SYNTHESIS POLYETHERSULFONE-BASED AMPHIPHILIC BLOCK COPOLYMERS & DEVELOPMENT OF SINGLE-ION CONDUCTORS FOR LITHIUM ION BATTERIES By Hui Zhao A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Chemistry 2012   ABSTRACT SYNTHESIS POLYETHERSULFONE-BASED AMPHIPHILIC BLOCK COPOLYMERS & DEVELOPMENT OF SINGLE-ION CONDUCTORS FOR LITHIUM ION BATTERIES By Hui Zhao Polyethersulfone (PES) membranes often foul easily because of their hydrophobicity, and addition of amphiphilic PES block copolymers to membrane formulations may help overcome this problem. This dissertation explores the synthesis and aggregation properties of relevant amphiphilic ABA block copolymers, where PES is the hydrophobic B block and poly(2hydroxyethyl methacrylate) or poly(2-hydroxypropyl methacrylate) are the 1 hydrophilic A blocks. H nuclear magnetic resonance (NMR) spectroscopy, Fourier transform infrared spectroscopy (FT-IR) and thermogravimetric 1 analysis (TGA) confirm the block copolymer synthesis, and H NMR spectra and TGA also provide consistent data on copolymer compositions. Aggregation behavior of the copolymers in solvent/non-solvent mixture is studied using NMR and dynamic light scattering (DLS). Lithium ion batteries are now ubiquitous, but concentration polarization is still a problem in the currently used battery electrolytes, especially in high-current applications. Immobilizing anions and having lithium cations contribute to most of the ionic conductivity is a good solution to address this problem. This work aims to create nanoparticle-containing electrolytes using + silica nanoparticles modified with polyanions that have Li as the counterion. Anion mobility is restricted by the polyanion polymer backbones, which is further immobilized nanoparticles were by the nanoparticles. synthesized by The surface polyelectrolyte-grafted atom transfer radical polymerization (ATRP) of monomers from initiator-grafted silica nanoparticles. To prepare a lithium-ion conductor, the polyelectrolyte-grafted nanoparticles were blended with polyethyleneglycol (PEG) oligomer. Because the anions are immobile, lithium is the only ion that conducts current. AC impedance shows that the best conductivity is from a Bis(trifluoromethanesulfonyl)imide (TFSI) analogue monomer around 10 -6 S/cm, which is in the same range as a monolayer-grafted silica nanoparticle system using similar TFSI analogue structure. A proposed model shows that the multilayer-grafted nanoparticles only have outermost layer of lithium cations accessible to the solution, because of the low solubility of polyelectrolytes in the PEG solvent. Direct modification of PEG via alkyne-azide or thiol-ene click chemistry as single lithium ion conductor. To make sure 1,2,3-triazole or sulfur structure from click chemistry is not impeding lithium transport, we synthesize 1,2,3-triazole and sulfur containing PEGs via step growth polymerization. Conductivity measurement of lithium perchlorate with triazole containing PEG or sulfur containing PEG shows similar data as the pure PEG, which proves that click chemistry could be applied in the development of single-ion conductors for Lithium Ion Batteries. Copyright by HUI ZHAO 2012 ACKNOWLEDGEMENTS Dr. Gregory Baker – for his patience, encouragement, advice and intellectual generosity Dr. Merlin Bruening – for much appreciated advice and discussions on my research project Dr. William Wulff, Dr. Milton Smith and Dr. Babak Borhan– for being my committee member Many people, past and present, in the lab, Qin Yuan, Sampa Saha, Tomas Jurek, Georgina Comiskey, Wen Yuan, Quanxuan Zhang, Heyi Hu, Yiding Ma, Zhe Jia, Greg Spahlinger, Salinda Wijeratne. Many people, in the department, in particular, Daniel Holms, Kathryn Severin for their help on the instruments. Finally, I would like to thank my family for their love and support. v  TABLE OF CONTENTS     LIST OF TABLES............................................................................................ viii LIST OF FIGURES ............................................................................................ x LIST OF SCHEMES ....................................................................................... xiv Chapter 1 Introduction .......................................................................................1 Part I: Synthesis of Polyethersulfone (PES)-Based Amphiphilic Block Copolymers .....................................................................................................1 Part II: Synthesis of Comb-Polyethyleneoxide (PEO) via Click ChemistryNew polymers for Possible Lithium Ion Conductors.......................................3 References .....................................................................................................8 Chapter 2 Synthesis of Polyethersulfone (PES)-Based Amphiphilic Block Copolymers ......................................................................................................13 Introduction...................................................................................................10 Polysulfone Materials and Their Application as Water Treatment Membranes ...............................................................................................10 Block Copolymers .....................................................................................13 Polysulfone-Based Amphiphilic Copolymers ............................................17 Self-assembly of Block Copolymers .........................................................22 Block Copolymers as Phase Inversion Membrane Materials...................24 Unique Aspects of Self-assembly of Polysulfone-Based Block Copolymers ..................................................................................................................27 Results and Discussion ................................................................................29 Purification of Hydroxypropyl Methacrylate (HPMA) Isomers ..................30 Synthesis of PES with Hydroxyl End Groups ...........................................33 Synthesis of a PES Macroinitiator and Use of a Model Compound to Explore the ATRP Catalytic System .........................................................36 Polymerization from a PES Macroinitiator ................................................42 GPC Analysis of HEMAn-PESm-HEMAn and HPMAn-PESm-HPMAn Block Copolymers .....................................................................................47 Infrared Spectra of the Block Copolymers ...............................................53 TGA of the Block Copolymers ..................................................................57 Critical Water Content (CWC) Values of PES and PES-based Block Copolymers...............................................................................................63 Aggregation of Block Copolymers in Dilute Solutions-NMR and DLS Studies ......................................................................................................72 vi  Conclusion ....................................................................................................81 Experimental Section ...................................................................................82 Appendix A ...................................................................................................91 References ...................................................................................................98 Chapter 3 Use of the Nanoparticles and Click Chemistry in the Development of Single-ion Conductors for Lithium Ion Batteries ............................................102 Introduction-Ion Conduction in Lithium Ion Batteries .................................102 Results and Discussion ..............................................................................106 Single-ion Conductors Containing Nanoparticles with Immobilized Anions ................................................................................................................106 Single-ion Conductors Prepared Using Nanoparticles Modified by Grafting of Polyanions ..........................................................................................109 Towards Click Chemistry for Synthesizing Single-ion Conductors with a High Density of Lithium Ion PEO ............................................................ 113 Conclusions ................................................................................................126 Experimental Section .................................................................................127 Appendix B .................................................................................................135 References .................................................................................................154   vii  LIST OF TABLES     Table 2.1. Different conditions for synthesis of PES and the molecular weights and PDI values of the resulting polymer. .........................................................33 Table 2.2. Block Copolymers with their Molecular Weights and PDI Values. ...... .........................................................................................................................46 Table 2.3. PDIs of various copolymers as determined from GPC data or calculated using equation (2) ........................................................................................... 52 Table 2.4. PolyHEMA Content in Copolymer Samples as Calculated from NMR Spectra and TGA under nitrogen .....................................................................60 Table 2.5. PolyHPMA Content in Copolymer Samples as calculated from NMR spectra and TGA. .............................................................................................63 Table 2.6. Commercial Polyethersulfones and their Molecular Weights, PDI Values and Terminal Groups. ...........................................................................64 Table 2.7. Solubility Parameters of Several Solvents and Polymers at 25 oC. .........................................................................................................................68 Table 2.8. Evolution of the proton NMR signals from HEMA22-PES34-HEMA22 with increasing of D2O content in the DMSO-d6 solvent. .............................................................................................................75 Table 2.9. DLS particle sizes and particle volume percentages for aggregates of HEMA22-PES34-HEMA22 formed in NMP/water mixtures. The mixtures contained 1 mg/mL HEMA22-PES34-HEMA22. ..............................................77 Table 2.10. DLS particle sizes and particle volume percentages for aggregates of HEMA22-PES34-HEMA22 formed in NMP/water mixtures. The mixtures contained 10 mg/mL HEMA22-PES34-HEMA22 ............................................79 viii  Table 2.11. DLS particle sizes and particle volume percentages for aggregates of HEMA22-PES34-HEMA22 formed in NMP/water mixtures. The mixtures contained 30 mg/mL HEMA22-PES34-HEMA22. ............................................80 Table 2.12. Polystyrene standards used to calibrate the results of GPC in DMF. .........................................................................................................................84 Table 3.1. Lithium transference numbers for several lithium salts in battery solvents. .........................................................................................................105 Table 3.2. Particle weight percentages and O/Li for electrolytes prepared from Si-C5NTfLi dispersed in PEGDME-500. ........................................................108 Table 3.3. Particle weight percentages and O/Li ratios for electrolytes prepared from Si-TfMALi dispersed in PEGDME-500. .................................................110   ix  LIST OF FIGURES     Figure 2.1. Equilibrium morphologies in AB diblock copolymers.....................24 Figure 2.2. Phase inversion membranes made from PSf (upper left: SEM top view, upper right: edge view, NMP solvent), and PES (bottom: SEM edge view DMAc solvent)..................................................................................................26   Figure 2.3. SEM images of membranes prepared by phase inversion of poly(styrene)-co-poly(4-vinylpyridine). Peinemann’s work. Edge view (left), top view (right). Scale bars correspond to 500 nm.. ..............................................27   Figure 2.4. Proton NMR 500 MHz spectra of an HPMA isomer mixture (bottom) and purified 2-hydroxypropyl methacrylate (top) in deuterated chloroform.....32   Figure 2.5. ProtonNMR 500 HMz spectra of bisphenol sulfone (bottom) and BisphenolS-I (top) in deuterated DMSO... .....................................................37   Figure 2.6. ATRP kinetic plot for polymerization of HEMA using BisphenolS-I as an initiator and CuBr/PMDETA as the catalyst in DMF, the initial monomer concentration was 2 M... ..................................................................................38   Figure 2.7. ATRP kinetic plot for polymerization of HEMA using ethyl bromoisobutyrate as an initiator and CuCl/CuCl2/PMDETA as the catalyst in NMP, the initial monomer concentration was 2 M... ........................................40   Figure 2.8. ATRP kinetic plot for polymerization of HEMA using ethyl BisphenolS-I as the initiator and CuCl/CuCl2/PMDETA as the catalyst in NMP, the initial monomer concentration was 2 M... ..................................................41   Figure 2.9. ATRP kinetic plot for polymerization of HEMA using a PES macroinitiator and CuCl/CuCl2/PMDETA as the catalyst in NMP, the initial monomer concentration was 1.3 M... ..............................................................42   Figure 2.10. Proton NMR 500 MHz spectra of (a) NMP, (b) HEMA26-PES42HEMA26 and (c) macroinitiator in DMSO-d6... ...............................................44   Figure 2.11. Gel-permeation chromatograms of (a) PES51 and (b) the copolymer HEMA9-PES51-HEMA9... ..............................................................47 x    Figure 2.12. Gel-permeation chromatograms of (a) PES42 (b) copolymer HEMA13-PES42-HEMA13 and (c) HEMA26-PES42-HEMA26... ................... 49   Figure 2.13. Gel-permeation chromatograms of (a) PES34 and (b) HEMA22PES34-HEMA22... ...........................................................................................50   Figure 2.14. An ABA block copolymer with monodisperse A block and polydisperse (PDI=2.0) B blocks... ..................................................................51 Figure 2.15. Gel-permeation chromatograms of (a) PES42, (b) HPMA12PES42-HPMA12 and (c) HPMA26-PES42-HPMA26... ...................................53 Figure 2.16. IR spectra of (a) PES, (b) polyHEMA and (c) HEMA22-PES34HEMA22... ........................................................................................................54   Figure 2.17. IR spectra of (a) HEMA9-PES51-HEMA9, (b) HEMA13-PES42HEMA13, (c) HEMA26-PES42-HEMA26 and (d) HEMA22- PES34-HEMA22.... .........................................................................................................................55   Figure 2.18. IR Spectra of (a) PES, (b) polyHPMA and (c) HPMA12-PES42HPMA12... ........................................................................................................56   Figure 2.19. IR spectra of (a)HPMA12-PES42-HPMA12 and (b) HPMA26PES42-HPMA26... ...........................................................................................57   Figure 2.20. TGA data for PES and several polyHEMA-co-PES-co-polyHEMA samples heated under air (samples were held at 120 oC until the weight was constant (~ 3 hours) before heating at 10 oC/min) (a) PES, (b) HEMA9-PES51HEMA9, (c) HEMA13-PES42-HEMA13, (d) HEMA26-PES42-HEMA26 and (e) HEMA22-PES34-HEMA22... ...........................................................................58   Figure 2.21. TGA data for PES, polyHEMA and polyHEMA-co-PES-copolyHEMA samples heated under nitrogen (samples were held at 120 oC until weight was constant (~ 3 hours) before heating at 10 oC/min). (a) PES, (b) polyHEMA, (c) HEMA9-PES42-HEMA9, (d) HEMA13-PES42-HEMA13, (e) HEMA26-PES42-HEMA26 and (f) HEMA22-PES34-HEMA22... ....................59   xi  Figure 2.22. TGA data for PES and polyHPMA-co-PES-co-polyHPMA samples heated under air (samples were held at 120 oC until the weight was constant (~ 3 hours) before heating at 10 oC/min) (a) PES, (b) HPMA12-PES42HPMA12 and (c) HPMA26-PES42-HPMA26... ...............................................61   Figure 2.23. TGA data for PES and polyHPMA-co-PES-co-polyHPMA samples heated under nitrogen (samples were held at 120 oC until weight was constant (~ 3 hours) before heating at 10 oC/min) (a) PES, (b) polyHPMA, (c) HPMA12PES42-HPMA12 and (d) HPMA26-PES42-HPMA26... ...................................62   Figure 2.24. Room-temperature critical water content (CWC) values of different PES materials in four solvents (NMP, DMF, DMSO, DMAc) (a) Ultrason 2020 P from BASF, (b) Ultrason 6020 P from BASF, (c) Veradel 3600 RP from Solvay, (d) Veradel 3000 RP from Solvay and (e) PES synthesized at MSU Mn=10,000, PDI=2.0. (Each data point is an average from three independent samples, and the data points often obscure the error bars.).. ..........    .................................................................................................................................................... 65    Figure 2.25. Room-temperature CWC values for several PES samples in NMP. (1) Ultrason 6020P, (2) Veradel 3000RP, (3) Ultrason 2020P, (4) Veradel 3600RP and (5) home-made PES. (Each data point is an average from three independent samples, and the data points often obscure the error bars.).. .........    .................................................................................................................................................... 70   Figure 2.26. Room-temperature CWC values in NMP for (a) PES, (b) HPMA26 -PES42-HPMA26 and (c) HEMA22-PES34-HEMA22. (Each data point is an average from three independent samples, and the data points often obscure the error bars.).. ...............................................................................................71   Figure 2.27. Proton NMR 500 MHz Spectra of HEMA22-PES34-HEMA22 in DMSO-d6/D2O co-solvents with varying amounts of water... ......................... 73   Figure 2.28. DLS size distributions for HEMA22-PES34-HEMA22 (1 mg/mL at 25 oC) dissolved in NMP and NMP/water co-solvents with different volume ratios... .............................................................................................................76   Figure 2.29. DLS size distributions for HEMA22-PES34-HEMA22 (10 mg/mL at 25 oC) dissolved in NMP and NMP/water co-solvents with different volume ratios... .............................................................................................................78   xii  Figure 2.30. DLS size distributions for HEMA22-PES34-HEMA22 (30 mg/mL at 25 oC) dissolved in NMP and NMP/water co-solvents with different volume ratios... .............................................................................................................80   Figure 3.1. Schematic diagram of a lithium ion battery containing a metal oxide cathode and a graphite anode. The figure also shows redox reaction during discharge... ....................................................................................................103   Figure 3.2. Concentration polarization during discharge of a lithium ion battery... .........................................................................................................104   Figure 3.3. Temperature-dependent conductivity of electrolytes containing different fractions of Si-C5NTfLi dispersed in PEGDE-500. These results were obtained by Fadi Asfour, and I repeated some of the measurements... .......................................................................................................................108   Figure 3.4. Temperature-dependent conductivity of electrolytes containing different fractions of Si-TfMALi dispersed in PEGDE-500. The various fractions of particles (see Table 3.3) lead to the different O/Li ratios shown in the figure... .......................................................................................................................110   Figure 3.5. Temperature-dependent conductivity for (a) Si-C5NTfLi at O/Li 425 and (b) Si-TfMALi at O/Li 32, both samples contain ~19 wt% modified particles... .......................................................................................................111   Figure 3.6. Kinetics of step-growth polymerization between dipropargyl and diazido tetraethylene glycol... ........................................................................122   Figure 3.7. Conductivities of mixtures of LiClO4 with pure PEO (black squares) and triazole-containing PEO (red circles) at different ratios of PEO to LiClO4. The O/Li ratios in the mixture were determined from the masses of LiClO4 and PEO added, and these ratios only include the O atoms from PEO. Measurements occurred at 90 oC... ...............................................................124   Figure 3.8. Conductivities of mixtures of LiClO4 with pure PEO (black squares) and thioether-PEO (red circles) at different ratios of PEO to LiClO4. The O/Li ratios in the mixture were determined from the masses of LiClO4 and PEO added, and these ratios only include the O atoms from PEO. Measurements occurred at 90 oC... ........................................................................................125     xiii  LIST OF SCHEMES       Scheme 1.1. Structure of the ABA amphiphilic block copolymers synthesized in this study... .....................................................................................................2   Scheme 1.2. Small molecule model (BisphenolS-I) initiator used to explore ATRP catalytic systems for synthesis of block copolymers. This molecule mimics the structure of PES macroinitiators... ...................................................2   Scheme 1.3. Grafting of polyelectrolytes to nanoparticles to create single-ion conductors... ......................................................................................................4   Scheme 1.4. Monolayer-modified silica nanoparticles for single ion conductors... ......................................................................................................6   Scheme 2.1. Chemical Structures of Bisphenol A Polysulfone (PSf) and Polyethersulfone (PES)... ................................................................................12   Scheme 2.2. Reactions invovled in ATRP... .....................................................14   Scheme 2.3. Possible architectures of copolymers synthesized from two monomers, A and B... ......................................................................................16   Scheme 2.4. Structure of the block copolymer synthesized by Jo... ............... 18   Scheme 2.5. PSf-based amphiphilic block copolymer prepared by Moore et al. for formation of membranes.............................................................................19   Scheme 2.6. PSf-based amphiphilic block copolymers synthesized by Wang el al for use as membrane additives... .................................................................20   Scheme 2.7. PSf-based graft copolymer prepared by Yi... .............................21   Scheme 2.8. PES-based graft copolymer synthesized by Yi et al... ...............21   Scheme 2.9. Protocol for making phase-inversion membranes......................25   xiv  Scheme 2.10. Selective reaction of triphenylmethyl chloride with the primary alcohol in hydroxypropyl methacrylate mixtures and subsequent isolationof 2-hydroxypropyl methacrylate..........................................................................30   Scheme 2.11. Synthesis of hydroxyl terminated polyethersulfone (PES)... ....34   Scheme 2.12. Mechanism of etherification in synthesis of PES... ..................35   Scheme 2.13. Synthesis of a macroinitiator from OH-terminated PES... ........36   Scheme 2.14. Synthesis of BisphenolS-I... ....................................................37   Scheme 2.15. Characteristic protons for calculation of copolymer composition from NMR spectra... .........................................................................................85   Scheme 2.16. Characteristic protons for calculation of copolymer composition from NMR spectra... .........................................................................................86   Scheme 3.1. Monolayer modified silica nanoparticle Si-C5NTfLi (silica nanoparticle derivatized with lithiated N-pentenyl triflouromethane sulfonimide)....................................................................................................106   Scheme 3.2. Method for determining the Li+ content of PEO-based electrolytes... ..................................................................................................107   Scheme 3.3. Silica nanoparticles prepared by grafting lithiated poly (trifluoromethane sulfonic aminoethyl-methacrylate) (Si-TfMALi) from the surface... ........................................................................................................109   Scheme 3.4. Proposed qualitative conformations of monolayer Si-C5NTfLi (top) and multilayer Si-TfMALi (bottom) at 30 oC and at 80 oC... ...........................112   Scheme 3.5. Proposed single-ion conductors prepared by synthesis of PEO containing alkene or alkyne groups and subsequent attachment of anions to these groups via click chemistry... .................................................................114   Scheme 3.6. Polysulfone structures synthesized by Bielawski’s et al for proton-conduction... .......................................................................................115   xv  Scheme 3.7. Polyacrylate structures prepared by Martwiset et al. for formation of membranes that exhibit proton conductivity at 200 oC... ...........................116   Scheme 3.8. Poly(ether ether ketone) prepared by Gao et al for reaction with 3-mercaptopropyltrimethoxysilane via click chemistry... ...............................117   Scheme 3.9. The thioether and triazole structures that result from (a) thiol-ene and (b) alkyne-azido click chemistry, respectively... .....................................117   Scheme 3.10. (a) Scheme of ideal lithium transport in an electrolyte material and (b) scheme of lithium transport if the click functionality (triazole or thioether) impedes Li+ transport... ..................................................................................118   Scheme 3.11. (a) Monomers synthesized to prepare triazole-containing PEO, (b) monomers to that will react to give sulfur-containing PEO... ...................119   Scheme 3.12. Two methods for synthesis of dipropargyl tetraethylene glycol... .......................................................................................................................120   Scheme 3.13. Synthesis of several step-growth polymerization monomers via the ditosyl derivative... ..................................................................................121   Scheme 3.14. Synthesis of diallyl tetraethylene glycol.. ...............................121   Scheme 3.15. Step-growth polymerization of diazido and dipropargyl tetraethylene glycol... .....................................................................................122   Scheme 3.16. Step-growth polymerization of dithiol and diallyl tetraethylene glycol... ...........................................................................................................123         xvi  Chapter 1. Introduction There are two parts to this dissertation: chapter 2 describes the synthesis of polyethersulfone (PES)-based block copolymers that are relevant to membrane modification, and chapter 3 explores potential single-ion conductors based on comb-polyethyleneoxide (PEO) materials. Such conductors are important elements of lithium ion batteries. Below I briefly summarize the two projects, and each chapter contains a more extensive introduction. Part I: Synthesis of Polyethersulfone (PES)-Based Amphiphilic Block Copolymers Polyethersulfone (PES) is widely employed in water-treatment membranes because of its chemical and thermal stability. However, the hydrophobic nature 2 of PES may cause severe fouling problems. The introduction of ABA block copolymers (Scheme 1.1) into PES membranes sometimes greatly improves their filtration properties. 3 A PES B block facilitates incorporation into the membrane, and hydrophilic A blocks can create more wettable surfaces that resist fouling. 1  O O S O hydrophobic block n polyethersulfone (PEO) m O m O O OH poly(hydroxylethyl methacrylate) polyHEMA O hydrophilic block OH poly(2-hydroxylpropyl methacrylate) polyHPMA Scheme 1.1. Structure of the ABA amphiphilic block copolymers synthesized in this study.  (For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation.)   Scheme 1.2. Small molecule model (BisphenolS-I) initiator used to explore ATRP catalytic systems for synthesis of block copolymers. This molecule mimics the structure of PES macroinitiators. Our synthesis of ABA copolymers relies on ATRP from PES macroinitiators. To explore atom transfer radical polymerization (ATRP) catalyst systems for polymerization, BisphenolS-I served as a model compound for the PES macroinitiator (Scheme 1.2). NMR characterization is much simpler with this 2  small molecule than a polymer. We tested different combinations of metal catalyst and ligand, and CuCl/CuCl2/N,N,N′,N′,N′′-pentamethyldiethylene triamine (PMDETA) proved effective for controlled polymerization of 2-hydroxyethyl methacrylate (HEMA) and 2-hydroxypropyl methacrylate (HPMA). This catalyst system enabled synthesis of polyHEMA-PESpolyHEMA and polyHPMA-PES-polyHPMA on a 10-20 g scale. Fourier transform infrared spectroscopy (FT-IR), thermogravimetric analysis (TGA) 1 and H NMR show successful synthesis of these block copolymers. Dynamic light scattering studies reveal copolymer aggregates (radii of 60 nm) in 20 vol% water in N-methylpyrrolidone. These copolymers may prove valuable in membrane modification, and we have delivered them to Pall Corporation for their investigation. Part II: Synthesis of Comb-Polyethyleneoxide (PEO) via Click ChemistryNew Polymers for Possible Lithium Ion Conductors  When a lithium ion battery discharges, electrons flow from the anode to the cathode through the external circuit. Simultaneously, inside the cell lithium cations formed at the anode migrate and intercalate into the cathode. Ideally, lithium ions carry all the current within the cell. Typical values of the Li + + transference number (tLi+, the fraction of the current carried by Li ) in electrolytes range from 0.2-0.3, however, which indicates the anion is the dominant species in carrying current. Since a low tLi+ limits a battery’s power density and often affects the chemical stability of electrolytes, development of 3  electrolytes with near-unity lithium-ion transference numbers is important. Scheme 1.3. Grafting of polyelectrolytes to nanoparticles to create single-ion conductors. 1 In research for my MS thesis, I investigated several nanoparticle systems + where Li is a counterion to anionic groups immobilized on grafted polymers (Scheme 1.3). Nanoparticle-containing electrolytes were prepared by mixing the purified particles and low-molecular weight (~500 g/mol) polyethylene glycol dimethyl ether (PEGDME-500). Movement of the anions was largely restricted because of the surface-anchored polymer backbones, so Li + became the only ion conducting current. The MS thesis explored four different polymer structures (Scheme 1.3) with the aim of improving the ionic conductivity of the nanoparticles/PEGDME-500 blend. The first grafted polymer, poly(lithium styrene sulfonate) was initially synthesized from sodium styrene sulfonate monomer, and lithium exchange 4  gave poly(lithium sulfonate styrene). In a second case, polyethylene glycol methyl ether methacrylate (PEGMA) was copolymerized with the styrene sulfonate, to facilitate transport of lithium cations and increase the miscibility of the particles with PEGDME-500. The third structure used phosphonates, which have two lithium counterions per monomer, to increase the lithium content. The room temperature conductivity of these electrolytes was 10 -7 S/cm. This is several orders of magnitude lower than the conductivity of electrolytes employed in current lithium ion batteries, and one or two orders of magnitude lower that current single-ion conductors. To further improve the conductivity, the fourth structure was inspired by lithium bis(trifluoromethane sulfonyl) imide (LiTFSI). High conductivity + correlates with the dissociation of Li from the anion, and therefore anions such as ClO4-, PF6- and bis(trifluoromethylsufonyl amide) (TFSI) are commonly used in electrolytes. We synthesized a polymerizable analogue of TFSI and grew the corresponding polymers from silica nanoparticles. The maximum conductivity, 10 -6 S/cm, occurred at an oxygen to lithium ratio of 32. The O in this ratio comes exclusively from ether oxygens in PEGDME-500, and Li comes from the amount of the lithium cations in the electrolytes. The value of O/Li is a common measure of the lithium concentration in the electrolytes. 5  Scheme 1.4. conductors. Monolayer-modified silica nanoparticles for single ion 4 We expected much higher conductivity than 10 -6 S/cm, because the polyelectrolyte constitutes 70~90 wt% of the polymer grafted particles. This high polyelectrolyte content should supply a large amount of free lithium cations to contribute to a high conductivity. A previous group member, Fadi Asfour, investigated monolayer-coated silica nanoparticles as single-ion conductors (Scheme 1.4), and these materials mixed with PEGDME-500 also have a conductivity of 10 -6 4 S/cm. A possible explanations for the low conductivity with a dense layer of polyelectrolyte on the particle surface is that not all of the lithium cations have access to the solvent (PEGDME-500). Most of the polyelectrolytes are buried near the silica surface and do not contribute to conductivity. We then planned to directly modify the PEO structure to obtain a single ion conductor via alkyne-azido or thiol-ene click chemistry. Chapter 3 describes synthesis of PEO derivatives with click functionalities in the polymer backbone. The conductivity with these materials mixed with salts is similar to data with pure PEO, suggesting that using click chemistry in the synthesis of lithium ion 6  electrolytes is feasible. The triazole and thioether groups do not limit conductivity. 7                      REFERENCES     8  REFERENCES 1. Zhao, H. Master Thesis, 2011, MSU. 2. Beyer, M.; Lohrengel, B.; Nghiem, L. D., Membrane fouling and chemical cleaning in water recycling applications. Desalination 2010, 250 (3), 977-981. 3. Zhou, H., Brunelle, J., Moore, D., R., Zhang, L., Misner, M., J., Chen, X., Ma, M., Block copolymer membranes and associated methods for making the same. US Patent, 123033 A1, 2011. 4. Asfour, F. Ph.D Dissertation, 2004, MSU. 9  Chapter 2. Synthesis of Polyethersulfone (PES)-Based Amphiphilic Block Copolymers 1. Introduction This chapter describes the synthesis of new polysulfone-based block copolymers and preliminary studies on their solubility and phase-segregation properties. To put this work in context, in this introduction I first discuss the importance of polysulfone materials in water-treatment membranes. Polysulfone block copolymers can serve as additives to make these membranes more hydrophilic and reduce fouling. Thus, section 1.2 describes block copolymers in general, and section 1.3 presents specific literature examples of the synthesis of polysulfone-based block copolymers. In section 1.4, I provide a brief description of the self-assembly of block copolymers, and section 1.5 of the introduction mentions a specific application of block copolymer self-assembly, the formation of nanoporous structures. Finally, section 1.6 discusses challenges in the self-assembly of polysulfone-based copolymers. 1.1. Polysulfone Materials and Their Application as Water Treatment Membranes Polysulfone (PSf) and polyethersulfone (PES) constitute an important class of engineering thermoplastics that is widely used to manufacture membranes with relatively high chemical, thermal and mechanical stability. Broadly speaking, polysulfone refers to all sulfone-containing polymers, so both bisphenol A polysulfone and polyethersulfone (PES) are polysulfone materials. 10  In all polysulfones, but especially PES, the aromatic rings in the polymer backbone are electronically deactivated by the adjacent sulfone (-SO2) groups. Additionally, the repeating aromatic rings cause steric hindrance to rotation around the polymer backbone, and both of these factors make PES unusually stable. Bisphenol A polysulfone or poly(oxy-1,4-phenylenesulfonyl-1,4-phenylene oxy-1,4-phenylene(1-ethylethylidene)-1,4-phenylene) (Scheme 2.1) has a o glass transition temperature (Tg) of around 185 C. Polyethersulfone (PES) or poly(oxy-1,4-phenylenesulfonyl-1,4-phenylene) has an even higher Tg of 220 o 1 C (Scheme 2.1). Because of its high Tg, PES can retain dimensional stability at temperature o 2 as high as 200 C. As a high-temperature-resistant resin, PES is also flame retardant, certified for UL94-V0 (burning stops within 10 seconds on a vertical 2 specimen; drips of particles allowed as long as they are not inflamed).   11  Scheme 2.1. Chemical Structures of Bisphenol A Polysulfone (PSf) and Polyethersulfone (PES). Membranes made of PES show wide pH tolerance (1 to 13); excellent resistance to oxidants, including chlorine employed in water treatment (i.e., < o 50 ppm); and stability at high temperature (operation at 75 C and limited o 3 exposure to temperature up to 125 C).   Because of this stability, PES is widely used to create water treatment membranes for applications such as wastewater purification and seawater desalination (removal of salt and other minerals from saline water). PES shows excellent chemical and water resistance, partly because of its hydrophobicity. However, this means that it also exhibits a low wettability. o o Water contact angles on PES range from 53 to 60 , and these values are o higher than contact angles on both cellulose (36.9 ) and aromatic polyamide o 4 (36.9 ) membranes. Recent research correlated low surface wettability with increased non-specific adsorption of naturally occurring organic matter (NOM) 5 on the membrane. Such adsorption is a major cause of chronic fouling of membranes during water treatment. Fouling causes a significant increase in hydraulic resistance and leads to a permanent flux decline or a need for higher 5 transmembrane pressures. Blending of hydrophobic PES with hydrophilic polymers should increase wettability and decrease fouling propensity. 6 However, blends suffer from long-term instability due to migration of the hydrophilic molecules to the 12  surface. Amphiphilic block copolymers can potentially increase hydrophilicity and avoid such instability. The next section describes some general methods for the synthesis of block copolymers, whereas section 1.3 gives specific examples of the synthesis of polysulfone-based block copolymers. 1.2. Block Copolymers The synthesis of block copolymers has been studied extensively, and 7 several authors reviewed this subject. Advances in polymer synthetic chemistry in recent decades, especially in controlled radical polymerization, have enabled access to a wide range of block copolymer compositions and architectures. In a conventional free radical polymerization, decomposition of an initiator generates a radical that starts the polymerization of a vinyl monomer (e.g. methylacrylate, methyl methacrylate and styrene). Because of the high reactivity of radicals with monomers, propagation is very rapid. Since initiation is not instantaneous, this rapid propagation makes the lengths of polymer chains inhomogeneous. Additionally, radicals in solution undergo coupling reactions that terminate chain growth and further broaden the molecular weight distribution. Attainment of narrower molecular weight distributions requires slower propagation rates and termination reactions through controlled or living polymerization techniques. Three criteria define such techniques: (1) fast initiation in which the polymer chains start to propagate at the same time; (2) 13  homogeneous propagation to ensure that the chains grow at the same rate; and (3) minimal termination. The case of zero termination corresponds to the definition of a ‘living’ polymerization. In reality, no system is completely “living”, but many are highly controlled. Atom Transfer Radical Polymerization (ATRP) is one of the most common controlled radical polymerization 8 techniques. Scheme 2.2 shows the ATRP mechanism. Scheme 2.2. Reactions invovled in ATRP. 8 In this scheme, PnX is the initiating alkyl halide/macromolecular species; kact is the rate constant of activation through radical formation; kdeact is the m rate constant for the reverse reaction, which gives a dormant chain; Mt represents the transition metal species in oxidation state m; and L is the metal-binding ligand. ATRP uses halide derivatives as initiators (bromo or chloro groups), and Cu(I) is the most widely used catalyst, although other metal species such as Ru(II), Fe(II), Cr(III), and Os(II) can also catalyze the 9,10 polymerization. Ligands (typically amine derivatives) chelate the Cu(I) to increase its solubility in organic solvents and tune its catalytic properties. Oxidation of the Cu(I)/L complex occurs with homo-cleavage of the carbon halide bond to form carbon radicals, and the radical initiates the polymerization. However, kact is typically two to four orders smaller than kdeact, so PnX serves 14  as a reservoir of radical initiators (Scheme 2.2), and the radical concentration at any given time is low. The propagation rate is kp[Pn*][M] (kp is the propagation rate constant, [Pn*] is the concentration of radicals, and [M] concentration of monomers). Equally important, the termination rate is kt[Pn*][Pn*] (kt is the termination reaction constant). Thus, the propagation rate is first order with respect to the radical concentration, but the termination rate is second order. The low radical concentration gives rise to a slow, controlled first-order propagation reaction, while the second-order termination reaction is nearly negligible. 15  Scheme 2.3. Possible architectures of copolymers synthesized from two monomers, A and B. Scheme 2.3 shows the common copolymer structures synthesized from two types of repeating units, A and B. In linear polymers A and B may appear randomly, in an alternating pattern or in blocks, and the resulting polymers are 16  termed random, alternating, and block copolymers, respectively. In some cases, blocks containing a single repeating unit may branch out from the polymer main chain to give a graft copolymer. If the blocks branch out from a common location on the main chain, the structure is a star copolymer. Because ATRP gives polymers with low polydispersity and minimal termination, it is particularly useful for synthesizing block copolymers by sequential synthesis of the different blocks. Amphiphilic block copolymers contain with both hydrophobic and hydrophilic blocks. Based on different categories of hydrophilic blocks, there are three kinds of amphiphilic block copolymers: non-ionic copolymers, such as poly(ethyleneoxide)-poly(propyleneoxide) (PEO-PPO), poly(ethyleneoxide)poly(oxybutylene) (PEO-PBO) and polyethyleneoxide-polystyrene (PEO-PS) diblock or triblock copolymers; ionic copolymers with polyacrylic acid (PAA) or polymethacrylic acid (PMAA); and copolymers containing monomers with nitrogen, such as poly(2-vinylpyridine) (P2VP) and poly(4-vinylpyridine) (P4VP), which could be quaternized to provide cationic blocks. Block copolymer materials are attractive for application as surfactants, foam stabilizers, flocculants, wetting agents, and dispersants. Micellization of the amphiphilic block copolymers can solublize active drugs to increase solubility 11 and circulation half life. 1.3. Polysulfone-Based Amphiphilic Copolymers A number of studies examined the synthesis of polysulfone-based 17  copolymers to modify the properties of polysulfone while retaining aspects of its high stability. Jo et al. prepared a polysulfone-based amphiphilic block 12 copolymer for potential use in a fuel cell membrane. Nafion (DuPont) is the most common material for polyelectrolyte membranes in proton exchange fuel cells, but its conductivity and mechanical stability deteriorate at high o temperature (over 100 C). Due to its excellent thermal stability, polysulfone might provide an attractive alternative material, although it is less chemically stable than Nafion. hydrophobic hydrophilic hydrophilic O O O S hydrophobic block O O CN p r m-r n m hydrophilic block SO3H (p, m and r are independent integers, and m is greater than r) Scheme 2.4. Structure of the block copolymer synthesized by Jo.   12 To prepare a polysulfone-based amphiphilic block copolymer, Jo et al. copolymerized styrene and acrylonitrile from a polysulfone macroinitiator. They subsequently partially sulfonated the polystyrene to provide the hydrophilic block (Scheme 2.4). Because of the high thermal stability of polysulfone, 18  electrolyte membranes prepared from the amphiphilic block copolymer o maintained their conductivity and mechanical stability at 130 C. hydrophobic hydrophilic hydrophilic O O S hydrophobic block O O OH O O 9 O O N O m O m polyHEMA polyPEGMA n O hydrophilic blocks m polyDMAEMA Scheme 2.5. PSf-based amphiphilic block copolymer prepared by Moore et al. 13 for formation of membranes. Moore et al. used polysulfone macroinitiators in polymerization of hydrophilic monomers such as 2-hydroxyethyl methacrylate (HEMA), polyethylene glycol methacrylate (PEGMA) and N, N-dimethylaminoethyl methacrylate (DMAEMA) 13 (Scheme 2.5). Synthesis occurred on a 100 g scale, and membranes made of the synthesized copolymers showed improved hydrophilicity and open pores that give minimum hydraulic resistance while maintaining the mechanical strength of PES. The membranes also showed reduced fouling. Wang et al. synthesized polysulfone-based amphiphilic ABA copolymerS(Scheme 2.6) using the hydrophilic monomers PEGMA and 19  3-O-methacryloyl-1,2:5,6-di-O-isopropylidene -D-glucofuranose (MAIpG). 14 The protected sugar moiety was acidolysized after polymerization to remove Scheme 2.6. PSf-based amphiphilic block copolymers synthesized by Wang for use as membrane additives. 14 isopropylidenyl groups and provide the hydrophilic block. They used the amphiphilic block copolymers as additives to improve the hydrophilicity and resistance to Bovin Serum Albumin (BSA) adsorption of polysulfone membranes. In another work, Yi et al. chloromethylated polysulfone (PSf-CH2Cl, Scheme 2.7), and used the chloromethyl groups to initiate subsequent ATRP of polyethylene glycol methacrylate (PEGMA) and produce the amphiphilic graft copolymer. 15 20  Scheme 2.7. PSf-based graft copolymer prepared by Yi. 15 Scheme 2.8. PES-based graft copolymer synthesized by Yi et al. 16 In a similar work, the same group synthesized a polysulfone-based graft copolymer using Reversible Addition Fragmentation Chain Transfer (RAFT) polymerization. In the synthesis of the PES backbone, the ratio of bis (3-amino-4-hydroxyphenyl) sulfone and bisphenol sulfone was 2/3. 16 The amino group in the synthesized PSf was converted to a RAFT chain transfer 21  agent, which participated in RAFT polymerization with N, N-dimethylaminoethyl methacrylate (DMAEMA) or N-isopropylacrylamide (NIPAAm) hydrophilic monomers to give the amphiphilic graft polymers (Scheme 2.8). 1.4. Self-assembly of Block Copolymers When dissolved at low levels in selective solvents, block copolymers tend to self-assemble into micelles. At even lower concentrations, there are not enough polymers to self-assemble, so the chains adsorb at the air-water or aqueous-organic solvent interface. When the copolymer concentration increases, more and more polymer chains adsorb at the interface until the concentration reaches a point where both solution and interface are saturated with polymer chains. Upon further addition of polymer chains to the solution, the copolymer self-assembles into micelles to reduce the free energy of the system. The selective solvent is good for only one of the blocks in the copolymer, so the insoluble blocks aggregate to give the core, while the soluble blocks form the corona of the micelle. Interaction of the soluble blocks and the solvent stabilizes the micelles. Equation (1) expresses the free energy of micellization, ∆G, ∆G =∆H - T∆S (1) where ∆H and ∆S are the enthalpy and the entropy of micellization. In the micellization of block copolymers in solution, ∆S is typically negative due to loss of entropy from localization of polymer molecules in the micelle. Moreover, 22  the solvent incompatible block is confined to the core, the solvent compatible block is confined to the corona, and the block copolymer junction is confined to the interfacial region between core and corona of the micelle. For micellization of copolymer chains to be a thermodynamically favorable process, the free energy change of the system upon micellization must decrease, i. e. ∆G<0, and ∆H must be negative. This readily occurs because the interaction of the nonselective block with the solvent is unfavorable. In a molten state or in polymer solutions at high concentrations, block copolymers can self-assemble into domains with sizes ranging from nanometer to micrometer scales. The size and structure of the self-assembly is governed by the polymer molecular weight, the block-size ratio and temperature. Block copolymers may adopt a disordered/homogeneous state at high temperature, but when the temperature decreases, the incompatibility of the constituent blocks increases, because the influence of combinatorial entropy decreases (T∆S decreases). At the order-to-disorder transition temperature (TODT), block copolymers form ordered mesophase structures. TODT depends on the volume fraction of one block, the total number of monomers in each polymer, and the segmental interaction described by the 17,18 Flory-Huggins interaction parameter χ. 23  17 Figure 2.1. Equilibrium morphologies in AB diblock copolymers. (Used from ref 17 with permission of Copyright 2012 Annual Reviews.) Self-consistent mean field theory describes the phase behavior of block copolymers. As the volume fraction of the A block increases, the microdomain changes from spheres, cylinders, to gyroid, and finally to lamellae (Figure 12 2.1). In recent years the gyroid phase has attracted attention because of its potential application in ionic conductive materials. While the matrix provides mechanical support, the bicontinuous gyroid phase can act as an ionically 19 ,20 conductive channel. 1.5. Block Copolymers as Phase Inversion Membrane Materials Phase-inversion has been widely used in the preparation of asymmetric 21,22,23 filtration membranes since the 1960s. (Asymmetric membranes contain a dense surface layer on a highly porous support of the same material.) Scheme 2.9 shows a typical phase-inversion process. Polymer materials are dissolved in a good solvent, such as tetrahydrofuran (THF), dimethylformamide (DMF), N-methylpyrrolidone (NMP) that is also miscible 24  with water. The concentration of the polymer is normally in the range of 20 to 35 wt% to maintain a high viscosity. The polymer solution is then cast on a glass substrate and solvent is allowed to evaporate for a short time (10 seconds to 1 minute). Finally, the substrate is immersed in a nonsolvent bath (typically water) at a pre-determined temperature. After oven drying, the film is detached from the substrate. viscous polymer solution cast on substrate allow the solvent to partially evaporate immerse in a nonsolvent bath dry the film Scheme 2.9. Protocol for making phase-inversion membranes. The exchange between solvent and nonsolvent introduces phase separation in the polymer layer, which leads to an asymmetric porous polymer membrane structure. Figure 2.2 shows the morphologies of phase-inversion membranes made from polysulfone materials. Because of the rapid solvent exchange near the membrane surface, the thin top layer is a structure of closely packed 25  polymeric spheres and serves as the functional layer for the separation process. The much thicker bottom structure with large elongated voids (macrovoids) provides the mechanical support, but little resistance to flow in 14 comparison to the skin layer. 14 Figure 2.2. Phase inversion membranes made from PSf 16 top view, upper right: edge view, NMP solvent), and PES (upper left: SEM (bottom: SEM edge view DMAc solvent), (used with permission of Elsevier from references 14 and 16, copyright ) When the membrane material is a block copolymer, the above-mentioned phase-inversion process can lead to structures that contain uniform nanopores in their interfacial layer. The most successful studies employed poly 24,25 (styrene)-block-poly (4-vinylpyridine) (PS-b-P4VP). This asymmetric membrane forms in a combination of phase-inversion and block copolymer self-assembly. In an appropriate solvent, the block copolymer self-assembles 26  into perpendicularly aligned cylinders at the membrane surface, because this is the thermodynamic equilibrium structure. When the membrane is immersed into the nonsolvent coagulation bath, solvent-nonsolvent exchange occurs immediately at the membrane surface to lock in the thermodynamic equilibrium structure (perpendicularly aligned cylinder morphology) at the interface. In contrast, the nonsolvent migrates slowly into the interior of the membrane, and the delayed solvent-nonsolvent exchange gives a non-equilibrium structure, similar to that in a conventional phase-inversion membranes. The narrow surface pore size distribution may lead to relatively sharp molecular weight cutoffs (Figure 2.3), whereas the minimal thickness of the interfacial layer can allow extremely high flux. In a recent publication, Peinemann and coworkers reported a remarkable flux of 890 L m -2 -1 h bar -1 24 through such membranes. Figure 2.3. SEM images of membranes prepared by phase inversion of poly(styrene)-co-poly(4-vinylpyridine). Peinemann’s work. Edge view (left), top view (right). 24 (Used with permission from ref 24, Copyright 2007 Nature Publishing Group.) 1.6. Unique Aspects of Self Assembly of Polysulfone-based Block Copolymers 27  We synthesized ABA triblock copolymers where the B block is PES and the A block is poly(2-hydroxyethyl methacrylate) (polyHEMA) or poly(2-hydroxy propyl methacrylate) (polyHPMA). We note that these copolymers present some unique features for self assembly. First, the PES block will have a relatively broad molecular weight distribution. Most of the previous research in block copolymer self assembly focused on model compound such as PPO-PEO and PS-PAA because those polymers are easy to synthesize with a narrow polydispersity index (PDI). The PDI value is the ratio of weight-average to number-average molecular weight (Mw/Mn). With polysulfone-based copolymers, a narrow PDI is difficult to achieve because polysulfones are obtained via polycondensation chemistry, which leads to a PDI of around 2. Although ATRP can give relatively monodisperse polyHEMA and polyHPMA, the condensation polymerization of PES will lead to a more polydisperse material. However, a recent study shows that narrow dispersity copolymers are not required for periodic nanoscale assembly. Widin and coworkers examined the self assembly of poly(styrene)-poly(1,4-butadiene)- poly(styrene) ABA triblock copolymers similar in structure to those described here. 26 Their ABA triblock copolymers contained a narrow dispersity A block (Mw/Mn ≤ 1.05) and a variable dispersity B block (Mw/Mn= 1.18−2.00). These copolymers show the same ordered structures regardless of the polydispersity of the B block, although the composition window for different ordered structures varies with PDI. We expect that if the PDI of a synthesized copolymer is below 1.5, 28  one can still achieve periodic nanoscale self-assembly. A second unique feature of PES-containing block copolymers is that PES is more rigid than PS and other well-studied hydrophobic polymer structures in the literature. The persistence length of this rigid polymer backbone will thus be larger than common PS and PMMA blocks, which have has a flexible backbone and coil structure. The extent to which rigidity will alter self-assembly is unknown. 2. Results and Discussion This section primarily describes the synthesis of several ABA block copolymers where PES is the hydrophobic B block and polyHEMA or polyHPMA serves as the hydrophilic A block. The first section describes purification of HPMA to obtain a single isomer for formation of polyHPMA blocks. Subsequent sections present the synthesis of PES and conversion of PES to a macroinitiator for polymerization of HEMA or HPMA. A model compound for the PES macroinitiator, BisphenolS-I (see below) is used to develop an appropriate ATRP catalytic system to synthesize block copolymers. Finally, we synthesized copolymers with a range of compositions. The new block copolymers are characterized using nuclear magnetic resonance (NMR) spectroscopy, gel permeation chromatography (GPC), Fourier transform infrared spectroscopy (FT-IR), thermogravimetric analysis (TGA) and dynamic light scattering (DLS). Measurements of critical water content (CWC), which is used to study the hydrophilicity of PES and synthesized block copolymers in 29  solvent/nonsolvent (water) mixtures, clearly show there is an apparent increase of hydrophilicity for the block copolymer. NMR spectroscopy and DLS indicate the formation of micellar structures. 2.1. Purification of Hydroxypropyl Methacrylate (HPMA) Isomers Hydroxypropyl methacrylate (HPMA) was purchased from Acros Organics, as an approximately 3:1 (2-hydroxypropyl methacrylate/1-hydroxypropan-2-yl methacrylate) mixture of the primary and secondary alcohols. The secondary alcohol (2-hydroxypropyl methacrylate) is the major product, and most commercially available HPMA has a similar composition. O O + OH 2-hydroxypropyl methacrylate O O Ph3CCl O O + Et3N, 50 oC, OH overnight OCPh3 OH 1-hydroxypropan-2-yl methacrylate O O Vacuum Distillation O O OH 2-hydroxypropyl methacrylate Scheme 2.10. Selective reaction of triphenylmethyl chloride with the primary alcohol in hydroxypropyl methacrylate mixtures and subsequent isolationof 2-hydroxypropyl methacrylate. 30  Purification of the isomers starts with reaction of the mixture with triphenylmethyl chloride (Scheme 2.10). The bulky triphenylmethyl group selectively protects the primary alcohol to allow isolation of 2-hydroxypropyl methacrylate by vacuum distillation. Triethylamine is a good base for this reaction, and in a typical purification process, 14.4 g of the isomer mixture gave 4.2 g of pure monomer. 31  1 Figure 2.4. H NMR 500 MHz spectra of an HPMA isomer mixture (bottom) and purified 2-hydroxypropyl methacrylate (top) in CDCl3. 1 The H NMR spectra in Figure 2.4 show that before purification, signals from protons on both isomers appear. After purification, only signals from 2-hydroxypropyl methacrylate are present. Peaks in the bottom spectrum 32  around 2.5 and 3.6 ppm correspond to the hydroxyl protons in the two isomers. The upper spectrum clearly demonstrates disappearance of the signals from 1-hydroxypropan-2-yl methacrylate and the successful purification of 2-hydroxypropyl methacrylate from the isomer mixture. 2.2. Synthesis of PES with Hydroxyl End Groups Temperature Monomer (mmol) ID Base Solvent c o Mn/PDI ( C) A B C 1 20 / 20 K2CO3 DMF 150 4,500/2.1 2 20 / 20 K2CO3 DMSO 180 5,200/1.9 3 20 / 20 KF DMSO 180 3,800/1.8 4 20 / 20 K2CO3 DMSO/Toluene 160 5,500/2.0 5 20 / 20 K2CO3 Sulfolane 220 4,700/1.8 6 20 / 20 K2CO3 DMAc/Toluene 140 7,800/1.8 7 / 20 20 K2CO3 DMAc/Toluene 140 10,800/2.0 8 / 20 20 K2CO3 DMAc/Toluene 140~170 14,700/2.1 a Monomer A: bis(4-chlorophenyl) sulfone d B: bis(4-fluorophenyl) sulfone C: 4,4’-sulfonydiphenol b c Reaction time: 24h Determined using GPC with polystyrene samples. d We did not consider signals from large aggregates in calculating these values. Table 2.1. Different conditions for synthesis of PES and the molecular weights and PDI values of the resulting polymer. a,b 33  d Table 2.1 shows the different conditions used to synthesize polyethersulfone (PES). 4,4’-sulfonylbisphenol (bisphenol S, monomer C) is reacted with bis(4-chlorophenyl) sulfone (monomer A) or bis(4-fluorophenyl) sulfone o (monomer B) in a dipolar aprotic solvent at high temperature (up to 285 C), e.g., Scheme 2.11.  K2CO3 or Na2CO3 generate phenate salts ‘in situ’. In a typical poly-condensation, the highest PDI of the final polymer product is around 2.0, but in most of the commercial products the PDI is as high as 4.0. A possible explanation for the lower polydispersity in our polymers is that low molecular weight oligomers are added to improve the rheology and processing properties in the commercial formulation. Table 2.1 shows that reactions using bis(4-fluorophenyl) sulfone give higher molecular weights than reactions with bis(4-chlorophenyl) sulfone.   Scheme 2.11. Synthesis of hydroxyl terminated polyethersulfone (PES). The weight-average molecular weights of the synthesized polymers (Mw) 34  are 20,000~30,000 and the PDIs are ~2 (Table 2.1). Weight average molecular weights are always higher than number average values (Mn). Compared to commercial PES, the synthesized PES has similar molecular weight and a lower PDI. To ensure that the chains terminate in phenol groups, we reacted the polymers with excess amount of 4,4’-sulfonydiphenol after chain formation (Scheme 2.11). In this way we were eventually able to cap both ends with an ATRP initiator and form ABA block copolymers. An SNAr mechanism is used to describe the etherification chemistry between bisphenol sulfone and O O S F F O O S Nu- F Nu F O O S F Nu + F- Scheme 2.12. Mechanism of etherification in synthesis of PES. bishalophenyl sulfone. The nucleophile in Scheme 2.12 represents a bisphenolate that attacks the bishalophenyl sulfone to give an anion intermediate, which in the final step goes through reductive elimination to yield 27 the ether product. The first step is generally quite slow and is rate determining, so any structure that facilitates the nucleophilic attack would give a higher reaction rate. This may explain why reactions using bis(4-fluorophenyl) sulfone give higher molecular weights than reactions with bis(4-chlorophenyl) 35  sulfone. 2.3. Synthesis of a PES Macroinitiator and Use of a Model Compound to Explore the ATRP Catalytic System   Scheme 2.13. Synthesis of a macroinitiator from OH-terminated PES. To synthesize PES macroinitiator (Scheme 2.13), hydroxyl terminal groups are allowed to react with bromoisobutyryl bromide. The resulting PES macroinitiator bears the bromoisobutyryl structure, which can initiate polymerization of HEMA or HPMA. Before moving on to the synthesis of block copolymers using this macroinitiator, we performed the reaction in Scheme 2.13. The purpose of this reaction is two-fold: first, to confirm that the reaction conditions for esterification yield an essentially quantitative reaction. In the synthesis of PES, we used an excess of bisphenol sulfone to ensure a phenol end group structure at both ends, but we must also convert all of these phenol end groups to the ester structure to synthesize the ABA block copolymer. Second, the product of the reaction in Scheme 2.14 (BisphenolS-I) served as a model compound to explore the chemistry to polymerize HEMA and HPMA blocks. BisphenolS-I is synthesized with two equivalents of bromoisobutyryl bromide and one equivalent of bisphenol sulfone in acetone in less than 3 36  hours, and the yield of this reaction is typically higher than 97%. O O S O OH HO + Br Et3N, Acetone Br O O S O Br O O O BisphenolS-I Scheme 2.14. Synthesis of BisphenolS-I. Figure 2.5. 1 H NMR 500 HMz spectra of bisphenol sulfone (bottom) and BisphenolS-I (top) in DMSO-d6. The signal at 2.5 ppm is due to DMSO-d6. Br Figure 2.5 shows the 1 H NMR spectra of bisphenol sulfone and BisphenolS-I. Upon reaction with bromoisobutyryl bromide, chemical shifts of the aromatic protons change from 7.70 ppm (Ha) and 6.90 ppm (Hb) to 8.10 ppm and 7.45 ppm, and a new peak at 2.0 ppm corresponds to the methyl groups of the bromoisobutyryl end group. This result shows that the conditions for esterification of the phenol groups are nearly quantitative and should also work for the polymer. Unfortunately, the large number of repeat units in the polymer make direct characterization of PES derivatization by NMR difficult.   Figure 2.6. ATRP kinetic plot for polymerization of HEMA using BisphenolS-I 38  as an initiator and CuBr/PMDETA as the catalyst in DMF, the initial monomer concentration was 2 M. As an analogue compound for the PES macroinitiator with the bromoisobutyryl functionality on both sides, BisphenolS-I allowed us to explore ATRP catalytic systems for HEMA polymerization. Since the PES macroinitiator is only soluble in polar solvents such as DMF, DMSO, DMAc and NMP, polymerization must also occur in theses solvents. BisphenolS-I was used to polymerize HEMA, using CuBr/PMDETA (N,N,N′,N′,N′′-pentamethyl diethylene triamine) as a catalyst, and NMP as the solvent at 40 °C. Some prior polymerizations of methacrylate monomers using macroinitiators occurred at 90-110 °C, but we did not use high temperatures to avoid auto-polymerization of the monomers. Figure 2.6 shows the polymerization kinetics for the CuBr/PMDETA/NMP system. Polymerization occurred rapidly during the first 30 min and then slowed suggesting significant termination of the propagating chains. The fast polymerization rate at initial stages might be due to the use of the CuBr catalyst, since C-Br bonds have a lower dissociation energy than C-Cl bonds and potentially gives a faster and less 28 controlled ATRP kinetics. The low conversion of monomers is also a problem, especially when using macroinitiators, since we desire the synthesis of block copolymers with high hydrophilic content. 39  Figure 2.7. ATRP kinetic plot for polymerization of HEMA using ethyl bromoisobutyrate as an initiator and CuCl/CuCl2/PMDETA as the catalyst in NMP, the initial monomer concentration was 2 M. To further explore ATRP of HEMA, we examined polymerization using a simple conventional initiator, ethyl bromoisobutyrate (Figure 2.7). We employed CuCl as a catalyst instead of CuBr, because the higher dissociation energy of the C-Cl (relative to C-Br) bond could give slower kinetics, as mentioned above. Additionally, 0.5 equivalents of CuCl2 were added to slow down and control the reaction. This polymerization was conducted as a small-scale reaction. The kinetic plot in Figure 2.7 indicates that with the 40  conventional initiator ATRP of HEMA is controlled with this catalyst system. The final conversion of the reaction is as high as 97% after 24 hours. With a controlled ATRP catalytic system in hand, we decide to use the model initiator BisphenolS-I to further test the controlled character of this polymerization chemistry. Using BisphenolS-I as the initiator, we employed a Cu(I)/Cu(II) Figure 2.8. ATRP kinetic plot for polymerization of HEMA using ethyl BisphenolS-I as the initiator and CuCl/CuCl2/PMDETA as the catalyst in NMP, the initial monomer concentration was 2 M. ratio of 1/1 to further decrease the reaction rate, considering that there are two initiating sites for each BisphenolS-I molecule. The kinetic plot in Figure 2.8 41  shows that this reaction gives a final conversion of around 96% at 24 hours, but it may still show a small slope decrease (some termination of polymerization) at around 100 minutes. Nevertheless, we felt that with the high conversion and reasonable reaction kinetics, these polymerization conditions were sufficient to proceed with HEMA polymerization from the macroinitiator. 2.4. Polymerization from a PES Macroinitiator Figure 2.9. ATRP kinetic plot for polymerization of HEMA using a PES macroinitiator and CuCl/CuCl2/PMDETA as the catalyst in NMP, the initial monomer concentration was 1.3 M. 42  The kinetic plot in Figure 2.9 shows that polymerization kinetics are well controlled (the plot is linear) when using a macroinitiator. The reaction reaches 92% conversion after 24 hours. Ideally to get a block copolymer with a desired hydrophilic block content, we could stop the polymerization at the time (determined from Figure 2.9) required for sufficient conversion of the hydrophilic monomer. In practice, we employ NMR spectroscopy to monitor the conversion in each reaction. (a) 1 Figure 2.10. H NMR 500 MHz spectra of (a) NMP 43  Figure 2.10 (cont’d) (b) (c) Figure 2.10. 1 H NMR 500 MHz spectra of (a) NMP, (b) HEMA26-PES42- HEMA26 and (c) macroinitiator in DMSO-d6. 1 Figure 2.10 shows H NMR spectra of the polymerization solvent NMP, 44  HEMA26-PES42-HEMA26 and macroinitiator. The macroinitiator spectrum (Figure 2.10c) shows a signal from the termini of the initiator structure (Hc), but this peak is too weak to deduce molecular weight. Comparison of spectra (a) and (b) shows that the small peaks at 1.9, 2.2, 2.7 and 3.3 ppm in the copolymer spectrum correspond to residual NMP solvent. This high-boilingpoint solvent is nearly impossible to remove from theses amphiphilic copolymers. Nevertheless, the spectrum of the copolymer (Figure 2.10b) contains the expected characteristic signals from both PES and polyHEMA blocks. The Ha and Hb signals stem from the aromatic protons in PES, whereas the He peak results from the methyl groups in the polymerized methacrylate structure These three sets of peaks are separated from other peaks in the spectrum and allow us to quantify the block copolymer composition. The ratio of polyHEMA repeat units to PES repeat units is simply 1/3 the area for the He peak divided by 1/4 the areas of the Ha or Hb peak. We then used the Mn value of the PES block (determined from GPC) to calculate an approximate composition of the copolymer. 45  Polymer Samples Scale (g) Mn of the starting PES (GPC) Mn Mn (NMR) (GPC) PDI (GPC) HEMA9-PES51-HEMA9 10 11,900 14,300 19,800 1.7 HEMA13-PES42-HEMA13 14 9,650 13,000 19,600 1.6 HEMA26-PES42-HEMA26 18 9,650 16,300 23,100 1.4 HEMA22-PES34-HEMA22 15 7,800 13,600 19,200 1.4 HPMA12-PES42-HPMA12 18 9,650 13,000 21,900 1.4 HPMA26-PES42-HPMA26 12 9,650 17,100 24,900 1.5 *End group signals were too weak to determine the PES Mn value from NMR spectra. The NMR Mn values assume that the PES molecular weight is that determined from GPC. Table 2.2. Block Copolymers with their Molecular Weights and PDI Values. Table 2.2 shows all the PES-based ABA block copolymer samples synthesized in this work, including 4 samples of polyHEMA-co-PES-copolyHEMA and 2 samples of polyHPMA-co-PES-co-polyHPMA. By optimizing the ATRP catalytic system in the previous section, we synthesized the copolymer on a scale of 10~20 g. To clarify the nomenclature of the copolymers, PES indicates the middle block (B block) is composed of polyethersulfone and HPMA or HEMA means that the A block is composed of 46  either polyHPMA or polyHEMA. The number after PES is number of PES repeating units in the B block, determined from GPC analysis of the synthesized PES samples; the number after HPMA or HEMA indicates the repeating units in theses blocks as determined from the NMR integration data for the copolymer. Appendices 1 to 6 contain the NMR spectra for all the block copolymers. 2.5. GPC Analysis of HEMAn-PESm-HEMAn and HPMAn-PESm-HPMAn Block Copolymers Figure 2.11. Gel-permeation chromatograms of (a) PES51 and (b) the copolymer HEMA9-PES51-HEMA9. Figure 2.11 shows gel-permeation chromatograms curves of PES and HEMA9-PES51-HEMA9. Curve a, which is the chromatogram of the PES 47  precursor used to synthesize HEMA9-PES51-HEMA9, shows a major peak corresponding to Mn of 11,900. The small peak at around 5.5 min suggests a Mn of 1,000,000, which corresponds to a small amount of aggregation of PES polymers in DMF. Curve b, the chromatogram of HEMA9-PES51-HEMA9, exhibits a major peak corresponding to an Mn of 19,800. Thus the Mn of the polymer increased from 11,900 to 19,800 upon addition of the HEMA blocks. This suggests that the molecular weight of each polyHEMA block is 3,450, but 9 units of HEMA have a molecular weight of only 1170. We used DMF as the GPC solvent and polystyrene chains as standards, and previous studies suggest that the molecular weights of poly-methacrylate polymers are 29 exaggerated by a factor of two in this GPC condition. This along with possible differences in PES and HEMA9-PES51-HEMA9 aggregation may account for the difference between the calculated GPC and NMR values for the length of the polyHEMA blocks. 48  Figure 2.12. Gel-permeation chromatograms of (a) PES42 (b) copolymer HEMA13-PES42-HEMA13 and (c) HEMA26-PES42-HEMA26. HEMA13-PES42-HEMA13 and HEMA26-PES42-HEMA26 were synthesized from the same PES precursor, which exhibits an Mn value of 9,650 based on the GPC peak centered around 7.6 min (Figure 2.12a). This precursor polymer also shows an aggregation peak at around 5 min in the chromatogram, and we did not use this peak in calculating the molecular weight. The major peaks from the unaggregated copolymers (Figure 2.12b and 2.12c) appear at smaller elution times than for the precursor, consistent with the expected increases in molecular weight. The copolymers also show large aggregation peaks at smaller elution times, and surprisingly the size of the aggregation peaks increases with the amount of HEMA in the polymer. Thus, with an increase in the hydrophilic content, more aggregation occurs. Mn values calculated based on the unaggregated peaks in Figure 2.12 increase from 9,650 for the PES 49  precursor to 13,000 and 16,300 for HEMA13-PES42-HEMA13 and HEMA26-PES42-HEMA26, respectively. As mentioned above, these values likely grossly overestimate the amount of HEMA in the polymer. Figure 2.13. Gel-permeation chromatograms of (a). PES34 and (b). HEMA22- PES34-HEMA22. HEMA22-PES34-HEMA22 was synthesized from the PES with a molecular weight of 7,800 (PES34), and has the highest hydrophilic content of any of the polyHEMA-containing copolymer, 46.2 wt% polyHEMA. Figure 2.13 shows that the major GPC peak for this copolymer shifts to a smaller elution time relative to the precursor, indicating a molecular weight increase from 7,800 to 13,600. Despite the high hydrophilic content of this copolymer, we did not observe significant aggregate peaks, probably because of the relatively low molecular weight of the PES. 50  Step-growth polymerizations, such as the synthesis of PES, typically yield PDI values around 2.0. As expected, the overall PDI decreased after formation of the copolymer (Table 2.2) because polyHEMA grown by ATRP has PDI values near 1.0. Also, increasing the polyHEMA content lowers the PDI of the final copolymer. 1.7 for HEMA9-PES51-HEMA9, 1.6 for HEMA13-PES42HEMA13, 1.4 for HEMA26-PES42-HEMA26 and 1.3 for HEMA22-PES34HEMA22. Equation (2) was used to correlate the PDIs of each block, with the 30 PDI of the entire block copolymer.  2    2 PDIAB=ωA (PDIA-1)+ωB (PDIB-1) + 1 (2) In this equation, PDIAB is the PDI of the copolymer, ωA and PDIA are the weight fraction and PDI of the A block, and ωB and PDIB are the weight fraction and PDI of the B block. Figure 2.14. An ABA block copolymer with monodisperse A block and polydisperse (PDI=2.0) B blocks. 51  Calculated PDI Polymer Sample PDI from GPC HEMA9-PES51-HEMA9 1.7 1.79 HEMA13-PES42-HEMA13 1.6 1.69 HEMA26-PES42-HEMA26 1.4 1.49 HEMA22-PES34-HEMA22 1.4 1.41 via equation (3) Table 2.3. PDIs of various copolymers as determined from GPC data or calculated using equation (2). Using equation (2), and assuming the polyHEMA blocks are monodisperse (PDIA=1.0), we calculated the PDIs of the four polyHEMA-co-PES-copolyHEMA samples (Table 2.3). In the calculation, we used PDI values from GPC for the PES block. The calculated and experimental results match very well for each sample. 52  Figure 2.15. Gel-permeation chromatograms of (a) PES42, (b) HPMA12- PES42-HPMA12 and (c) HPMA26-PES42-HPMA26. Similar to copolymers with HEMA, Figure 2.15 shows that the HPMA12-PES42-HPMA12 and HPMA26-PES42-HPMA26 copolymers have increased molecular weights compared to PES, with the major peak shifted to smaller elution times. Aggregation peaks also appear for these copolymers. As mentioned above, GPC with polystyrene standards gives much higher molecular weights than NMR analysis. 2.6. Infrared Spectra of the Block Copolymers 53  Figure 2.16. IR spectra of (a) PES, (b) polyHEMA and (c) HEMA22- PES34-HEMA22. Figure 2.176 shows IR spectra of PES, polyHEMA and the copolymer HEMA22-PES34-HEMA22. In the spectrum of PES, sulfone groups (SO2) give -1 -1 rise to strong antisymmetric (1369-1290 cm ) and symmetric (1170-1120 cm ) stretching modes. The PES backbone contains para--substituted benzene -1 structures that show an absorption maximum around 817 cm due to the out-of-plane CH bending. Although the PES we prepared has phenol end groups, there are not corresponding OH stretching bands between 3500 and -1 3200 cm . Because the phenol only exists at the chain terminus, the small amount of phenol groups does not give a strong signal in the IR spectrum. In the IR spectrum of polyHEMA, the hydrogen-bonded OH stretching mode of the hydroxyl group in HEMA repeating units gives rise to a strong broad 54  -1 band between 3500 and 3200 cm ; the CH2 antisymmetric stretch appears -1 between 2940 and 2915 cm ; the strong C=O stretching band shows up near -1 1750 cm , and the symmetric C-O-C stretch at low frequency absorbs -1 between 1160 and 1000 cm . The IR spectrum of HEMA22-PES34-HEMA22 is essentially a sum of the PES and polyHEMA spectra. Figure 2.17. IR spectra of (a) HEMA9-PES51-HEMA9, (b) HEMA13-PES42- HEMA13, (c) HEMA26-PES42-HEMA26 and (d) HEMA22-PES34-HEMA22. Figure 2.17 shows IR spectra of the four polyHEMA-co-PES-co-polyHEMA copolymer samples synthesized in this work. The fraction of polyHEMA in the copolymer increases from a to d. In each sample, the spectrum contains characteristic signal s from both PES (1369-1290 cm -1 and 1170-1120 cm -1 from sulfone groups, 817 cm-1 from para-substituted benzene structures) and polyHEMA (strong broad band between 3500 and 3200 cm-1 from hydroxyl 55  -1 groups, 1750 cm from C=O carbonyl groups) blocks. The four spectra were -1 normalized using the peak at 817 cm , which corresponds to the out-of-plane CH bending on the para-substituted benzene structure in the PES block. On going from spectrum a to d, that intensity of the broad peak between 3500 and 3200 cm -1 (OH strectch of HEMA), and the absorbance at 1750 cm -1 (C=O stretch in HEMA) increase, reflecting increasing polyHEMA content. Figure 2.18. IR Spectra of (a). PES, (b). polyHPMA and (c). HPMA12-PES42- HPMA12. Figure 2.18 presents the IR spectra of PES, polyHPMA and HPMA12-PES42-HPMA12. Similar to polyHEMA, In the IR spectrum of polyHPMA, the hydroxyl group gIves rise to a strong broad band between -1 3500 and 3200 cm ; the CH2 antisymmetric stretch leads to a band between -1 -1 2940 and 2915 cm ; the C=O stretching band appears near 1750 cm , and the symmetric C-O-C stretch at low frequency absorbs between 1160 and 56  -1 1000 cm . The IR spectrum of HEMA12-PES42-HEMA12 is essentially a sum of the PES and polyHPMA spectra as expected.   Figure 2.19. IR spectra of (a)HPMA12-PES42-HPMA12 and (b) HPMA26- PES42-HPMA26.   Figure 2.19 shows IR spectra of the two polyHPMA-co-PES-co-polyHPMA copolymer samples synthesized in this work. As with polyHEMA-co-PES-co-1 polyHEMA, the two spectra were normalized using the peak at 817 cm , which corresponds to out-of-plane CH bending on the para-substituted benzenes in the PES block. The intensity of the broad peak between 3500 and 3200 cm -1 -1 (hydroxyl group in HPMA) and the C-O stretch (1750 cm ) are larger for the copolymer containing more HPMA, consistent with the NMR characterization. 2.7. TGA of the Block Copolymers 57  Figure 2.20. TGA data for PES and several polyHEMA-co-PES-co-polyHEMA o samples heated under air (samples were held at 120 C until the weight was o constant (~ 3 hours) before heating at 10 C/min) (a) PES, (b). HEMA9-PES51HEMA9, (c) HEMA13-PES42-HEMA13, (d) HEMA26-PES42-HEMA26 and (e) HEMA22-PES34-HEMA22. Figure 2.20 shows TGA of PES and the five polyHEMA-co-PES-copolyHEMA copolymers heated under air. In each case, the sample was held at o 120 C for around 3 hours before starting the heating. All polyHEMA containing o copolymers, show a three-stage weight loss. The weight loss before 250 C probably corresponds partly to loss of the NMP, which could not be removed from the copolymer. In the second stage, the weight loss between 250 and 500 o C corresponds to degradation of the polyHEMA block. The final stage of o weight loss between 500 and 700 C results from degradation of the PES block, as the PES sample (curve a) shows major degradation only in this temperature 58  region. In all samples, the weight loss curve ends at zero weight retention, which indicates that there is almost no residual metal catalyst (copper) in the copolymer. The fraction of polyHEMA in the copolymers increases from b o through e, so the fraction of the weight loss that occurs from 500 to 700 C decreases from b through e because of the lower PES content when samples contain more polyHEMA.   Figure 2.21. TGA data for PES, polyHEMA and polyHEMA-co-PES-copolyHEMA samples heated under N2 (samples were held at 120 o C until o weight was constant (~ 3 hours) before heating at 10 C/min). (a) PES, (b) polyHEMA, (c) HEMA9-PES42-HEMA9, (d) HEMA13-PES42-HEMA13, (e) HEMA26-PES42-HEMA26 and (f) HEMA22-PES34- HEMA22. Figure 2.21 shows TGA of the five polyHEMA-co-PES-co-polyHEMA copolymers and PES heated under N2. In each case, the sample was held at 59  o 120 C for around 3 hours before starting the heating scan. Curve a is the weight loss of pure PES, which eventually carbonizes under nitrogen and has a final weight retention of around 40 wt%. This PES sample was synthesized in the lab and has a Mn of 10,000 and a PDI of 2.0. TGA of of pure polyHEMA o (curve b) shows a complete weight loss before 550 C, and the major o degradation occurs between 350 and 550 C. The polyHEMA sample was synthesized via free radical polymerization, and GPC analysis shows Mn = 57,400 and PDI = 3.6. For the polyHEMA-co-PES-co-polyHEMA copolymers, based on the TGA of pure PES and polyHEMA, all of the polyHEMA in the copolymer should be gone at the end of the analysis, and the PES content should have 40% weight retention. Accordingly, the PES content should be the final weight retention divided by 0.4, and polyHEMA and residual solvent should make up the rest of the polymer. Sample polyHEMA content from NMR polyHEMA content from TGA* HEMA9-PES51-HEMA9 11.0 wt% 15.4 wt% HEMA13-PES42-HPMA13 17.0 wt% 16.7 wt% HEMA26-PES42-HEMA26 30.0 wt% 33.3 wt% HEMA22-PES34-HEMA22 46.2 wt% 53.0 wt% *Calculations are described in experimental section. Table 2.4. PolyHEMA Content in Copolymer Samples as Calculated from NMR 60  Spectra and TGA under N2. Table 2.4 shows the compositions of the four polyHEMA-co-PES-copolyHEMA samples as calculated from both NMR and TGA. The results from the two characterization techniques are quite consistent.   Figure 2.22. TGA data for PES and polyHPMA-co-PES-co-polyHPMA o samples heated under air (samples were held at 120 C until the weight was o constant (~ 3 hours) before heating at 10 C/min) (a) PES, (b) HPMA12PES42-HPMA12 and (c) HPMA26-PES42-HPMA26. Figure 2.22 shows TGA curves of the two polyHPMA-co-PES-co-polyHPMA copolymers and PES heated under air. As mentioned above, curve a shows a two-stage degradation of PES. The two polyHPMA-containing copolymers show a similar 3-stage weight loss pattern to polyHEMA-co-PES-copolyHEMA. In all samples, the TGA curve ends at zero weight retention, again indicating that there is little residual metal catalyst (copper) in the copolymer 61  sample. The fraction of polyHPMA in the copolymers increases from b to c, so o the fraction of the weight loss that occurs from 500 to 700 C decreases from b to c because of the lower PES content when samples contain more polyHPMA. Figure 2.23. TGA data for PES and polyHPMA-co-PES-co-polyHPMA o samples heated under N2 (samples were held at 120 C until weight was o constant (~ 3 hours) before heating at 10 C/min) (a) PES, (b) polyHPMA, (c) HPMA12-PES42-HPMA12 and (d) HPMA26-PES42-HPMA26. Figure 2.23 shows TGA of PES, polyHPMA and polyHPMA-co-PES-copolyHPMA with heating under N2. Curve b shows that the weight loss of pure o polyHPMA is complete before 550 C. The polyHPMA sample was synthesized via free radical polymerization, and GPC analysis shows Mn = 34,646 and PDI = 1.8. As mentioned above, PES retains 40% of its weight, presumably due to carbonization. Thus, similar to polyHEMA-co-PES-co-polyHEMA, we can 62  determine the fraction of polyHPMA in the copolymers from the final weight retention in the TGA. Sample polyHPMA content from NMR polyHPMA content from TGA HPMA12-PES42-HPMA12 26.3 wt% 33.5 wt% HPMA26-PES42-HPMA26 43.3 wt% 43.0 wt% *Calculations are described in experimental section. Table 2.5. PolyHPMA Content in Copolymer Samples as calculated from NMR spectra and TGA. Table 2.5 shows the calculated compositions of the two polyHPMA-coPES-co-polyHPMA samples, as determined from NMR and TGA data. Again, the results from the two characterization techniques are reasonably consistent. 2.8. Critical Water Content (CWC) Values of PES and PES-based Block Copolymers In one process to prepare micelles using amphiphilic block copolymers, the copolymers are first dissolved in a good solvent for both blocks, for example N, N-dimethylformamide (DMF), dimethylsulfoxide (DMSO), N, N-dimethyl acetamide (DMAc) or methylpyrrolidone (NMP). In a subsequent step, deionized water is added slowly and the quality of the resulting co-solvent for the PES block gradually decreases. At a certain level of water, the PES blocks start to associate/aggregate to form micelles in the solution and the solution becomes visually turbid. This transition is abrupt, and I further stirred the 63  mixture for another 30 min to make sure the turbidity maintains. The fraction of deionized water at which the micellization process starts is defined as the 25,26 critical water content (CWC). Terminal Sample Source a Mw Ultrason 2020P BASF 27,000-37,000 2.5-3.0 OMe, Cl Ultrason 6020P BASF 46,000-55,000 3.0-4.0 OMe, Cl BASF 45,000 3.0 OH, Cl Solvay 21,000 3.0 OH, Cl Solvay 43,000-50,000 3.0 OH, Cl Ultrason E 2020P SR Micro Veradel 3600 RP Veradel 3000 RP a Values from GPC characterization PDI b Structure b Characterized using titration by suppliers Table 2.6. Commercial Polyethersulfones and their Molecular Weights, PDI Values and Terminal Groups. Although CWC is characteristic micellization for the block copolymers, for pure PES, the CWC corresponds to the point where the polymer separates from the mixture. The CWC of the hydrophobic block (PES) plays a major role in the block copolymer micellization process. Figure 2.24 shows the o room-temperature (20 C) CWC data for five different PES materials: Ultrason 2020P, Ultrason 6020P, Veradel 3600RP, Veradel 3000RP and home-made PES in four different solvents (NMP, DMF, DMSO, DMAc). Table 2.6 presents the composition of the commercial polymers, and the PES we synthesized has 64  terminal hydroxyl groups, a molecular weight (Mn) of 10,000 and a PDI of 2.0. (a). (1) DMSO, (2) DMF, (3) DMAc, (4) NMP. (b) (1) DMSO, (2) DMF, (3) DMAc, (4) NMP. Figure 2.24. Room-temperature critical water content (CWC) values of different PES materials in four solvents (NMP, DMF, DMSO, DMAc). 65  Figure 2.24 (cont’d) (c) (1) DMSO, (2) DMF, (3) DMAc, (4) NMP. (d) (1) DMSO, (2) DMF, (3) DMAc, (4) NMP. 66  Figure 2.24 (cont’d) (e) (1) DMSO, (2) DMF, (3) DMAc, (4) NMP. As Figure 2.24 shows, there is a linear relation between CWC data and the logarithm of the polymer concentration, which can be written as equation (3) CWC = -Alogc0 + B (3) where c0 is the initial copolymer concentration and A and B are constants for a specific copolymer. At the critical water content, the value of the copolymer concentration is the critical micelle concentration if micellization occurs. Figure 2.24 shows that in each polymer sample, the order of CWC values in different organic solvents is NMP>DMAc>DMF>DMSO. This could be explained by the solubility parameters in Table 2.7. The solubility parameter (δ) gives a numerical estimate of the degree of interaction between materials and is a good indicator of solubility. Materials with similar values of δ are more likely to be miscible. 67  d Solvent δ (MPa 1/2 p ) δ (MPa 1/2 h ) δ (MPa 1/2 ) δ (MPa 1/2 a 18.0 12.3 7.2 22.9 a 17.4 13.7 11.3 24.8 a 18.4 16.4 10.2 26.6 a 16.8 11.5 10.2 22.7 a 15.5 16.0 42.4 47.9 19.3 0 7.0 20.3 19.6 10.8 9.2 24.2 15.1±0.7 11.9±0.3 18.8±0.4 ) 26.9±0.5 NMP DMF DMSO DMAc Water a PSu (Udel) b PES c polyHEMA a. Brandrup, J.; Immergut, E. H.; Grulke, E. A. Polymer Handbook, 4 th ed., vol. 3. New York: Wiley-Interscience; 1999. b. Hansen, C. M. Hansen Solubility Parameters: A User’s Handbook, 2 nd ed., Boca Raton, Fla: CRC Press; 2007. c. Caykara, T.; Ozyurek, C.; Kantoglu, O.; Guven, O. J. Polym. Sci. B: Polym. Phy. 2002, 40, 1995-2003. o Table 2.7. Solubility Parameters of Several Solvents and Polymers at 25 C. There are three components that contribute to a total solubility parameter: a dispersion force component (δd), a hydrogen bonding component (δh) and a polar component (δp). The square of the overall solubility parameter (δ) is the sum of the squares of the component solubility parameters. 2 2 2 δ = δd + δh + δp 2 Table 2.7 shows the solubility parameters of solvents and polymers used in 68  this study. Both NMP (δ=22.9 MPa 1/2 ) and DMAc (δ=22.7 MPa solubility parameter than PES (δ=24.2 MPa (δ=47.9 MPa 1/2 1/2 1/2 ) have lower ). Initially, addition of water ) will increase the co-solvent solubility parameter, making it closer to that of PES and increasing solubility of PES in the co-solvent. However, further addition of the water will increase the solubility parameter of the co-solvent beyond that of PES to decrease PES solubility. DMF (24.8 MPa 1/2 1/2 ) has a solubility parameter that is very close to PES (24.4 MPa ), so addition of water only decreases the PES solubility in DMF; DMSO (26.6 MPa 1/2 ) has a larger solubility parameter than PES and thus this solvent requires the least amount of water to precipitate PES. Figure 2.25. Room-temperature CWC values for several PES samples in NMP. (1) Ultrason 6020P, (2) Veradel 3000RP, (3) Ultrason 2020P, (4) Veradel 3600RP and (5) home-made PES. (Each data point is an average from three 69  independent samples, and the data points often obscure the error bars.) Figure 2.25 compares CWC values of different polymers in NMP. The CWC values decrease in the order home-made PES>Veradel 3600RP>Ultrason 2020P>Veradel 3000RP>Ultrason 6020P. CWC depends on temperature, the nature of the organic solvent, and polymer concentration and molecular weight. Veradel 3600RP and Ultrason 2020P have similar molecular weights that are smaller than those of Veradel 3000RP and Ultrason 6020P, so Veradel 3600RP and Ultrason 2020P have larger CWC values than Veradel 3000RP and Ultrason 6020P. Most of the termini in Veradel polymers are OH, whereas Ultrason polymers have OMe and Cl terminal groups. The larger portion of OH end groups increase the CWC of Veradel polymers relative to Ultrason when molecular weights are equal. Our synthesized PES has a low PDI (~2) relative to the commercial available materials (PDI=3~4). The broad molecular weight distribution in the commercial polymers means that a portion of the sample has very high molecular weight and this fraction of the polymer chains should precipitate most readily from water/organic co-solvents. Thus the CWC value is smaller for the commercial polymers than for our PES. 70  Figure 2.26. Room-temperature CWC values in NMP for (a) PES, (b) HPMA26-PES42-HPMA26 and (c). HEMA22-PES34-HEMA22. (Each data point is an average from three independent samples, and the data points often obscure the error bars.) Figure 2.26 shows room-temperature CWC data for the pure PES, HPMA26-PES42-HPMA26 and HEMA22-PES34-HEMA22 in NMP. Both copolymers have higher CWC values than PES, due to the higher hydrophilicity of the copolymers. Although HPMA26-PES42-HPMA26 has a similar hydrophilic polymer content (43.3 wt% polyHPMA) to HEMA22-PES34HEMA22 (46.2 wt% polyHEMA), it has smaller CWC values than HEMA22PES34-HEMA22. HPMA and HEMA differ only by a methyl group, but HEMA is a more hydrophilic monomer than HPMA. 2.9. Aggregation of Block Copolymers in Dilute Solutions—NMR and DLS Studies 71  The previous sections show that we can successfully synthesize polyHEMA-co-PES-co-polyHEMA and polyHPMA-co-PES-co-polyHPMA ABA block copolymers on scales of 10-20 g. Moreover, the block copolymers precipitate or form micelles upon addition of water to organic solvents. In the following discussion, we use HEMA22-PES34-HEMA22 to study the copolymer aggregation behavior in solvents that are selective for one of the blocks. Initially, a good solvent (DMSO-d6 in NMR studies, NMP in DLS studies) for both hydrophobic and hydrophilic blocks is dissolves the polymer, and then we add water to induce aggregation of the block copolymer. Since PES is the hydrophobic block, water/DMSO or water/NMP co-solvents will selectively dissolve the polyHEMA block. 1 H NMR spectroscopy can conveniently reveal aggregation. Because the PES block aggregates while the polyHEMA remains dissolved, signals from PES should decrease with increasing water content while signals from polyHEMA should remain relatively constant. 72    Chemical Structure of HEMA22-PES34-HEMA22   DMSO-d6      DMSO-d6/D2O 19:1 (volume ratio) Figure 2.27. 1 H NMR 500 MHz Spectra of HEMA22-PES34-HEMA22 in DMSO-d6/D2O co-solvents with varying amounts of water. 73  Figure 2.27 (cont’d)   DMSO-d6/D2O 9:1 (volume ratio)    DMSO-d6/D2O 4:1 (volume ratio) Figure 2.27 shows the NMR spectra of HEMA22-PES34-HEMA22 in DMSO-d6 or mixtures of DMSO-d6 and D2O with different volume ratios. The peaks between 0.7 and 0.9 ppm (He) are the characteristic signals of methyl groups in polymerized methacrylate monomers, so all of the four spectra are normalized using this peak whose integral is set to 3.0. The two peaks at 74  7.2-7.5 ppm (Hb) and 7.8-8.1 ppm (Ha) stem from the aromatic protons in the PES block, and each of the these peaks should correspond to 4 protons per repeat unit. Solvent (v/v) Integral of He Integral of Hb Integral of Ha (polyHEMA) (PES) (PES) DMSO-d6 3.0 2.71 2.62 DMSO-d6/D2O (19/1) 3.0 2.16 2.26 DMSO-d6/D2O (9/1) 3.0 0.75 0.84 DMSO-d6/D2O (4/1) 3.0 0.30 0.40 1 Table 2.8. Evolution of the H NMR signals from HEMA22-PES34-HEMA22 with increasing of D2O content in the DMSO-d6 solvent. See the structure in Figure 2.27 for proton assignments. Signals were normalized to that of He. Table 2.8 shows integrals of three sets of peaks obtained with co-solvents at different ratios. Clearly, as the D2O content in the solvent increases, NMR signals corresponding to the PES block decrease, presumably because the PES block aggregates to form the core of a micelle. The semi-solid state micelle core loses its NMR sensitivity due to magnetic gradients, dipolar broadening, and slow relaxation in the micelle core. The hydrophilic polyHEMA block, on the other hand, form the corona part of the micelle and stays mostly solvated. The signal to noise ratios in the NMR spectra are similar in pure DMSO-d6, DMSO-d6/D2O (19:1) and DMSO-d6/D2O (9:1), because the 75  copolymer is quite soluble in these three different co-solvents. However, when the D2O content increases to 20% by volume, the copolymer become insoluble and the signal to noise ratio in the spectrum declines. Visually the NMR solution in the tube also becomes turbid. DLS provides values for the sizes of aggregates in solution. Because of the high CWC values (Figure 2.24), we selected NMP/water co-solvents to study the aggregation behavior over a wide range of water contents. HEMA22-PES34-HEMA22 was first dissolved in NMP, and deionized water was added dropwise to reach desired NMP/water ratio. The solution was further stirred for 30 min before DLS characterization. Figure 2.28. DLS size distributions for HEMA22-PES34-HEMA22 (1 mg/mL at o 25 C) dissolved in NMP and NMP/water co-solvents with different volume ratios. 76  Solvent Radii at (v/v) Peak 1 (nm) NMP Volume % of aggregates Radii at Volume % of aggregates In Peak 1 Peak 2 (nm) 3.7 99.8% X X NMP/water (19/1) 4.9 99.8% X X NMP/water (9/1) 7.1 96.5% X X NMP/water (4/1) 59.6 34.7% 2,581 65.3% In Peak 2 Table 2.9. DLS particle sizes and particle volume percentages for aggregates of HEMA22-PES34-HEMA22 formed in NMP/water mixtures. The mixtures contained 1 mg/mL HEMA22-PES34-HEMA22. Figure 2.28 and Table 2.9 show the particle size distributions that form in mixtures containing 1 mg/mL of HEMA22-PES34-HEMA22 in NMP/water co-solvents. The particle size distribution in pure NMP (curve a, Figure 2.28) shows a single peak centered around 3.7 nm. This corresponds to a fully dissolved polymers, since NMP is a good solvent for both PES and polyHEMA blocks. Curve b shows the size distribution when the polymer is dissolved in NMP/water (with a 19/1 volume ratio). With this small amount of water, the solvent mixture is still a good solvent for the copolymer, so the DLS measurement shows a fully dissolved state with monomodal radii centered around 4.9 nm. At a water content of 10 % by volume (curve c, Figure 2.28), the copolymer remains solvated, but the most common particle radii increases to 7.1 nm. As the solvent becomes poorer and some molecules begin 77  interacting, the dissolved molecule sizes increases by a factor of two on going from pure NMP to NMP with at 10 vol% water. Below 10 vol% water, a few aggregates with radii of 1,000-2,000 nm also exist, but they constitute less than 5 vol% of the total aggregates. Finally when the water content reaches 20 vol% (curve d, Figure 2.28), all of the copolymer form micelles with a bimodal size distribution. Aggregates with radii centered around 59.6 nm and 2,000 nm account for 34.7 vol% and 65.3 vol% of the particles, respectively. Figure 2.29. DLS size distributions for HEMA22-PES34-HEMA22 (10 mg/mL at o 25 C) dissolved in NMP and NMP/water co-solvents with different volume ratios. 78  Solvent Radii at (v/v) Peak 1 (nm) NMP Volume % of aggregates Radii at Volume % of aggregates In Peak 1 Peak 2 (nm) 3.6 99.8% X X NMP/water (19/1) 4.0 100% X X NMP/water (9/1) 5.9 96.3% X X NMP/water (4/1) 56.3 100% X X In Peak 2 Table 2.10. DLS particle sizes and particle volume percentages for aggregates of HEMA22-PES34-HEMA22 formed in NMP/water mixtures. The mixtures contained 10 mg/mL HEMA22-PES34-HEMA22. Figure 2.29 and Table 2.10 show the particle size distributions that form in mixtures containing 10 mg/mL of HEMA22-PES34-HEMA22 in NMP/water co-solvents with different volume ratios. Even at this polymer concentration, the copolymer is fully dissolved in pure NMP solvent and NMP with 5 vol% and 10 vol% water. The particle sizes are essentially the same as in mixtures with 1 mg/mL HEMA22-PES34-HEMA22 (compare Tables 2.9 and 2.10). Finally when the water content reaches 20 vol% (curve d, Figure 2.29), the particle radii center around 56 nm. We are not sure why the higher polymer concentration leads to a monomodal size distribution while the 1 mg/mL solution shows a bimodal size distribution in 20 vol% water. The higher polymer concentration may lead to a higher concentration of micelles, which overwhelms the signals from a few large aggregates. 79  Figure 2.30. DLS size distributions for HEMA22-PES34-HEMA22 (30 mg/mL at o 25 C) dissolved in NMP and NMP/water co-solvents with different volume ratios. Solvent Radii at (v/v) Peak 1 (nm) NMP Volume % of aggregates Radii at Volume % of aggregates In Peak 1 Peak 2 (nm) 2.9 100% X X NMP/water (19/1) 2.8 100% X X NMP/water (9/1) 5.8 94.5% X X NMP/water (4/1) 61.3 100% X X In Peak 2 Table 2.11. DLS particle sizes and particle volume percentages for aggregates of HEMA22-PES34-HEMA22 formed in NMP/water mixtures. The mixtures contained 30 mg/mL HEMA22-PES34-HEMA22. Figure 2.30 and Table 2.11 show the particle size distribution in mixtures 80  containing 30 mg/mL of HEMA22-PES34-HEMA22 in NMP/water co-solvents with different volume ratios. The size distributions are essentially the same as those with 10 mg/mL HEMA22-PES34-HEMA22, although the radii are a little smaller with 5 vol% water. Most notably, regardless of the copolymer concentrations (1 mg/mL, 10 mg/mL or 30 mg/mL), the aggregated state with 20 vol% water content contains micelles with a radii around 60 nm. 3. Conclusion ABA block copolymer samples were synthesized on of 10~20 g scale, including 4 samples of polyHEMA-co-PES-co-polyHEMA and 2 samples of polyHPMA-co-PES-co-polyHPMA. Using a model compound 4,4'-sulfonylbis (4,1-phenylene) bis(2-bromo-2-methyl-propanoate) as initiator, we selected CuCl/CuCl2/PMDETA in NMP as the catalytic system for the polymerization of HEMA and HPMA monomers from PES macroinitiators. FT-IR spectra, TGA 1 and H NMR spectra show successful synthesis of these block copolymers. Critical Water Content (CWC) studies demonstrate that the hydrophilicity of synthesized copolymers increases with the amount of polyHEMA or polyHPMA 1 in the block copolymers as expected. Both DLS and H NMR studies show that addition of non-solvent (water) to the copolymer solution causes aggregation of the block copolymer, particularly the PES block. DLS indicates that with 20 vol% water micellar the aggregates have radii of 60 nm, regardless of the copolymer concentration. Such copolymers may prove useful in modification of polysulfone membranes. 81  4. Experimental Section 4.1. Materials Phthalimide, potassium derivative (500 g, 99%, Acros Organics), triethylamine (J. T. Baker), N, N, N’, N’, N’’-pentamethyldiethylenetri amine (PMDETA, Aldrich, 99%), 4-fluorophenyl sulfone (100 g, 99%, Acros Organics), 4, 4’-sulfonyldiphenol (500 g, 98%, Aldrich), bis(4-chlorophenyl) sulfone (500 g, 98%, Aldrich), α-bromoisobutyryl bromide (500 g, 98%, Aldrich), N, N-dimethylacetamide (2.5 L, 99%, Alfa Aesar), 1-methyl-2-pyrrolidinone (4 L, ACS grade, 99.0+%, Alfa Aesar), postassium carbonate (500 g, ACP Chemicals), 4-(dimethylamino) pyridine (100 g, 99% prilled, Aldrich), sulfolane (500 mL, Kodak) and pyridine (500 mL, ACS grade ≥99%, CCI Chemical) were used as received. Cu(I)Br (99.999%, Aldrich) was purified using a saturated aqueous NaBr solution. Hydroxypropyl methacrylate (1 kg, 97+%, mixture of isomers, stabilized, Acros Organics), 2-Hydroxyethyl methacrylate (500 mL, 200-220 ppm monomethyl ether hydroquinone as inhibitor, Aldrich, 97%) were passed through a basic alumina column before use. ULTRASON E 6020P (100 g), E 2020 P (100 g) and 2020P SR Micro (100 g) were received as free samples from BASF. Veradel 3000 RP and Veradel 3600 RP were received as free samples from Solvay Specialty Polymers. Aluminum oxide (1 kg, activated, o basic, ~150 mesh, 58 Å, Aldrich) was heated in an oven overnight at 120 C for activation. 4.2. Characterization 82  1 H NMR and 13 C NMR spectra were collected using a Varian UnityPlus-500 spectrometer using the residual proton signals from the solvent as the chemical shift standard. Thermogravimetric analysis (TGA) were carried out in air or under N2 using a Perkin-Elmer TGA 7 instrument at a heating rate of 10 o o C/min. Samples were held at 120 C until the weight stabilized (~3 h) before starting the heating process. FTIR spectra were collected with a Mattson Galaxy 300 spectrometer. FTIR samples were made by mixing the polymer with KBr and forming pellets. DLS studies were carried out on a Malvern o Zetasizer Nanoseries ZEN3600 instrument at 30 C. Standards and Conditions of Gel Permeation Chromatography (GPC) Analysis This GPC instrument contains Water 1515 Isocratic HPLC Pump, Water 717 Plus Autosampler, Waters 2414 Refractive Index Detector and Water 2487 Dual λ Absorbance Detector. This instrument uses DMF as solvent and PLgel o o 10 μm mixed-B as column, 70 C at column and 50 C at RI detector. DMF was used as the eluting solvent at a flow rate 1 mL/min, and monodisperse polystyrene standards were used to calibrate the molecular weights. The concentration of the polymer solutions used for GPC measurements was 1 mg/mL. 83  Standard Elution time (min) 6.3 7.1 7.7 8.4 PSI Molecular weight (Mn) 401,340 45,730 12,860 2,727 Elution time (min) 5.9 6.8 7.4 8.3 PSII Molecular weight (Mn) 931,780 95,800 24,150 3,680 Table 2.12. Polystyrene standards used to calibrate the results of GPC in DMF. The molecular weight and PDI of each polymer sample were determined by GPC. Two sets of polymer standards, PSI and PSII, were employed and results are shown in Table 2.12. The effective calibration range for GPC analysis is between 5.9 min (Mn of 931,780) and 8.4 min (Mn of 2,727). Calculation of polyHEMA content in Copolymers Using TGA and NMR Data The polyHEMA content from TGA was calculated using Equation (4), (4) o where W120~300 is the weight loss between 120 and 300 C (we assume this mostly corresponds to the unremoved solvent in the copolymer sample) and Wretention is the final weight retention of the sample. The value of Wretention should exclusively come from the PES block, since pure PES has a weight retention of around 40%, whereas polyHEMA and polyHPMA show no weight 84  retention at the end of TGA. Based on integration of the proton peaks in NMR, we obtained polyHEMA or polyHPMA and PES contents in the copolymer.   Scheme 2.15. Characteristic protons for calculation of copolymer composition from NMR spectra.   As shown in Scheme 2.15, to calculate polyHEMA content from 1H NMR, we compared signals from protons in the methyl groups from methacrylate (Ha, 3 protons per repeating units, 0.7~0.9 ppm), with signals from the aromatic protons at the ortho position to sulfone group (Hb, 4 protons per repeating unit, 7.4~7.5 ppm). Based on integrations of these two sets of peaks, polyHEMA content could be calculated according to equation (5).                         (5)                Where IHa is the integral of NMR peaks at 0.7-0.9 ppm (the molecular weight of the HEMA repeating unit is 130 g/mol) and IHb is the integral of the NMR peak at 7.4-7.5 ppm (the molecular weight of the PES repeating unit is 232 g/mol). Calculation of polyHPMA content in Copolymers Using TGA and NMR Data 85  The polyHPMA content from TGA was calculated using Equation (4). Similar to polyHEMA-containing copolymers, NMR could also be used to obtain the polyHPMA content.   Scheme 2.16 Characteristic protons for calculation of copolymer composition from NMR spectra. As shown in Scheme 2.16, polyHPMA gives signals from methyl groups (Ha) from methacrylate at 0.7-0.9 ppm, as well as from the side-chain methyl group (Ha’) at 0.9-1.2 ppm. The integration of the combined peaks corresponds to 6 protons. We compared this integral with that for the aromatic protons at the ortho position (Hb) to the sulfone group (4 protons per repeating unit, 7.4-7.5 ppm). Based on integrations of these sets of peaks, we calculated the polyHPMA content according to equation (6).                           (6)                 IHa+Ha’ is the integral of the NMR peaks from 0.7-1.1 ppm (the molecular weight of the HPMA repeating unit is 144 g/mol) and IHb is the integral of the NMR peak at 7.4-7.5 ppm (the molecular weight of the PES repeating unit is 232 g/mol). 86  Synthesis of PES with Hydroxyl End Groups. Bisphenol sulfone (26.00 g, 0.104 mol) and bisfluorophenyl sulfone (26.92 g, 0.104 mol 1:1 molar ratio) were added to a 500 mL 3-neck flask charged with N, N-dimethylacetamide (240 mL) and toluene (100 mL). Potassium carbonate (2 equivalents, 0.21 mol) was added, and then the solution was purged with nitrogen for 30 min. The reaction was heated to 145 °C, and a Dean-Stark trap attached to the flask removed residual water from the reactor. After 12 hours at 145 °C, bisphenol sulfone (1 g, 4 mmol) was added to the solution to cap the polymer with phenolic groups, and after an additional hour the polymer was precipitated by addition into 4M HCl. The white precipitate was dispersed in boiling methanol (350 mL) for 4 hours to extract residual monomer and salts from the polymer. The polymer was recovered by filtration, and the extraction was repeated with fresh methanol. After drying under vacuum at 60 °C overnight, the product was a white fine powder, and 48 g of product was isolated (99% yield). Synthesis of the PES Macroinitiator. Polyethersulfone (50 g, 5 mmol end groups, calculated using molecular weight from GPC assuming hydroxyl end groups) was dissolved in a mixture of DMF/DMSO (400 mL, 1:1 volume). After the polymer was dissolved, 20 equivalents of triethylamine (10.1 g, 0.1 mol) and bromoisobutyryl bromide (22.9 g, 0.1 mol) were added at 0 °C. The solution was heated to 60 °C and stirred with a mechanical stirrer for 24 hours. Upon addition to 4 M aqueous HCl, the macroinitiator precipitated as a gel. After poring off the solvent, the product was immersed in 300 mL boiling 87  methanol (sometimes more than once), and after 4h, the gel-like material solidified. The solid material was broken into small particles, and methanol extraction was performed two times using a Soxhlet extraction setup. After the final wash, the product was a fine light yellow powder. The product was dispersed in 400 mL deionized water at 60 °C and vigorously stirred for 1 h before filtration, this dispersion-filtration process was repeated twice before the solid was dried under vacuum at 60 °C overnight. Synthesis of 4,4'-sulfonylbis(4,1-phenylene) bis(2-bromo-2-methyl- propanoate) (BisphenolS-I). In a 100 mL round bottom flask, 2.55 g bisphenol sulfone (0.01 mol) and 2.1 g triethylamine (0.021 mol) were added to 50 mL dry acetone. The flask was put on an ice bath to give a solution temperature of around 0 °C. Bromoisobutyryl bromide (2.7 mL, 0.022 mol) was added to the solution slowly through a septum (typical addition time is around 5 min). Large amounts of white/light yellow salts formed in the course of addition. After the addition, the ice bath was removed and the solution was allowed to warm to room temperature. After 2 h at room temperature, 50 mL of deionized water was added to dilute the reaction solution, and the product was extracted with dichloromethane (3 x 40 mL). The combined dichloromethane layer was washed with deionized water (3 x 30 mL), dried with magnesium sulfate, and the solvent was removed using rotary evaporation. The white solid was dried in vacuum to give 5.2 g (96% yield). Purification of 2-Hydroxypropyl Methacrylate. A 200-mL round bottom flask 88  was charged with 14.4 g commercial hydroxypropyl methacrylate isomer mixture (0.1 mol), 27.8 g trityl chloride (0.1 mol), 10.1 g triethylamine (0.1 mol) and 300 mL acetone. The solution was heated to 50 °C in an oil bath and allowed to react overnight. Solvent and residual triethylamine were removed by rotary evaporation, and vacuum distillation at 90 °C at 80 mtorr gave the unreacted secondary alcohol isomer (4.2 g, yield 29.2%). Synthesis of ABA block copolymers: polyHEMA-co-PES-co-polyHEMA. In a 250 mL Schlenk flask, the PES macroinitiator (15 g, 1.5 mmol), HEMA (20 g, 0.154 mol) and PMDETA (0.60 mL, 2.9 mmol) were dissolved in NMP (120 mL). (The stirred mixture occasionally required heating to obtain a homogeneous solution.) After two freeze-pump-thaw cycles, the flask was refilled with nitrogen, CuCl2 (0.108 g, 0.8 mmol) was added and the solution was stirred until the Cu species dissolved. After two freeze-pump-thaw cycles, the flask was refilled with nitrogen and CuCl (0.078 g, 0.8 mmol) was added to the flask under N2. The flask was heated in a 40 °C oil bath. The 1 reaction was monitored by H NMR and when the polymerization reached the desired HEMA conversion, the solution was opened to the air to quench polymerization reaction. The polymer was precipitated by addition of the solution into 300 mL of saturated disodium ethylenediaminetetraacetate (EDTA) solution, and the copolymer was recovered as a fine powder. Suction filtration required 5-6 hours to recover the product. The air-dried product was dissolved in DMSO, precipitated in water, and after a second precipitation from DMSO, 89  the polymer was dried under vacuum at 90 °C for 24 h. Synthesis of polyHPMA-co-PES-co-polyHPMA occurred similarly. Free Radical Polymerization of Hydroxyethyl Methacrylate (HEMA). A Schlenk flask was charge with 15 mL dry THF, 4 g of HEMA (0.031 mol) and 0.43 g azobisisobutyronitrile (2.6 mmol). After one freeze-pump-thaw cyle, the flask was put onto a 70 °C oil bath. After an overnight reaction, NMR spectroscopy of the reaction solution showed that the conversion was 100%, and the flask was opened to air to quench the reaction. PolyHEMA precipitated onto the bottom of the flask, and the solid product was dried under vacuum at 70 °C. 3.56 g product was recovered (90% yield).                 90                      APPENDIX A     91        Appendix A1. 1 H NMR 500 MHz spectrum of HEMA9-PES51-HEMA9 in DMSO-d6. 92    1 Appendix A2. H NMR 500 MHz spectrum of HEMA13-PES42-HEMA13 in DMSO-d6. 93    1 Appendix A3. H NMR 500 MHz spectrum of HEMA26-PES42-HEMA26 in DMSO-d6. 94    1 Appendix A4. H NMR 500 MHz spectrum of HEMA22-PES34-HEMA22 in DMSO-d6. 95    1 Appendix A5. H NMR 500 MHz spectrum of HPMA12-PES42-HPMA12 in DMSO-d6. 96  1 Appendix A6. H NMR 500 MHz spectrum of HPMA26-PES42-HPMA26 in DMSO-d6. 97                        REFERENCES   98  REFERENCES 1. Meier-Haack, J.; Vogel, C.; Butwilowski, W.; Lehmann, D., Sulfonated poly(ether sulfone)s for fuel cells by solvent-free polymerization. Pure and Applied Chemistry 2007, 79 (11), 2083-2093. 2. Locatelli, F. R., C.; Tetta, C. , Polyethersulfone: membranes for multiple clinical Applications. Karer AG: Basel, 2003. 3. Zeeman; Zydney, A. L. M. D., INC. New York, Microfiltration and ultrafiltration: principles and applications. 1996. 4. Perry, R. H. G., D. W.; Eds McGraw-Gill: New York, 1997, Perry's chemical engineer's handbook, 7th ed. 1997. 5. Beyer, M.; Lohrengel, B.; Nghiem, L. D., Membrane fouling and chemical cleaning in water recycling applications. Desalination 2010, 250 (3), 977-981. 6. Peyravi, M.; Rahimpour, A.; Jahanshahi, M.; Javadi, A.; Shockravi, A., Tailoring the surface properties of PES ultrafiltration membranes to reduce the fouling resistance using synthesized hydrophilic copolymer. Microporous and Mesoporous Materials 2012, 160 (0), 114-125. 7. (a) Schacher, F. H.; Rupar, P. A.; Manners, I., Functional block copolymers: nanostructured materials with emerging applications. Angewandte Chemie-International Edition 2012, 51 (32), 7898-7921; (b) Moad, G.; Rizzardo, E.; Thang, S. H., Radical addition-fragmentation chemistry in polymer synthesis. Polymer 2008, 49 (5), 1079-1131; (c) Braunecker, W. A.; Matyjaszewski, K., Controlled/living radical polymerization: features, developments, and perspectives. Progress in Polymer Science 2007, 32 (1), 93-146. 8. Matyjaszewski, K. W., J.-S. 1996. 9. Matyjaszewski, K., Atom transfer radical polymerization (ATRP): current Status and future perspectives. Macromolecules 2012, 45 (10), 4015-4039. 10. di Lena, F.; Matyjaszewski, K., Transition metal catalysts for controlled 99  radical polymerization. Progress in Polymer Science 2010, 35 (8), 959-1021. 11. Owen, S. C.; Chan, D. P. Y.; Shoichet, M. S., Polymeric micelle stability. Nano Today 2012, 7 (1), 53-65. 12. Jo, W. H. K., T.; Hwang, I. C.; Kwon, N. H. Novel amphiphilic block copolymer, method for manufacturing the same and polymer electrolyte membrane using the same. 2010. 13. Zhou, H., Brunelle, Daniel, J., Moore, D., R., Zhang, L., Misner, M., J., Chen, X., Ma, M., Block copolymer membranes and associated methods for making the same. US Patent, 123033 A1, 2011. 14. Wang, J. Y.; Xu, Y. Y.; Zhu, L. P.; Li, J. H.; Zhu, B. K., Amphiphilic ABA copolymers used for surface modification of polysulfone membranes, Part 1: Molecular design, synthesis, and characterization. Polymer 2008, 49 (15), 3256-3264. 15. Yi, Z. A.; Zhu, L. P.; Xu, Y. Y.; Zhao, Y. F.; Ma, X. T.; Zhu, B. K., Polysulfone-based amphiphilic polymer for hydrophilicity and fouling-resistant modification of polyethersulfone membranes. Journal of Membrane Science 2010, 365 (1-2), 25-33. 16. Yi, Z.; Zhu, L. P.; Cheng, L.; Zhu, B. K.; Xu, Y. Y., A readily modified polyethersulfone with amino-substituted groups: Its amphiphilic copolymer synthesis and membrane application. Polymer 2012, 53 (2), 350-358. 17. Bates, F. S.; Fredrickson, G. H., Block copolymer thermodynamics - theory and experiment. Annual Review of Physical Chemistry 1990, 41, 525-557. 18. Fredrickson, G. H.; Bates, F. S., Dynamics of block copolymers: theory and experiment. Annual Review of Materials Science 1996, 26, 501-550. 19. Kim, J. K.; Yang, S. Y.; Lee, Y.; Kim, Y., Functional nanomaterials based on block copolymer self-assembly. Progress in Polymer Science 2010, 35 (11), 1325-1349. 20. Ichikawa, T.; Yoshio, M.; Hamasaki, A.; Kagimoto, J.; Ohno, H.; Kato, T., 3D Interconnected ionic nano-channels formed in polymer films: self-Organization and polymerization of thermotropic bicontinuous cubic liquid crystals. Journal of the American Chemical Society 2011, 133 (7), 2163-2169. 21. Nunes, S. P., Recent advances in the controlled formation of pores in membranes. Trends in Polymer Science 1997, 5 (6), 187-192. 100  22. Smolders, C. A.; Reuvers, A. J.; Boom, R. M.; Wienk, I. M., Microstructures in phase-inversion membranes .1. formation of macrovoids. Journal of Membrane Science 1992, 73 (2-3), 259-275. 23. Amirilargani, M.; Mohammadi, T., Synthesis and characterization of asymmetric polyethersulfone membranes: effects of concentration and polarity of nonsolvent additives on morphology and performance of the membranes. Polymers for Advanced Technologies 2011, 22 (6), 962-972. 24. Peinemann, K. V.; Abetz, V.; Simon, P. F. W., Asymmetric superstructure formed in a block copolymer via phase separation. Nature Materials 2007, 6 (12), 992-996. 25. Nunes, S. P.; Sougrat, R.; Hooghan, B.; Anjum, D. H.; Behzad, A. R.; Zhao, L.; Pradeep, N.; Pinnau, I.; Vainio, U.; Peinemann, K. V., Ultraporous films with uniform nanochannels by block copolymer micelles assembly. Macromolecules 2010, 43 (19), 8079-8085. 26. Widin, J. M.; Schmitt, A. K.; Schmitt, A. L.; Im, K.; Mahanthappa, M. K., Unexpected Consequences of block polydispersity on the self-Assembly of ABA triblock copolymers. Journal of the American Chemical Society 2012, 134 (8), 3834-3844. 27. Percec, V.; Clough, R. S.; Grigoras, M.; Rinaldi, P. L.; Litman, V. E., Reductive dehalogenation versus substitution in the polyetherification of 4,4'-dihalodiphenyl sulfones with bisphenolates. Macromolecules 1993, 26 (14), 3650-3662. 28. Tang, W.; Kwak, Y.; Braunecker, W.; Tsarevsky, N. V.; Coote, M. L.; Matyjaszewski, K., Understanding Atom transfer radical polymerization: Effect of ligand and initiator structures on the equilibrium constants. Journal of the American Chemical Society 2008, 130 (32), 10702-10713. 29. Xia, J.; Matyjaszewski, K., Controlled/“living” radical polymerization. atom transfer radical polymerization using multidentate amine ligands. Macromolecules 1997, 30 (25), 7697-7700. 30. Lynd, N. A.; Meuler, A. J.; Hillmyer, M. A., Polydispersity and block copolymer self-assembly. Progress in Polymer Science 2008, 33 (9), 875-893.       101  Chapter 3. Use of Nanoparticles and Click Chemistry in the Development of Single-ion Conductors for Lithium Ion Batteries This chapter describes research aimed at developing single-ion conductors for lithium ion batteries. The introduction first discusses why single-ion conductors are important for decreasing concentration polarization and related voltage losses during battery discharge. Subsequently, I first present work on nanoparticles coated with immobilized anions as part of single-ion conductors. This work follows up previous studies by Fadi Asfour and shows that nanoparticles with grafted polyanions have a higher lithium conterion weight fraction than nanoparticles coated with only monolayers of anion. However, because the conductivity of these nanoparticle-PEO electrolytes is not high enough for battery applications, subsequent work aims toward using click chemistry to modify PEO directly and create single-ion conductors. These studies show that the presence of triazole or thioether groups does not limit conductivity. Thus, click chemistry may provide a useful method for introducing immobile anions in PEO to create single-ion conductors. 1. Introduction- Ion Conduction in Lithium Ion Batteries In discharge of a lithium ion battery, electrons flow from the anode to the cathode through the external circuit, while inside the cell, lithium cations formed at the anode migrate to the cathode and intercalate into the cathode material with reduction at the cathode (Figure 3.1). The electrolyte usually + contains a lithium salt dissolved in a solvent, and ideally Li ions are the only current carriers in this region. When anions carry a significant fraction of the 102  current, concentration polarization occurs, and the battery loses voltage. 1 Figure 3.1. Schematic diagram of a lithium ion battery containing a metal oxide cathode and a graphite anode. The figure also shows redox reaction during discharge. Figure 3.2 illustrates the concentration polarization. At the anode, only part of the Li + formed by oxidation migrates away to carry current, so Li + accumulates. Additionally, anions migrate toward the anode and collect there. + At the cathode, more Li is intercalated with reduction than can migrate to the + cathode due to current, so Li is depleted. Anions migrate away from the cathode and are also depleted in this region. These accumulation and depletion zones decrease the potential drop at the cathode-solution interface 103  and increase the potential drop at the anode-solution interface to decrease the battery voltage. If the discharge is slow enough, diffusion will help dissipate the concentration polarization, but at fast discharge, concentration polarization is a serious problem. Moreover, the low conductivity in the depleted region near the cathode increases the ohmic potential drop, which leads to heating and an additional loss of battery voltage. The effective voltage during discharge is V (cell voltage) - Vp (polarization voltage), and the voltage needed to charge the cell is V+Vp. Of course, power output is directly proportional to discharge voltage. Figure 3.2. Concentration polarization during discharge of a lithium ion battery.  The lithium ion transference number (tLi+) is the fraction of current in solution + carried by Li . For electrolytes with only one cation and one anion, the sum of the cation and anion transference numbers must equal 1. Typical Li + transference numbers in PEO-based electrolytes range from 0.2 to 0.4 (Table 3.1), showing that the anion is the dominant current-carrying species. Thus, concentration polarization during discharge may be severe as more Li+ ions 104  form at the anode than can migrate away, and more Li + enters the cathode + than migrates there. Electrolytes with Li transference numbers close to 1 are highly desirable for decreasing concentration polarization. Lithium Salt Polyether (or solvent) O/Li ratio o tLi+ ( C) 20 30 0.23 a 24 30 0.21 20 70 0.32 1M 25 0.38±0.04 PEO1000 LiCF3SO3 PEO1000 LiBF4 PEO500 LiPF6 PC/EC/DMC b Temperature a LiClO4 a d b c Poly(ethyleneoxide) with Mn of 1,000. Poly(ethyleneoxide) with Mn of 500. c PC, propylene carbonate; EC, ethylene carbonate; DMC, dimethyl carbonate; 1:1:1 is volume ratio. d only the ether oxygens in PEO are considered. Table 3.1. Lithium transference numbers for several lithium salts in battery 1 solvents. + One way to obtain single-ion (Li ) electrolytes is to attach anions to + polymers through covalent bonds, so the Li counterion is the only mobile ion. As long as all anions are tethered to the polymer and immobile, tLi+ should be 2,3,4 unity. Another method for immobilizing anions involves grafting to nanoparticles. In this work I began following up on our previous studies with nanoparticles coated with monolayers of anions. To increase the amount of Li 105  + per nanoparticles, I grafted polyanions to the particles. Unfortunately, the conductivity of polyelectrolytes containing these modified particles was impractically low, so I also began examining methods to introduce anions directly into PEG. 2. Results and Discussion 2.1. Single-ion Conductors Containing Nanoparticles with Immobilized Anions.   Scheme 3.1. Monolayer modified silica nanoparticle Si-C5NTfLi (silica nanoparticle derivatized with N-pentenyl lithiated triflouromethane 5 sulfonamide). Scheme 3.1 shows the monolayer-modified nanoparticles (Si-C5NTfLi) that our group mixes with PEO to create single-ion conductors. Fadi Asfour initially 5 developed and studied these particles. We employ silica nanoparticles with diameters of around 12, and a 5-carbon spacer separates the trifluoromethane 5 sulfonamide from the nanoparticle surface. To prepare electrolytes, we disperse the nanoparticles in polyethylene oxide dimethylether (Mw=500 g/mol) (PEGDME-500). Since the nanoparticles and hence the anions are immobile, + Li is the only species that conducts current. 106  Scheme 3.2. Method for determining the Li + content of PEO-based electrolytes. In following up on previous studies of these conductors, I first determined the amount of Li in the modified nanoparticles. Quantitative analysis of the metal content in the TGA residue by using inductively coupled plasma optical emission spectrometry (ICP-OES) gives an estimate of the number of anions bound to the nanoparticles. The lithium in the residue should be Li2O or Li2CO3, and both are water soluble. The TGA residue was stirred in deionized water for 12 h, and filtered prior to the ICO-OES analysis. The combined TGA/ICP-OES analysis of the modified particle shows a lithium content of 1.2 x 10 -3 g Li/g particles. We prepared electrolytes with various O/Li ratios by dispersing the Si-C5NTfLi in PEGDME-500, and Table 3.2 presents the O/Li ratio in the electrolytes. 107  O/Li ratio a Particle Content (wt%) 425 320 230 185 18.5 23.6 30.0 35.0 o Conductivity at 30 C (S/cm) 1.3x10 a -6 8.4x10 -7 1.2x10 -6 -7 5.3x10 The O/Li ratio does not include any O from the Si-C5NTfLi particles. Table 3.2. Particle weight percentages and O/Li for electrolytes prepared from Si-C5NTfLi dispersed in PEGDME-500.   Figure 3.3. Temperature-dependent conductivity of electrolytes containing different fractions of Si-C5NTfLi dispersed in PEGDE-500. These results were obtained by Fadi Asfour, and I repeated some of the measurements. 5 Figure 3.3 shows the temperature-dependent conductivities of electrolytes o with different particle contents. The conductivities at 30 C (1000/T=3.3) are -6 around 10 S/cm, and conductivity increases by less than an order of 108  o o magnitude on gong from 30 C (1000/T=3.3) to 80 C (1000/T=2.83). These low conductivities may stem from the low lithium content in the electrolyte materials. As Table 3.2 shows, the O/Li ratio is between 200 and 400 for all the + electrolytes. Increasing the particle content adds Li to the electrolyte, but does not change the conductivity by more than a factor of 3. At high particles contents, the lack of ionic conductivity through the interior of the nanoparticles may decrease the overall conductivity. Thus improvements in conductivity will require increases in the amount of Li per g of particle. 2.2. Single-ion Conductors Prepared Using Nanoparticles Modified by Grafting of Polyanions   Scheme 3.3. Silica nanoparticles prepared by grafting lithiated poly (trifluoromethane sulfonic aminoethyl-methacrylate) (Si-TfMALi) from the 1 surface. Scheme 3.3 shows our strategy to increase the Li content in modified silica nanoparticles. Growth of a poly(trifluoromethane sulfonimide aminoethyl+ methacrylate) (TfMALi) from the particles yields many sites for Li counterions. 1 (I describe the synthesis of these particles in my MS thesis. ) TGA-ICP -2 characterization of the grafted particles shows a lithium content of 2.0 x 10 Li/g Si-TfMALi particles. Compared to the 1.2 x 10 109  -3 g g Li/g Si-C5NTfLi, Si-TfMALi contains an order of magnitude more Li per g of particles. Mixing Si-TfMALi particles with PEGDME-500 thus yields high O/Li ratios compared to those in dispersions of Si-C5NTfLi (compare Tables 3.2 and 3.3). O/Li ratio a Particle Content (wt%) o 16 32 64 96 33.2 19.9 11.1 7.7 Conductivity at 30 C (S/cm) 7.5x10 a 1 -7 3.2x10 -6 2.1x10 -6 -7 3.1x10 The O/Li ratio does not include any O from the Si-TfMALi particles. Table 3.3. Particle weight percentages and O/Li ratios for electrolytes prepared from Si-TfMALi dispersed in PEGDME-500. Figure 3.4. Temperature-dependent conductivity of electrolytes containing different fractions of Si-TfMALi dispersed in PEGDE-500. The various fractions 1 of particles (see Table 3.3) lead to the different O/Li ratios shown in the figure. 110  Figure 3.4 shows the temperature-dependent conductivities for dispersions of Si-TfMALi in PEGDME-500 as a function of the O/Li ratios in the dispersions. The conductivities, which are extracted from impedance spectroscopy data, are roughly linear with 1/T, consistent with thermally o activated transport. Conductivities at 30 C are on the order of 10 -6 S/cm, similar to values with Si-C5NTfLi/PEGDEM-500. Figure 3.5. Temperature-dependent conductivity for (a) Si-C5NTfLi at O/Li 425 and (b) Si-TfMALi at O/Li 32, both samples contain ~19 wt% modified particles. Figure 3.5 compares temperature-dependent conductivities for Si-C5NTfLi/PEGDME-500 (O/Li=425) and Si-TfMALi/PEGDME-500 (O/Li=32) dispersions. Both composite electrolytes have particle fractions of around 19 wt%, but the polyelectrolyte-modified particles contain much more Li. The system with a monolayer of anions on the particles (Si-C5NTfLi/PEGDME-500) 111  exhibits a 2-fold increase in conductivity on going from 30 to 80 °C. At 30 °C the polyectrolyte system, Si-TfMALi/PEGDME-500, has a conductivity similar to that of Si-C5NTfLi/PEGDME-500 (a difference of about 0.5 log units), but at 80 °C, the conductivity of Si-TfMALi/PEGDME-500 is a full order of magnitude higher. Scheme 3.4 presents a proposed explaination for these results. Monolayer + + Li Li Li+ + Li+ Li Li+ + Li Li+ Li+ Li+ Li+ Li+ Li+ Li+ + Li Li+ Li+ Li+ -6 o 1.3 x 10 S/cm at 30 C -6 o 2.8 x 10 S/cm at 80 C Grafted polymer Li+ Li+ Li+ Li+ + Li Li Li+ Li+ + Li+ + Li+ Li Li+ Li+ Li+ Li+ Li+ Li+ -6 o 3.2 x 10 S/cm at 30 C = Li+ -5 o 2.9 x 10 S/cm at 80 C O N S CF3 O Scheme 3.4. Proposed qualitative conformations of monolayer Si-C5NTfLi o o (top) and multilayer Si-TfMALi (bottom) at 30 C and at 80 C. The conformation of the monolayer does not change significantly with increasing temperature, so the conductivity of Si-C5NTfLi/PEGDME-500 increases marginally on going from 30 to 80 °C. At 80 °C, the polyTfMALi chains are not highly soluble in PEGDME-500, so only the exterior lithium ions 112  are accessible to the PEGDME-500 matrix for conductivity. Increasing the temperature increases the solubility or mobility of the polyelectrolyte in + PEGDME-500, and more Li becomes available to the matrix for conductivity. o Thus at 80 C, the conductivity of Si-TfMALi/PEGDME-500 reaches 3x10 -5 S/cm. The conductivity of this system is an order of magnitude lower than the 6 values of typical electrolytes used in Li ion batteries, but the increase in + conductivity at high temperature suggests that if more Li is accessible to PEGDME-500, we may be able to create competitive single-ion conductors. 2.3. Towards Click Chemistry for Synthesizing Single-ion Conductors with a High Density of Li+ in PEO + To increase the concentration of Li in poly(ethyleneoxide) (PEO) while still maintaining single-ion conductivity, we aim to eventually graft anionic groups directly to the PEO backbone (Scheme 3.5). The anionic groups should be + immobile, and most of the Li will be available for carrying current. Below I describe some previous studies of anion insertion via click chemistry and then present studies aimed at determining whether the triazole or thioether linkers + alter Li conductivity. 113  Scheme 3.5. Proposed single-ion conductors prepared by synthesis of PEO containing alkene or alkyne groups and subsequent attachment of anions to these groups via click chemistry. 2.3.1. Prior Use of Click Chemistry for Creating Proton-Conducting Membranes. To the best of my knowledge, no studies reported using click chemistry to + prepare Li electrolytes. However, several groups employed click chemistry to + synthesize proton-conducting materials. Since proton and Li conduction are similar, in the following section I first review the used of click chemistry 7 8 (alkyne-azide or thiol-ene ) for post-functionalization of polymers to create proton-conductive membranes. 114  NaO3S O N N N O S O O N N O n N N O O NN N O S O N N n O NaO3S   Scheme 3.6. Polysulfone structures synthesized by Bielawski’s et al for 9 proton-conduction. Especially relevant to this work, Bielawski et al. synthesized polysulfone-based polyanions for proton-conductive membranes in methanol 9 fuel cells. Scheme 3.6 shows the structure of these polymers, in which azide-containing polysulfone was partially functionalized with sulfonate side chains via alkyne-azide click chemistry; the rest of the azide groups served as crosslinking points to prepare a gel-like material. The performance of fuel cells containing click-functionalized polysulfone membranes was comparable to that 115  with cells containing Nafion-based membranes, with a maximum power output 2 of 130 mW/cm.   Scheme 3.7. Polyacrylate structures prepared by Martwiset et al. for formation o 10 of membranes that exhibit proton conductivity at 200 C. Martwiset et al. synthesized polyacrylates containing different numbers of 10 1H-1,2,3-triazole groups (Scheme 3.7). They did not use click chemistry for post-functionalization of the polymers, but the 1H-1,2,3-triazole groups served as proton donors (Scheme 3.7) in proton-conducting membrane. In an o anhydrous environment at 200 C, such membranes exhibited a maximum conductivity of 17.5 µS/cm, which is comparable to the conductivity of Nafion under these conditions. 116  Scheme 3.8. Poly(ether ether ketone) prepared by Gao et al for reaction with 3-mercaptopropyltrimethoxysilane via click chemistry. 11 Gao et al. used thiol-ene chemistry to attach the mercaptopropyl 11 trimethoxysilane to a poly(ether ether ketone) derivative (Scheme 3.8). Subsequent crosslinking catalyzed by hydrochloric acid gave a polymer/silica hybrid material with high dimensional and oxidative stabilities. Although the proton conductivity of this material is lower than that of the original sulfonated poly(ether ether ketone), presumably due to cross-linking, the material maintained relatively high conductivity (10 -2 S/cm at room temperature), 5% o and 10% weight loss temperature of the membrane are higher than 280 C, these properties made this series of membranes a good alternative for proton exchange membranes in fuel cell application. 2.3.2. Synthesis of PEO-Containing Click Functionalities   Scheme 3.9. The thioether and triazole structures that result from (a) thiol-ene and (b) alkyne-azido click chemistry, respectively. The above examples of the use of click chemistry to prepare 117  proton-conductive polymer electrolyteS suggest that the resulting structures (either thiol-ethers or 1,2,3-triazoles, Scheme 3.9) are not obstacles to proton transport. However, I found no previous studies that employed click + chemistry to functionalize polymers for Li conduction. (a)      (b) Scheme 3.10. (a) Scheme of ideal lithium transport in an electrolyte material and (b) scheme of lithium transport if the click functionality (triazole or thioether) + impedes Li transport. Scheme 3.10 illustrates a possible challenge of the click structure to Li + conductivity. The X-axis represents the lithium transport path, and the Y-axis shows energy barriers for lithium transport. If the click structure has little or no effect on lithium transport, lithium will encounter no additional barriers to migration, as shown in Scheme 3.10 (a). The 1,2,3-triazole and sulfide 118  structures that result from click chemistry, could potentially associate with Li + to decrease mobility. Click structures, particularly triazole, may make the polymer back bone more rigid and give a high Tg that could reduce chain + 12 mobility and decrease Li conductivity. In these cases, high energy barrier will appear across the whole lithium transport path, as shown in Scheme 3.10 (b). Finally, the solubility of Li + in the polymer may decrease because of unfavorable association with the click functionality.13 Before applying click chemistry in this field, we decided to examine the influence of the click + structure on Li transport in PEO, and this required synthesis of PEO containing triazoles and thioethers. Dipropargyl and diazido tetraethylene glycol (Scheme 3.11a) will react to give triazole-containing PEO, and diallyl and dithiol tetraethylene glycol (Scheme 3.11b) will form sulfur-containing PEO, so we synthesized these reactants.   Scheme 3.11. (a) Monomers synthesized to prepare triazole-containing PEO, (b) monomers to that will react to give sulfur-containing PEO. 119  Ts C r. t. l/ K OH Scheme 3.12. Two methods for synthesis of dipropargyl tetraethylene glycol. Reaction of propargyl bromide and tetraethylene glycol successfully gives dipropargyl tetraethylene glycol in one step (Route 1, Scheme 3.12), but the propargyl bromide starting material is relatively expensive ($58 for 50 g of an 80 wt% solution in toluene). More importantly this material is potentially explosive and quite hazardous. To enable large scale synthesis while avoiding use of the hazardous starting material, we developed Route 2 (Scheme 3.12), which uses ditosyl tetraethylene glycol as an intermediate and reaction with propargyl alcohol. We synthesized the target molecule using Route 2 at a similar yield to Route 1, around 60%. The tosylate derivative is a very useful and convenient intermediate to covert hydroxyl groups to other functionalities. 14 Conversion of the hydroxyl group to tosylate using tosyl chloride is quantitative for many substrates, the reaction condition is mild, and extraction purification usually gives pure product. In fact, we synthesized three of the four compounds for click reactions using the tosylate intermediate (Scheme 3.13). The only drawback to this method is that tosyl chloride is toxic. 120  Scheme 3.13. Synthesis of several step-growth polymerization monomers via the ditosyl derivative. Scheme 3.14. Synthesis of diallyl tetraethylene glycol. In principle, the method to synthesize dipropargyl tetraethylene glycol in Scheme 3.13 could be further modified to obtain diallyl tetraethylene glycol. However, allyl bromide is a relatively cheap commercial material ($35 for 250 mL, Aldrich) and is not hazardous, so we synthesized diallyl tetraethylene glycol directly from tetraethylene glycol in the one step reaction shown in Scheme 3.14. 121    Figure 3.6. Kinetics of step-growth polymerization between dipropargyl and diazido tetraethylene glycol.   Scheme 3.15. Step-growth polymerization of diazido and dipropargyl tetraethylene glycol. Step-growth polymerization of diazido and dipropargyl tetraethylene glycol (Scheme 3.15) was conducted using pre-synthesized Cu(PPh3)Br catalyst. Figure 3.6 shows that after 4 h the reaction conversion plateaus around 80%. Polymer was isolated by precipitation after adding the reaction 122  solution into ether, and GPC analysis of the polymer product gave Mw=8.1x10 4 and PDI=3.0.   Scheme 3.16. Step-growth polymerization of dithiol and diallyl tetraethylene glycol. Step-growth polymerization of dithio and diallyl tetraethylene glycol was conducted via a thermally activated process (Scheme 3.16). GPC analysis of the polymer product gave Mw=7,400 and PDI=1.1. 2.3.3. Comparison of the Conductivity of Li Salts in PEO and PEO with Click Functionalities. To determine if the triazole functionality affects the conductivity of Li salts in PEO, we compared the conductivity of LiClO4 dissolved in a commercially 5 available polyethylene oxide (PEO, MV=10 ), with the conductivity of the polymer obtained from polymerization of diazido and dipropargyl 4 tetraethylene glycol (Mw=8.1 x 10 ). Mv is calculated from the viscosity of a polymer solution. Because polymer chains with higher molecular weights make solutions much more viscous than chains with smaller molecular weights, Mv > 123  Mw, which is greater than Mn. Figure 3.7. Conductivities of mixtures of LiClO4 with pure PEO (black squares) and triazole-containing PEO (red circles) at different ratios of PEO to LiClO4. The O/Li ratios in the mixture were determined from the masses of LiClO4 and PEO added, and these ratios only include the O atoms from PEO. o Measurements occurred at 90 C. PEO with this large molecular weight is a solid with crystalline regions. Previous studies show that the majority of current conduction occurs through amorphous regions, so we determined conductivity with liquid polymers at 90 o o C. The melting point of PEO is around 60 C. Figure 3.5 compares the conductivity of electrolytes prepared with pure PEO and with PEO containing 5 triazole groups. In the pure PEO material (Mv=10 ), as the O/Li ratios + decreases from 128 to 64, conductivity increases due to more Li and ClO4- 124  species in the mixture. The conductivity reaches a peak value of 0.039 S/cm. Further increases in the LiClO4 concentration (O/Li ratios between 64 and 16) do not increase conductivity. Previous work shows that at high salt concentrations ions pair or aggregate, reducing the number and overall 15,16 mobility of the ions. Most importantly, electrolytes made with pure PEO and PEO containing the triazole structure have similar conductivities. Thus, the triazole groups that result from alkyne-azido click chemistry do not greatly + impede Li transport. Figure 3.8. Conductivities of mixtures of LiClO4 with pure PEO (black squares) and thioether-PEO (red circles) at different ratios of PEO to LiClO4. The O/Li ratios in the mixture were determined from the masses of LiClO4 and PEO added, and these ratios only include the O atoms from PEO. Measurement o occurred at 90 C. 125  Figure 3.8 compares the conductivities of electrolytes made from pure PEO and PEO containing thioether groups. A commercial poly(ethyleneoxide) (PEO) with a molecular weight (Mw= 8,000) similar to that of the thioether PEO served as the control polymer for these experiments. In this case, conductivity generally increases with increasing concentrations of LiClO4 (decreasing O/Li ratios). Comparison of Figures 3.7 and 3.8 suggests that the conductivities are generally a factor of 2 or so higher in the lower molecular weight PEO (Figure 3.6). The conductivities with pure PEO are about the same as in the PEO prepared with thiol-ene click chemistry, suggesting that the thioethers do + not impede Li transport. At a specific temperature, ionic conductivity is proportional to the number of carriers, the charge of the carriers and their mobility. Both cations and anions contribute to the ionic conductivity data measured by AC impedance spectroscopy. Thus, the data in Figure 3.7 and 3.8 reflect the mobility of both lithium cations and perchlorate anions. Previous studies show that with LiClO4 in PEO, 70% - 80% of the conductivity is due to perchlorate anions. It might be + possible that the triazole of thioether species disrupts conductivity from Li and + not from ClO4-, but since Li contributes only 20% - 30% of the conductivity, we do not detect it. However, given that the data show no evidence of a trend in lower conductivity with triazole or thioethers, any effect would likely be small. 3. Conclusions The conductivities of single-ion conductors based on nanoparticles with 126  grafted polyanions and nanoparticles with mono-layers of anions are similar, despite the larger number of lithium counterions accompanying the polyanions. This suggests that most of the lithium counterions in the polyanion-grafted particles do not conduct current, probably due to the low solubility of polyelectrolyte in PEO. However, electrolytes with polyanion-modified nanoparticles show relatively large increases in conductivity as temperature + increases, suggesting that more Li may be available at high temperature due to conformational changes. To create single-ion conductors with higher conductivity, we are considering grafting anions to PEO using click chemistry. This will avoid problems with anion solubility in PEO. However, the triazole or thioether groups in click linkers might impede conductivity. To examine this, we polymerized tetraethylene oxides using click chemistry and determined the conductivity of mixtures of LiClO4 with these polymers. Triazole and thioether-containing PEO were synthesized with one triazole or thioether for every tetraethylene oxide repeating unit. Conductivities were essentially the same for LiClO4 mixed with pure PEO and with the new polymers, suggesting + that triazole and thioether groups do not impede Li transport. 4. Experimental Section 4.1. Materials Sodium azide (500 g, Alfa Aesar, 99%), propargyl bromide (125 g, 80 wt% solution in toluene, Aldrich), potassium carbonate (500 g, ACP Chemicals), 127  p-toluenesulfonyl chloride (500 g, ACP Chemicals), thiourea (1 kg, Aldrich), allyl bromide (1 L, reagent grade 97%, Aldrich), poly(ethyleneoxide) (250 g, Mv=100,000, Aldrich), poly (ethyleneoxide) (500 g, Mn=8,000, Aldrich), sodium hydride (100 g, 60% dispersion in mineral oil, Aldrich), tetraethylene glycol (1 kg, 99%, Aldrich), lithium perchlorate (100 g, ≥99%, Aldrich) and triethylene glycol (1 kg, 99%, Aldrich) were used as received. Cu(I)Br (99.999%, Aldrich) was purified using saturated aqueous NaBr solution. THF was distilled over sodium metal and benzophenone. 4.2. 1 Characterization H and 13 C NMR spectra were collected using a Varian UnityPlus-500 spectrometer in CDCl3 with the residual proton signals from the solvent serving as the chemical shift standard. Conductivity data were collected from an HP 4192A LF Impedance Analyzer scanning from 5 Hz to 13 MHz with an applied voltage of 10 mV. The sample cell contains two steel disks that serve as symmetric electrodes separated by a Teflon collar containing electrolyte with a radius of 0.61 cm and a thickness of 0.0175 cm. Electrolytes were equilibrated at a pre-determined temperature for at least 10 minutes before measurement. The molecular weights of polymers were determined by gel permeation chromatography (GPC) using two PL-gel 20 m Mixed A columns and a Waters R401 Differential Refractometer detector at room temperature. THF served as the eluting solvent at a flow rate of 1 mL/min, and mono-disperse polystyrene standards were used to calibrate the molecular 128  weights. The concentration of the polymer solutions used for GPC measurements was 1 mg/mL. FT-IR spectra were collected in a Mattson Galaxy 300 spectrometer. FT-IR samples were made by mixing with KBr, grounded and pressed to pellets. Liquid materials were spread on a KBr pellet and directly characterized in the instrument. Quantitative Li analysis was performed using a Varian 710-ES ICP Optical Emission Spectrometer. Six standard solutions (LiCl concentration of 0, 0.05, 0.5, 1, 2 and 4.5 ppm in 2 wt% HNO3) were used to prepare calibration curves. Concentrations determined using emission at wavelengths of 460.289 nm, 670.783 nm, 610.365 nm were essentially the same 4.3. Synthesis Synthesis of Tris (triphenylphosphine) copper(I) bromide Cu(PPh3)3Br: 6 g of triphenylphosphine (23 mmol) was dissolved in 100 mL boiling methanol in a 250 mL flask. After addition of 1.43 g CuBr2 (10 mmol), the product formed as a white precipitate, which was collected by filtration and dried under o vacuum at 90 C overnight. The reaction gave 4.2 g product (62.3% yield ) o 17 with a melting point of 169 C. Synthesis of ditosyl tetraethylene glycol: Tetraethylene glycol (5.1 g, 0.05 mol) was dissolved in 60 mL of THF in a 250 mL round bottom flask. Twenty g of tosyl chloride (0.105 mol) was added and the mixture was cooled in an ice bath. In a separate Erlenmeyer flask, 25 g of KOH was dissolved in 30 mL of deionized water. This KOH solution was transferred to an addition funnel and 129  slowly added into the 250 mL flask over 20 min. After the addition, the ice bath was removed, and the reaction was allowed to proceed at room temperature overnight. Most of THF was removed by rotary evaporation, and the product was extracted from the residual solution with dichloromethane (3 x 100 mL). The organic layer was washed with deionized water (4 x 200 mL) and dried over MgSO4. The solvent was removed by rotary evaporation, and the product was dried under vacuum to give 23.8 g of product as a viscous liquid (94% 1 yield). See Appendix 1 for the H NMR spectrum. Synthesis of diazido tetraethylene glycol: Ditosyl tetraethylene glycol (10.04 g, 0.02 mol), sodium azide (3.9 g, 0.06 mol) and 150 mL DMF were o mixed in a 500 mL flask at 90 C overnight. After vacuum distillation to remove DMF, the residual mixture was dissolved in 70 mL deionized water and the product was extracted into dichloromethane (3 x 100 mL). The organic layer was washed three times with 200 mL of deionized water and dried over MgSO4. After rotary evaporation of solvent, the product was purified by silica column chromatography (hexane/EtOAc=1:1), to give 3 g of product as a light 1 orange oil (61.5% yield). See Appendix 4 for the H NMR spectrum. Synthesis of dipropargyl tetraethylene glycol: Route 1 (using tetraethylene glycol and propargyl bromide): In a 250 mL 3-neck flask under N2, 5.68 g tetraethylene glycol (28.4 mmol) was dissolved in 60 mL dry THF prior to gradual addition of 5.68 g sodium hydride (0.142 mol) while stirring. After 20 minutes, the 3-neck flask was placed in an ice bath, and 130  a solution containing 12.52 g of propargyl bromide (91 mmol) and 0.022 g 18-crown-6 (0.08 mmol) in 30 mL dry THF was slowly added to the flask through an addition funnel over 10 minutes. The ice bath was then removed, and the reaction was allowed to proceed at room temperature overnight. Subsequently, 30 mL of deionized water was added, and the mixture was stirred for 40 minutes to quench the reaction. Most of the THF was removed by rotary evaporation, and the product was extracted from the residual solution with dichloromethane (3 x 150 mL). The combined organic layer was dried over MgSO4 and column chromatography (silica, hexane/ethyl acetate 1/4) gave 3.3 g pure product (12.8 mmol, 45.2% yield). Route 2 (using ditosyl tetraethylene glycol and propargyl alcohol): In a 300 mL beaker 7.5 g NaH in mineral oil (0.1875 mol) was added to 150 mL pentane and stirred with a magnetic bar for 10 min. The NaH in the suspension was allowed to settle, and a 50 mL syringe was used to carefully remove the supernatant. The above process was repeated to give purified NaH. Next, 10.5 g propargyl alcohol (0.1875 mol) was dissolved in 200 mL dry THF in a 500 mL round bottom flask, and 7.5 g of NaH was slowly transferred into this flask under N2. After 30 min of stirring at room temperature, 20 g ditosyl o tetraethylene glycol was added, and the flask was heated at 60 C overnight. Fifty mL of deionized water was added to the flask, and after 10 min, most of the THF was removed by rotary evaporation. The product was extracted in dichloromethane (3 x 100 mL) and washed with deionized water (4 x 200 mL) 131  before drying over MgSO4. The solvent was removed by rotary evaporation, and the product was purified by silica column chromatography (hexane/EtOAc=1:1) to give 7 g of product (65% yield) as a light yellow oil. See 1 Appendix 7 for the H NMR spectrum. Alkyn-azido Step growth polymerization: 1.318 g of diazido tetraethyleen glycol (5.4 mmol), 1.458 g of diacetylene tetraethylene glycol (5.4 mmol) and 0.05 g Cu(PPh3)3Br (0.01 equiv.) were dissolved in 20 mL of chloroform in a o 100 mL Schlenk flask and allowed to react at 60 C for 24 h, After cooling to room temperature, this solution was added dropwise to 200 mL diethyl ether with strong stirring. Product was precipitated as a light brown viscous oil, which was dissolved in chloroform and precipitated in ether again. The product was dried under vacuum (1.2 g, 43% yield). GPC (THF solvent, PS standard) 4 4 1 showed Mn=2.73 x 10 , Mw=8.11 x 10 , PDI=3.0. See Appendix 15 for the H NMR spectrum. Synthesis of dithiol tetraethylene glycol: In a 250 mL round bottom flask, 20.1 g ditosyl tetraethylene glycol (0.04 mol), 40 mL of ethanol, and 6.1 g of thiourea (0.08 mol) were added to 30 mL of deionized water. The solution was purged with nitrogen for 15 minutes, and the flask was fitted with a condenser. After reflux for 15 h, a solution of 4 g NaOH (0.1 mol) in 40 mL deionized water was added, and after another 2 h of reflux, the flask was placed in an ice bath. Fifteen mL of concentrated HCl solution was added to neutralize the solution, and the product was then extracted in dichloromethane (3 x 50 mL). The 132  combined dichloromethane solution was dried over MgSO4, solvent was removed by rotary evaporation, and vacuum distillation gave 4 g of dithiol tetraethylene glycol (0.025 mol, 62%) as colorless oil. See Appendix 12 for the 1 H NMR spectrum. Synthesis of diallyl tetraethylene glycol: In a 500 mL round bottom flask, 11.6 g of dihydroxyl tetraethylene glycol (0.06 mol) and 14 g of powderized KOH (0.25 mol) were dissolve in 200 mL of acetone. After placing the flask in an ice bath, 10 mL of allyl bromide (0.12 mol) was slowly added over 10 minutes. The ice bath was then removed, and the mixture was allowed to react at room temperature overnight. Acetone was removed using rotary evaporation, and the product was dissolved in 300 mL dichloromethane. Deionized water was used to wash the dichloromethane solution (3 x 80 mL), the organic layer was dried using MgSO4, and solvent was removed by rotary evaporation. Silica column chromatography (hexane/ethyl acetate 9/1) gave 1 10.84 g (0.042 mol, 70% yield) of product. See Appendix 10 for the H NMR spectrum. Thiol-ene Step growth polymerization: 6.46 g diallyl tetraethylene glycol (0.024 mol), 5.36 g dithiol tetraethylene glycol (0.024 mol) were dissolved with 100 mL dichloromethane in a 250 mL round bottom flask. The solution was o purged with nitrogen for 30 minutes and was done in a 50 C oil bath for 24 h. To work up the reaction, the mixture was put under rotovac to remove residual solvent. The product was dissolved in 100 mL chloroform and added dropwise 133  to 200 mL diethyl ether with strong stirring. Product was obtained as a light yellow viscous oil, which was dissolved in chloroform and precipitated in ether again. Polymer product was dried under vacuum as viscous oil (11.5 g, 97.3%). GPC: Mn=6,700, Mw=7,400, PDI=1.1. See Appendix 14 for the spectrum.             134  1 H NMR                   APPENDIX B     135  1 Appendix B1. H NMR spectrum of ditosyl tetraethyleneglycol in CDCl3. 136    Appendix B2. 13 C NMR spectrum of ditosyl tetraethyleneglycol in CDCl3. 137  Appendix B3. FT-IR spectrum of ditosyl tetraethyleneglycol. 138    1 Appendix B4. H NMR spectrum of diazido tetraethyleneglycol in CDCl3. 139    Appendix B5. 13 C NMR spectrum of diazido tetraethyleneglycol in CDCl3. 140    Appendix B6. FT-IR spectrum of diazido tetraethyleneglycol. 141    1 Appendix B7. H NMR spectrum of dipropargyl tetraethyleneglycol in CDCl3. 142    Appendix B8. 13 C NMR spectrum of dipropargyl tetraethyleneglycol in CDCl3. 143    Appendix B9. FT-IR spectrum of dipropargyl tetraethyleneglycol. 144    1 Appendix B10. H NMR spectrum of diallyl tetraethyleneglycol in CDCl3. 145    Appendix B11. FT-IR spectrum of diallyl tetraethyleneglycol. 146    1 Appendix B12. H NMR spectrum of dithiol tetraethyleneglycol in CDCl3. 147    Appendix B13. 13 C NMR spectrum of dithiol tetraethyleneglycol in CDCl3. 148  1 Appendix B14. H NMR spectrum of dithiol tetraethyleneglycol in CDCl3. 149    Appendix B15. FT-IR spectrum of dithiol tetraethyleneglycol. 150  Appendix B16. FT-IR spectrum of thioether-PEO. 151    1 Appendix B17. H NMR spectrum of triazole-PEO in CDCl3. 152    Appendix B18. FT-IR spectrum of triazole-PEO. 153                        REFERENCES     154  REFERENCES 1. Zhao, H. Single ion conductors based on polyelectrolyte grafted nanoparticles. Michigan State University, 2011. 2. Matsumoto, K.; Endo, T., Synthesis of networked polymers with lithium counter cations from a difunctional epoxide containing poly(ethylene glycol) and an epoxide monomer carrying a lithium sulfonate salt moiety. Journal of Polymer Science Part A: Polymer Chemistry 2010, 48 (14), 3113-3118. 3. Sun, X.-G.; Kerr, J. B., Synthesis and Characterization of Network Single Ion conductors based on comb-branched polyepoxide ethers and lithium bis(allylmalonato)borate. Macromolecules 2005, 39 (1), 362-372. 4. Snyder, J. F.; Hutchison, J. C.; Ratner, M. A.; Shriver, D. F., Synthesis of comb polysiloxane polyelectrolytes containing oligoether and perfluoroether side chains. Chemistry of Materials 2003, 15 (22), 4223-4230. 5. Asfour, F. Molecularly Reinforced Polymers and Self-assembled Nanocomposites for Secondary Lithium Batteries. Michigan State University, 2004. 6. (a) Aravindan, V.; Gnanaraj, J.; Madhavi, S.; Liu, H. K., Lithium-Ion conducting electrolyte salts for lithium batteries. Chem.-Eur. J. 2011, 17 (51), 14326-14346; (b) Meyer, W. H., Polymer electrolytes for lithium-ion batteries. Adv. Mater. 1998, 10 (6), 439. 7. Kolb, H. C.; Finn, M. G.; Sharpless, K. B., Click chemistry: Diverse chemical function from a few good reactions. Angew Chem Int Edit 2001, 40 (11), 2004. 8. Hoyle, C. E.; Bowman, C. N., Thiol–ene click chemistry. Angewandte Chemie International Edition 2010, 49 (9), 1540-1573. 9. Norris, B. C.; Li, W.; Lee, E.; Manthiram, A.; Bielawski, C. W., ‘Click’-functionalization of poly(sulfone)s and a study of their utilities as proton conductive membranes in direct methanol fuel cells. Polymer 2010, 51 (23), 5352-5358. 155  10. Martwiset, S.; Yavuzcetin, O.; Thorn, M.; Versek, C.; Tuominen, M.; Coughlin, E. B., Proton conducting polymers containing 1H-1,2,3-triazole moieties. Journal of Polymer Science Part A: Polymer Chemistry 2009, 47 (1), 188-196. 11. Gao, S.; Zhao, C.; Na, H., Chemically stable hybrid polymer electrolyte membranes prepared by silane-crosslinking and thiol-ene click chemistry. Journal of Power Sources 2012, 214 (0), 285-291. 12. Lonergan, M. C.; Nitzan, A.; Ratner, M. A.; Shiriver, D. F., Dynamically disordered hopping, glass transition, and polymer electrolytes. The Journal of Chemical Physics 1995, 103, 3253-3261. 13. Johansson, P.; Tegenfeldt, J.; Lindgren, J., Modelling amorphous lithium salt–PEO polymer electrolytes: ab initio calculations of lithium ion–tetra-, pentaand hexaglyme complexes. Polymer 1999, 40 (15), 4399-4406. 14. Chen, Y.; Baker, G. L., Synthesis and Properties of ABA Amphiphiles. The Journal of Organic Chemistry 1999, 64 (18), 6870-6873. 15. Linford, R. G., Application of electroactive polymers. Chapman & Hall: London, 1993. 16. Wen, S. J.; Richardson, T. J.; Ghantous, D. I.; Striebel, K. A.; Ross, P. N.; Cairns, E. J., FTIR characterization of PEO+LiN(CF3SO2)(2) electrolytes. J Electroanal Chem 1996, 408 (1-2), 113-118. 17. Gujadhur, R.; Venkataraman, D.; Kintigh, J. T., Formation of aryl nitrogen bonds using a soluble copper(I) catalyst. Tetrahedron Letters 2001, 42 (29), 4791-4793. 156