20 0.25 0.30 50000
40000
30000 Mn 20000
10000
0 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Conversion
Figure 10.3: Results from the ATRP controlled polymerization of MMA using PMDETA as a ligand and a ratio of Monomer:CuBr:lig:E-2-BIB was 1000:1:1.5:1 at 60oC was used. Straight line represents theoretical molecular weight 10. Appendix 2 263
10.4. Polymerization of tert-butyl acrylate
Tert-butyl acrylate was polymerized in the following manner: t-BA (0.0769 moles, 15g), CuBr (2.34x10-4 moles, 0.0336g) and E-2-BIB (1.31x10-4 moles, 0.0256g) were weighed into a Schenck vial and degassed by sparging with nitrogen for 30 minutes. PMDETA (4.67x10-4 moles, 0.0810g) was injected via a degassed syringe and the polymerization was started by immersion into an oil bath at 60oC and 90oC. 10. Appendix 2 264
0.6
0.4 ln(1/1-x)
0.2
0.0 0 5 10 15 20 25 30 Time (h) Temperature 60oC 90oC
1.6 1.4
1.2 pdi 80000 1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 70000 60000 50000 40000 Mn 30000 20000 10000 0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Conversion Temperature 60oC 90oC
Figure 10.4: Results from the ATRP controlled polymerization of t-BA; Effect of temperature. A Ratio of Monomer:CuBr:lig:E-2-BIB was 1000:1:2:1 was used. Straight line represents theoretical molecular weight 11. References 265
11. References
(1) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149-2154. (2) Hudson, D. J. Comb. Chem. 1999, 1, 403. (3) Hudson, D. J. Comb. Chem. 1999, 1, 333. (4) Gysen, H. M.; Meloen, R. H.; Barteling, S. J. J. Proc. Natl. Acad, Sci. USA 1984, 81, 3998. (5) Gysen, H. M.; Rodda, S. J.; Mason, T. J.; Tribbick, G.; Schoofs, P. G. J. Immunol. Methods 1987, 102, 259. (6) Wang, S. S. J. Am. Chem. Soc. 1973, 95, 1328. (7) Gerritz, S. W.; Norman, M. H.; Lee, A. B.; Berman, J.; Bigham, E. C.; Bishop, D. H.; Drewry, D. T.; Garrison, D. T.; Heyer, D.; Hodson, S. J.; Kakel, J. A.; Linn, J. A.; Marron, B. E.; Nanthakumar, S. S.; Navas, F. J. J. Comb. Chem. 2003, 5, 110. (8) Rasoul, F.; Ercole, F.; Pham, Y.; Bui, C. T.; Wu, Z.; James, S. N.; Trainor, R. W.; Wickham, G.; Maeji, N. J. Biopolymers 2000, 55, 207-216. (9) Li, W. B.; Yan, B. J. Org. Chem 1998, 63, 4092. (10) Tomoi, M.; Ford, W. T. J. Am. Chem. Soc. 1981, 103, 3821. (11) Groth, T.; Grotli, M.; Meldal, M. J. Comb. Chem. 2001, 3, 461. (12) Walsh, D. P.; Pang, C.; Parikh, P. B.; S., K. Y.; Chang, Y. T. J. Comb. Chem. 2002, 4, 204. (13) Vaino, A. R.; Janda, K. D. J. Comb. Chem. 2000, 2, 579. (14) Bayer, E.; Dengler, M.; Hemmasi, B. Int. J. Pept. Protein Res. 1985, 25, 178. (15) Mimotopes www.mimotopes.com.au Vol. 2005. (16) Kato, K.; Uchida, E.; Kang, E. T.; Uyama, Y.; Ikada, Y. Progress in Polymer Science 2003, 28, 209-259. (17) Minko, S.; Muller, M.; Motornov, M.; Mitschke, M.; Grundke, K.; Stamm, M. J. Am. Chem. Soc. 2003, 125, 3896. (18) Yosikawa, S.; Tsubokawa, N. Polym. J. 1996, 28, 317. (19) Charlesby, A. Atomic radiation and polymers; Pergamon press ltd.: Oxford, 1960; Vol. 1. (20) Kabanov, V. Y.; Kudryavtsev, V. N. High energy chemistry 2002, 37, 1. (21) Denaro, A. R.; Jayson, G. G. Fundamentals of radiation chemistry; Butterworths and Co: London, 1972. (22) Bevington, J. C.; Eaves, D. E. Nature 1956, 178, 1112. (23) Ballantine, D. S.; Glines, D. S.; Adler, A.; Metz, D. J. J. Polym. Sci. 1959, 34, 419. (24) Ceresa, R. J. Block and graft copolymers; Universities Press: Belfast, 1962. (25) Schelz, S.; Schuhler, N.; Richmond, T.; Oelhafen, P. Thin sold films 1995, 266, 133. (26) Yang, W.; Ranby, B. Macromolecules 1996, 29, 3308. (27) Lee, H. J.; Nakayama, Y.; TMatsuda, T. Macromolecules 1999, 32, 6989. (28) Quinn, J. F.; Barner, L.; Barner-Kowollik, C.; Rizzardo, E.; Davis, T. P. Macromolecules 2002, 35, 7620. (29) You, Y. Z.; Hong, C. Y.; Bai, R.; Pan, C. Y.; Wang, J. S. Macromol. Chem. Phys. 2002, 203, 477. (30) Meng, C.; Tsin-chi, Y. J. Polym. Sci Part A: Polymer Chem 1999, 37, 325. (31) Lee, J. H.; Kim, D. S.; Lee, Y. H. In Irradiation of polymers fundamentals and technological applications; Clough, R. L.; Shalaby, S. W., Eds.; ACS, 1999. 11. References 266
(32) Chang-Do, I.; Son-Ki, I. J. Membrane Sci. 1995, 98, 89. (33) Mishra, M. K. Handbook of radical vinyl polymerization; Marcel Dekker: New York, 1998. (34) Moad, G.; Solomon, D. H. The Chemistry of Free Radical Polymerization; Pergamon: Oxford, 1995. (35) Stevens, M. P. Polymer Chemistry: An Introduction, 2nd ed.; Oxford University Press: New York, 1990. (36) Odian, G. Principles of Polymerization, 3rd ed.; Wiley: New York, 1991. (37) Zammit, M. D.; Davis, T. P.; Haddleton, D. M.; Suddaby, K. G. Macromolecules 1997, 30, 1915. (38) Cowie, J. M. G. Polymers: Chemistry and Physics of Modern Materials; Blackie Academic and Professional: London, 1991. (39) Russell, G. T.; Gilbert, R. G.; Napper, D. H. Macromolecules 1992, 25, 2459. (40) Buback, M.; Kowollik, C. Macromolecules 1998, 31, 3211. (41) Buback, M.; Kowollik, C.; Kurz, C.; Wahl, A. Macromol. Chem. Phys. 2000, 201, 464. (42) Pavjucenko, V. N.; Kolosova, T. O.; Lovjagina, L. D.; Kerzkovskaja, V. V.; Gromov, E. V.; Vylegzanina, K. A.; I., E. E.; S., I. S. Acta Polymerica 1983, 34, 399. (43) Burczyk, A. F.; O'Driscoll, K. F.; Rempel, G. L. J. Polym. Sci Chem 1984, 22, 3255. (44) Barson, C. A. Comprehensive Polymer Science; Pergamon Press: Oxford, 1989. (45) Bamford, C. H. The kinetics of vinyl polymerization by radical mechanisms; Butterworth: London, 1958. (46) Deady, M.; Mau, A. W. H.; Moad, G.; Spurling, T. H. Makromol. Chem. 1993, 194. (47) Gridnev, A. A.; Ittel, S. D. Macromolecules 1996, 29. (48) Olaj, O. F.; Vana, P.; Zoder, M.; Kornherr, A.; Zifferer, G. Macromolecular Chemistry: Rapid Communications, 2000, 21, 913. (49) Heuts, J. P. A.; Gilbert, R. G.; Radom, L. Macromolecules 1995, 28, 8771. (50) Bukack, M.; Busch, M.; Kowollik, C. Macromol. Theory simul. 2000, 9, 442. (51) O'Driscoll, K. F. Comprehensive Polymer Science; Pergamon Press: Oxford, 1989. (52) Buback, M.; Egorov, M.; Kowollik, C.; Kaminsky, V. Macromol. Theory simul. 2001, 10, 209. (53) Kapfenstein-Doak, H. M.; Barner-Kowollik, C.; Davis, T. P.; Schweer, J. Macromolecules 2001, 34, 2822. (54) Staudinger, H. Chem. Ber. 1920, 53, 1073. (55) Szwarc, M. L., M.; Milkovich, R. Nature 1956, 178, 1168. (56) Szwarc, M. L., M.; Milkovich, R. J. Am. Chem. Soc. 1956, 78, 2656. (57) Fieser, L. F.; Fieser, M. Organic Chemistry; D.C. Heath and Company: Boston, 1944. (58) http://tooldoc.wncc.edu/polymers.htm. (59) Becker, M. L.; Remsen, E. E.; Wooley, K. L. J. Polym. Sci Part A: Polymer Chem 2001, 39, 4152. (60) Harth, E.; Horn, B. V.; Lee, V. Y.; Germack, D. S.; Gonzales, C. P.; Miller, R. D.; Hawker, C. J. J. Am. Chem. Soc. 2002, 124, 8653. (61) Holzinger, D.; Kickelbick, G. J. J. Polym. Sci Part A: Polymer Chem 2002, 40, 3858. 11. References 267
(62) Hong, S. C.; Jia, S.; Teodorescu, M.; Kowalewski, T.; Matyjaszewski, K.; Gottfried, A. C.; Brookhart, M. J. Polym. Sci Part A: Polymer Chem 2002, 40, 2736. (63) Narrainen, P.; Pascual, S.; Haddleton, D. M. J. Polym. Sci Part A: Polymer Chem 2002, 40, 439. (64) Webster, O. W. Science 1991, 251, 887. (65) Otsu, T.; Yoshida, M. Macromolecular Chemistry: Rapid Communications, 1982, 3, 127-132. (66) Turner, S. R.; Blevins, R. W. Macromolecules 1990, 23, 1856. (67) Otsu, T.; Yoshida, M.; Kuriyama, A. Polym. Bull. 1982, 7, 45. (68) Darling, T. R.; Davis, T. P.; Fryd, M.; Gridnev, A. A.; Haddleton, D. M.; Ittel, S. D.; Matherson, R. R. J.; Moad, G.; Rizzardo, E. J. Polym. Sci Part A: Polymer Chem 2000, 38, 1706. (69) Gridnev, A. A.; Ittel, S. D.; Fryd, M.; Wayland, B. B. Organometallics 1993, 12, 4871. (70) Davis, T. P.; Kukulj, D.; Haddleton, D. M.; Maloney, D. R. Trends in polymer science, 1995; Vol. 3. (71) Szwarc, M.; van Beylan, M. Ionic Polymerisation and Living Polymers; Chapman and Hall: New York, 1993. (72) Jin, G. T. Y., W. T. Journal of Polymer Materials 1994, 11, 63. (73) Yu, J. M. D., P.; Jerome, R. Macromolecules 1997, 30, 6536. (74) Yu, J. M. D., P.; Jerome, R. Macromolecules 1997, 30, 4984. (75) Yu, Y. S. Z., L. F.; Eisenberg, A. Langmuir 1997, 13, 2578. (76) Cazzniga, L. C., R. E. Macromolecules 1991, 24, 5817. (77) Dryfuss, M. P. D., P. Polymer 1965, 6, 93. (78) Vosfi, D. T., A. V. J. Polym. Sci Chem 1965, 3, 3261. (79) Bawn, C. E. H. B., R. M.; Ledwith, A. Polymer 1965, 6, 95. (80) Lin, C. H.; Xiang, J. S.; Matyjaszewski, K. Macromolecules 1993, 26, 2785. (81) Hasebe, T. K., M.; Sawamoto, M. Macromolecules 1996, 29, 6100. (82) Webster, O. W.; Hertler, W. R.; Sogah, D. Y.; Farnham, W. B.; Rajanbabu, T. V. J. Am. Chem. Soc. 1983, 105, 5706. (83) Miyamoto, M.; Sawamoto, M.; Higashimura, T. Macromolecules 1984, 17, 265. (84) Sawamoto, M.; Fujimori, J.; Higashimura, T. Macromolecules 1987, 20, 916. (85) Faust, R.; Kennedy, J. P. Polym. Bull. 1986, 15, 317. (86) Mishra, M. K.; Kennedy, J. P. Polym. Bull. 1987, 17, 7. (87) Mishra, M. K.; Kennedy, J. P. Macromol. Sci., Chem. 1987, A24, 933. (88) Aida, T.; Maekawa, Y.; Asano, S.; Inoue, S. Macromolecules 1988, 21, 1195. (89) Inoue, S.; Aida, T. Makromol. Chem. 1986, 6, 217. (90) Matyjaszewski, K. ACS Symposium Series, 1998. (91) Matyjaszewski, K.; Davis, T. P. Handbook of radical polymerization; Wiley- Interscience, 2002. (92) Matyjaszewski, K., Ed. Controlled/Living Radical Polymerization - Progress in ATRP, NMP and RAFT.; American Chemical Society.: Washington DC, 2000. (93) Hawker, C. J.; Bosman, A. W.; Harth, E. Chemical Reviews 2001, 3661. (94) Otsu, T. Journal of Polymer Science, Part A: Polymer Chemistry 2000, 38, 2121-2136. (95) Okawara, S.; Nakai, T.; Morishita, K.; Imoto, E. Kogyo Kagaku Zasshi 1964, 67, 2108. (96) Ajayagosh, A.; Francis, R. J. J. Am. Chem. Soc. 1999, 121, 6599. (97) Opresnik, M.; Sebenik, A. Polym. Int. 1995, 36, 13. 11. References 268
(98) Doi, T.; Matsumoto, A.; Otsu, T. J. Polym. Sci. Part A Polymer Chem 1994. (99) Sebenik, A. Prog. Polym. Sci. 1998, 23, 875. (100) Otsu, T. Polym. Bull. 1986, 16, 277. (101) Solomon, D. H.; Rizzardo, E.; Cacioli, E. In Eur. Pat. Appl., 1985; Vol. 102. (102) Georges, M. K.; Veregin, R. P. N.; Kazamaier, P. M.; Hamer, G. K. Macromolecules 1993, 26, 2987. (103) Benoit, D.; Grimaldi, S.; Finet, J. P.; Tordo, P.; Fontanille, M.; Gnanou, Y. Advances in Free Radical Polymerizations: Washington, 1998; Vol. 14. (104) Benoit, D.; Chaplinski, V.; Braslau, R.; Hawker, C. J. J. Am. Chem. Soc. 1999, 121, 3904. (105) Sawamoto, M.; Kato, M.; Kamigatto, M.; Higashimura, T. Macromolecules 1995, 28, 1721-1722. (106) Wang, J. S.; Matyjaszewski, K. Macromolecules 1995, 28, 7901-7910. (107) Wang, J. S.; Matyjaszewski, K. J. Am. Chem. Soc. 1995, 117, 5614. (108) Sawamoto, M.; Kamigaito, M.; Ando, T. Chemical Reviews 2001, 101, 3689- 3745. (109) Katani, Y.; Kamigaito, M.; Sawamoto, M. Macromolecules 1999, 32, 6877. (110) Kotani, Y.; Kamigaito, M.; Sawamoto, M. Macromolecules 2000, 33, 3543. (111) Davis, K. A.; Matyjaszewski, K. Macromolecules 2000, 33, 4039. (112) Grimnaud, T.; Matyjaszewski, K. Macromolecules 1997, 30, 2216. (113) Percec, V.; Barboiu, B. Macromolecules 1995, 28, 7970. (114) Teodorescu, M.; Matyjaszewski, K. Macromol. Rapid Commun. 2000, 21, 190. (115) Xia, J. H.; Zhang, X.; Matyjaszewski, K. Macromolecules 1997, 30. (116) Matyjaszewski, K.; jo, S. M.; Paik, H.; Gaynor, S. G. Macromolecules 1997, 30, 6398. (117) Ashford, E. J.; Naldi, V.; O'Dell, R.; Billingham, N. C.; Armes, S. P. Chem. Commun 1999, 1285. (118) Matyjaszewski, K.; Xia, J. H. Chemical Reviews 2001, 101, 2921-2990. (119) Xia, J. H.; Zhang, X.; Matyjaszewski, K. Macromolecules 1999, 32, 4484. (120) Angot, S.; Ayres, N.; Bon, S. A. F.; Haddleton, D. M. Macromolecules 2001, 34, 768. (121) Ueda, J.; Kamigaito, M.; Sawamoto, M. Macromolecules 1998, 31, 6762. (122) Beers, K. L.; Kern, A.; Matyjaszewski, K. Polym. Pepr. (Am. Chem. Soc. Div. Polym. Chem.) 1997, 38, 695. (123) Husseman, M.; Malmstrom, E. E.; McNamara, M.; M., M.; Mecerreyes, D.; Benoit, D. G.; Hedrick, J. L.; Mansky, P.; Huang, E.; Russell, T. P.; Hawker, C. J. 32 1999. (124) Blomberg, S.; Ostberg, S.; Harth, E.; Bosman, A. W.; Van Horn, B.; Hawker, C. J. Journal of Polymer Science Part A-Polymer Chemistry 2002, 40, 1309. (125) Matyjaszewski, K.; Teodorescu, M.; Miller, P. J.; Peterson, M. L. Journal of Polymer Science, Part A: Polymer Chemistry 2000, 38, 2440-2448. (126) Huan, K.; Bes, L.; Haddleton, D. M.; Khoshdel, E. J. Polym. Sci Part A: Polymer Chem 2001, 39, 1833. (127) Bon, S. A. F.; Steward, A. G.; Haddleton, D. M. Journal of Polymer Science Part A-Polymer Chemistry 2000, 38, 2678-2686. (128) Fischer, H. J. Am. Chem. Soc. 1985, 108, 3925. (129) Fischer, H. Chemical Reviews 2001, 101, 3581. (130) Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina, J.; Jeffery, J.; Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, G.; Moad, C. L.; Rizzardo, E.; Thang, S. H. Macromolecules 1998, 31, 5559. 11. References 269
(131) Le, T. P. T.; Moad, G.; Rizzardo, E.; Thang, S. H. Chem. Abstr. 1998, 128, 115390. (132) Le, T. P. T.; Moad, G.; Rizzardo, E.; Thang, S. H. In In. PCT Int. Appl., 1998. (133) Chong, B. Y. K.; Le, T. P. T.; Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecules 1999, 32, 2071. (134) Quirk, R. P.; Lee, B. Polym. Int. 1999, 27, 359. (135) Matyjaszewski, K.; Mueller, A. H. E. Polym. Prepr. 1997, 38, 6. (136) Stenzel, M. H.; Cummins, L.; Roberts, E.; Davis, T. P.; Vana, P.; Barner- Kowollik, C. Macromol. Chem. Phys. 2003, 204, 1160. (137) Destarac, M.; Bzducha, W.; Taton, D.; Gauthier-Gillaizeau, I.; Zard, S. Z. Macromol. Rapid Commun. 2002, 23, 1049. (138) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Chong, Y. K.; Moad, G.; Thang, S. H. Macromolecules 1999, 32, 6977. (139) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Krstina, J.; Moad, G.; Postma, A.; Thang, S. H. Macromolecules 2000, 33, 243. (140) Rizzardo, E.; Chiefari, J.; Mayadunne, R. T. A.; Moad, G.; Thang, S. H. ACS. Symp. Ser. 2000, 768, 278. (141) Taton, D.; Wilczewska, A. Z.; Destarac, M. Macromol. Rapid Commun. 2001, 22, 1497. (142) Barner-Kowollik, C.; Quinn, J. F.; Nguyen, T. L. U.; Heuts, J. P. A.; Davis, T. P. Macromolecules 2001, 34, 7849. (143) Moad, G.; Anderson, A. G.; Ercole, F.; Johnson, C. H.; Kristina, J.; Moad, C. L.; Rizzardo, E.; Spurling, T. H.; Thang, S. H. In Controlled radical polymerization.; Matyjaszewski, K., Ed.; American Chemical Society: Washington D. C., 1997. (144) Barner-Kowollik, C.; Davis, T. P.; Heuts, J. P. A.; Stenzel, M. H.; Vana, P.; Whittaker, M. J. J. Polym. Sci Part A: Polymer Chem 2003, 41, 365. (145) Moad, G.; Chiefari, J.; Chong, B. Y. K.; Krstina, J.; Mayadunne, R. T. A.; Postma, A. Polym. Int. 2000, 49, 993. (146) Vana, P.; Quinn, J. F.; Davis, T. P.; Barner-Kowollik, C. Aust. J. Chem. 2002, 55, 425. (147) Lind, S. C.; Bardwell, D. C. Science 1924, 60, 364. (148) Tazuke, S.; Kimura, H. Makromol. Chem. 1978, 179, 2603. (149) Chapiro, A. Radiation chemistry of polymeric systems; Interscience Publishers: New York, 1962. (150) Westlake, J. F.; Huang, R. Y. J. Polym. Sci Part A: Polymer Chem 1972, 10, 1443. (151) Braks, J. G.; Mayer, G.; Huang, R. Y. Journal of Applied Polymer Science 1980, 25, 449. (152) Garnet, J. L.; Ng, L. T.; Viengkhou, V. Nucl. Instr. and Meth. in Phys. Res. B 1997, 131, 260. (153) Chaplin, R. P.; Gamage, N. J. W.; Garnet, J. L. Radiat. Phys. Chem. 1995, 46, 949. (154) Matsuzaki, K.; Ichijo, C.; Taiichi, K. Makromol. Chem. 1981, 182, 2919. (155) Swallow, A. J. Radiation Chemistry: An introduction.; Longman Group: London, 1973. (156) Quinn, J. F.; Barner, L.; Rizzardo, E.; Davis, T. P. Journal of Polymer Science, Part A: Polymer Chemistry 2001, 40, 19-25. (157) Quinn, J. F.; Rizzardo, E.; Barner, L.; Barner-Kowollik, C.; Davis, T. P. Division of Polymer Chemistry Preprints 2002. 11. References 270
(158) Hua, D.; Xiao, J.; Bai, R.; Lu, W.; Pan, C. Y. Macromol. Chem. Phys. 2004, 205. (159) You, Y. Z.; Bai, R.; Pan, C. Y. Macromol. Chem. Phys. 2001, 202, 1980. (160) Hong, C. Y.; You, Y. Z.; Bai, R.; Pan, C. Y.; Borjihan, G. J. Polym. Sci Chem 2001, 39, 3934. (161) Bai, R.; Pan, C. K.; You, Y. Z. Macromol. Rapid Commun. 2001, 22, 315. (162) Bai, R.; You, Y. Z.; Zhong, P.; Pan, C. Y. Macromol. Chem. Phys. 2001, 202, 1970. (163) Hua, D.; Zhang, J.; Bai, R.; Weiqi, L.; Pan, C. Y. Macromol. Chem. Phys. 2004, 205, 1125. (164) Miwa, Y.; Yamamoto, K.; Sakaguchi, M.; Shimada, S. Macromolecules 1999, 32, 8234-8236. (165) Miwa, Y.; Yamamoto, K.; Sakaguchi, M.; Shimada, S. Macromolecules 2001, 34, 2089. (166) Gridnev, A. A.; Ittel, S. D. Chemical Reviews 2001, 3611. (167) Enikolopyan, N. S.; Smirnov, B. R.; Ponomarev, G. V.; Bel'govskii, I. M. J. Polym. Sci Chem 1981, 19, 879. (168) Hawker, C. J. J. Am. Chem. Soc. 1994, 116, 11185. (169) Hodges, J. C.; Harikrishnan, L. S.; Ault-Justus, S. Journal of Combinatorial Chemistry 2000, 2, 80-88. (170) Donovan, M. S.; Lowe, A. B.; Sumerlin, B. S.; McCormick, C. L. Macromolecules 2000, 35, 4123. (171) Stenzel-Rosenbaum, M. H.; Davis, T. P.; Fane, A. G.; Chen, V. J. Polym. Sci Part A: Polymer Chem 2001, 39, 2777. (172) Kato, K.; Kamigaito, M.; Sawamoto, M.; Higashimura, T. Macromolecules 1995, 28, 1721. (173) Shen, Y. Q.; Zhu, S. P. Macromolecules 2001, 34, 8603-8609. (174) Stehling, U. M.; Malmstrom, E. E.; Waymouth, R. M.; Hawker, C. J. Macromolecules 1998, 31, 4396. (175) Zheng, G.; Stover, D. H. Macromolecules 2003, 36, 1808. (176) Huang, X.; Doneski, L. J.; Wirth, M. J. J. Anal. Chem. 1998, 70, 4023. (177) Holmberg, S.; Holmlund, P.; Wilen, C.-E.; Kallio, T.; Sundholm, G.; Sundholm, F. Journal of Polymer Science, Part A: Polymer Chemistry 2002, 40, 591-600. (178) Edmonsdson, S.; Huck, W. T. S. J. Mater. Chem. 2004, 14, 730. (179) Jones, D. M.; Huck, W. T. S. Advanced Materials (Weinheim, Germany) 2001, 13, 1256-1259. (180) Schellekens, M. A. J.; Klumperman, B.; Van der Linde, R. Macromol. Chem. Phys. 2001, 202, 1595. (181) Barner, L. Aust. J. Chem. 2003, 56, 1091. (182) Skaff, H.; Emrick, T. Angewandte Chemie int. ed. 2004, 43, 5383. (183) Barner, L.; Li, C.; Hoa, X.; Stenzel, M. H.; Barner-Kowollik, C.; Davis, T. P. J. Polym. Sci Part A: Polymer Chem 2004, 42, 5067. (184) Chen, Y. W.; Wang, W. C.; Yu, W. H.; Kang, E. T.; Neoh, K. G.; Vora, R. H.; Ong, C. K.; Chen, L. F. J. Mater. Chem. 2004, 14, 1406. (185) Ayres, N.; Haddleton, D. M.; Shooter, A. J.; Pears, D. A. Macromolecules 2002, 35, 3849-3855. (186) De Brouwer, H.; Schellekens, M. A. J.; Klumperman, B.; Monteiro, M. J.; German, A. L. J. Polym. Sci Part A: Polymer Chem 2000, 38, 3596. 11. References 271
(187) Wu, D.-C.; Zou, Y.-F.; Pan, C.-Y. Chinese Journal of Polymer Science 2002, 20, 525-530. (188) Hong, C. Y.; You, Y. Z.; Pan, C. Y. Chem. Mater. 2005, 17, 2247. (189) Wang, W. P.; Pan, C. Y. J. Polym. Sci Part A: Polymer Chem 2003, 41, 2715. (190) AhToy, A. L. C. In Chemical Engineering and Industrial Chemistry; University of NSW: Sydney, 2002; p 252. (191) Draganic, I. G.; Holm, N. W.; Maul, J. E. In Danish Atomic Energy commission, 1961; pp 1-25. (192) Morison, D. A.; University of New South Wales: Sydney, 2001. (193) Coote, M. L.; University of New South Wales: Sydney, 1999. (194) Sanayei, R.; O'Driscoll, K. F.; Rudin, A. Chromatography of polymers - Characturization by SEC and FFF.; American Chemical Society: Washinton, 1993. (195) Suddaby, K. G.; Sanayei, R.; O'Driscoll, K. F.; Rudin, A. Makromol. Chem. 1993, 194, 1965. (196) Skoog, D. A.; Leary, J. J. Principles of instrumental analysis.; Saunders College Publishing, 1992. (197) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometic identification of organic compounds.; John Wiley and sons, Inc.: New York, 1991. (198) Rasoul, F.; Ercole, F.; Pham, Y.; Bui, C. T.; Wu, Z.; James, S. N.; Trainor, R. W.; Wickham, G.; Maeji, N. J. Biopolymers 2000, 55, 207. (199) Mitchell, A. R.; Kent, S. B. H.; Engelhard, M.; Merrifield, R. B. J. Org. Chem 1978, 43, 2845. (200) Perera, M. C. S. Mimotopes SOP: FMOC loading determination of polystyrene. (201) http://www.cem.com/lifescience/deprotect.asp. Vol. 2005. (202) Perrier, S.; Barner-Kowollik, C.; Quinn, J. F.; Vana, P.; Davis, T. P. Macromolecules 2002, 35, 8300-8306. (203) Barner-Kowollik, C.; Quinn, J. F.; Nguyen, T. L. U.; Heuts, J. P. A.; Davis, T. P. Macromolecules 2001, 34, 7849. (204) Ciampolini, M.; Nardi, N. Inorg. Chem. 1966, 5, 41-44. (205) Queffelec, J.; Gaynor, S. G.; Matyjaszewski, K. Macromolecules 2000, 33, 8629-8639. (206) McMurray. J. Organic Chemistry, 4th ed.; International Thomson Publishing Company, 1995. (207) Paik, H.; Gaynor, S. G.; Matyjaszewski, K. Macromol. Rapid Commun. 1998, 19, 47. (208) Gaynor, S. G.; Matyjaszewski, K. ACS. Symp. Ser. 1998, 685, 396. (209) Fonagy, T.; Ivan, B.; Szesztay, M. Macromol. Rapid Commun. 1998, 19, 479. (210) Hong, S. C.; Pakula, T.; Matyjaszewski, K. Polym. Met. Sci. Eng. 2001, 84, 767. (211) Liu, S.; Sen, A. A. Polym. Prepr. 2000, 41, 1573. (212) Liu, S.; Sen, A. Macromolecules 2000, 33, 5106. (213) Huang, X.; Wirth, M. J. J. Anal. Chem. 1997, 69, 4577. (214) Huan, X.; Wirth, M. J. Macromolecules 1999, 32, 1694. (215) Matyjaszewski, K.; Miller, P. J.; Shukla, N.; Immaraporn, B.; Gelman, A.; Luokala, B. B.; Siclovan, T. M.; Kickelbick, G. J.; Vallant, T.; Hoffmann, H.; Pakula, T. Macromolecules 1999, 32, 8716. (216) Kim, J. B.; Bruening, M. L.; Baker, G. L. Journal of the American Chemical Society 2000, 122, 7616-7617. 11. References 272
(217) Huang, W.; Baker, G. L.; Bruening, M. L. Angewandte Chemie int. ed. 2001, 40, 1510. (218) Zhang, Z. B.; McCormick, C. L. J. applied polymer science. 1997, 66, 307. (219) Brandrup, J.; Immergut, E. H.; Grulke, E. A. Int. Polymer Handbook; Wiley- Interscience: New York, 1999. (220) Jayachandran, K. N.; Takacs-Cox, A.; Brooks, D. E. Macromolecules 2002, 35, 4247-4257. (221) Xiao, D. Q.; Wirth, M. J. Macromolecules 2002, 35, 2919-2925. (222) Huang, J.; Pintauer, T.; Matyjaszewski, K. J. Polym. Sci Part A: Polymer Chem 2004, 42, 3285. (223) Felder, R. M.; Rousseau, R. W. Elementry principles of chemical process; John Wiley & Sons: Singapore, 1986. (224) Cheng, G.; Hu, C. H.; Ying, S.-K. Macromol. Rapid Commun. 1999, 20, 303. (225) Wang, W.; Yan, D.; Jiang, X.; Detrembleur, C.; Lecomte, P.; Jerome, R. Macromol. Rapid Commun. 2001, 22, 439. (226) Xia, J. H.; Matyjaszewski, K. Macromolecules 1999, 32, 5199-5202. (227) Ucan, R.; Tuncam, U.; Hizal, G. J. Polym. Sci Part A: Polymer Chem 2003, 41, 2019. (228) Ziegler, M., J.; Matyjaszewski, K. Macromolecules 2001, 34, 415. (229) Fukuda, T.; Goto, A.; Ohno, K. Macromol. Rapid Commun. 2000, 21, 151. (230) Goto, A.; Fukuda, T. Macromol. Chem. Phys. 2000, 201, 2138. (231) Hawthorne, D. G.; Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecules 1999, 32, 5457. (232) Oae, S. Organic Sulfur Chemistry: Structure and Mechanism; CRC Press: UK, 1991. (233) Du, F. S.; Zhu, M. Q.; Gui, H. Q.; Li, Z. C.; Li, F. M.; Kamichi, M.; Kajiwara, A. Macromolecules 2002, 35, 6739. (234) Engstrom, M.; Vahtras, O.; Agren, H. Chem. Phys. Letters 2000, 328, 483. (235) Barner-Kowollik, C.; Quinn, J. F.; Morsley, D. R.; Davis, T. P. J. Polym. Sci Part A: Polymer Chem 2001, 39, 1353. (236) Monteiro, M. J.; Hodgson, H.; De Brouwer, H. J. Polym. Sci Part A: Polymer Chem 2000, 38, 3864. (237) Monteiro, M. J.; De Brouwer, H. Macromolecules 2001, 34, 349. (238) Kwak, Y.; Goto, A.; Tsujii, Y.; Murata, Y.; Komatsu, K.; Fukuda, T. Macromolecules 2002, 35, 3026. (239) Barner-Kowollik, C.; Vana, P.; Quinn, J. F.; Davis, T. P. J. Polym. Sci Part A: Polymer Chem 2002, 40, 1058. (240) Wang, A. R.; Zhu, S. J. J. Polym. Sci Part A: Polymer Chem 2003, 41, 1553. (241) Coote, M. L.; Radom, L. J. Am. Chem. Soc. 2003, 125, 1490. (242) AhToy, A. L. C.; Vana, P.; Davis, T. P.; Barner-Kowollik, C. Macromolecules 2004, 37, 744-751. (243) Feng, X. S.; Pan, C. Y. Macromolecules 2002, 35, 4888. (244) Shinoda, H.; Matyjaszewski, K. Macromol. Rapid Commun. 2001, 22, 1176. (245) Zheng, Q.; Pan, C. Y. Macromolecules 2005. (246) Hao, X.; Nilsson, C.; Jesberger, M.; Davis, T. P.; Stenzel, M. H.; Barner- Kowollik, C. J. Polym. Sci Part A: Polymer Chem 2004, 42, 5877. (247) He, T.; Zou, Y.-F. J. Polym. Sci Part A: Polymer Chem 2002, 40, 3867. (248) He, T.; Zou, Y.-F.; Pan, C. Y. Polymer 2002, 34, 138. (249) Qin, D. L.; Pan, C. Y. Macromol. Chem. Phys. 2004, 205, 2097. (250) Yuan, J. Y.; Pan, C. Y. European Polymer Journal 2002, 38, 2069. 11. References 273
(251) Tao, L.; Luan, B.; Pan, C. Y. Polymer 2003, 44, 1013. (252) Quinn, J. F.; Barner, L.; Rizzardo, E.; Davis, T. P. Journal of Polymer Science Part A-Polymer Chemistry 2002, 40, 19. (253) Quinn, J. F.; Rizzardo, E.; Davis, T. P. Chem. Commun 2001, 1044. (254) Borsig, E. Pure Appl. Chem. 1999, A 36, 1699. (255) Berg, R. H.; Almdal, K.; Pedersen, W. B.; Holm, A.; Tam, J. P.; Merrifield, R. B. J. Am. Chem. Soc. 1989, 111, 8024. (256) Maeji, N. J.; Valerio, R. M.; Bray, A. M.; Campbell, R. A.; Geysen, H. M. Reactive polymers 1994, 22, 203. (257) Perera, M. C. S.; Hill, D. J. T., 4th ed.; Brandrup, J.; Immergut, E. H.; Grulke, E. A., Eds.; Wiley-Interscience: New York, 1999. (258) Minko, S.; Gafijchuk, G.; Sidorenko, A.; Voronov, S. Macromolecules 1999, 32, 4525. (259) Stenzel, M. H.; Davis, T. P. J. Polym. Sci Part A: Polymer Chem 2002, 40, 4498. (260) Luzinov, I.; Minko, S.; Senkovsky, V.; Voronov, A.; Hild, S.; Marti, O.; Wilke, W. Macromolecules 1998, 31, 3945. (261) Renil, M.; Meldal, M. Tetrahedron Letters 1996, 37, 6185-6188. (262) Rademann, J.; Grotli, M.; Meldal, M.; Bock, K. Journal of the American Chemical Society 1999, 121, 5459-5466. (263) Ercole, F.; FitzGerald, M.; Perera, S.; Ede, N. J.; Riley, P.; Campbell, R. A. J. applied polymer science. 2003, 89, 3371. (264) Minati, L.; Biffis, A. Chem. Commun. 2005, 8, 1034. (265) El Sawy, N. M.; Al Sagheer, F. European Polymer Journal 2001, 37, 161. (266) Schilli, C. M.; Zhang, M.; Rizzardo, E.; Thang, S. H.; Chong, B. Y. K.; Edwards, K.; Karlsson, G.; Muller, A. H. E. Macromolecules 2004, 37, 7861. (267) Peppas, N. A. Hydrogels in medicine and parmacy: Volume 1: Fundamentals; CRC Press Inc.: Boca Ration, Florida, 1986; Vol. 1. (268) Midha, S.; Nijakowski, T. R. In WO Pat. 9851261. (269) Midha, S.; Nijakowski, T. R. In WO Pat. 9851261. (270) Donovan, M. S.; Sanford, T. A.; Lowe, A. B.; Sumerlin, B. S.; Mitsukami, Y.; McCormick, C. L. Macromolecules 2002, 35, 4570-4572. (271) Donovan, M. S.; Lowe, A. B.; Sumerlin, B. S.; McCormick, C. L. Macromolecules 2002, 35, 4123-4132. (272) Loiseau, J.; Doerr, N.; Suau, J. M.; Egraz, J. B.; Llauro, M. F.; Ladaviere, C. Macromolecules 2003, 36, 3066. (273) Hohn, W.; Tieke, B. Macromol. Chem. Phys. 1997, 198, 703-715. (274) Postma, A.; Davis, T. P.; Moad, G.; O'Shea, M. S. Macromolecules 2005, 38, 5371-5374. (275) Chapiro, A.; Wahl, P. Compt. Rend. 1954, 238, 1803. (276) Frechet, J. M. J.; Murthy, N. In 424/486; 424/488; 424/184.1; 514/59; 514/44: US, 2003. (277) Venkatesh, R.; Yajjou, L.; Koning, C. E.; Klumperman, B. Macromol. Chem. Phys. 2004, 205, 2161. (278) Stenzel, M. H.; Davis, T. P.; Fane, A. G. J. Mater. Chem. 2003, 13, 2090. (279) Luo, Y.; Gu, H. Macromol. Rapid Commun. 2005, 27, 21. (280) Von Werne, T. A.; Germack, D. S.; Hagberg, E. C.; Sheares, V. V.; Hawker, C. J.; Carter, K. R. J. Am. Chem. Soc. 2002, 125, 3838.