SYNTHESIS AND PROPERTIES OF SUBSTITUTED POLYLACTIDES AND SURFACE FUNCTIONALIZATION OF SMALL SILICA NANOPARTICLES By Quanxuan Zhang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Chemistry - Doctor of Philosophy 2013 ABSTRACT SYNTHESIS AND PROPERTIES OF SUBSTITUTED POLYLACTIDES AND SURFACE MODULATION OF SMALL SILICA NANOPARTICLES By Quanxuan Zhang Polylactides have been widely studied as a degradable material in environmental and biomedical fields due to their biodegradability and biocompatibility. However, the applications of polylactide has been limited by its high crystallinity, hydrophobicity and lack of functionalization; so strategies that provide routes to functionalized polyesters are necessary to produce materials with the desired physical and chemical properties. Development of those strategies has been one of the major efforts to expand the application of polylactides, especially as degradable delivery vehicles. A series of propargyloxy lactic acid derivatives using poly(ethylene glycol) with different lengths as spacers between propargyl group and lactic acid were prepared with high yield via hydrolysis of cyanohydrin from inexpensive starting materials without using column purification. The versatility of the developed method was also tested by the survival of alkene group in the synthesis of allyloxy lactic acid. An azido lactic acid was also prepared to act as an alternative approach to render click function to polylactides. This allows for easy access of building up a library of clickable α-hydroxy acids to prepare clickable substituted polylactides. Post-polymerization modification via click chemistry provides a convenient route to functionalized polylactides with a variety of new properties. We prepared clickable polylpropargyloxylactide that have pendant propargyl groups available for the attachment of chemical functionality using click chemistry without degradation of polymer backbones. The clicked degradable polylactides have thermo-responsive properties with lower critical solution temperatures (LCST) from room temperature to 68 o C. The prepared amphiphilic copolymers, mPEG-b-polypropargyloxylactide were self- assembled into micelles which were core cross-linked to improve the structural stability of the micelles with better controlled release of cargo molecules and response to the reducing reagent. The relationship between the glass transition temperature (Tg) of polylactide and its backbone structure was studied. The direct attachment of rigid and bulky substituents can be used to tune the Tg and thermal properties of polylactides and broaden the scope of their physical properties. Silica nanoparticles (SiNPs) have attracted much attention due to their chemical inertness, low or non toxicity, excellent thermal stability and especially, easy access of surface functionalization. A facile grafting route was developed to prepare stably monodispersed SiNPs in aqueous media via click chemistry. Different sized SiNPs were grafted with a broad range of surface chemistries (carboxylic acids, amines, alcohols and methoxy-terminated polyethylene glycol chains). The resulted SiNPs were purified by dialysis and stably dispersed in water and in vitro cell culture media with narrow size distributions and no aggregation. ACKNOWLEDEMENTS I would like to sincerely thank my advisor, Dr.Greg Baker, for his guidance and support during my graduate study. I value tremendously his training and encouragements in all these years. I would also like to thank my current advisor, Dr. William Wulff, for his scientific guidance and support when serving on my committee, especially during the finalization of my graduate work here at MSU. I would also like to thank Dr. Xuefei Huang, Dr. Mitch Smith and Dr. Christina Chan for their invaluable and helpful discussions and suggestions for my research. Thanks to Dan Holmes, Kermit Johnson, Kathy Severin, Richard Staples, Lijun Chen and Xudong Fan for their help on the instruments and valuable discussions. I would like to thank former and current Baker group members Feng, Qin, Tom, Hui, Gina, Wen, Heyi, Zhe, Greg, Salinda and Olivia for their friendship and profitable discussions. I would also like to thank all of my friends I have met here, for making my life at MSU enjoyable. Finally, I would like to thank my wife Hong for her love and support, as well as our lovely daughter, Aimee. I would also like to thank my parents, sisters, brothers-in-law, and all other family members for their support and encouragement during my study. iv TABLE OF CONTENTS LIST OF TABLES .......................................................................................................... viii LIST OF FIGURES .......................................................................................................... ix LIST OF SCHEMES ...................................................................................................... xvi LIST OF ABBREVIATIONS ........................................................................................... xix CHAPTER 1 INTRODUCTION .................................................................................. 1 1.1 Structure and Properties of Polylactides ...................................................... 1 1.2 Polymerization Methods and Mechanism to Polylactides ............................ 2 1.2.1 Polycondensation to Polylactides .......................................................... 2 1.2.2 Ring Opening Polymerization to Polylactides ........................................ 3 1.2.2.1 Coordination-insertion Mechanism ................................................ 3 1.2.2.2 Organo Catalysis Mechanism ........................................................ 5 1.2.2.3 Cationic Mechanism ...................................................................... 8 1.2.2.4 Anionic Mechanism ........................................................................ 9 1.2.2.5 Enzymatic Mechanism ................................................................. 10 1.3 Functionalization of Polylactides ................................................................ 11 1.3.1 Initiating/End Group Modification Approach ........................................ 12 1.3.2 Monomer Derivative Approach ............................................................ 19 1.3.3 Post-polymerization Modification Approach ........................................ 33 1.3.3.1 Direct Post-polymerization Modification Approach ....................... 33 1.3.3.2 Traditional Synthetic Coupling Approach via Side Functional Groups ......................................................................................... 35 1.3.3.3 Click Chemistry Approach via Side Functional Groups ................ 38 1.4 Smart Polylactide Materials ....................................................................... 44 1.5 Conclusion ................................................................................................. 46 CHAPTER 2 BUILDING UP A LIBRARY OF CLICKABLE α-HYDROXY ACIDS TO CLICKABLE SUBSTITUTED POLYLACTIDES ............................................................. 48 2.1 Introduction ................................................................................................ 48 2.2 Results and Discussion ............................................................................. 50 2.2.1 Synthesis of Propargyl Glycolic Acid ................................................... 50 2.2.1.1 Synthesis of Propargyl Glycolic Acid via an Organoalane ........... 52 2.2.1.2 Synthesis of Propargyl Glycolic Acid via the Enolate of a Malonate and Propargyl Tosylate ................................................................ 53 2.2.2 Synthesis of Propargyloxy Lactic Acid via Hydrolysis of a Cyanohydrin56 2.2.3 Synthesis of Poly(ethylene glycol) Monopropargyl Ether Lactic Acid via Hydrolysis of a Cyanohydrin ............................................................... 60 2.2.3.1 Synthesis of Poly(ethylene glycol) Monopropargyl Ether Lactic Acid (23, n=1,2,3,4) ..................................................................... 61 v 2.2.4 Synthesis of Octa(ethylene glycol) Monopropargyl Ether Lactic Acid (n=8) ................................................................................................... 65 2.2.4.1 Synthesis of Tetraethylene Glycol Monobenzyl Ether (29) .......... 66 2.2.4.2 Synthesis of Monotetrahydropyran-protected Tetraethylene Glycol Monotosylate (35) ........................................................................ 67 2.2.4.3 Synthesis of Octa(ethylene Glycol) Monobenzyl Ether (36) and the Lactic Acid (40, n=8) .................................................................... 68 2.2.5 Synthesis of Poly(ethylene glycol) Monopropargyl Ether Lactic Acid (50, n≈23) ........................................................................................... 70 2.2.6 Synthesis of Allyloxy Lactic Acid (54) via Hydrolysis of Cyanohydrin .. 71 2.2.7 Synthesis of Azido Lactic Acid (56) ..................................................... 72 2.3 Conclusion ................................................................................................. 73 2.4 Experimental Section ................................................................................. 73 CHAPTER 3 SYNTHESIS OF CLICKABLE POLYLACTIDES AND THEIR SMART PROPERTIES 91 3.1 Introduction ................................................................................................ 91 3.2 Results and Discussion ............................................................................. 93 3.2.1 Monomer Synthesis ............................................................................ 93 3.2.2 Bulk Polymerization............................................................................. 96 3.2.2.1 Kinetics of Bulk Polymerization .................................................... 96 3.2.2.2 Thermal Properties of Homopolymers ....................................... 103 3.2.3 Copolymers of P58, P59 and L-Lactide............................................. 106 3.2.4 Thermo-responsive Properties of mPEG/Alkyl-Grafted Polymers from P59 via Click Chemistry .................................................................... 114 3.2.5 Formation of Core Cross-linked Biodegradable Micelles and the Modulation of Redox-Responsive Properties of PEG Block Copolymers of 58 and 59 via Click Chemistry ...................................................... 123 3.3 Conclusion ............................................................................................... 142 3.4 Experimental Section ............................................................................... 143 CHAPTER 4 SYNTHESIS OF POLYSPIROLACTIDES ........................................ 155 4.1 Introduction .............................................................................................. 155 4.2 Results and Discussion ........................................................................... 156 4.2.1 Monomer Synthesis .......................................................................... 156 4.2.2 Solution Polymerization ..................................................................... 157 4.3 Conclusion ............................................................................................... 168 4.4 Experimental Section ............................................................................... 169 CHAPTER 5 SURFACE FUNCTIONALIZATION OF SMALL SILICA NANOPARTICLES VIA CLICK CHEMISTRY .............................................................. 175 5.1 Introduction .............................................................................................. 175 5.2 Results and Discussion ........................................................................... 176 5.2.1 Synthesis of Clicking Reagents ......................................................... 176 5.2.2 SiNPs Surface Modification with mPEG Azides (76) ......................... 178 5.2.3 SiNPs Surface Modification with Different Functionalities ................. 186 vi 5.2.4 Click Surface Modification of Different Sized SiNPs with Azido Carboxylic Acid (83) ......................................................................... 189 5.2.5 In Vitro Colloidal Stability Studies of SiNPs Modified with Different Surface Functionalities ..................................................................... 192 5.3 Conclusion ............................................................................................... 196 5.4 Experimental Section ............................................................................... 196 APPENDIX .................................................................................................................. 206 REFERENCES ............................................................................................................ 239 vii LIST OF TABLES Table 1. Propargyl glycolic acid (1) after treatment with different reductants/bases ...... 56 Table 2. Deprotection of 9 with various acids and solvents at room temperature ......... 58 Table 3. Hydrolysis of 11 at various concentrations of H2SO4, temperatures and solvents ......................................................................................................................... 60 a Table 4. Bulk Polymerizations of 58 and 59 at 130 oC ............................................... 101 Table 5. Mn and Tg of copolymers (P58-co-L-LA, P59-co-L-LA) ................................ 107 a Table 6. Block copolymers of mPEG (initiator) and monomers 58 and 59. ............... 125 Table 7. CMC and contact angles of polymers P59 and P59B, P59C and P59D ........ 126 Table 8. Properties of UCMS and CMs ....................................................................... 128 Table 9. Hydrodynamic diameters and PDI of UCMs and CMs in different solvent ..... 140 Table 10. Solution polymerization rates of 72a, 72b and 72c. .................................... 161 a Table 11. Solution polymerizations of 72a, 72b and 72c at -20 oC. ........................... 164 Table 12. Surface grafting of silica nanoparticles (25 nm)........................................... 182 Table 13. Surface grafting of SiNPs (25 nm) with different functionalities. .................. 187 Table 14. Surface grafting of different sized SiNPs with carboxylic acid (83). ............. 190 viii LIST OF FIGURES Figure 1. Examples of functionalized polylactides through initiating group approach .... 14 Figure 2. Examples of functionalized polylactides via end group modification .............. 18 Figure 3. Substituted polylactides and their Tg .............................................................. 22 Figure 4. Structure of poly(ethylene glycol) monopropargyl ether lactic acid ................ 61 Figure 5. Structure of octa(ethylene glycol) monopropargyl ether lactic acid ................ 65 1 Figure 6. H NMR spectra of 58, P58A, 59 and P59A. Slovent: CDCl3. ....................... 95 Figure 7. Molecular drawing of 59 with 50% thermal ellipsoids from X-ray crystallography. “For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this thesis (or dissertation).” ............. 96 Figure 8. Percent conversion for the bulk polymerization of 58 and 59. Conditions: 130 o C, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. ............................................................................... 98 Figure 9. Bulk polymerization kinetics for 58 and 59. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. ..................................................................................... 99 Figure 10. The evolution of molecular weight (Mn) from GPC (▲, P58; ■, P59) and PDI (∆, P58; □, P59) with time for bulk polymerization of 58 and 59. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. ................................................................................... 100 Figure 11. GPC curves of polymers from bulk polymerization of 58 with different [M]/[I] (130 oC). ...................................................................................................................... 102 Figure 12. GPC curves of polymers from bulk polymerization of 59 with different [M]/[I] (130 oC). ...................................................................................................................... 102 Figure 13. TGA curves of polymers from the bulk polymerization of 58 with different [M]/[I] (130 oC). ............................................................................................................ 104 Figure 14. TGA curves of polymers from the bulk polymerization of 59 with different [M]/[I] (130 oC). ............................................................................................................ 104 Figure 15. DSC traces of polymers from the bulk polymerization of 58 with different [M]/[I]. (second heating scan, 10 oC/min in N2) ........................................................... 105 ix Figure 16. DSC traces of polymers from the bulk polymerization of 59 with different [M]/[I]. (second heating scan, 10 oC/min in N2) ........................................................... 106 Figure 17. TGA curves of P58-co-L-LA with different feed ratios of L-LA to 58. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. ................................................... 108 Figure 18. TGA curves of P59-co-L-LA with different feed ratios of L-LA to 59. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. ................................................... 108 Figure 19. DSC curves of P58-co-L-LA with different feed ratios of L-LA to 58. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. ................................................... 110 Figure 20. DSC curves of P59-co-L-LA with different feed ratios of L-LA to 59. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. ................................................... 111 Figure 21. Plot of 1/Tg versus weight percentage of 58 and 59 in their copolymers. (Solid lines were generated using Fox equation (4) based on the measured Tg values of homopolymers) ........................................................................................................... 111 Figure 22. 125 MHz 13 C NMR carbonyl region of homopolymers PLLA, P58 and copolymer P58-co-L-LA. Solvent: CDCl3. .................................................................... 113 Figure 23. 125 MHz 13 C NMR carbonyl region of homopolymers PLLA, P59 and copolymer P59-co-L-LA. Solvent: CDCl3. .................................................................... 114 1 Figure 24. H NMR spectra of P59 and its click-grafted polymers P60 (P59-g-m3PEG), P61 (P59-g-C10H21) and P62 (P59-g-C10H21/m3PEG, 2 : 1 feed ratio of m3PEGN3 : C10H21N3). Solvent: CDCl3. ........................................................................................ 117 Figure 25. FT-IR spectra of P59 and its click-grafted polymers P60 (P59-g-m3PEG), P61 (P59-g-C10H21) and P62 (P59-g-C10H21/m3PEG, feed ratio of m3PEGN3 and C10H21N3 is 2 : 1). ...................................................................................................... 118 Figure 26. Plots of absorbance as a function of temperature for aqueous solutions (~3 mg/mL, in D2O) of P60 and P62. The solid lines are the non-linear curve fit to the data. .................................................................................................................................... 120 Figure 27. Plot of LCST as a function of mole fraction of m3PEG in polymers P60 and P62. The solid line is the linear fit to the data. ............................................................. 120 Figure 28. Visualization of thermo-responsive properties of P60 (P59-g-m3PEG), P62 (P59-g-C10H21/m3PEG, 2:1 and 7 : 2 feed ratio of m3PEGN3 : C10H21N3) at x temperatures below (left) and above (right) their LCST. Solvent: D2O........................ 121 Figure 29. DLS of P60 (P59-g-m3PEG), P62 (P59-g-C10H21/m3PEG, 2:1 and 7 : 2 feed ratio of m3PEGN3 : C10H21N3) at temperatures below (filled symbol) and above (open symbol) their LCST. Solvent: D2O. ................................................................... 121 1 Figure 30. H NMR spectra of P60 (P59-g-m3PEG), P62 (P59-g-C10H21/m3PEG, 7 : 2 feed ratio of m3PEGN3 : C10H21N3) at temperatures below and above their LCST. Solvent: D2O. The broken line denotes the significant changes OF chemical shift...... 123 1 Figure 31. H NMR spectra of UCMs prepared from P59B, P59C and P59D in D2O. 128 Figure 32. Visualization of UCMs’ integrity stability prepared from P58B, P59B, P59C and P59D in D2O without (a) and with (b) addition of aqueous CuSO4. ..................... 130 Figure 33. Hydrodynamic diameters of prepared UCMs and CMs versus aging time (Cross-linker: 63). ........................................................................................................ 133 1 Figure 34. H NMR spectra of UCMP59D and CMP59D (cross-linker: 63) in DMSO-d6. .................................................................................................................................... 134 Figure 35. DLS size distribution of hydrodynamic diameters of (a) UCMs and (b) CMs in water before and after 10-fold dilution with DMF (Cross-linker: 63). ........................... 135 Figure 36. Continual release profiles of Nile red from NR-loaded uncross-linked micelles and cross-linked micelles. ........................................................................................... 137 Figure 37. DLS size distribution of both uncross-linked and cross-linked micelles before and after nile red release............................................................................................. 137 Figure 38. DLS size distribution of (a) CM*P59C and (b) CM*P59D in water before and after 10-fold dilution with DMF with or without 10 nM DTT (Cross-linker:64)............... 141 Figure 39. DLS size distribution of (a) CM**P59C and (b) CM**P59D in water before and after 10-fold dilution with DMF with or without 10 nM DTT (Cross-linker: 63:64 = 2:8). ............................................................................................................................. 141 1 Figure 40. H NMR spectra of TBD in CD2Cl2 at different time and temperature. ...... 159 Figure 41. Conversions for solution polymerization of 72a, 72b and 72c. Conditions: -20 o C in CH2Cl2, [M] = 1 M, [M]/[cat]/[I] = 100:1:1 for 72a and 72b; [M]/[cat]/[I] = 100:1.5:1 for 72c. ........................................................................................................................ 161 Figure 42. Solution polymerization kinetics of 72a, 72b and 72c. Conditions: -20 oC in xi CH2Cl2, [M] = 1 M, [M]/[cat]/[I] = 100:1:1 for 72a and 72b; [M]/[cat]/[I] = 100:1.5:1 for 72c. ............................................................................................................................. 162 Figure 43. The evolution of molecular weight (Mn) from GPC and PDI with time for 72a, 72b and 72c. Conditions: -20 oC in CH2Cl2, [M] = 1 M, [M]/[cat]/[I] = 100:1:1 for 72a and 72b; [M]/[cat]/[I] = 100:1.5:1 for 72c. ............................................................................ 163 Figure 44. GPC profiles of P72a with different [M]/[I]. ................................................. 164 Figure 45. GPC profiles of P72b with different [M]/[I]. ................................................. 165 Figure 46. GPC profiles of P72c with different [M]/[I]. ................................................. 165 Figure 47. TGA curves of P72a, P72b and P72c. Conditions: [M] = 1 M, [M]/[I] = 300:1 for 72a, 72b and 72c, -20 oC in CH2Cl2....................................................................... 167 Figure 48. DSC scans of P72a, P72b and P72c. Conditions: [M] = 1 M, [M]/[I] = 300:1 for 72a, 72b and 72c, -20 oC in CH2Cl2....................................................................... 168 Figure 49. Normalized FT-IR spectra of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different mPEG azides (76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000). .................................................................. 181 Figure 50. Normalized TGA curves of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and click-modified silica particles with different mPEG azides (76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000). .......................................................... 183 Figure 51. DLS profiles (size distribution by intensity) of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different mPEG azides (76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000) after 4 h sonication at 60 oC. . 185 Figure 52. Dispersion stability of silica particles (SiNP25) click-modified with different mPEG azides (25 nm; 76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000) in aqueous solution. ........................................................................................................ 186 Figure 53. Structural stability of click-grafted surface layer of silica particles modified with alkynyl silane (SiNP25/74) and mPEG azide (SiNP25/76a, 4PEG) during 4 h sonication at 60 oC. ..................................................................................................... 186 Figure 54. Normalized FT-IR spectra of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different surface functionalities (76a: 4PEGOMe; 78: 4PEG-OH; 79: 4PEG-NH2; 83: 4PEG-COOH). ........................................... 188 Figure 55. DLS profiles (size distribution by intensity) of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different functional azides (76a: 4PEG-OMe; 78: 4PEG-OH; 79: 4PEG-NH2; 83: 4PEG-COOH).................................. 189 xii Figure 56. Normalized FT-IR spectra of different sized SiNPs click-modified with 83 (4PEG-COOH after 4 h sonication. Core diameter: SiNP8, 8 nm; SiNP25, 25 nm; SiNP45, 45 nm; SiNP70, 70 nm. ................................................................................. 191 Figure 57. DLS profiles (size distribution by intensity) of different sized SiNPs clickmodified with 83 (4PEG-COOH) after 4 h sonication. Core diameter: SiNP8, 8 nm; SiNP25, 25 nm; SiNP45, 45 nm; SiNP70, nm. ............................................................ 192 Figure 58. Size distribution of SiNPs (SiNP25/76a, -OMe terminated) when incubated in media at 0, 21, 42 and 68 h. ........................................................................................ 194 Figure 59. Size distribution of SiNPs (SiNP25/78, -OH terminated) when incubated in media at 0, 21, 42 and 68 h. ........................................................................................ 194 Figure 60. Size distribution of SiNPs (SiNP25/83, -COOH terminated) when incubated in media at 0, 20, 42 and 68 h. ........................................................................................ 195 Figure 61. Size distribution of SiNPs (SiNP25/79, -NH2 terminated) when incubated in media at 0, 21, 42 and 68 h. ........................................................................................ 195 1 13 1 13 1 13 Figure A 1. H and Figure A 2. H and Figure A 3. H and C NMR spectra of propargyl glycolic acid (1) in CDCl3. ............ 207 C NMR spectra of propargyloxy lactic acid (12) in CDCl3. ........ 208 C NMR spectra of poly(ethylene glycol) monopropargy lether lactic acid (23, n =1) in CDCl3. ............................................................................................. 209 1 Figure A 4. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (23, n =2) in CDCl3. ............................................................................................. 210 1 Figure A 5. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (23, n =3) in CDCl3. ............................................................................................. 211 1 Figure A 6. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (23, n =4) in CDCl3. ............................................................................................. 212 1 Figure A 7. H and 13 C NMR spectra of tetraethylene glycol monobenzyl ether (29) in CDCl3. ......................................................................................................................... 213 1 Figure A 8. H and 13 C NMR spectra of monotetrahydropyran-protected tetraethylene glycol monotosylate (35) in CDCl3. ............................................................................. 214 1 Figure A 9. H and 13 C NMR spectra of octa(ethylene glycol) monobenzyl ether (36) in xiii CDCl3. ......................................................................................................................... 215 1 Figure A 10. H and 13 C NMR spectra of 38 in CDCl3................................................ 216 1 Figure A 11. H and 13 C NMR spectra of octa(ethylene glycol) monopropargyl ether lactic acid (40) in CDCl3. ............................................................................................. 217 1 Figure A 12. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (50) in CDCl3. ............................................................................................. 218 1 13 1 13 1 13 1 13 Figure A 13. H and Figure A 14. H and Figure A 15. H and Figure A 16. H and C NMR spectra of allyloxy lactic acid (54) in CDCl3. ............... 219 C NMR spectra of azido lactic acid (56) in DMSO-d6.............. 220 C NMR spectra of hetero-monomer (58) in CDCl3. ................. 221 C NMR spectra of homo-monomer (59) in CDCl3. .................. 222 1 Figure A 17. H NMR spectra of polymer P58A in CDCl3. .......................................... 223 1 Figure A 18. H NMR spectra of polymer P59A in CDCl3. .......................................... 223 1 Figure A 19. H NMR spectra of polymer P58B in CDCl3. .......................................... 224 1 Figure A 20. H NMR spectra of polymer P59B in CDCl3. .......................................... 224 1 Figure A 21. H NMR spectra of polymer P59C in CDCl3. .......................................... 225 1 Figure A 22. H NMR spectra of polymer P59D in CDCl3. .......................................... 226 Figure A 23. Water contact angle of glass substrate (a); P58 film (b); P58B film (c); P59 film (d); P59B film (e); P59C film (f); P59D film (g); P60 film (h). ................................. 227 Figure A 24. TEM of UCMP58B. Scale bar: 100 nm. .................................................. 228 Figure A 25. TEM of UCMP59B Scale bar: 50 nm. ..................................................... 228 Figure A 26. TEM of UCMP59C Scale bar: 100 nm. ................................................... 229 Figure A 27. TEM of CMP59C from cross-linker 63. Scale bar: 50 nm. ...................... 229 Figure A 28. TEM of CM*P59C from cross-linker 64. Scale bar: 50 nm. ..................... 230 Figure A 29. TEM of UCMP59D. Scale bar: 200 nm. .................................................. 230 xiv Figure A 30. TEM of CMP59D from cross-linker 63. Scale bar: 100 nm. .................... 231 Figure A 31. TEM of CM*P59D from cross-linker 64. Scale bar: 200 nm. ................... 231 1 13 1 13 1 13 Figure A 32. H and Figure A 33. H and Figure A 34. H and C NMR spectra of monomer (72a) in CDCl3. .......................... 232 C NMR spectra of monomer (72b) in CDCl3. .......................... 233 C NMR spectra of monomer (72c) in CDCl3. .......................... 234 1 Figure A 35. H NMR spectra of polymer (P72a) in CDCl3. ........................................ 235 1 Figure A 36. H NMR spectra of polymer (P72b) in CDCl3. ........................................ 235 1 Figure A 37. H NMR spectra of polymer (P72c) in CDCl3. ........................................ 236 Figure A 38. TEM images of SiNP25 (a); SiNP25/74 (b); SiNP25/76a (c); SiNP25/78 (d); SiNP25/83 (e); SiNP25/79 (f). ..................................................................................... 237 xv LIST OF SCHEMES Scheme 1. Mechanism of polycondensation of lactic acid to PLA ................................... 3 Scheme 2. Proposed Coordination-insertion mechanism for lactide ring-opening polymerization ................................................................................................................. 5 Scheme 3. Proposed mechanism of DMAP-catalyzed lactide ring opening polymerization ................................................................................................................. 6 Scheme 4. Bifunctional thiourea-tertiary amine catalyst and its proposed intermediate .. 8 Scheme 5. Proposed cationic polymerization mechanism .............................................. 9 Scheme 6. Proposed anionic polymerization mechanism ............................................. 10 Scheme 7. Strategies to functionalized Polylactides. “For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation.”.................................................................................................................. 12 Scheme 8. Examples of chain-end clickable polylactides capable of further modification through initiating group approach .................................................................................. 16 Scheme 9. General synthesis of functionalized lactide monomer derivative ................. 19 Scheme 10. General preparation routes to substituted polylactides ............................. 20 Scheme 11. Preparation of hydroxy group tethered polylactides .................................. 24 Scheme 12. Poly(α-hydroxy acids) (Polylactides) via O-carboxyanhydride routes........ 25 Scheme 13. Preparation of amine tethered polylactides ............................................... 26 Scheme 14. Preparation of carboxylic acid tethered polylactides ................................. 27 Scheme 15. Synthesis and polymerization of poly(ethylene glycol) monomethyl ethermodified polyglycolides. ................................................................................................ 29 Scheme 16. Castillo’s monomer derivative approach to poly(ethylene glycol) monomethyl ether-modified PLA ................................................................................... 30 Scheme 17. Synthesis of allyl functionalized polylactides ............................................. 31 Scheme 18. Synthesis of alkynyl functionalized polylactides ........................................ 32 Scheme 19. Passerini-type condensation route to versatile functionalized lactide monomers ..................................................................................................................... 32 xvi Scheme 20. Functionalization of bulk polylactide via direct post-polymerization modification using free radical initiator .......................................................................... 35 Scheme 21. Functionalization of hydroxylated polylactide via EDC coupling ................ 36 Scheme 22. Functionalization of hydroxylated polylactide via DCC coupling ................ 37 Scheme 23. Post-polymerization modification of polylactides via tetrazine-norbornene click chemistry ............................................................................................................... 39 Scheme 24. Post-polymerization modification of polylactides via thiol-ene click chemistry ....................................................................................................................... 40 Scheme 25. Baker’s synthesis of functionalized polylactides via Cu(I) catalyzed click chemistry ....................................................................................................................... 41 Scheme 26. Rubinshtein’s synthesis of functionalized polylactides via Cu(I) catalyzed click chemistry ............................................................................................................... 42 Scheme 27. Zhang’s synthesis of functionalized polylactides via both Cu(I) catalyzed and thiol-yne click chemistry ......................................................................................... 43 Scheme 28. General synthetic routes to substituted lactide monomers ........................ 49 Scheme 29. Synthetic route for propargyl glycolide and its polymer ............................. 51 Scheme 30. Failed synthetic procedures conducted to synthesize 1 ............................ 51 Scheme 31. Epoxide ring opening by organoalane in the synthesis of 1 ...................... 52 Scheme 32. Synthesis of propargyl glycolic acid from diethyl acetamidomalonate ....... 54 Scheme 33. Synthesis of propargyloxy lactic acid (12) ................................................. 58 Scheme 34. Synthesis of poly(ethylene glycol) monopropargyl ether lactic acid (23, n=1,2,3,4) ...................................................................................................................... 64 Scheme 35. Synthesis of tetraethylene glycol monobenzyl ether (29) .......................... 66 Scheme 36. Synthesis of monotetrahydropyran-protected tetraethylene glycol monotosylate (33).......................................................................................................... 68 Scheme 37. Synthetic route to octa(ethylene glycol) monopropargyl ether lactic acid (40) ................................................................................................................................ 69 Scheme 38. Synthesis of poly(ethylene glycol) monopropargyl ether lactic acid (50, n≈ 23) ................................................................................................................................. 70 xvii Scheme 39. Synthesis of allyloxy lactic acid (54) .......................................................... 72 Scheme 40. Synthesis of azido lactic acid (56) ............................................................. 73 Scheme 41. Synthesis of propargyloxy lactides and their polymers .............................. 94 Scheme 42. Click functionalization of P59 with alkyl and PEG azides ........................ 116 Scheme 43. Preparation of UCMs and CMs via click chemistry and their response to environment change. ................................................................................................... 132 Scheme 44. Synthesis of spirolactide monomers and polyspirolactides ..................... 157 Scheme 45. Synthesis of linear clicking reagents. ...................................................... 178 Scheme 46. Surface modification of silica nanoparticles using click chemistry. .......... 180 xviii LIST OF ABBREVIATIONS TBBA 4-tert-Butylbenzylalcohol br Broad CM Cross-linked micelle CMC Critical micelle concentration d Doublet δ Chemical shift DCC N,N'-Dicyclohexylcarbodiimide dd Doublet of doublet DLS Dynamic light scattering DMAP 4-Dimethylaminopyridine DMF N,N-Dimethyl-formamide DMSO-d6 Deuterated dimethyl sulfoxide DSC Differential scanning calorimetry FT-IR Fourier transform infrared spectroscopy GPC Gel permeation chromatography HOBT Hydroxybenzotriazole J Coupling constant LCST Lower critical solution temperature m Multiplet Mn Number average molecular weight m.p. Melting point xix mPEG Poly(ethylene glycol) monomethyl ether Mw Weight average molecular weight MWCO Molecular weight cutoff HRMS High resolution mass spectroscopy NMR Nuclear magnetic resonance p Pentet PDI Polydispersity index PEG Poly(ethylene glycol) PLA polylactide q Quartet RPMI 1640 media Roswell Park Memorial Institute 1640 media rt Room temperature s Singlet SiNP Silica nanoparticles Sn(Oct)2 Tin(II)-2-ethylhexanoate t Triplet TBD Triazabicyclodecene TEM Transmission electron microscopy Tg Glass transition temperature Tm Melting temperature TGA Thermal gravimetric analysis THF Tetrahydrofuran xx UV-Vis Ultraviolet-Visible xxi CHAPTER 1 INTRODUCTION 1.1 Structure and Properties of Polylactides Polylactide (PLA) is a biodegradable thermoplastic derived from lactic acid. Both polylactide and poly(lactic acid) are abbreviated as PLA since they have the same chemical structure, and since they only differ from each other in means of production. PLA can be prepared either by ring-opening polymerization (ROP) of lactides or by condensation polymerization of lactic acid which are obtained from economical 1 fermentation of corn, beet-sugar, cane-sugar etc. PLA has high mechanical properties, thermal plasticity, fabric ability, excellent biodegradability and biocompatibility. Notably these polymers are not only biodegradable (the aliphatic polyester backbone is intrinsically sensitive to water and heat) but also bio-assimilative, since the lactic acids generated by hydrolysis under physiological conditions are nontoxic components and can be eliminated via the Krebs cycle as water and carbon dioxide. 2 Glass transition temperature (Tg) is another important physical characteristic of polymers since it defines the useful temperature range of polymers for many applications. 2,3 Heating polymers above their Tg affects many physical properties and causes increased permeability, loss of dimensional stability, and increased resilience. Due to fairly low Tg of PLA (50 – 60 C), crystalline poly(L-lactide) (PLLA) has been used in structural applications such as fibers and packaging, sutures, 7,8 4 orthopedic and bone fragment fixations, 9 5,6 absorbable and scaffolds for tissue engineering. Rubbery polylactides (lower Tg) at 1 room temperature degrade faster, and are preferred for applications such as controlled 10,11,12 drug delivery and sustained drug release. 1.2 Polymerization Methods and Mechanism to Polylactides 1.2.1 Polycondensation to Polylactides Polycondensation of lactic acids in the presence of acid catalysts yields PLA and water as byproduct. The mechanism (Scheme 1) is the step-growth esterification polymerization by removing water to favor equilibrium. 13 Recently, this approach has gained some attention and it has been reported that binary catalysts comprised of metal compounds activated by co-catalysts are more effective catalysts in polycondensation. Although the polycondensation of L-lactic acid using a single catalyst (titanium(IV) butoxide) was reported to give PLA with high molecular weight up to 130,000 Da, the elongated reaction time (> 40 h) and wide PDI (>3.5) hindered its application. 14 The combination of SnCl2 and p-toluenesulfonic acid (TSA) in melt polycondensation of Llactic acid resulted in PLA with molecular weight up to 100,000 Da with low PDI (~1.5) and little racemization. 15 Using triphenylphosphine (PPh3) as a co-catalyst, polycondensation of lactic acid catalyzed by SnCl2 could similarly provide PLA with high molecular weight and low PDI. 16 Although this approach is relatively simple and inexpensive, it has several disadvantages. It is difficult to control the molecular weight, the polydispersity index (PDI), the PLA end groups, and the syntheses of copolymers with controlled sequences. As a consequence, there is increasing interest in methods that allow for the preparation of PLA polymers in a more reproducible and controlled 2 fashion. Scheme 1. Mechanism of polycondensation of lactic acid to PLA 1.2.2 Ring Opening Polymerization to Polylactides The ring opening polymerization of lactide allows for a much higher control of lactide polymerization and remains by far the most widely used method for the synthesis of well-defined PLA materials. Mainly, there are five well-accepted polymerization mechanisms for lactide which are coordination-insertion, organo-catalysis, cationic, anionic and enzymatic mechanisms. The coordination-insertion and organocatalytic polymerizations are undoubtedly the most efficient and general methods reported so far for the ring opening polymerization of lactides. 1.2.2.1 Coordination-insertion Mechanism There are a variety of catalysts which are studied and utilized in the effective polymerization of lactide, such as Al, Zn and Sn derivatives. Among them, tin(II) bis(2ethylhexanoate) (Sn(Oct)2) is the most widely used complex as catalyst for the preparation of PLA in industry and research due to its commercial availability, easiness of handling, low toxicity, and good solubility in common organic solvents and in melt monomers. It is highly active (from minutes to a few hours in bulk at 140-180 °C) and 5 6 allows for the preparation of high-molecular-weight polymers (up to 10 or even 10 Da 3 in the presence of a alcohol co-initiator) with really low PDI. The mechanism is as shown in Scheme 2. 17 Two molecules of methanol coordinate to Sn(Oct)2 with retention of the two octanoate moieties through hydrogen bonds formed between alcohol and octanoate ligands. Then lactide coordinates to the complex and induces a proton migration from methanol to the nearby octanoate ligand. Subsequently, the formed methoxy group on tin nucleophilically attacks the carbonyl of the coordinated lactide followed by ring opening to form the ring opening product with a hydroxyl end group which coordinates with the catalyst and allows for chain propagation. Several other research groups have also suggested that the alcohol added as co-initiator might substitute at least one octanoate group in a rapid equilibration, and the resulting Snalkoxide is then acting as the true initiator of the polymerization process, which also falls into the coordination-insertion mechanism. 18,19 Ring-opening polymerization of lactides with organometallic catalysts at high temperatures or with long reaction times often leads to both inter- and intra- molecular transesterification reactions resulting in an increase in PDI of the polylactides. 4 Scheme 2. Proposed Coordination-insertion mechanism for lactide ring-opening polymerization 1.2.2.2 Organo Catalysis Mechanism Although Sn(Oct)2 has been approved by the U.S. FDA, the toxicity related to most tin compounds is still one considerable drawback in biomedical applications. Thus organo-catalyzed lactide polymerization has been widely studied due to promising applications in the biomedical field and is considered to be one of the viable alternatives to organometallic approaches. Organo catalysts are easy to prepare, handle, and are typically nontoxic, and thus have generated much attention from both industry and academia. Hedrick and co-workers reported the first use of the organic catalyst 4(dimethylamino)pyridine (DMAP) in the ring opening polymerization of lactide which 20 provided well controlled molecular weight and low PDI (~1.1). Later on, other super bases, (DBU), such as 1,8-Diazabicyclo[5.4.0]undec-7-ene 5 21 1,5,7- Triazabicyclo[4.4.0]dec-5-ene (TBD) 22 and sparteine 23 were also reported to have excellent catalytic activity to afford polylactide with high yield, controlled molecular weight and narrow PDI. Taking DMAP as an example (Scheme 3), it serves to activate a carbonyl of the lactide ring and then nucleophilic attack of an alcohol gives a ring opened structure with an ester functionality on one end and a hydroxyl functionality on the other that reacts with the next lactide to continue chain propagation. Hedrick and coworkers also claimed that a secondary alcohol functionality is the best compromise for propagating lactide ROP without inducing undesirable transesterification of ringopened products. This is supported by model reactions where primary alcohols react not only with lactide but also with ring-opened esters, whereas secondary alcohols only react with lactide and are inert toward ring opened products. This selectivity is responsible for the living character of nucleophilically catalyzed lactide polymerization. 24 Scheme 3. Proposed mechanism of DMAP-catalyzed lactide ring opening polymerization N-Heterocyclic carbenes (NHC) have showed high catalytic activity in ring opening polymerization of lactide 25 through a mechanism similar to that for DMAP by nucleophilically activating the lactide monomer, achieving near quantitative conversions for target degrees of polymerization (feed ratio: 30 to 200) within 24 hours with narrow PDI. The living nature of the polymerization was proved by charging additional lactide to 6 the system to afford further PLA chain growth with a final molecular weight close to theoretical value. 26,27 This type of catalyst also enables the selective construction of complex polymer architectures. For example, in the presence of alcohols/co-initiators, linear polylactides are generated using N-heterocyclic carbene catalysts, whereas in the absence of alcohols, N-heterocyclic carbene catalysts result in cyclic polylactides with high molecular weight and narrow PDI. Hedrick and coworkers also utilized nucleophilic tertiary phosphines as a metal-free catalyst for the ring-opening polymerization of lactides. 28 The lactide was activated by the Lewis basic phosphine towards attack by the nucleophilic alcohol. However, tertiary phosphine catalysts only exhibit high activity in bulk polymerization at high temperature. The activity in solution is fairly low with polymerization requiring up to several weeks to achieve high conversion. A bifunctional thiourea−tertiary amine catalyst was reported by Hedrick and coworkers in 2005 for living polymerization of lactide. 29 Polylactides with predictable molecular weights and extremely narrow PDI ( 1.05) were produced under mild and highly selective polymerization conditions, which is the characteristic of a living polymerization. The bifunctional nature of the catalyst was proposed to be involved in the simultaneous activation of the initiating/propagating alcohol by an acid−base interaction with the tertiary amine moiety and the activation of the carbonyl of the lactide monomer via hydrogen bonding by the thiourea moiety of the catalyst (Scheme 4). 7 Scheme 4. Bifunctional thiourea-tertiary amine catalyst and its proposed intermediate 1.2.2.3 Cationic Mechanism There are limited strong acids or carbenium ion donors capable of initiating a cationic polymerization of lactides, such as trifluoromethanesulfonic acid (HOTf) and methyl trifluoromethanesulfonate (MeOTf). 30 According to the fact that poly(L-lactide) with methyl ester end groups exhibit 100% retention of the optical purity of the lactide 1 monomer as determined by H NMR analysis and optical rotation measurements when using either MeOTf or HOTf as the initiator. Thus, the cationic mechanism was proposed to occur via a protonation or alkylation of the carbonyl O-atom of lactide with the consequence that the O–CH bond is subject to electrophilic activation. This activated O-CH bond is cleaved by the nucleophilic attack of another monomer which occurs repeatedly until a termination step is reached, ruling out the possibility that the polymerization occurs via cleavage of the acyl-oxygen bond (Scheme 5). 8 Scheme 5. Proposed cationic polymerization mechanism However, it was reported 31 that the cationic copolymerization of lactide and glycolide was far from a living polymerization under the reported cationic polymerization conditions. By varying monomer-to-initiator ratios from 50 to 400, the resulting polymer samples showed very similar viscosities (0.15 - 0.27 dl/g), which indicates the nonliving nature of the polymerization. 1.2.2.4 Anionic Mechanism The effective initiators used for anionic polymerization of lactides are alkali metals and alkali metal oxides, etc. The polymerization may proceed either by a living or a nonliving process depending on the reaction conditions, initiators, and monomers. The reaction may be initiated by either deprotonation of lactide nucleophilic attack on the carbon of the carbonyl group 33 32 (Scheme 6 (a)) or (Scheme 6 (b)) followed by chain propagation resulting in formation of linear polylactide. These two initiation steps can be differentiated by end group analysis of the resulted PLA by whether or not initiator fragments are incorporated in the polymer. The reported experimental results strongly support both pathways depending on the basicity of the initiators. 34 Initiators formed in situ (such as metal alkoxides) can increase the efficiency of ring opening 9 polymerization compared to their commercial materials. The racemization and transesterification can also be minimized by carefully choosing the right initiators. Scheme 6. Proposed anionic polymerization mechanism Although higher activities can be anticipated for anionic initiators that typically display strong nucleophilic and/or basic character, the deleterious contributions from trans-esterification and racemization reactions are not neglectable when trying to achieve well controlled polymerizations when naked or loosely bonded anionic species are involved. 1.2.2.5 Enzymatic Mechanism The use of enzymes as tools in in vitro catalysis is increasing interest. Enzymes have been recently used in ring opening polymerization to make polyesters. 35 Enzymes offer an advantage over conventional catalysts, namely, selectivity and biocompatibility. Most commerically available lipases have proved to be active in ring opening polymerization of lactide. 36 , 37 The lipase-catalyzed polymerization of lactides is believed to share a mechanism similar to the organo catalysts discussed above. The 10 catalytic site of the lipase has a serine residue which activates the lactide to form an acyl-enzyme intermediate, and this intermediate suffers nucleophilic attack by terminal hydroxy group of the propagating polymer chain leading to chain propagation and regeneration of the enzyme. 38 However, the reported PLAs from enzyme catalysis have either low molecular weights, relatively wide polydispersity indexes or elongated polymerization times of days, which currently limits the application of PLAs formed from enzyme catalysts. 1.3 Functionalization of Polylactides However, due to the low solubility of polylactides in water, the applications of polylactides in an aqueous environment have been greatly restricted. 39 In addition, the lack of functional group diversity in the backbones and side chains in polylactide makes further modification extremely difficult. So functionalization of polylactides has been a major effort in expanding the application of polylactides, especially as degradable delivery vehicles. Three strategies have been developed to introduce functionality in the polylactide scaffold: the initiating/end group modification approach, the monomer derivative approach and the post-polymerization modification approach (Scheme 7). All three routes allow the incorporation of functional groups onto the polymer backbone or side chains or to introduce access points for further modification to tune the physical properties of PLA. The first approach has been widely investigated, and since it involves the conversion of lactide to polylactides, it is often an integral part of the other two approaches as well. 11 Scheme 7. Strategies to functionalized Polylactides. “For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation.” 1.3.1 Initiating/End Group Modification Approach Due to the limited scope of their applications, the properties of polylactides have been thoroughly modified and improved by the initiating/end group modification method. A variety of functionalities have been incorporated into polylactides either using different functional initiators during the polymerization or introducing different functionalities into the ω-hydroxyl group by treating with reactive functional molecules. Polyethylene glycol-co-polylactide (PEG-co-PLA) block copolymers have gained much attention due to their amphiphilic properties, their ability to form nanoparticles and 12 their application as delivery vehicles (Figure 1 (a)). Li and coworkers prepared di-block and tri-block copolymers of PEG-co-PLA using mono- or di- hydroxy PEG as initiators. 40 The hydrophilicity of these copolymers was dramatically improved and hydrogels could be formed in water for both di-block and tri-block copolymers. Longer PEG segments could prevent the crystallization of PLA segments in the copolymer. PEG-PLA copolymers were also formulated into micelles and their degradation behavior 41 was also studied. Micelles made from the dialysis method are more stable than those from the direct dissolution method which showed a collapsed micellar structure. The composition of PLA-PEG copolymers greatly affects the degradation of micelles. Micelles with longer hydrophilic PEG blocks exhibit more size changes due to the less compact structure compared to copolymers with longer PLA segments. Polysaccharides (TMS protected Pulluan) and a monosaccharide (benzyl protected α-glucose) have also been introduced into polylactides (Figure 1 (b)) to improve hydrolytic degradability and surface wettability. 42,43 As expected, the higher hydrophilicity due to the presence of saccharides in the copolymer resulted in faster degradation. Amino, halide, pyrene (fluorescent) and other functionalities were also reported to have been incorporated into polylactides either through direct initiation of polymerization or post-deprotection of the polylacitdes (Figure 1 (c)). 44,45,46 13 Figure 1. Examples of functionalized polylactides through initiating group approach Another method for introducing functionality via initiating group modification is to first install a moiety that is capable of undergoing “click” chemistry, thus enabling further modification through super easy and efficient chemistry. If the polylactide chain end is caped with 1,6-heptadiyne, which it can be further cyclo-polymerized to afford brush polymers containing a semiconducting polyene backbone (Scheme 8 (a)). The incorporation of an alkyne into polylactides was not affected by the presence of terminal alkyne group and PLA was obtained with well controlled molecular weight and narrow PDI. 47 However, it was reported that the alkyne function could be affected in the 14 presence of Sn(Oct)2 probably due to oxidative coupling. This deleterious result was avoided by adding N,N,N′,N″,N″-pentamethyldiethylenetriamine whose steric hindrance prevents interaction with the propargyl groups and affords polylactide with higher chainend functionality (f = 86%) and low polydispersity (PDI = 1.22) (Scheme 8 (b)). 48 Propargyl alcohol initiated polylactides formation has also been used to prepare materials for click modification with other azido-polymers or small organic azides to prepare amphiphilic polymers promoting the study of their self assembly behavior in aqueous media (Scheme 8 (c)). 49,50 Thio-ene click chemistry (Michael addition of thiols) was also utilized to functionalize polylactides by capping a thiol at the polymer chain end. Molla and coworkers reported versatile chain end modification of polylactide containing pyridyl disulfide group which can be quantitatively reduced to a thiol by DTT. 51 This versatile methodology includes thiol−ene, thiol−maleimide, and thiol−acrylate “click” reactions, as well as chain-end modification by thiol−disulfide exchange reactions (Scheme 8 (d)). Thus the functionalities of the resulted polylactides can be greatly modified depending upon the nature of the attached molecules or polymers. Recently, one tetrazine-norbornene click modification was also applied in a mild and efficient manner to norbornene-initiated polylactide formation without 52 degradation of the polymer backbone. 15 Scheme 8. Examples of chain-end clickable polylactides capable of further modification through initiating group approach The ω-hydroxy group in polylactide chain ends is very important for end group modification to modulate the polymer’s properties, most commonly via activated esterification to protect the hydrolytic PLA backbone. Stupp and coworkers prepared self-assembled polylactides initiated by cholesterol (Chol) to investigate the selfassembly behavior and the interactions of the resulting polylactides with cells. They further modified the Chol-PLA by reacting the ω-hydroxy group with different functionalized carboxylic acids to afford new ordered PLA scaffolds. The bulk Chol-PLA material could self-assemble to give layered structures through proper orientation of cholesterol moiety with the spacing distance corresponding to the short PLA chain 53 length. End group modification of the secondary alcohol terminus of Chol-PLA was also 16 examined with the rodcoil L-lysine dendrons (3 generations, G1, G2 and G3) which could be coupled in high yield by double esterification (DIPC/DPTS) while maintaining the sensitive PLA part of the backbone without cleavage (Figure 2). The resulting amphiphilic molecules displayed very different self assembly behavior compared to their precursor (Chol-PLA). They exhibit very poor ordering in the dry state, which consisted of a lamellar self assembled structure with periodicities that depend on the degree of hydration (< 50%) and discrete nanosized aggregates at high degree of hydration (> 54 50%). Significant improvement in water wettability, enhanced adhesion of 3T3 mouse calvaria cells and greater population growth rates were observed after coating the selfassembled layer structure onto poly(L-lactic acid) fibers. 55 These rodcoil dendron biomaterials could be of potential interest as temporary molecular scaffolds for cell and tissue engineering (Figure 2). 17 Figure 2. Examples of functionalized polylactides via end group modification Furthermore, the secondary alcohol terminus of the Chol-PLA oligomers allows for the opportunity to introduce additional bioactive substituents such as cholesterol, indomethacin (an anti-inflammatory drug), pyrene and rhodamine B (fluorescent labels for bioimaging purposes) and various α-amino acids, which would dramatically increase the applications of PLA in the biomedical field. 56 18 However, initiating/end group modification can only modulate the properties of PLA through two sole sites leaving the polylactide backbone inaccessible. Thus modification methods that lead to greater diversity are desired. 1.3.2 Monomer Derivative Approach Since only a few -hydroxy acids, such as lactic acid, 2-hydroxyisocaproic acid and -phenyllactic acid, can be isolated via fermentation, access to most substituted hydroxy acids has to rely on synthetic preparation from other materials. Thus the substituted lactide monomers would be prepared via dimerization of those substituted hydroxy acids which requires highly diluted organic solutions under acidic conditions, and the dimerization step typically gives low isolated yields (Scheme 9). The resulting substituted polylactides produced from these substituted lactides have comb-like structures which can dramatically alter the physical properties of polylactides depending on the length and category of the attached functional group (FG 1), such as thermal and solubility properties, while retaining the degradability of the polylactide backbone. Scheme 9. General synthesis of functionalized lactide monomer derivative Substituted polylactides have been prepared by polymerizing cyclic monomers derived from the corresponding -hydroxy acids to give comb like structures. The preparation of this type of substituted polylactides typically begins by Grignard additions to oxalylic esters or by hydrolysis of cyanohydrins to afford -hydroxy acids both of 19 which require harsh conditions (Scheme 10). However, due to the demanding requirements of some functional groups that would be of interest to introduce into polylactides, some mild reaction conditions were also developed in order to prepare a variety of functionalized polylactides and these will be discussed below. Scheme 10. General preparation routes to substituted polylactides Baker and coworkers addressed the functionalization of PLA and studied the effect of both alkyl and aryl substitution of PLA on their thermal properties by preparing a series of alkyl and aryl groups tethered to the polylactide backbone. 57 There are three factors that affect the glass transition temperature (Tg) of polymers. The backbone rigidity, the rotation barrier along the backbone and the chain-chain interaction between the backbones all contribute to the Tg changes of polymers. By incorporating alkyl or aryl groups with different lengths or ring structures within the polylactides, the glass 20 transition temperature is dramatically affected by altering both the rotation barrier and chain-chain interaction as shown by the examples in Figure 3. It is believed that the rotation barrier would increase upon increasing the steric hindrance of the substituents attached to the backbone, and thus Tg would increase. This trend is clearly shown in Figure 3 (f and g), where the incorporation of rigid phenyl and cyclohexyl rings onto the polylactide backbone dramatically increases the Tg of PLA from 55 oC to ~100 oC. A similar trend was also observed for h and i where bigger substituents were attached to polylactide backbone. The steric effect of tethered side chains decreases as the bulky groups move farther from the polymer backbone, thus, the Tg of h and k dropped significantly compared to f and j, respectively. This is thought to be due to the induction of a higher rotation barrier when the branch in the substituent is closer to the backbone. When increasing the length of the side chain from a methyl group on the lactide to ethyl, isopropyl, isobutyl, n-hexyl and n-octyl groups, one would expect a higher Tg for the polylactides based on the discussion above since longer alkyl chains are larger than a methyl group. However, the Tg of those substituted polylactides (b, c, d, j and k) were 58,59,60,61 reported to have a lower Tg (-46 to 50 oC) compared to polylactide (a, 55 oC). This clearly points out the importance of chain-chain interactions (dipole-dipole interaction) of polylactides in determining their Tg. By increasing the lengths of the side chains, the dipole-dipole interactions between polymer chains decrease, and thus Tg decreases, because the steric effects are overridden by a screening of the ester groups by the linear alkyl chains in the polymer backbone. This screening effect on dipole21 dipole interaction by alkyl groups was confirmed by the Tg of substituted polylactide (e) which comprised of both n-hexyl and cyclohexyl groups. The Tg of polylactide is only 13 o C which is much closer to the Tg of c (ΔT=50 oC) than g (ΔT= 85 oC). Figure 3. Substituted polylactides and their Tg 22 Although a variety of the substituted polylactides mentioned above were prepared and their thermal properties were investigated, functionalization of these polylactides is very limited due to the inert nature of the alkyl or aryl groups. So substituted lactide monomers with a great variety of functional groups were synthesized to provide highly functionalized polylactides, such as hydroxy, carboxylic acid, amine, saccharides and polyethylene glycol functions. This was done via protection-deprotection methods before and after polymerization due to the limited functional group tolerance in the polymerization step of polylactide preparation. The introduction of hydroxy functions (-OH) tethered to the polylactide backbone has been reported by several groups through the monomer derivative approach. One of the most common starting materials is O-benzyl-L-serine which was converted to an αhydroxy acid via diazotization for use in preparing homo- or hetero-substituted lactide dimers and thus in turn to various polylactide homo- or co-polymers (Scheme 62,63,64 11). Notably, the monomer derivative approach usually requires extremely pure monomer to satisfy the polymerization conditions. The physical properties of hydroxy group tethered polylactides were reported to have significantly changed glass transition temperatures and degrees of crystallinity. They also have been shown to serve as delivery vehicles for sustained protein release with retained protein structural integrity while polymers degradation. 65 The tethered hydroxy groups also provide an opportunity for further grafting biotin/peptides to enhance the bio-applications, such as cell adhesion, and even in polymer grafting to produce more complex structures. 23 Scheme 11. Preparation of hydroxy group tethered polylactides Another synthetic approach to prepare poly(α-hydroxy acids) with hydroxy groups tethered to the backbone is the polymerization of O-benzyl-L-serine carboxyanhydride, which is worth discussing here due to the fact that it provides the same polylactide structural polymer as that obtained from lactide via an alternative synthetic route. This can also be applied to the preparation of other poly(α-hydroxy acids) with functionalized tethers. 66 Starting from O-benzyl-L-serine, the synthesized α-hydroxy acid was converted to O-carboxyanhydride by phosgene followed by ring opening polymerization to afford protected polymers (Scheme 12). After the removal of the benzyl group, degradable, water-soluble poly(α-hydroxy acids) bearing pendant hydroxyl groups were obtained with excellent cell compatibility, suggesting their potential for use as novel materials in constructing drug delivery systems and as hydrogel scaffolds in tissue engineering. 24 Scheme 12. Poly(α-hydroxy acids) (Polylactides) via O-carboxyanhydride routes Amine tethered polylactides were also reported by Gerhard et al. via ring opening polymerization of carbamate-protected monomer derived from lysine followed by quantitative deprotection under acidic condition (HBr/AcOH). The deprotection was clean and resulted in no degradation of the backbone, which was confirmed by NMR 62 analysis (Scheme 13). Several groups have investigated the preparation of polylactides tethered with carboxylic acid functions which were targeted to address solubility issues. As shown in Scheme 14., starting from either malic acid or aspartic acid, homo- and heteromonomers were both prepared and polymerized to the corresponding protected polylactides. 67 , 68 Catalytic hydrogenolysis using Pd/C provided the carboxylic acid tethered polylactides, which are water-soluble. More recently, Gerhardt and coworkers also reported similar substituted polylactides by polymerizing hetero-monomers with the carboxylic acid function protected by benzyl ester where glutamic acid was the starting material. 62 Deprotection was clean and quantitative without affecting the polymer backbone as determined by NMR analysis. However, GPC results were not reported due to the poor solubility of functionalized polylactides. 25 Scheme 13. Preparation of amine tethered polylactides 26 Scheme 14. Preparation of carboxylic acid tethered polylactides Polyethylene glycol (PEG) was also tethered to the polylactide backbone to tune the physical properties through the monomer derivative approach. Recent reports have shown that introducing polyethylene glycol groups into hydrophobic polymers render them hydrophilic. 69,70 In addition, the thermo-responsive property of these polymers makes them ideal for drug delivery systems, smart surfaces, and bioseparations. These materials exhibit a solution-gel transition (a transition from a clear solution to a cloudy mixture) when heated through the lower critical solution temperature (LCST), which corresponds to the entropically driven expulsion of solvating water molecules from the 27 polymer. Recent research on thermoresponsive materials, especially those that contain mPEG segments, suggests that tuning the hydrophobic/hydrophilic balance in materials can shift the LCST over a broad temperature range. Baker and coworkers have synthesized and polymerized lactide derivatives with 1-4 ethylene glycol segments using Sn(Oct)2. 71 , 72 The synthetic approach to the monomers uses poly(ethylene glycol) monomethyl ether (mPEG, 1-4 ethylene glycol segments), 1,6-dibromohexane and diethyl oxalate to generate the corresponding hydroxy acid, which was dimerized to the PEG-substituted lactide. The bulk polymerization at 130 oC afforded the alkyl/PEG substituted amphiphilic polylactides (Scheme 15). Polymers with short oligo(ethylene oxide) chains (1 or 2 ethylene glycol repeat units) are more hydrophilic than polylactide but nonetheless insoluble in water, whereas polymers having 3 or 4 ethylene glycol repeat units are water-soluble. Aqueous solutions of polymers with 3 and 4 ethylene oxide repeat units in each pendent chain are thermo-responsive, with LCSTs at 19 and 37 °C, respectively, which clearly indicates their promise in applications as thermo-triggered materials in the biomedical field. More recently, Castillo et al. reported another approach to poly(ethylene glycol) (PEG) side-chain functionalized lactide analogues in four steps from commercially 73 available L-lactide (Scheme 16). The key step in the synthesis is a “click” reaction (1,3-dipolar cycloaddition) between mPEG-azides and a highly strained spirolactide– heptene monomer (Norbornene-lactide) 74 and occurs with high conversion. The ring- opening polymerization was affected with triazabicyclodecene (TBD) and gave welldefined tri- and hepta(ethylene glycol)–poly(lactide)s with molecular weights above 10 28 kDa and PDI around 2. mPEG–poly(lactide) (PLA) with mPEG chain Mn 2000 was also prepared, but GPC analysis showed a bimodal profile indicating the presence of starting macromonomer. Reduced MC3T3-E1 cell adhesion using tri(ethylene glycol)– poly(lactide) indicated the addition of mPEG chains to PLA modulates biological responses compared to the unfunctionalized PLA due to the nonspecific protein adsorption-resistant nature of PEG. Scheme 15. Synthesis and polymerization of poly(ethylene glycol) monomethyl ethermodified polyglycolides. 29 Scheme 16. Castillo’s monomer derivative approach to poly(ethylene glycol) monomethyl ether-modified PLA The general goal for the synthesis of substituted polylactides via the monomer derivative approach is to prepare novel monomers, not only for tethering new functional groups to the polylactide backbones but also for providing further access to modified polylactides via simple, efficient and versatile functionalization of existing functions within the polylactides themselves via simple and efficient chemistry. Alkenyl, azido and alkynyl functional groups have been incorporated into polyplactides via the monomer derivative approach, which enables subsequent efficient click chemistry on these polymers for further functionalization. Leemhuis et al. have prepared allyl functionalized lactide monomers in relative high yield (Scheme 17). Bulk polymerization at 110 oC afforded polymers in high conversion, high yields and good control over molecular weight, although the polydispersity index was moderate (~2.0). 30 75,76 The major issue in the synthesis of functionalized lactide monomers is the low isolated yield (typically 2040%) in the dimerization step due to the unfavorable ring formation, which wastes considerable amount of the precious α-hydroxy acid. So overcoming this issue in the dimerization step or developing a new approach to avoid this step would be highly desirable. The only example in the literature that avoids the dimerization step in the synthesis of substituted lactides is that of Jing et al. who prepared a tricyclic lactide (norbornene-lactide) from commercially-available lactide in high yield (>90%, Scheme 16). 74 Lactide was brominated, and following dehydrobromination, a Diels-Alder reaction to construct the norbornene-lactide monomer which was polymerized to give polymers with narrow PDI (1.1) using TBD as catalyst at -20 oC. The polymer can be further modified through click reactions, 52 which will be discussed later. Scheme 17. Synthesis of allyl functionalized polylactides Baker and coworkers synthesized and polymerized an alkynyl modified lactide monomer to click with different small functional molecules after polymerization (Scheme 18). The α-hydroxy acid was synthesized via a Reformatsky reaction between propargyl bromide in the presence of activated zinc and ethyl glyoxylate followed by mild hydrolysis in water. Then this α-hydroxy acid was converted to monomer which was polymerized to give homopolymers with good molecular weight control and narrow PDI. 77 31 Scheme 18. Synthesis of alkynyl functionalized polylactides Recently, a novel and versatile synthetic procedure to a variety of side chain functionalized lactide monomers was reported by Rubinshtein and coworkers using a Passerini-type condensation reaction. 78 The multicomponent reaction of isocyanide, aldehyde, and water afforded the N-acylindole of a substituted lactic acid in good yield in a one-pot fashion. Then, functionalized lactide monomers were prepared through hydrolysis of the N-acylindole under mild conditions followed by lactide ring construction in moderate yield (Scheme 19). Scheme 19. Passerini-type condensation route to versatile functionalized lactide monomers 32 1.3.3 Post-polymerization Modification Approach As shown in Scheme 7 (c) and discussed above, the monomer derivative approach to functionalize polylactides involves multi-step synthetic procedures to prepare the new substituted lactide monomers which consume a great deal of energy and time. Moreover, the functionality that is initially installed into the monomers must be compatible with the polymerization conditions. In contrast to the monomer derivative approach, post-polymerization approach offers the distinct advantage that a series of polymers with different functionalities tethered to the polymer backbone can in principle be accessed from a single monomer. Having a single, simple procedure for placing a broad range of pendant functional groups onto polylactide substrates is highly desirable. The ideal post-polymerization modification approach should utilize a chemical process that not only provides a simple synthesis of functional monomers but also does not degrade the polymer backbone. It should also be compatible with a wide range of functional groups, solvents, and conditions to achieve efficient and quantitative tethering. 1.3.3.1 Direct Post-polymerization Modification Approach Due to the labile nature of the polylactide backbone towards degradation, there are limited literature reports regarding the direct post-polymerization modification of polylactides. The reported examples indicate that this approach is plausible for surface 79 modification of polylactide films deposited on substrates. Carlislf et al. used a pulsed plasma deposition strategy to attach a thin polyallyl amine layer onto polylactide films for tissue culture application. The polyethylene glycol-adhesion peptides (RGD or 33 YIGSR) were covalently grafted to the thin polyallyl amine layer to mimic the in vitro extracellular matrix and thus enhance the hepatocyte adhesion. single-step, nondestructive, and versatile technique for modification of a polylactide films was reported. 81 80 More recently, a the covalent surface PLA films were subjected to three different vinyl monomers in the vapor phase:  acrylamide, maleic anhydride, and Nvinylpyrrolidone. The solvent-free surface grafting was induced by photoinitiation by benzophenone under UV irradiation. The extent of surface grafting and the surface wettability increases upon elongating the UV treatment time. The grafted carboxylic acid groups remain functional and can subsequently be modified to provide altered desired properties. The in vitro degradation rate was enhanced by covalently grafting to the surface. 82 However, neither of the strategies discussed above didn’t present the structural stability information of the polylactide films after pulsed plasma deposition or UV irradiation due to the difficult detachment of the surface layer and their low content for analysis. Fortunately, Pan et al. have addressed the structural stability by measuring the Mn of polylactide materials before and after radical induced direct post-modification of bulk polylactide with maleic anhydride. 83 Radical reactions initiated by benzoyl peroxide, a radical initiator, provided for the covalent grafting of maleic anhydride to PLA (Scheme 20). The resulting bonded succinic anhydride could be ring-opened to give carboxylic acid and amine side chains via amidation with 1,6-hexanediamine. However, in the process, the modified PLA backbone significantly degraded, which was indicated by the loss of up to 20% of the original Mn. These results suggest that the direct post- 34 polymerization modification of PLA may not fulfill the general requirements of postpolymerization modifications, although it is a simple and convenient method for surface modification to tune the physical properties. Scheme 20. Functionalization of bulk polylactide via direct post-polymerization modification using free radical initiator 1.3.3.2 Traditional Synthetic Coupling Approach via Side Functional Groups Traditional synthetic methods have been applied to the post modification of polylactide since many functional groups, such as alcohol, amine and carboxylic acid, can be revealed after protection/deprotection strategies as discussed in monomer derivative approach. Those protected functional groups provide the opportunities for further access to modified polymers with desired unique properties. As shown in Scheme 21, hydroxylated polylactides with benzyl protected hydroxyl groups can be deprotected and coupled with succinic anhydride to afford carboxylated polylactides 35 which can be further coupled with biotin or RGD-peptide in the presence of EDC. This provides the opportunity for the fabrication of a variety of bioactive ligands onto PLA to improve biocompatibility. 64 Scheme 21. Functionalization of hydroxylated polylactide via EDC coupling Hydroxylated polylactides have also been prepared by the hydroboration-oxidation of poly(lactide-co-diallylglycolide)Error! Reference source not found.. The alkeneunctionalized PLA could be oxidized by 9-borabicyclo[3.3.1]nonane (9-BBN) and hydrogen peroxide to provide primary alcohol functions along the polymer chain without 36 interrupting the backbone. Subsequently, the hydroxylated polylactides could be coupled with different fatty acids using DCC coupling to afford polylactides with different 84 long aliphatic chains again without backbone degradation (Scheme 22). Functionalized polylactides with carboxylic acids as pendent groups were also reported by du Boullay and coworkers through polymerization of O-carboxyanhydride derived from γ-benzyl L-glutamic acid in two steps. 85 The polymerization proceeded under mild condition using DMAP as the catalyst to produce well controlled polylactides. The pendent carboxylic acid groups protected with benzyl esters were readily recovered after hydrogenolysis. This procedure could provide well controlled homo, block and random polylactide polymers which cannot be achieved via polymerization of similar lactide derivatives probably due to their lower activity. Scheme 22. Functionalization of hydroxylated polylactide via DCC coupling 37 1.3.3.3 Click Chemistry Approach via Side Functional Groups Click chemistry is a chemical process introduced by K. Barry Sharpless in 2001 86 and describes chemical reactions that generate products quickly, reliably and in quantitative yield by joining small building blocks under mild conditions. Click chemistry is not limited to alkyne-azide 1,3-dipolar cycloadditions and can be extended to other highly efficient reactions, such as nucleophilic substitution, 90,91 Michael additions, as well as Diels–Alder 92 87 , 88 radical additions, 52 and retro-Diels–Alder 89 reactions.. Norbornene pendant groups along the backbone of polylactides can be functionalized to give amine-functional PLAs and its copolymers by the addition of substituted 3,6-di-2-pyridyl-1,2,4,5-tetrazine via sequential and retro Diels-Alder reaction without degradation of the polymer backbone. 52 The tetrazine-norbornene click chemistry is fast, atom ecominical, catalyst free, air insensitive, and quantitative, and hence fulfils all of the requirements of the click concept. Due to the alkenyl side groups, it should be able to be subjected to alkene-azide, thiol-ene click chemistry as well for further post modification. However, there are no such reports in the literatures for modifying polylactides. They have been recently documented in the orthogonal postmodification of polycarbonate. 93 38 Scheme 23. Post-polymerization modification of polylactides via tetrazine-norbornene click chemistry An allyl-functionalized polylactide prepared by Chen and coworkers was subjected to UV-induced thiol-ene click reaction with (diethylamino)ethanethiol hydrochloride (DEAET) in CDCl3 using 2,2′-dimethoxy-2-phenylacetophenone (DMPA) as the photoinitiator at room temperature (Scheme 24). 76,94,95 To tune the polymer charge densities, the feed ratio of alkene, thiol and DMPA was altered and well-defined cationic polylactides (CPLA) with different amounts of tertiary amine groups were successfully synthesized. The CPLAs possess remarkable hydrolytic degradability with relatively low cytotoxicity. Prostate cancer cells can readily take up the CPLA- interleukin-8 siRNA nanoplexes, which results in significant - interleukin-8 gene silencing. 39 94 Scheme 24. Post-polymerization modification of polylactides via thiol-ene click chemistry Alkyne-azide click chemistry has been applied to functionalize polylactides by several groups with quite different synthetic approaches involving either alkyne or azide substituted polylactides. 77,78, 96 Recently, Baker and coworkers introduced an acetylene modified lactide derivative capable of post-polymerization modification by Cu(I) catalyzed click chemistry to functionalize the polylactides with alkyl and poly(ethylene glycol) monomethyl ether branches under mild condition (Scheme 25). Their method showed good control over the degree of functionalization and no backbone degradation under the new click conditions was observed. The resulting functionalized polylactides exhibit thermo-responsive properties with the LCST 25-65 oC 40 depending on the mPEG/alkyl ratio in the polylactides. 77 When more hydrophilic mPEG segment was incorporated into polylactides, higher LCST was observed. Scheme 25. Baker’s synthesis of functionalized polylactides via Cu(I) catalyzed click chemistry Both alkyne and azide functionalized polylactides were reported by Rubinshtein and coworkers using similar versatile synthetic routes (Scheme 26). The monomers were synthesized using a Passerini-type condensation reaction with broad functional group tolerance. The resulting polylactides were able to undergo click chemistry in the presence of Cu(I) catalyst while retaining the polymer structure. 78 This new methodology represents a significant step towards accessing biodegradable and biocompatible materials with improved functional and potentially tunable properties that may find utility in a variety of applications. 41 Scheme 26. Rubinshtein’s synthesis of functionalized polylactides via Cu(I) catalyzed click chemistry Polymerization of O-carboxyanhydride is an alternative procedure to produce polylactide polymers with exact same backbone as polylactides from lactide monomers. Zhang et al. synthesized an O-carboxyanhydride bearing an alkyne moiety to prepare clickable polylactides which could be further functionalized through either Cu(I) catalyzed click chemistry 96 or thiol-yne photo click chemistry. 97 As shown in Scheme 27, ring opening polymerization of O-carboxyanhydride resulted in amphiphilic clickable 42 polylactide which self assembled into micelles in water via nano precipitation method. The formed micelles was cross-linked in the cores using cross-linkers (1,4diazdobutane and bis(azidoethyl) disulfide) via click chemistry. Core cross-linking not only improved the structural stability of the micelles, but also allowed controlled and triggered release of cargo molecules due to the fact that the disulfide cross-linkers can be cleaved by the reducing reagent DTT (dithiothreitol). Thiol-yne photo click chemistry of polylactides with 2-aminoethanethiol hydrochloride (AET) under UV irradiation provided side-chain aminated polylactides which displayed excellent cell penetration and gene delivery properties with low toxicities. Scheme 27. Zhang’s synthesis of functionalized polylactides via both Cu(I) catalyzed and thiol-yne click chemistry 43 1.4 Smart Polylactide Materials Smart polymeric materials respond with a dramatic physical property change to a small change in their environment. Environmental stimuli include temperature, light, chemicals, and pH etc. “Smart” stimuli-sensitive materials from either synthetic or natural polymers can be used to solve biological problems such as drug delivery, biosensor design, tissue engineering and protein folding. The versatility and untapped potential of smart polymeric materials makes them one of the most exciting interfaces of chemistry and biology. Of all stimuli-responsive polymers, temperature-responsive (thermo-responsive), 77, 98 light-responsive 99 and redox-responsive 96, 100 polymers have been studied intensely because of the easiness to handle the corresponding small environmental change, such as temperature increase from room temperature to steady human body temperature at 37 oC and the common interconversion of thiol and disulfide in biological responses. Post-polymerization modification of polymers via click chemistry provides advantages over traditional methods to smart polymeric materials, especially the functionalization of biodegradable polylactides due to the low chemical yield associated with synthesizing substituted lactide monomers and the extreme purity of the monomer that is required for the sequential polymerization. Post-modification with click chemistry in the preparation of smart polylactide materials would allow for the simple preparation of a series of functionalized polylactides from single clickable polylactides, which would improve the synthesis efficiency for specific smart property screening. It can also be used to tune the ratio of incorporation of different functionalities into the polylactides to modulate the stimuli-response at the desired level for a specific application. Both 44 thermo-responsive 71,72,77,101 and redox-responsive 96,102 properties of functionalized polylactides have drawn much attention and have been reported recently due to the interest in the biodegradable and stimuli-responsive properties of polylactides. Thermo-responsive polymers exhibit a unique property which features a lower critical solution temperature (LCST) and/or an upper critical solution temperature (UCST). The LCST is a fascinating phenomenon found for various polymer solutions. At temperatures below the LCST, the polymer is completely miscible in water due to extensive hydrogen bonding interactions with the surrounding water molecules, whereas at temperatures above the LCST a phase separation occurs because hydrogen bonding with water is disrupted, and intra- and intermolecular hydrogen bonding/hydrophobic interactions of the polymer chains dominate, which results in a transition in solubility. 103 The ratio of hydrophilic and hydrophobic segments incorporated onto the polymer backbone can be used to tune LCST. 77,104,105 Higher LCST can be achieved by increasing the hydrophilic nature of the polymers due to the increased overall hydrogen bonding; whereas incorporating more hydrophobic groups lowers the LCST which could be rationalized by the disruption of hydrogen bonding and 106 higher hydrophobic interaction. Redox-responsive polymers, especially thiol-sensitive polymers, are of particular interest, because the thiol-disulfide exchange reaction is one of the most important reactions in many biological processes including protection of important cellular components from reactive oxygen species. Polylactides with disulfide bonds are redoxresponsive because the disulfide bonds can be reversibly converted to thiol groups in 45 the presence of reducing agents (such as Dithiothreitol (DTT)) via redox chemistry or by an excess of other thiol compounds via disulfide exchange. The result is that the original disulfide functionalized polylactides would undergo partial structural disruption and induce changes in certain physical properties. For example, Zhang et al. have prepared cross-linked polylactides nanoparticles with cross-linkers including disulfide crosslinkers. The cross-linked nanoparticles showed high morphologic stability, and they displayed controlled releas of entrapped cargo molecules upon treatment with DTT with increased nanoparticle size due to the thiol-disulfide exchange reaction. 96 This clearly indicates the redox-responsive property of disulfide functionalized polylactides. The most abundant reducing agent in cell, glutathione (GSH) has an intracellular concentration of 10 mM, 1000 times higher than that in the cellular exterior. 107 The development of redox-responsive and biodegradable polylactides as delivery vehicles to deliver and release therapeutics upon entry into cells due to the significant concentration difference of GSH is very promising in biomedical research. However, at this time there is limited research on delivery behavior using redox-responsive polylactides prepared from click post-modification of side chains of polylactides. 1.5 Conclusion Functionalized polylactides greatly enhance the application of PLA in biomedical and structural applications. Three approaches to functionalized substituted polylactides have been intensely studied to explore their novel physical properties resulting from the functionalization and their applications in biomedical field. The polymerization mechanisms from substituted lactides to functionalized polylactides were also discussed 46 to address the choice of polymerization when new monomers are prepared. These new functionalized polylactides fall in the range of smart polymeric materials which would expand their promise in a variety of application. In this thesis, we addressed the functionalization of polylactides via monomer derivative approach and post-polymerization modification approach. A series of novel and clickable α-hydroxy acid containing terminal alkyne groups and polyethylene glycol units were synthesized via a facile synthetic procedure without requiring column purification. We prepared the alkynyl substituted lactide monomers and the bulk polymerization of these monomers afforded the clickable polylactides. The postmodification of the resulting polylactides via click reactions provided thermo-responsive and redox responsive properties to polylactides. The spirolactide monomers were also prepared and polymerized into the corresponding polyspirolactides which displayed different thermal properties for different ring substituents attached to polylactide backbones. 47 CHAPTER 2 BUILDING UP A LIBRARY OF CLICKABLE α-HYDROXY ACIDS TO CLICKABLE SUBSTITUTED POLYLACTIDES 2.1 Introduction At the present time, there are two kinds of monomers, substituted lactides and O- carboxyanhydrides which can be polymerized to give polylactides via ring opening polymerization. Theoretically, both of them can be prepared from either ring construction using substituted α-hydroxy acids as starting materials or direct functionalization of lactides or O-carboxyanhydrides (Scheme 28). The direct functionalization of lactides or O-carboxyanhydrides would be more advantangeous since it does not require additional synthesis of specific substituted α-hydroxy acids followed by ring construction which typically results in poor yield. However, due to the labile nature of both monomers under acidic or basic condition, there is only one reported method detailed in several reports to directly functionalize commercially52,73,74,108 available lactides in high yields (55-90%). Commercially-available L-lactide was brominated, dehydrohalogenated and the resulting double bond subjected to a Diels-Alder reaction with cyclopentadiene to construct a norbornene-lactide monomer. Although the polymerization of this monomer could provide alkenyl-substituted polylactides with narrow PDI, the polymerization in days with triazabicyclodecene (TBD) as the only known active catalyst gave poor molecular weight control and thus is not plausible to produce reliable and reproducible polymer for further applications. Due to the limitation of direct functionalization methods, the monomer synthesis for ring opening polymerization to polylactides is always based on dimerization through the 48 reaction between α-hydroxy acid and bromoacyl chloride derivatives followed by cyclization to give asymmetrical lactide monomers (hetero-monomer), or direct dimerization of α-hydroxy acids to symmetrical lactide monomers (homo-monomer). The first method usually gives low to moderate yields while the latter one always occurs with poor to low yields. So the synthesis of substituted lactide monomers is always the limiting step in the preparation of functionalized polylactides. Based on the discussion above, versatile methodologies which can directly functionalize commerically available lactides or dramatically improve the yield of the dimerization step from α-hydroxy acids to its lactide monomers would be highly desirable. Scheme 28. General synthetic routes to substituted lactide monomers Given the fact there is a lack of new methodologies to prepare substituted lactide 49 monomers in high yield, the pursuit of simple but versatile synthetic procedures to prepare a variety of α-hydroxy acids with different functional groups in high yield was considered to be the most practical way to produce a sufficient supply of monomers for the ultimate preparation of functionalized biodegradable polylactides. We are interested in building up a library of substituted α-hydroxy acids functionalized with different clickable moieties which can be converted to their lactide monomers and then to functionalized polylactides for exploring a variety of click post-modification process and the evaluation of the resulting polylactides for their potential in biological applications. 2.2 Results and Discussion 2.2.1 Synthesis of Propargyl Glycolic Acid Our group was the first to develop a procedure to synthesize propargyl glycolic acid 1 to prepare alkyne functionalized polylactides for post-modification via click chemistry. 77 As shown in Scheme 29, a Reformatsky-type reaction of propargyl bromide with ethyl glyoxylate in the presence of activated zinc gives ethyl 2-hydroxy-3-butynoate, which was hydrolyzed in water to provide α-hydroxy acid 1 in 42% overall yield. Besides the low overall yield of 1, propargyl bromide is very costly and is shock sensitive and subject to explosive thermal decomposition, which makes the preparation of 1 via this method impractical. Several other synthetic routes to 1 were examined in order to find a more convenient procedure that does not use propargyl bromide. However, all of the routes indicated in Scheme 30 were determined not to be feasible. 50 109 Scheme 29. Synthetic route for propargyl glycolide and its polymer Scheme 30. Failed synthetic procedures conducted to synthesize 1 51 2.2.1.1 Synthesis of Propargyl Glycolic Acid via an Organoalane The intrinsic nucleophilicity of organoalanes is well below that of organolithium and magnesium compounds, which enables them to be better nucleophiles in reactions with base sensitive substrates. 110 Also, because of the Lewis acidic nature of aluminium, organoalane can coordinate with the epoxide oxygen leading to activation of the epoxide and transfer of the alkyl group and the opening of the epoxide ring. 111 However, ethynylaluminum compounds are unstable because of the acidic nature of the ethynyl proton, thus the use of a protecting group, such as TMS, is necessary. Scheme 31. Epoxide ring opening by organoalane in the synthesis of 1 Trimethylsilyl acetylene (2), prepared from acetylene and trimethylsilyl chloride according to literature procedures, 112 , 113 was deprotonated by butyl lithium and converted to trimethylsilylethynyldiethylaluminum by reaction with diethylaluminum chloride. Then methyl glycidate (3) was added. The epoxide ring opening was completed after stirring overnight (>90% yield) with no evidence for byproduct formation. 52 The product (4) was directly deprotected in MeOH in the presence of K2CO3 to give the propargylglycolic acid (1, 55% yield) where the methyl ester was hydrolyzed in the acidic workup (Scheme 31). However, at least 2 equivalent of the alanae is required to give high conversion; one equivalent of alane presumably coordinates to the carbonyl oxygen 114 activating and another equivalent of alane facilitates the ring-opening process after the epoxide. 115 , 116 , 117 An alternative reaction to prepare diethylethynylaluminium was also conducted where equal molar amounts of trimethylsilyl acetylene and triethylaluminium were heated to reflux in toluene. 118 However, a complex mixture was obtained with no obvious formation of product 1 after the addition of epoxide (3) and workup. However, this procedure is highly costly at about 1400$/mol compared to 600$/mol for the route from propargyl bromide (Scheme 29), since both trimethylsilyl acetylene and diethylaluminium chloride are expensive, although no propargyl bromide was involved. 2.2.1.2 Synthesis of Propargyl Glycolic Acid via the Enolate of a Malonate and Propargyl Tosylate Two procedures discussed above suffered either from the use of expensive and explosive propargyl bromide or the low pKa of an acetylene proton which can be deprotonated under strong basic conditions. Malonate have a low pKa around 13 and can be easily deprotonated by mild bases such as sodium/potassium alkoxides without deprotonating a terminal alkyne. Thus it should be possible to alkylate a malonate containing a terminal alkyne function. Moreover, the resulting alkylated malonate can be converted to a carboxylic acid by simple decarboxylation upon heating. Propargyl 53 tosylate, 119 a safe analogue of propargyl bromide, can be prepared from propargyl alcohol and tosyl chloride. Conveniently, both malonates and propargyl alcohol starting materials are commercially available in bulk quantities and can be used to synthesize propargyl glycolic acid (1). As shown in Scheme 32, the enolate of diethyl acetamidomalonate (5) will react with 120 propargyl tosylate in dioxane under reflux to give 6 in 95% yield. The final glycolic acid (1) was obtained in a sequential one-pot reaction with moderate yield (41%) from the combination of 4 steps, hydrolysis of ester (6), decarboxylation, hydrolysis of amide, and diazotization of the resulted amino acid (propargylglycine 7) using NaNO2 under 121,122 acidic conditions. Scheme 32. Synthesis of propargyl glycolic acid from diethyl acetamidomalonate Propargyl glycolic acid (1) prepared by this method rapidly decomposes under high vacuum after extraction from aqueous solution with ether (Table 1). This was surprising 54 since 1 prepared by other method does not decompose upon isolation. 77 The possible reason for this decomposition is that there might be certain oxidants present in the ether extract (verified by a rapid color change of KI-starch test paper, which destroys the product during concentration. However, neither the addition of urea (reacting with HNO2) nor the passing of the crude product through a short silica gel column can stop the decomposition. So several reductants were examined which presumably would react with any possible oxidants. No product 1 was detected in the filtrate from an ether solution after addition of Na2SO3 to an ether solution of 1 and filtration to remove the solid. However, 1 can be recovered in low yield (<15%) as a light yellow solid after extracting the acidified aqueous solution of Na2SO3 with ether. Apparently, acid 1 can be converted to its sodium salt which precipitates from ether through an acid-base reaction with Na2SO3. This shows that it is not convenient to use Na2SO3 as a reductant to remove the possible oxidants. When the reductant Na2S2O3 was added to an ether solution of acid 1, both sulfur and acid 1 were obtained as yellow solid mixture after the removal of ether under vacuum. Dissolution of this mixture in water, filtration and removal of the water gave pure acid product 1 in 30% yield. However, pure product 1 could be obtained as light yellow solid in 41% yield by adding Na2S2O5 solid to an ether solution of acid 1 followed by simple filtration, concentration and sublimation under vacuum. This indicates that Na2S2O5 solid may remove any possible oxidants and provide product 1 as solid with a relatively uncomplicated procedure. Based on these results, we conclude that the reducing salts added to an ether solution of 1 probably act 55 as bases to remove the possible acidic oxidants (like HNO2, HNO3 and H2SO4) and enable the product 1 to form as solid without decomposition under vacuum. Despite the moderate yield, the overall cost to synthesize 1 mol of acid 1 is 220$, much lower than other procedures. Table 1. Propargyl glycolic acid (1) after treatment with different reductants/bases a Reductant/Base None Na2SO3 Na2S2O3 Na2S2O5 Compounds b in ether 1 No 1 Sulfur + 1 1 Appearance c of dry 1 Decomposition black Light yellow d solid Yellow solid Light yellow solid a Solid reductants/bases were added into ether extract containing 1 and stirred for 12 hours. b Compounds in ether extract before and after treatment with reductants/bases followed by filtration. c Product 1 obtained after removal of ether after treatment with reductants/bases followed by filtration. d The product 1 was recovered after extracting the acidified aqueous solution of solid Na2SO3. However, several factors, like the use of dioxane as solvent, the requirement for toxic chemicals and especially the low atom economy, render this procedure unfeasible for a practical synthesis of the target acid 1. 2.2.2 Synthesis of Propargyloxy Lactic Acid via Hydrolysis of a Cyanohydrin Propargyl alcohol contains two functional groups, a terminal alkyne and a hydroxy group and it is possible to employ the hydroxy group as nucleophile under weak basic condition without affecting the terminal alkyne function. Bromoacetaldehyde diethyl acetal has a good leaving group (bromo group) and the protected aldehyde which can be later converted to a cyanohydrin. The 56 use of propargyl alcohol and bromoacetaldehyde diethyl acetal as starting materials to synthesize a cyanohydrin containing terminal alkyne would be able to provide a general and economic synthetic route to the corresponding α-hydroxy acid since both starting materials are widely used in industry and can be supplied in large quantities at low prices. However, the hydrolysis 123 of cyanohydrins containing a terminal alkyne group is rare in the literature. The synthesis of propargyloxy lactic acid (12) is outlined in Scheme 33. An excess of propargyl alcohol was deprotonated by NaH in THF and the resulting alkoxide will nucleophilically attack the bromo group of the bromoacetaldehyde diethyl acetal (8) to afford propargyloxyacetaldehyde diethyl acetal (9) as pale red liquid in 92% yield without affecting the terminal alkyne (Further purification can be done through vacuum distillation, but this is not necessary). The optimized hydrolysis condition for the acetal 9 to aldehyde 10 was determined to be a biphasic mixture of trifluoroacetic acid (TFA), water and dichloromethane (1:1:4), which require vigorous stirring to achieve quantitative conversion (Table 2, entries 8 and 9). The resulting aldehyde 10 was used without further purification. H2SO4 128 124 , 125 , 126 p-Toluenesulfonic acid (TsOH), HCl, 127 and acetic acid were not effective in the deprotection of diethyl acetal 9 (Table 2, entries 1-7). 57 Scheme 33. Synthesis of propargyloxy lactic acid (12) Table 2. Deprotection of 9 with various acids and solvents at room temperature Entry Acid Solvent Time / h Conversion (%) 1 TsOH (10%) H2O 20 0 2 TsOH (10%) THF-H2O 5 < 5% 3 AcOH (25%) CH2Cl2-H2O (4:1) 36 < 10% 4 HCl (2M) Ether-H2O 20 30% 5 HCl (2M) THF-H2O 20 35% 6 HCl (2M) AcOH-H2O 20 < 20% 7 H2SO4 (2M) AcOH-H2O 36 40% 8 TFA (2%) CH2Cl2-H2O (4:1) 36 < 10% 9 TFA (16%) CH2Cl2-H2O (4:1) 12 a 100% a 1 Conversion was measured by H NMR. The cyanation of aldehyde 10 was performed under acidic condition and the resulting aqueous solution of 11 was extracted continuously with CH2Cl2 to give cyanohydrin 11 in high yield (91%) and was used without further purification. The 58 hydrolysis of cyanohydrin 11 was first tried in aqueous HCl (2, 6 and 11 M) and resulted in decomposed starting material (not shown in Table 3). Considering the fact that the alkyne function can survive in refluxing 2 M H2SO4 (Scheme 32), hydrolysis of 11 at various concentrations of H2SO4 and at various temperatures was conducted in a mixture methanol and water to find the best condition (Table 3). The results of hydrolysis at different temperatures and concentrations of H2SO4 showed that hydrolysis can proceed much faster at higher reaction temperature and higher concentration of H2SO4. However, high temperatures also gave slightly lower yields (entries 7-10). Changing the solvent ratio between MeOH and H2O did not show an effect on the yields (entries 9 and10). So the optimized conditions for hydrolysis of 11 were with 5~6 M H2SO4 at 70 oC for 24 h at (entries 3 and 4). The product consists of a mixture of the α-hydroxy acid and its methyl ester and was continuously extracted with ether and then hydrolyzed in 2 M NaOH for 12 h at room temperature followed by extraction with ether to remove mineral oil (from first step) and other organic impurities. After acidifiication back to pH<1, the final product was extracted with ether again and concentrated to give 12 as a light yellow solid with an overall yield of 66% from 8 (4 steps) which can be further decolorized with charcoal in water to give a white solid. These studies indicate that this procedure can be used to synthesize sufficient quantities of alkyne functionalized α-hydroxy acid 12 with low cost and convenient reaction conditions. 59 Table 3. Hydrolysis of 11 at various concentrations of H2SO4, temperatures and solvents Entry Temp (oC) [H2SO4] Time (M) MeOH/H2O a (h) Conv. Yield b 12 1 1:1 3.6 160 98% 78% 2 1:1 4 31 94% 79% 1:1 5 25 91% 76% 4 1:1 6 26 100% 81% 5 1:1 6 48 - 76% 6 1:1 4 24 91 80% 1:1 5 12 99 78% 3:1 2 65 84% 63% 1:1 5 16 98% 67% 3:1 5 18 100% 65% 3 7 70 80 8 9 98 10 a 1 Conversions were measured by H NMR; b c 1 Yield of 12 were measured by H NMR by adding an internal standard, triphenylmethane after basic and acidic workup; Isolated yield after basic and acidic workup followed by extraction with ether. c 2.2.3 Synthesis of Poly(ethylene glycol) Monopropargyl Ether Lactic Acid via Hydrolysis of a Cyanohydrin PEG is a biologically inert polymer which has been extensively used to alter the physical properties of polymer complexes, especially as electrolyte and therapeutic agents. Conjugation with PEG can confer many beneficial properties, such as increased solubility, reduced interaction with serum and proteins and thus significantly increasing 129 the half-life in vivo. Polylactides functionalized with hydrophilic PEG and/or hydrophobic alkyl chains as pendant groups have been reported to have thermo71,72,77 responsive properties and to influence the interactions of cells with polymer 60 73 materials . They are prepared either by the monomer derivative approach or postpolymerization modification via click chemistry. Incorporating both terminal alkyne and polyethylene glycol chain pendant groups onto polylactides can enable functionalization of the polymer by click chemistry and enable PLA novel physical properties, such as thermo-responsive property. So a series of α-hydroxy acids (poly(ethylene glycol) monopropargyl ether lactic acids) (Figure 4) with different PEG unit lengths were synthesized to expand the library of clickable and thermo-responsive polylactides and their preparation will be discussed below. Figure 4. Structure of poly(ethylene glycol) monopropargyl ether lactic acid 2.2.3.1 Synthesis of Poly(ethylene glycol) Monopropargyl Ether Lactic Acid (23, n=1,2,3,4) The synthesis of poly(ethylene glycol) monopropargyl ether lactic acid (n=1,2,3,4) starts from the corresponding oilgo(ethylene glycol) with 1 - 4 glycol units, all of which are commercially available. The synthetic strategy is based on the coupling of monoprotected oligo(ethylene glycol), propargyl 61 alcohol and protected bromoacetaldehyde (8) to generate a cyanohydrin which can be hydrolyzed to αhydroxy acids of the type 23. As shown in Scheme 34, the oligo(ethylene glycol) (13, n=1,2,3,4) were desymmetrized and monoprotected with trityl chloride to afford 14 as white solids (n=1,2) or viscous liquid (n=3,4) in high yield (>90%) which were used without further purification. 130 , 131 Ditrityl protected oligo(ethylene glycol)s were obtained as minor side products with 14, but were not removed since it was anticipated that it could be easily removed in derivatized forms while purifying the final α-hydroxy acids and this will be discussed later. Coupling of the propargyl group with 14 is not plausible due to the fact that deprotection of trityl group will employ a palladium catalyzed hydrogenolysis and due to possible side reactions of the terminal alkyne when NaH is used in subsequent steps. So the monotriytl protected PEG (14) was coupled with protected bromoacetaldehyde (8) to quantitatively give 15 which does not require further purification. After workup, excess 8 was simply removed under high vacuum at 40 oC overnight. However, due to the low solubility of the alkoxide of 14 (n=1), the coupling between ethylene glycol monotrityl ether (14, n=1) and 8 only proceeds to 66% conversion after refluxing for 48 h and this cause problems in the purification. This problem was solved by using the commerically available ethylene glycol monobenzyl ether (16) which was converted to pure 17 quantitatively following the same protocol used in the synthesis of 15. The selective deprotection of trityl group of 15 and 17 in both THF and CH2Cl2 were not effective or required very high catalyst loading and long reaction times to give high conversion. However, the deprotection in an alcohol, such as ethanol, gave 100% conversion without affecting the acetal. So both 15 and 17 were selectively deprotected by hydrogenolysis in ethanol catalyzed by 62 palladium on carbon (≥ 2%) at 1400 psi H2. 131 The resulting alcohol 18 was then carefully dried under vacuum to remove ethanol. However, the ethanol cannot be completely removed (5 - 10% left). If it is completely removed, the protected aldehyde will undergo alcohol exchange to give a mixture of compounds. This mixture can be converted to 18 by mixing with ethanol under high pressure in a steel autoclave reactor. However, the small amount of ethanol will not affect subsequent reactions since the byproduct (propargyl ethyl ether) from ethanol can be easily removed at the end. The 131 tosylation of 18 followed by coupling with propargyl alcohol provided alkyne functionalized acetal 20 in quantitative yield. The deprotection of 20 by trifluoroacetic acid in CH2Cl2 and water gave aldehyde 21 which was subjected to cyanation directly without further purification to provide 22. After continuous extraction of 22 from an aqueous solution and removal of CH2Cl2, cyanohydrin 22 was hydrolyzed in 6M H2SO4 in a mixture of MeOH and water then followed by extraction with CH2Cl2 and with aqueous NaOH treatment to hydrolyze methyl ester. Then the mixture was diluted and extracted with CH2Cl2 to remove all nonionic organic impurities, such as poly(ethylene glycol) dipropargyl ether, propargyl ethyl ether and mineral oil. After it was acidified with 4 M HCl, the pure target α-hydroxy acids (23, poly(ethylene glycol) monopropargyl ether lactic acids) were obtained via extraction as light yellow or pale red viscous liquid in 4060% yields. 63 Scheme 34. Synthesis of poly(ethylene glycol) monopropargyl ether lactic acid (23, n=1,2,3,4) Notably, the overall synthesis discussed above doesn’t involve purification by column chromatography which is one of the advantages of this procedure and thus the synthesis can be easily scaled up. This also indicates that the step involving hydrolysis of cyanohydrin is compatible with the overall synthesis of terminal alkyne functionalized 64 lactic acids and should be general and useful. 2.2.4 Synthesis of Octa(ethylene glycol) Monopropargyl Ether Lactic Acid (n=8) Figure 5. Structure of octa(ethylene glycol) monopropargyl ether lactic acid Oligo(ethylene glycol)s with up to 4 glycol units (n=1,2,3,4) are available commerically in large scales and thus during the desymmetrization and monofunctionalization one can afford to waste these inexpensive diols. However, when extending the number of repeating units of the oligo(ethylene glycol) to greater than 5, it is not acceptable to start with oligo(ethylene glycol) in the synthesis of the corresponding α-hydroxy acid due to the extreme high cost. So it is necessary to synthesize these poly(ethylene glycol)s with an efficient route not involving complicated purifications, such as column chromatographic separations. Herein, we describe a simple procedure to produce longer well-defined bifunctional octa(ethylene glycol) (n=8) by introducing orthogonal protecting/functional groups to two shorter oligo(ethylene glycol)s and then coupling of the two for the purpose of synthesizing monopropargyl ether lactic acid (Figure 5). This method could potentially be used to synthesize various 65 longer poly(ethylene glycol)s with exact repeating units (n>8). 2.2.4.1 Synthesis of Tetraethylene Glycol Monobenzyl Ether (29) Scheme 35. Synthesis of tetraethylene glycol monobenzyl ether (29) The synthesis of tosylate 26 started from the commercially available diethylene glycol monochlorohydrin (24) which was quantitatively protected with dihydropyran 132 (DHP) to afford 25 which was isolated by drying under high vacuum at 50 oC. Then the chloro group of 25 was substituted with sodium acetate in DMF followed by 133 hydrolysis and tosylation to provide the pure protected tosylate 26 (60%, 4 steps). Thus diethylene glycol monobenzyl ether (27) was coupled with 26 under basic conditions to provide 28 that contained negligible impurity (yield: 95%). 134 This indicates one efficient coupling step to prepare elongated bifunctional PEG chains with high conversion and high yield, which does not require column chromatographic 66 separation due to the negligible amount of by-product from the reaction. After heated in a 160 oC oil bath under high vacuum, the THP protecting group of 28 was efficiently removed to give the target compound tetraethylene glycol monobenzyl ether (29) in high yield of 90%. 2.2.4.2 Synthesis of Monotetrahydropyran-protected Tetraethylene Glycol Monotosylate (35) Tetraethylene glycol (30) was monoprotected with dihydropyran (DHP) using a large excess of 30 in THF to give a mixture of 31 and 32 (8:1) which was quantitatively tosylated to tosylates 33 and 32 (Scheme 36). 134 After complete cleavage of the THP protecting groups of 33 and 32 under acidic conditions in MeOH, the crude product was concentrated and redissolved in CH2Cl2 and washed with aqueous NaCl solution to give the pure tetraethylene glycol monotosylate (34) as colorless liquid in 87% yield. The desired product 33 was achieved by protecting 34 using DHP under acidic condition (74%). Although di-THP protected tetraethylene glycol (32) was an impurity in the preparation of 31, it can be simply removed by washing with water after deprotection due to the highly hydrophilic nature of tetraethylene glycol. The overall procedure allows for the synthesis of pure tetraethylene glycol monotosylate (34) in a simple and efficient manner. 67 Scheme 36. Synthesis of monotetrahydropyran-protected tetraethylene glycol monotosylate (33) 2.2.4.3 Synthesis of Octa(ethylene Glycol) Monobenzyl Ether (36) and the Lactic Acid (40, n=8) As shown in Scheme 37, tetraethylene glycol monobenzyl ether (29) and 33 were coupled together via a substitution reaction followed by deprotection of the THP group by heating at 55 oC under high vacuum to give pure octa(ethylene glycol) monobenzyl ether (36). The feed ratio of 29 to 33 (1:1.1 in our case) was found to be crucial to get 36 in a form which can be easily purified. Then tosylate 37 was synthesized from the coupling of 8 and 36, and subsequent deprotection of the benzyl group and tosylation gave 37 at 91% yield (3 steps). Upon coupling with propargyl alcohol, 37 was quantitatively converted to 38 (90%) which was deprotected to give an aldehyde that was converted to cyanohydrin 39. Then final pure product (40) was obtained in 65% 68 yield after hydrolysis of the cyanohydrin, hydrolysis of the methyl ester by NaOH, extraction from acidic solution and washing with 1 M HCl solution to remove all impurities. The washing of 40 using 1 M HCl was intended to wash out some of the organic byproducts that are more hydrophilic than 40 in water. Scheme 37. Synthetic route to octa(ethylene glycol) monopropargyl ether lactic acid (40) 69 2.2.5 Synthesis of Poly(ethylene glycol) Monopropargyl Ether Lactic Acid (50, n≈23) Scheme 38. Synthesis of poly(ethylene glycol) monopropargyl ether lactic acid (50, n≈23) Using a procedure similar to that for the preparation of tetraethylene glycol monotosylate (34, Scheme 36), poly(ethylene glycol) (41, n≈23) was mono-protected by 70 DHP to give 42 as major product with bis-THP ether 43 as a minor product which can be carried through the synthesis and removed as its derivative (47) at the end (Scheme 38). The THP protection of 41 gives a ~10:1 mixture of 42 and 43 in 85% total yield after washing their dichloromethane solution with water for 40 times. Coupling 8 with 42 under basic condition followed by deprotection of the THP group and then tosylation resulted in the formation of desired product 44 containing 45 as minor product (92% total yield). The tosylate 44 was coupled with propargyl alcohol to install the alkyne function and give 46 in 86% yield. Acetal hydrolysis and cyanation gave cyanohydrin 48 (95% yield) which was hydrolyzed in 6 M H2SO4 to afford a mixture of 47 and 49. After hydrolyzing the methyl ester, hexane/CH2Cl2 (1:1) was used to extract the aqueous solution twice to remove mineral oil, 47, and some other nonionic impurities. The aqueous solution of 49 was acidified and extracte with CH2Cl2. The organic solutions was washed with 1M HCl solution 3 times to provide pure poly(ethylene glycol) monopropargyl ether lactic acid (50, n≈23, 74% yield) as white wax-like solid after drying under vacuum. 2.2.6 Synthesis of Allyloxy Lactic Acid (54) via Hydrolysis of Cyanohydrin The versatility of the synthetic route to substituted lactic acids described above which involves a cyanohydrin hydrolysis was tested by the preparation of allyloxy lactic acid (54), since alkene is feared to undergo unexpected side reactions, such as H2O addition under acidic conditions and polymerization by light or heat. As outlined in Scheme 39, the synthesis of allyloxy lactic acid (54) was prepared following a 71 procedure similar to prepare propargyloxy lactic acid (12, Scheme 33) using allyl alcohol and acetal (8) as starting materials. The coupling between 8 and allyl alcohol gave 51 in 76% yield after vacuum distillation. After hydrolysis to aldehyde 52 under acidic condition, cyanohydrin 53 was prepared as a pale red oil (>100%, with acetic acid) and isolated by continuous extraction. Following hydrolysis under acidic and basic conditions allyloxy lactic acid (54) was extracted from acidic solution in 55% yield (4 steps). Scheme 39. Synthesis of allyloxy lactic acid (54) 2.2.7 Synthesis of Azido Lactic Acid (56) Azide functionalized polylactides provide an alternative approach to render click function through Cu(I) catalyzed (CuAAC) or strain-promoted alkyne-azide cycloadditions (SPAAC). The azido lactic acid (56) was prepared as viscous liquid (59% yield) from methyl glycidate (3) and sodium azide in the presence of NH4NO3 (Scheme 40)under reflux followed by hydrolysis of the methyl ester with 1M aqueous NaOH 135 solution. 72 Scheme 40. Synthesis of azido lactic acid (56) 2.3 Conclusion With the goal of development of a convenient synthetic process for preparation of propargyl glycolic acid (1), two synthetic processes were explored. The disadvantages of both processes promoted us to search for new substituted lactic acids and appropriate methods for their synthesis. Propargyloxy lactic acid was proposed as a clickable polylactide precursor and was synthesized from inexpensive starting materials in high yield with a method that involves a cyanohydrin hydrolysis. A series of propargyloxy lactic acids derivatives incorporating poly(ethylene glycol) units of different lengths as spacers between the propargyl group and the lactic acid were prepared. The synthesis featured by poly(ethylene glycol) desymmetrization and coupling methods without using column chromatographic purification. The versatility of the developed method was also tested for its tolerance of an alkene group in the synthesis of allyloxy lactic acid. Finally, azido lactic acid was also prepared to act as an alternative approach to incorporate click functions to polylactides. 2.4 Experimental Section Materials. Butyl lithium (2.5 M in hexane), diethylaluminum chloride, propargyl alcohol, allyl alcohol, bromoacetaldehyde diethyl acetal, oilgo(ethylene glycol) (n=1,2,3,4), dihydropyran, dioxane, sodium hydride, trityl chloride and palladium on 73 activated charcoal (10%) were purchased from Sigma-Aldrich and used as received. Potassium tert-butoxide, diethyl acetamidomalonate, ethylene glycol monobenzyl ether, diethylene glycol monobenzyl ether, and 2-(2-chloroethoxy) ethanol were purchased from Alfa-Aesar and used directly. THF, dioxane and toluene were dried over sodium and distilled before use. All other reagents and solvents were ACS grade and used as received unless specified. Characterization. Melting points were measured on Electrothermo® melting point 1 apparatus. H NMR and 13 C NMR were recorded on a Varian 300 MHz or VXR-500 MHz or 600 MHz instrument in CDCl3 unless otherwise noted. CDCl3 was used as the 1 internal standard for both H NMR ( = 7.24) and 13 C NMR ( = 77.0). Synthesis of trimethylsilylacetylene (2). To a 500-mL 2-necked flask (flame-dried) was added 5.25 g (0.218 mol) of magnesium which was dried under vacuum overnight. After addition of 30 mL of dry THF and a piece of I2, 15 mL (0.20 mol) of dry bromoethane in 100 mL of dry THF was added to the magnesium dropwise. After refluxing for 40 min and then cooling down to room temperature, the fresh Grignard reagent was titrated following Knochel’s method 136 and the concentration was determined to be 0.76 M. To a flame-dried 1000-mL 3-necked flask equipped with an addition funnel, reflux condenser, and a gas-inlet diffusion tube connected to an acetylene cylinder through two traps immersed into dry ice-acetone bath, 140 mL of dry THF was added and saturated with acetylene for 40 min in a 15 oC water bath. Then the freshly-made Grignard reagent was added dropwise. After addition of all the ethyl Grignard to the 74 flask, the acetylene was continuously introduced for another 3 hours. The concentration of ethynylmagnesium bromide was 0.38 M after titration and was used directly. Then 14 mL (0.11 mol) of trimethylsilyl chloride in dry THF was added rapidly to ethynylmagnesium bromide (260 mL, 0.100 mol) and the water bath was kept at 15 oC for another 30 min. The reaction mixture was refluxed for 1 h. Then the liquid mixture was distilled and the distillate was washed with water. The liquid was distilled to give 2 1 as liquid (11.7 g, 60%). b.p. 50-57 oC. H NMR (500 MHz, CDCl3): δ 2.34 (s, 1H), 0.17 (s, 9H). Synthesis of methyl glycidate (3). To a 500-mL flask containing 146 mL (0.120 mol) of bleach solution, 8.6 g (0.10 mol) of methyl acrylate was added in one portion at 0 oC, and the resulted biphasic mixture was stirred vigorously for 35 min. Then the solution was slowly heated from room temperature to 35 oC for 1.5 h. After cooling with ice bath, the product was extracted with CH2Cl2 and the combined organic layers were dried over sodium sulfate and concentrated to give slightly yellow oil. Vacuum distillation 1 gave the pure colorless liquid 3, (4.0 g, 40%). b.p. 71-78 oC/60 Torr). H NMR (500 MHz, CDCl3): δ 3.74 (s, 3H), 3.39 (dd, 1H, J = 2 and 4 Hz), 2.90 (dd, 2H, J = 2 and 6 Hz). Synthesis of propargyl glycolic acid methyl ester (4) via organoalanate. To a flame-dried 500 mL round bottom flask was added 20 mL of 2.5 M (50 mmol) of n-BuLi solution, then 7.6 mL (50 mmol) of trimethylsilyl acetylene (2) in 15 mL of toluene was added dropwise in 30 min at -78 oC under N2. After stirring the resulted mixture for another 10 min under -78 oC and 15 min in ice bath, diethylaluminium chloride in 75 hexane (50 mL, 50 mmol) was added dropwise and stirred for 30 min at rt. Then the flask was cooled in ice bath, 5.2 mL (50 mmol) of methyl glycidate (3) in 5 mL of toluene was added dropwise and stirred for another 24 h. After quenching with saturated NH4Cl solution, the product was extracted with ether (4x75 mL), dried over Na2SO4 and concentrated to give 4.5 g product (4), 90%. This product was used without further 1 purification. H NMR (500 MHz, CDCl3) δ 4.3 (q, 1H, J = 5 Hz), 3.8 (s, 3H), 2.97 (d, 1H, J = 2 Hz), 2.7 (t, 2H, J = 5 Hz), 2.0 (s, 1H), 0.15(s, 9H); 13 C NMR (125 MHz, CDCl3) δ -1 174.4, 100.5, 88.3, 67.0, 52.7, 26.4, 0.16; IR 3474, 2959, 2179, 1743, 1441, 1250 cm . Synthesis of propargyl glycolic acid (1) via organoalanate. A 250-mL flask was charged with 3 g of ester (4), 25 g of anhydrous K2CO3 and 150 mL of MeOH and stirred for 24 h at rt. Then 100 mL of water was added and the acid (1) was extracted with ether (3x40 mL), and the solvent was removed under vacuum to give light yellow 1 oily product (1) (0.8 g, 55%). H NMR (500 MHz, CDCl3) δ 4.35 (t, 1H, J = 4.5 Hz), 2.69-2.82 (m, 2H), 2.08 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 177.1, 77.5, 72.1, 68.4, 24.5. Synthesis of propargyl tosylate. A 1L round bottom flask was charged with 11.2 mL (0.200 mol) of propargyl alcohol, 50 g (0.26 mol) of tosyl chloride and 200 mL of ether. After cooled in a -5 oC ice-salt bath, NaOH (40 g, 1.0 mol) was added to the above stirred solution in 6 portions at -5 oC and stirred for another 4 h at rt. After the suspension was poured to cold water and extracted with ether (2x50 mL), the ether layer was dried over Na2SO4 and concentrated under vacuum to give pure propargyl 76 tosylate 37.0 g, 84%. The product was purified under vacuum distillation before use 1 (100-105 oC/45 mtorr). H NMR (500 MHz, CDCl3) δ 7.77 (d, J = 8 Hz, 2H), 7.32 (d, J = 8.5Hz, 2H), 4.65 (d, J = 2.5Hz, 2H), 2. 48 (t, J = 2.5Hz, 1H), 2.47 (s, 3H). Synthesis of diethyl α-propargyl-α-acetamidomalonate (6) from malonate and popargyl tosylate. To a solution of 105 g (0.480 mol) of diethyl acetamidomalonate (5) in 1.41 L of dioxane was added 61 g (0.54 mol) of potassium t-butoxide in 650 mL of dioxane dropwise via cannula over 2 h at rt. Then the suspension was warmed to 50 oC for 2 h followed by adding 83 mL (0.49 mol) of propargyl tosylate in 170 mL of dioxane dropwise at 50 oC over 1h. After the mixture was refluxed for 6 h, it was cooled and filtered to remove the solid. The filtrate was dried over Na2SO4, concentrates and dried under vacuum overnight to give solid diethyl α-propargyl-α-acetamidomalonate (6), 124 1 g, 100%. m.p. 85-88 oC. H NMR (500 MHz, CDCl3) δ 6.91 (s, 1H), 4.24 (dd, J = 3, 7.5 Hz, 4H), 3.26 (d, J = 2.5 Hz, 2H), 2.04 (s, 3H), 1.94 (t, J = 2.5 Hz, 1H), 1.24 (t, J = 7.5 Hz, 6H); 13 C NMR (125 MHz, CDCl3) δ 169.7, 166.9, 78.6, 71.8, 65.6, 63.2, 24.1, 23.2, 14.3. Synthesis of propargyl glycolic acid (1) from malonate and popargyl tosylate. A 3-L round bottom flask was charged with 110 g (0.430 mol) of 6, 1.2 L of H2SO4 (2M) and heated to reflux for 36 h until full conversion to 2-aminopent-4-ynoic acid (7) which was used directly without further purification. Then 5 equiv. of NaNO2 in water (40%) was added dropwise to the solution of amino acid (7) at 0 oC followed by addition of another 3 equiv. of NaNO2 in water (40%). After stirring for 40 h at rt, urea in diluted HCl 77 was added dropwise until the mixture didn’t make KI-Starch test paper blue or purple. Then the resulted solution was extracted with ether continuously for 30 h. The ether layer was treated with Na2S2O5 and stirred for 12 h, then filtered to remove the Na2S2O5. The ether layer was concentrated and dried under vacuum to give 22.8 g of crude propargyl glycolic acid (48%). Later, the product was purified by sublimation to 1 give 19 g of acid (1) (41%). H NMR (500 MHz, CDCl3) δ 4.35 (t, 1H, J= 4.5 Hz), 2.68 (m, 2H), 2.08 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 177.5, 77.7, 71.9, 68.5, 24.4. Synthesis of propargyloxyacetaldehyde diethyl acaetal (9). Propargyl alcohol (134 mL, 2.25 mol) in 200 mL of THF was added to NaH (97 g, 60%, 2.4 mol) in 1.1 L of THF dropwise over 1h at 0 oC and then stirred at rt for another 1h. 8 (234 mL, 1.50 mol) in 100 mL of THF was added to the mixture prepared above and the resulted suspension was refluxed for 4 days to achieve 96% conversion. Then the suspension was filter and the filtrate was concentrated and dissolved in 500 mL of dichloromethane which was washed with aqueous NaCl solution (1x200 mL). The dichloromethane layer was dried over sodium sulfate, concentrated and dried under vacuum for 24 h to give 1 pale red liquid, 9 (237 g, 92%). H NMR (500 MHz, CDCl3) δ 4.65 (t, 1H, J= 5 Hz), 4.21 (d, 2H, J = 2 Hz), 3.70 (m, 2H), 3.58 (m, 4H), 2.43 (t, 1H, J = 2 Hz) 1.22 (t, 6H, J = 10 Hz). Synthesis of propargyloxy lactic acid (12). Acetal (9, 222 g, 1.29 mol) in 180 mL of dichloromethane was added to a mixture of TFA (230 mL), water (230 mL) and dichloromethane (700 mL) dropwise and stirred for 3 days until full conversion. Then 78 dichloromethane was removed and 450 mL of acetic acid was added. After cooled in ice bath, KCN (256 g, 3.90 mol) in 420 mL of water was added dropwise and stirring 1 continued for 18 h until no aldehyde signal in H NMR. The resulted cyanohydrin (11) was extracted continuously with ether to provide a mixture of 11 and acetic acid (644 1 mL, 91% yield from H NMR). The resulted solution was cooled in ice bath followed by the addition of concentrated H2SO4 and methanol (Final concentration of H2SO4 was 6 M). After heating for 24 h at 70 oC, the resulted dark red solution was continuously extracted with ether to give dark liquid after removing all acetic acid. Then the acid mixture was hydrolyzed by addition of 2M NaOH solution. Extraction with dichloromethane removes all impurities, such as mineral oil. After acidifying with 4 M H2SO4, the final acid (12) was obtained via continuous extraction as light yellow solid, 1 144 g, 66% (4 steps). m.p. 58-61 oC. H NMR (500 MHz, CDCl3) δ 4.44 (t, 1H, J= 4 Hz), 4.26 (d, 2H, J= 2 Hz), 3.92 (m, 2H), 2.50 (t, 1H, J = 2 Hz); 13 C NMR (125 MHz, CDCl3) δ 176.14, 78.64, 76.75, 70.59, 70.17, 58.87; IR 3286, 2933, 2122, 1735, 1358, 1203, - 1129, 1104, 1044; HRMS(m/z) calcd for C6H7O4 143.0343 [M-H] , found 143.0344. General Procedure of Synthesis of oligo(ethylene glycol) (n=1,2,3,4) monotrityl ether (14). A 500 mL three-neck round-bottom flask was charged with oligo(ethylene glycol) (n=1,2,3,4) (3 mol) and pyridine (30 mL) and heated to 55 °C, then powdered trityl chloride (42 g, 0.15 mol) was added to the reaction mixture under vigorous stirring. After being stirred at 55 °C for 18 h, the suspension was poured into water and extracted with toluene (5x100 mL). The combined toluene layer was washed with 0.3 M 79 HCl (2x600 mL), aq. NaOH (1x600 mL) and water (5x600 mL), dried over sodium sulfate, concentrated to give white solid or viscous liquid. Ethylene glycol monotrityl ether (n=1): white solid (5% ditrityl ether), 42 g, 91%, 1 m.p. 95-98 oC. H NMR (300 MHz, CDCl3) δ 7.50 (m, 6H), 7.30 (m, 9H), 3.80 (t, 2H, J = 6 Hz), 3.32 (t, 2H, J = 6 Hz), 2.03 (s, br. 1H). Diethylene glycol monotrityl ether (n=2): white solid (8% ditrityl ether), 48 g, 1 92%, m.p. 108-110 oC. H NMR (500 MHz, CDCl3) δ 7.50 (m, 6H), 7.29 (m, 9H), 3.623.82 (m, 6H), 3.30 (t, 2H, J = 6 Hz), 2.06 (s, br. 1H). Triethylene glycol monotrityl ether (n=3): viscous liquid (4% ditrityl ether), 57 g, 1 98%. H NMR (300 MHz, CDCl3) δ 7.51 (m, 6H), 7.32 (m, 9H), 3.64-3.81 (m, 10H), 3.31 (t, 2H, J = 6 Hz). Tetraethylene glycol monotrityl ether (n=4): viscous liquid (9% ditrityl ether), 62 g, 1 95%. H NMR (300 MHz, CDCl3) δ 7.50 (m, 6H), 7.30 (m, 9H), 3.60-3.80 (m, 14H), 3.29 (t, 2H, J = 6 Hz), 2.50 (s, br. 1H). General Procedure of Synthesis of 15 and 17. 15 and 17 were synthesized according to the procedure as described for 9. 1 17 (n=1): pale liquid, 96 g, 91%. H NMR (600 MHz, CDCl3) δ 7.31 (m, 5H), 4.62 (t, 1H, J = 6 Hz), 4.54 (s, 2H), 3.50-3.70 (m, 10H), 1.19 (t, 6H, J = 6 Hz), 2.50 (s, br. 1H). 1 15 (n=2): pale red liquid, 208 g, 100%. H NMR (600 MHz, CDCl3) δ 7.50 (m, 6H), 7.25-7.36 (m, 9H), 4.67 (t, 1H, J = 5 Hz), 3.55-3.80 (m, 12H), 3.27 (t, 2H, J = 5 Hz), 1.23 (t, 6H, J = 5 Hz). 80 1 17 (n=3): dark red liquid, 229 g, 91%. H NMR (500 MHz, CDCl3) δ 7.50 (m, 6H), 7.23-7.37 (m, 9H), 4.66 (t, 1H, J = Hz), 3.54-3.77 (m, 16H), 3.27 (t, 2H, J = 5 Hz), 1.24 (t, 6H, J = 7 Hz). 1 17 (n=1): light brown liquid, 276 g, 92%. H NMR (600 MHz, CDCl3) δ 7.52 (m, 6H), 7.22-7.36 (m, 9H), 4.68 (t, 1H, J = Hz), 3.52-3.78 (m, 20H), 3.30 (t, 2H, J = 5 Hz), 1.28 (t, 6H, J = 7 Hz). General Procedure of Synthesis of 20. A high-pressure autoclave reactor with a glass insert was charged with tritylate 15 or 17, 2% of catalyst (10% palladium on carbon) and 150 mL of ethanol. Hydrogenolysis was carried out with effective stirring at room temperature under 1400 psi of H2 for 48 h. Upon completion of the reaction, the catalyst (which can be reused) and triphenylmethane were removed via filtration. The filtrate was concentrated and dried under vacuum carefully to give a mixture of 18, triphenyl methane and ethanol (<10%) which was used without further purification. Then this mixture was tosylated by TsCl in THF/H2O to give tosylate 19 quantitatively as light yellow liquid. Propargyl alcohol was then coupled with tosylate 19 under basic condition in THF to provide 20 with high yield (containing byproducts from impurities). 1 20 (n=1): light yellow liquid, 35 g, 95%. H NMR (600 MHz, CDCl3) δ 4.62 (t, 1H, J = 6 Hz), 4.18 (d, 2H, J = 2.5 Hz), 3.65-3.71 (m, 6H), 3.50-3.58 (m, 4H), 2.40 (t, 1H, J = 2.5 Hz), 1.19 (t, 6H, J = 6 Hz). 1 20 (n=2): pale liquid, 48 g, 92%. H NMR (600 MHz, CDCl3) δ 4.59 (t, 1H, J = 6 Hz), 4.16 (d, 2H, J = 2.5 Hz), 3.57-3.68 (m, 10H), 3.42-3.56 (m, 4H), 2.38 (t, 1H, J = 2.5 Hz), 81 1.17 (t, 6H, J = 6 Hz). 1 20 (n=3): pale red liquid, 62 g, 94%. H NMR (600 MHz, CDCl3) δ 4.62 (t, 1H, J = 6 Hz), 4.19 (d, 2H, J = 2.5 Hz), 3.60-3.73 (m, 14H), 3.48-3.58 (m, 4H), 2.41 (t, 1H, J = 2.5 Hz), 1.20 (t, 6H, J = 6 Hz). 1 20 (n=4): light brown liquid, 81 g, 90%. H NMR (600 MHz, CDCl3) δ 4.64 (t, 1H, J = 6 Hz), 4.20 (d, 2H, J = 2.5 Hz), 3.62-3.74 (m, 18H), 3.52-3.61 (m, 4H), 2.43 (t, 1H, J = 2.5 Hz), 1.22 (t, 6H, J = 6 Hz). General Procedure of Synthesis of oligo(ethylene glycol) monopropargyl ether lactic acid (23, n=1,2,3,4). Following similar procedure as described in the synthesis of 11, protected acetal 20 was hydrolyzed in a mixture of TFA, water and dichloro methane (1:1:4) followed by cyanation in acetic acid/water solution to give cyanohydrin 22 which was subjected to hydrolysis directly. To a solution of 22 in acetic acid and water (from hydrolysis and cyanation steps), was added a mixture concentrated sulfuric acid and methanol with final concentration of sulfuric acid adjusted to 6M in ice bath. After full 1 conversion monitored by H NMR, the resulted dark solution was diluted, extracted with dichloromethane, concentrated to remove acetic acid under vacuum. Then the acid mixture was hydrolyzed by 2M NaOH solution. Extraction with dichloromethane removes all impurities, such as mineral oil, propargyl ethyl ether and oligo(ethylene glycol) dipropargyl ether. After acidifying with 4 M HCl, the final acids (23, n=1,2) was obtained via continuous extraction and concentrated under vacuum as dark red liquids. The final acid (23, n=3,4) was further purified by washing their solution in dichloromethane with 1 M HCl three times to give pure products as dark red liquids. 82 1 23 (n=1): dark red liquid, 20 g, 53% (8 steps). H NMR (600 MHz, CDCl3) δ 5.05 (br. 1H), 4.38 (t, 1H, J = 5 Hz), 4.22 (d, 2H, J = 2.5 Hz), 3.82-3.92 (m, 2H), 3.70-3.76 (m, 4H), 2.48 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 174.5, 79.1, 75.0, 72.1, 70.8, 70.1, 68.9, 58.4; IR 3286, 2922, 2878, 2122, 1746, 1446, 1355, 1248, 1203, 1110, + 1038; HRMS(m/z) calcd for C8H13O5. 189.0763 [M+H] , found 189.0754. 1 23 (n=2): dark red liquid, 60 g, 56% (8 steps). H NMR (600 MHz, CDCl3) δ 5.55 (br. 1H), 4.37 (t, 1H, J = 5 Hz), 4.22 (d, 2H, J = 2.5 Hz), 3.80-3.92 (m, 2H), 3.64-3.76 (m, 8H), 2.47 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 174.1, 79.2, 74.9, 72.3, 71.0, 70.3, 70.2, 70.1, 70.0, 58.4; IR 3269, 2916, 2878, 2117, 1746, 1460, 1352, 1253, - 1201, 1104, 1035; HRMS(m/z) calcd for C10H15O6 231.0869 [M-H] , found 231.0868. 1 23 (n=3): dark red liquid, 38.6 g, 49% (8 steps). H NMR (600 MHz, CDCl3) δ 4.34 (t, 1H, J = 5 Hz), 4.21 (d, 2H, J = 2.5 Hz), 3.78-3.89 (m, 2H), 3.59-3.74 (m, 12H), 2.45 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 174.26, 79.28, 74.73, 77.29, 70.93, 70.41, 70.38, 70.32, 70.15, 70.11, 68.90, 58.22; IR 3264, 2911, 2878, 2117, 1746, - 1457, 1358, 1250, 1203, 1104; HRMS(m/z) calcd for C12H19O7 275.1131 [M-1] , found 275.1129. 1 23 (n=4): dark red liquid, 48 g, 46% (8 steps). H NMR (600 MHz, CDCl3) δ 4.35 (t, 1H, J = 5 Hz), 4.23 (d, 2H, J = 2.5 Hz), 3.81-3.93 (m, 2H), 3.60-3.79 (m, 16H), 2.46 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 173.47, 79.42, 74.66, 72.25, 70.92, 70.56, 70.51, 70.48, 70.45, 70.23, 70.22, 70.06, 68.91, 58.35; IR 3258, 2916, 2878, 83 2122, 1749, 1457, 1355, 1297, 1250, 1107, 947; HRMS(m/z) calcd for C14H23O8 - 319.1393 [M-H] , found 319.1390. Synthesis of tetraethylene glycol monobenzyl ether (29). Alcohol 24 (187 g, 1.50 mol) was mixed with dihydropyran (1.3 equiv., 166 g, 2.00 mol) at room teperature and stirred for 10 mins at rt. Then 20 drops of concentrated HCl was added and stirred for another 14 h. Upon completion, 15 g of potassium carbonate and 25 g of sodium sulfate were added to remove HCl and water followed by filtration and drying under vacuum at 1 55 oC to give 310 g of colorless liquid (25), 98%. H NMR (500 MHz, CDCl3) δ 4.63 (t, 1H, J = 5 Hz), 3.82-3.90 (m, 2H), 3.72-3.80 (m, 2H), 3.69 (m, 2H), 3.57-3.65 (m, 3H), 3.49 (m, 1H), 1.76-1.87 (m, 1H), 1.66-1.75 (m, 1H), 1.45-1.62 (m, 4H). Then 25 was put into the flask containing NaI (0.1 equiv.), sodium acetate (738 g, 6 equiv.) and 2.5 L of DMF at 110 oC for 12 h while stirring. After removing the solid, the filtrate was concentrated, dissolved in 1.4 L of ethyl acetate and washed with water (5x 400 mL). After drying, concentrating, the dark red liquid was directly hydrolyzed under basic condition (5 M aq. NaOH solution) to give oily product, 175 g. Then this oily product was mixed with TsCl (262 g, 1.38 mol) in THF/water mixture followed by the addition of 362 g of KOH in water. After stirring overnight, the tosylate (26) was extracted with dichloromethane, dried over sodium sulfate and concentrated under vacuum to give 1 light brown liquid, 309 g, 60% (3 steps). H NMR (500 MHz, CDCl3) δ 7.78 (d, 2H, J = 5 Hz), 7.32 (d, 2H, J = 5 Hz), 4.57 (t, 1H, J = 5 Hz), 4.15 (t, 1H, J = 5 Hz), 3.75-3.88 (m, 2H), 3.66-3.70 (m, 2H), 3.58 (m, 2H), 3.41-3.54 (m, 2H), 2.43 (s, 3H), 1.76-1.88 (m, 1H), 1.66-1.75 (m, 1H), 1.45-1.60 (m, 4H); ); 13 C NMR (125 MHz, CDCl3) δ 144.59, 132.96, 84 129.64, 127.77, 98.80, 77.25, 76.99, 76.74, 70.52, 69.13, 68.49, 66.43, 62.07, 30.40, 25.25, 21.43, 19.32. Then 27 (195 g, 0.997 mol) in THF (320 mL) was added dropwise to ice-colded THF (1.8 L) containing NaH (1.05 equiv., 25.1 g, 60%) and NaI (2.5 g) and stirred for another 3 h. Then 26 (1.02 equiv., 350 g, 1.012 mol) in THF (1.2 L) was added over 4 h and stirred overnight. After refluxing for 24 h, the solid was filtered and concentrated. Then it was dissolved in hexane and washed with water followed by decolorization by charcoal to give slight reddish liquid, 396 g, which was directly subjected to be dried under high 1 vacuum at 160 oC to provide pure tetraethylene glycol monobenzyl ether (29). H NMR (500 MHz, CDCl3) δ 7.36 (m, 5H), 4.59 (s, 2H), 3.62-3.75 (m, 16H), 2.57 (br., 1H); 13 C NMR (125 MHz, CDCl3) δ 138.23, 128.29, 127.69, 127.53, 73.20, 72.45, 70.57-70.62 (br.), 70.35, 69.42, 61.72. Synthesis of monotetrahydropyran-protected tetraethylene glycol monotosylate (35). To a flask charged with tetraethylene glycol (30, 970 g, 5 mol), dichloromethane (3 L) and TsOH (6 g) was added dihydropyran (109 mL, 1.15 mol) dropwise at rt. After stiring for 24 h, the reaction mixture was washed with aqueous NaCl solution (6x1 L) and dried over sodium sulfate to give a mixture of 31 and 32 after drying under vacuum, 276 g (8:1). Then this mixture (190 g) was reacted with TsCl (196 g, 1.03 mol) in THF/water mixture while adding saturated KOH solution. The product was extracted with dichloromethane and concentrated to give 258 g of light yellow viscous liquid (32 and 33).Then this mixture was dissolved in 450 mL of methanol containing 1 g of TsOH and stirred for 36 h. After removing methanol, it was dissolved in 85 dichloromethane (1 L) and washed with aqueous NaCl/HCl solution (5x300 mL). Then the organic layer was concentrated and dried under vacuum to give yellow liquid (34), 198 g. Then pure 34 (198 g, 0.6 mol) was protected with dihydropyran (71 mL, 0.78 mol) in dry THF (570 mL) in the presence of TsOH. After removing THF and excess of dihydropyran, it was washed with 10% NaCl solution (3x300 mL), dried over sodium sulfate and concentrated under vacuum to give a dark red liquid (35), 244 g, 74% yield 1 (4 steps). H NMR (500 MHz, CDCl3) δ 7.82 (d, 2H, J = 10 Hz), 7.36 (d, 2H, J = 10 Hz), 4.65 (t, 1H, J = 5 Hz), 4.18 (t, 1H, J = 5 Hz), 3.85-3.92 (m, 2H), 3.58-3.75 (m, 13H), 3.52 (m, 1H), 2.47 (s, 3H), 1.80-1.90 (m, 1H), 1.68-1.77 (m, 1H), 1.48-1.66 (m, 4H); 13 C NMR (125 MHz, CDCl3) δ 144.71, 133.07, 129.77, 127.95, 98.94, 70.76, 70.64, 70.58, 70.54, 69.20, 68.67, 66.63, 62.22, 30.57, 25.42, 21.60, 19.49. Synthesis of octa(ethylene glycol) monopropargyl ether lactic acid (40). Benzyl ether (29, 71 g, 0.25 mol) in THF (200 mL) was added to NaH (1.2 equiv., 12.5 g, 60%), NaI (2 g) in THF (1.3 L) dropwise and stirred for 2 h. Then 35 (1.1 equiv., 121.7 g, 0.275 mol) in THF (300 mL) was added dropwise over 1 h at 0 oC. After refluxing for 12 h, the solid was removed via filtration and the filtrate was and dissolved in dichloromethane. After washed with aqueous NaCl/NaOH solution (1x300 mL) and dried over sodium sulfate, the organic layer was concentrated and dried and vacuum to give reddish viscous liquid, 147 g. The obtained liquid was mixed with TsOH (1 g) and 500 mL of anhydrous methanol and stirred for 36h. The methanol was removed and the oily product was dissolved in dichloromethane which was washed with saturated NaCl/NaOH solution (4x250 mL). After removing the dichloromethane, the product was 86 heated to 55 oC under high vacuum for 12h to provide 121 g of pure octa(ethylene 1 glycol) monobenzyl ether (36) as pale viscous liquid quantitatively. H NMR (500 MHz, CDCl3) δ 7.36 (m, 5H), 4.59 (s, 2H), 3.60-3.77 (m, 32H), 2.60 (t, 1H, J =5 Hz); 13 C NMR (125 MHz, CDCl3) δ 138.25, 128.28, 127.66, 127.51, 73.18, 72.45, 70.45-70.65 (br.), 70.32, 69.41, 61.69. Then bifunctional 38 was synthesized from the coupling of 36 (121 g, 0.26 mol) and 8 following similar procedure to synthesize 20 from 14 to provide 101 g of 38 as amber 1 liquid, 90% (4 steps). H NMR (600 MHz, CDCl3) δ 4.59 (t, 1H, J = 6 Hz), 4.17 (d, 2H, J = 2.5 Hz), 3.47-3.72 (m, 38H), 2.40 (t, 1H, J = 2.5 Hz), 13 C NMR (125 MHz, CDCl3) δ 101.16, 79.64, 74.47, 71.88, 70.45-70.65 (br.), 70.38, 69.08, 62.25, 58.36, 15.32. Octa(ethylene glycol) monopropargyl ether lactic acid (40) was synthesized following the procedure as described in the synthesis of 23 from 20 as light yellow viscous liquid, 1 58.0 g, 65% (3 steps). H NMR (500 MHz, CDCl3) δ 4.34 (t, 1H, J = 5 Hz), 4.21 (d, 2H, J = 2.5 Hz), 3.79-3.91 (m, 2H), 3.60-3.77 (m, 32H), 2.45 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 173.61, 79.66, 74.55, 72.42, 71.07, 70.3-70.6 (br.), 70.18, 69.10, 58.38; IR 3241, 2878, 2117, 1746, 1457, 1352, 1297, 1250, 1107, 950; HRMS(m/z) - calcd for C22H39O12 495.2442 [M-H] , found 495.2437. Synthesis of poly(ethylene glycol) monopropargyl ether lactic acid (50, n≈23). Poly(ethylene glycol) (41, 250 g, 0.250 mol) was stirred with dihydropyran (13 mL, 0.125 mol) for 4 h in 3 L of dichloromethane containing 0.5 g of TsOH. Then the mixture was 87 washed with water (60x500 mL), dried over sodium sulfate, concentrated to give white 1 solid (mixture of 42 and 43), 106 g, 85%. H NMR (500 MHz, CDCl3) δ 7.82 (d, 2H, J = 10 Hz), 7.36 (d, 2H, J = 10 Hz), 3.40-3.90 (m, 93H), 2.61 (t, 0.6H), 1.75-1.87 (m, 1H), 1.65-1.75 (m, 1H), 1.45-1.62 (m, 4H); The this mixture (106 g, 0.110 mol) was mixed with NaH (6.06 g. 60%, 0.160 mol) in THF (410 mL) and stirred for 40 mins before adding 8 (33.6 g, 0.260 mol). After refluxing for 30 h, the solid was removed via filtration and the filtrate was concentrated to dissolve in dichloromethane. The organic layer was washed with water, dried over sodium sulfate, concentrated and dried under high vacuum for 48 h to provide a mixture of 43 and 44, 112 g, 92%. Then the resulted mixture (112 g, 0.110 mol) was dissolved in a mixture of TsOH (1 g), ethanol (450 mL) and dichloromethane (70 mL). After stirring for 24 h, 3 g of K 2CO3 was added before the solvents were removed. Then it was dissolved in dichloromethane and washed with aqueous NaCl/NaOH solution (1x450 mL). The organic layer was dried over sodium sulfate, concentrated and dried under high vacuum at 50 oC for 6 h to give 105 g of white sold. This solid was then reacted with TsCl (72 g, 0.40 mol) in THF/water in the presence of KOH. After workup following similar procedure used before, 120 g of a mixture of 44 and 45 were obtained. Then this solid product mixture in 570 mL of THF was added to the suspension prepared from propargyl alcohol (17 g) and NaH (10 g) in 400 mL of THF. After refluxing for 15 h, 80 ml of water was added and THF was removed followed by the addition of 400 mL of CHCl3 and 500 mL of NaCl solution (10%). Then the product was extracted with CHCl3 to give 112 g of a mixture of 46 and 1 47 after drying under high vacuum at 52 oC, 86%. H NMR (500 MHz, CDCl3) δ 4.54 (t, 88 1H, J = 5 Hz), 4.12 (d, 2H, J = 2.5 Hz), 3.45-3.68 (m, 95H), 2.38 (t, 1H, J = 2.5 Hz), 1.13 (t, 6H, J = 2.5 Hz). Then poly(ethylene glycol) monopropargyl ether lactic acid (50, n≈23) was synthesized as white solid (86 g, 74%, 3 steps) using similar procedure to 1 synthesize 23 from 20 as discussed above. H NMR (500 MHz, CDCl3) δ 4.34 (t, 1H, J = 5 Hz), 4.21 (d, 2H, J = 2.5 Hz), 3.79-3.91 (m, 2H), 3.60-3.77 (m, 32H), 2.45 (t, 1H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 173.41, 79.67, 74.50, 72.42, 71.06, 70.30-70.65 (br.), 70.17, 69.11, 58.39; IR 3252, 2872, 2111, 1749, 1463, 1355, 1300, 1248, 1112, 955. Synthesis of allyloxy lactic aicd (54). The synthesis of 54 used exactly same procedure as described in the synthesis of propargyloxy lactic acid (12). Allyloxyacetadehyde diethyl acetal (51): vacuum distillation, 46 oC/2 Torr., colorless 1 liquid, 198 g, 76%. H NMR (600 MHz, CDCl3) δ 5.84 (m, 1H), 5.22 (m, 1H), 5.12 (m, 1H), 4.58 (t, 1H, J = 6 Hz), 3.98 (m, 2H), 3.65 (m, 2H), 3.52 (m, 2H), 3.42 (d, 2H, J = 6 Hz), 1.15 (t, 6H, J = 6 Hz); 13 C NMR (150 MHz, CDCl3) δ 134.52, 117.01, 101.08, 72.28, 70.55, 62.12, 15.22. Allyloxy cyanohydrin (53): crude product was extracted from aqueous solution, no 1 further purification, 145 g, 100%. H NMR (500 MHz, CDCl3) δ 5.92 (m, 1H), 5.30 (m, 2H), 4.61 (t, 1H, J = 5 Hz), 4.14 (m, 2H), 3.73 (m, 2H), 3.52 (m, 2H); 13 C NMR (150 MHz, CDCl3) δ 133.31, 118.49, 118.15, 72.65, 70.35, 60.73. 1 Allyloxy lactic aicd (54): light yellow liquid, 120 g, 55%. H NMR (500 MHz, CDCl3) δ 5.83-5.93 (m, 1H), 5.17-5.33 (m, 2H) 4.40 (t, 1H, J = 5 Hz), 4.08 (m, 2H), 3.74-3.85 (dq, 89 2H, J = 5 Hz, J = 10 Hz); 13 C NMR (125 MHz, CDCl3) δ 175.97, 133.72, 118.06, 72.62, 70.80, 70.23; IR 3407, 3093, 2977, 2933, 2872, 2500-3500 (br.), 1744, 1424, 1352, - 1206, 1123, 1041, 931; HRMS(m/z) calcd for C6H9O4 145.0511 [M-H] , found 145.0499. Synthesis of azido lactic aicd (56). To a solution of methyl glycidate (3) (26 g, 0.26 mol) in a mixture of 224 mL of MeOH and 11.2 mL of water was added 55 g (0.84 mol) of NaN3 and 24 g (0.37 mol) of NH4NO3. After refluxing overnight, methanol was removed and methy ester (55) was extracted with CH2Cl2. After removal of CH2Cl2, the crude product (55) was purified by vacuum distillation to give 19.2 g of colorless liquid, 1 51%. b.p. 47-50 oC/70 mTorr., H NMR (500 MHz, CDCl3) δ 4.36 (t, 1H, J= 3 Hz), 3.80 (s, 3H), 3.63-3.45 (dq, 2H, J = 3, 13 and 85 Hz), 3.24 (br. s, 1H); 13 C NMR (125 MHz, CDCl3) δ 172.6, 70.4, 53.7,50.1; IR 3470, 2959, 2110, 1743, 1442, 1224; HRMS, cal. + 168.0385 [M+Na] , found 168.0351. The above methyl ester (55, 16 g, 0.11 mol) was hydrolyzed overnight in a mixture of 1 M NaOH solution (180 mL) and methanol (200 mL). After removal of methanol, the solution was acidified and the product was extracted continuously with ether to give 8.5 1 g of light yellow liquid after dried under vacuum, 59%. H NMR (500 MHz, DMSO-d6) δ 5.90 (br. s), 4.20 (q, 2H, J= 4 and 6 Hz), 3.80 (s, 3H), 3.40 (dq, 2H, J = 4, 13 and 35 Hz); 13 C NMR (125 MHz, CDCl3) δ 173.7, 70.4, 54.2; IR 3360, 2514, 2075, 1732, 1443, - 1271; HRMS, calcd for C3H4N3O3 130.0253 [M-1] , found, 130.0208. 90 CHAPTER 3 SYNTHESIS OF CLICKABLE POLYLACTIDES AND THEIR SMART PROPERTIES 3.1 Introduction Polylactides have been extensively studied for biomedical applications due to their biodegradability and biocompatibility. However, their application is limited by their hydrophobic nature and the lack of chemical functionality along the polymer backbone to support further modification. The monomer derivative approach typically involves complex and tedious synthetic procedures, requires careful control of conditions in postpolymerization modifications to avoid backbone degradation, and limits the versatility of pendant functional groups. So having a single procedure that allows the introduction of a family of pendant functional groups onto a polylactide polymer would be highly desirable. Polymeric micelles have been extensively studied as drug delivery carriers because they can enhance drug stability and solubility, elongate in vivo circulation time, and target specific tissues. 137 “Smart” stimuli-responsive materials can be designed to mimic nature in solving biological problems, such as those associated with biological self-response and self-defense. The versatility and untapped potential of smart polymeric materials makes them one of the most exciting new tools at the interface of chemistry and biology. Conventional polymeric micelles disintegrate at concentrations below their critical micelle concentrations (CMCs), which results in premature release of the encapsulated cargo before exerting specific functions. One effective approach to solve this problem is to stabilize micellar structures by cross-linking. 91 138 Approaches to cross-linking in the shell of micelles have been reported and the resulting micelles were more stable than uncross-linked micelles (UCMs) upon dilution. 139 However, cross- linking of micelles with hydrophilic shells may affect the surface mobility, and requires a highly diluted solution to avoid undesired intermicellar cross-linking. So core crosslinked micelles (CMs) have been developed as an effective alternative. Most of the existing UCMs are based on nondegradable polymeric cores such as polyacrylate or polyacrylamide. 140 Considering the requirement for biocompatibility and biodegradability in clinical applications, CMs composed entirely of polyesters would be ideal candidates as drug delivery vehicles. The choice of the cross-linkers is also important because they can render specific properties to CMs, such as stimuli responsive properties, and thus facilitate the release of encapsulated drugs. Because of its high selectivity, reliability, and tolerance to a broad range of functional groups and reaction conditions, “click” chemistry, specifically the copper(I)-mediated 1,3-dipolar cycloaddition of azides and alkynes, is an excellent strategy for elaborating polymer architectures. “Click” chemistry has been used for the preparation of block copolymers, 141 dendrimers, 142 and for the introduction of pendant and terminal functional groups into various polymers including polyesters. 143 Polylactides with clickable side chains would enable them to be efficiently modified with different functionalities through click chemistry under mild conditions and thus impart polylactides with novel solution behavior which would enhance their biomedical application. Thus, we prepared and polymerized propargyloxy-functionalized lactide monomers. The subsequent polymerization of lactide monomers (58 and 59) provided the homopolymer of lactide monomers, random copolymers with lactide as well as 92 block copolymer with poly(ethylene glycol) (mPEG) that have pendant propargyl groups available for the attachment of functionalities using click chemistry. 3.2 Results and Discussion 3.2.1 Monomer Synthesis As shown in Scheme 41, hetero- and homo- monomers (58 and 59) of propargyloxylactic acid were synthesized for the purpose of making polylactide polymers. Hetero-monomer (58) was synthesized as a colorless viscous liquid or a waxy solid by cyclization of 57 under basic conditions which was prepared from 12 and bromoacetyl bromide in 39% yield after flash column chromatography. Homo-monomer (59) was prepared in 32% yield from 12 in refluxing toluene with azeotropic removal of water and then purified by column chromatography on silica to as a white solid. Notably, the purification on silica gel should be completed quickly to avoid decomposition of the 1 monomer. The H NMR spectra of 58 and 59 (shown in Figure 6) show significant chemical shift change for the methine protons (~5.2 ppm for d and i) of both monomers, and are the characteristic of methine protons in lactide rings. The stereochemistry of 59 as a meso (R, S) stereoisomer was also confirmed by an X-ray study (Figure 7). The latide 59 was only isolated as trans isomer which avoids any stereochemistry effect on the polymerization of monomer. 93 Scheme 41. Synthesis of propargyloxy lactides and their polymers 94 1 Figure 6. H NMR spectra of 58, P58A, 59 and P59A. Slovent: CDCl3. 95 C(3A) O(1A) C(4A) C(5A) C(6A) C(1A) O(2) C(6)C(5) C(4) O(3A) C(2) O(3) O(2A) C(2A) O(1) C(1) C(3) Figure 7. Molecular drawing of 59 with 50% thermal ellipsoids from X-ray crystallography. “For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this thesis (or dissertation).” 3.2.2 Bulk Polymerization 3.2.2.1 Kinetics of Bulk Polymerization Due to the low solubility of the monomers 58 or 59 in organic solvents, as well as the low solubility of the polymers in organic solvents, the solution polymerization of 58 or 59 in toluene or THF proceeded very slowly, which leads to poor molecular weight control. This is also true when copolymerization of 58 or 59 with L-lactide (L-LA) was attempted. So the bulk polymerization process without solvent was used to polymerize 58 and 59. Bulk polymerizations of 58 and 59 were performed at 130 °C and were catalyzed by Sn(Oct)2 using 4-tert-butylbenzyl alcohol (TBBA) as the initiator in sealed polymerization bulbs, yielding polymers, P58A and P59A (Scheme 41). The catalyst-toinitiator ratio was 1:1 for all polymerizations, and the monomer-to-initiator ([M]/[I]) ratio 96 was varied from 25:1 to 200:1 to provide different molecular weight materials. The 1 extent of conversion of the monomer to polymer was calculated by comparing the H NMR integration of the methine protons in the monomer at 5.2 ppm with those in the polymer at 5.5 ppm (Figure 6). The ring-opening polymerizations of lactides are 144 typically first-order in monomer and can be expressed as Rp = -d[M]/dt = kp [M][cat] (1) where [M] and [cat] are the concentration of the monomer and the catalyst, respectively, and kp is the rate constant for propagation. Assuming that the catalyst concentration is constant, integration of (1) provides equation (2) -ln([M]t/[M]0) = kp[cat]0 t (2) where the subscripts 0 and t refer to initial concentrations at time = t. Data from such a living polymerization should be linear when plotting-ln([M]t/[M]0 versus time t with a slope kp[cat]0. Lactide polymerizations run to high conversions often deviate from linearity due to the establishment of an equilibrium between the propagation and depropagation steps, leaving an equilibrium monomer concentration [M]e. 145 To simplify evaluation of experimental data, equation (2) is revised to account for [M]e by subtracting [M]e from [M]0, giving equation (3). -ln{([M]t - [M]e)/([M]0 - [M]e)}=kp[cat]0t (3) The [M]e was determined from the limiting monomer conversion reached during the polymerization at 130 oC. As shown in Figure 8, 3% and 6% of 58 and 59 were 97 determined to be left in the system, respectively, and with these numbers equation (3) was used as [M]e to plot the kinetics of bulk polymerization for both 58 and 59. As shown in Figure 9, a linear relationship between -ln{([M]t - [M]e)/([M]0 - [M]e)} and polymerization time t was observed which indicates that the bulk polymerization of propargyloxy substituted lactides exhibits a kinetic pattern similar to that of lactide, and that the ring-opening polymerization of 58 and 59 is controlled and facile. The slope of the linear fitting which is the apparent rate constant values (kp[cat]0) for 58 and 59 clearly indicates that monomer 58 undergoes polymerization much faster than 59 presumably due to the reduced steric hindrance present in 58. 100 Conversion / % 90 80 70 60 50 40 58 59 0 25 50 75 100 125 150 175 200 Time / min Figure 8. Percent conversion for the bulk polymerization of 58 and 59. Conditions: 130 o C, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. 98 -ln(([M]t-[M]e)/([M]0-[M]e)) 2.8 2.6 2.4 2.2 2.0 1.8 58 59 0 10 20 Time / min 30 40 Figure 9. Bulk polymerization kinetics for 58 and 59. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. Figure 10 shows the evolution of the number-average molecular weight (Mn) and polydispersity index (PDI) from gel permeation chromatography (GPC) with time during the bulk polymerization. As the time increases and the percent conversion increases (Figure 8), Mn increases, reaches a maximum at high conversion (<90%) and then drops. The unexpected high Mn of the P59A at polymerization time t=180 mins is probably due to the cross-linking of the alkyne group at high temperature which will be discussed in next section. The PDI is in the range of 1.2-1.5 when conversions are below 90% and then rapidly increases when conversions get higher (Figure 8 and Figure 10). This behavior has been reported to be typical characteristic of ring-opening polymerization of lactides. 58,61 Mn increases linearly with a narrow PDI while 99 conversion increases at low conversion due to the living nature of the ring-opening polymerization. However, intermolecular or intramolecular transesterification will compete with propagation at high conversion (>90% in our case) and results in an increase of PDI and a change in Mn. 25 4 P58, Mnx10-3 P58, PDI P59, Mnx10-3 20 3 P59 PDI 15 PDI Mn x 10-3 145 10 2 5 0 0 40 80 120 Time / min 160 1 200 Figure 10. The evolution of molecular weight (Mn) from GPC (▲, P58; ■, P59) and PDI (∆, P58; □, P59) with time for bulk polymerization of 58 and 59. Conditions: 130 o C, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. The results of the bulk polymerization of 58 and 59 at different ratios of [M]/[I] were collected and are listed in Table 4. All results showed that Xn (number of repeating units) calculated from 1 H NMR spectra are close to their theoretical values after correcting for conversion by NMR at low [M]/[I] ratios; however, Xn calculated from the 1 H NMR at high [M]/[I] ratios shows lower values compared to the theoretical values 100 after correcting for conversion and this is probably due to the low integration accuracy of 1 H NMR at high molecular weight and with increased amounts of transesterification at longer polymerization times as evidenced by the slightly higher PDI. Mn from GPC is listed to show that the molecular weight increases with increasing [M]/[I] ratios; however, the data are not matching to the ratios of [M]/[I] due to the calculation accuracy of GPC software in our lab. This is consistent with GPC profiles which shift to shorter elution times with increasing [M]/[I] ratios as shown in Figure 11 and Figure 12, since large polymers elute out of GPC faster by traveling less in size exclusion chromatography (SEC). 146 The PDI as determined from GPC is around 1.4-1.6 for all of the polymers and is not affected by the GPC calculation. a Table 4. Bulk Polymerizations of 58 and 59 at 130 oC entry M [M]/[I] t/min 1 58 30 2 58 3 b c Mn x 10 -3 e 8 86 26 25 6.9 1.40 50 15 89 44 45 10.8 1.42 58 100 30 88 88 75 18.2 1.51 4 58 200 30 86 172 126 27.9 1.59 5 59 25 30 90 23 22 9.9 1.48 6 59 50 40 87 44 48 10.1 1.46 7 59 100 60 80 80 65 11.4 1.55 8 59 200 100 85 170 135 17.2 1.60 b PDI e Xn a Xn d Conv.(%) 1 c Using Sn(Oct)2 as the catalyst and TBBA as initiator. Measured by H NMR. d 1 Calculated from the [M]/[I] and corrected for conversion. Calculated from H NMR using end group analysis. e g/mol, measured by GPC in THF. 101 30:1 50:1 100:1 200:1 12 13 14 15 16 Elution time / min 17 18 Figure 11. GPC curves of polymers from bulk polymerization of 58 with different [M]/[I] (130 oC). 25:1 50:1 100:1 200:1 12 13 14 15 16 17 Elution time / min 18 Figure 12. GPC curves of polymers from bulk polymerization of 59 with different [M]/[I] (130 oC). 102 3.2.2.2 Thermal Properties of Homopolymers The thermal stability of the polymers P58 and P59 are shown in Figure 13 and Figure 14. The onset decomposition temperature (5% weight loss) for P58 starts at ~196 oC for [M]/[I] at a 30:1 feed ratio and gradually increases to ~230 oC when increasing [M]/[I] to 100:1 and 200:1. A similar trend was observed for P59 but with higher temperature values. The onset decomposition temperature (5% weight loss) for P59 starts at ~210 oC for [M]/[I] at a 25:1 feed ratio and gradually increases to ~255 oC upon increasing [M]/[I] to 100:1 and 200:1. This indicates that the thermal stability of P59 is higher than P58. This behavior is more evidenced in the 100% decomposition temperature with 600 oC for P58 and 750 oC for P59: a nearly 150 o C difference. This might be due to the pendant terminal alkyne groups which might act as a stabilizing agent if they undergo cross-linking upon heating. Indeed, alkyne groups are known to 147,148,149 act as cross-linkers where this is used to make materials flame-retardant. The increased number of pendant propargyl groups on P59 may result in increased cross-linking upon heating. The thermal decomposition of P59 is delayed and shows a plateau from 335 – 620 oC with much higher polymer weight retention than P58, and this is followed by a second rapid decomposition (P59, Figure 14). Notably, the weight retention at the plateau after cross-linking of P59 is doubled compared to that of P58, This indicates that the propargyl group content accounts for the doubled weight retention at the plateau. 103 100 30:1 50:1 100:1 200:1 Weight Retention / % 80 60 40 20 0 100 200 300 400 500 o Temperature / C 600 700 Figure 13. TGA curves of polymers from the bulk polymerization of 58 with different [M]/[I] (130 oC). 100 25:1 50:1 100:1 200:1 Weight Retention / % 80 60 40 20 0 100 200 300 400 500 600 o Temperature / C 700 800 Figure 14. TGA curves of polymers from the bulk polymerization of 59 with different [M]/[I] (130 oC). 104 DSC traces of polymers P58 and P59 are shown in Figure 15 and Figure 16. P58 has a glass transition temperature (Tg) around 21 oC at a low [M]/[I] ratio of 30:1 and the Tg increases to ~ 25 oC upon increasing the molecular weight of the polymers. Due to the flexible chains (propargyloxy groups) along the polymer backbone which disrupts backbone-backbone interaction, the Tg decreases fairly rapidly compared to that of polylactide which has a Tg of 55-60 oC. This is even more obvious when homopolymer P59 is considered. The Tg observed from the DSC for P59 ranges from ~3 to 9 oC depending on the molecular weight. No melting/crystallizing transitions were observed for both P58 and P59 in either the heating or cooling scans, which indicates the amorphous state of the polymers. o 21.3 C o Heat flow 22.1 C o 25.3 C o 26.6 C -40 -20 0 20 40 o Temperature / C 30:1 50:1 100:1 200:1 60 Figure 15. DSC traces of polymers from the bulk polymerization of 58 with different [M]/[I]. (second heating scan, 10 oC/min in N2) 105 o 2.7 C o Heat flow 7.9 C o 8.3 C o 8.9 C 25:1 50:1 100:1 200:1 -60 -40 -20 0 20 o Temperature / C 40 60 Figure 16. DSC traces of polymers from the bulk polymerization of 59 with different [M]/[I]. (second heating scan, 10 oC/min in N2) 3.2.3 Copolymers of P58, P59 and L-Lactide Copolymerization is a common strategy for synthesizing polymers with properties different from the corresponding homopolymers since it can be used to control the density and position of functional groups along the polymer backbones. A series of copolymers (P58-co-L-LA, P59-co-L-LA) of 58, 59 and L-lactide were prepared in the melt of a mixture of L-lactide (L-LA) and either 58 or 59 (M) under the same conditions as was used for preparing homopolymers, i.e. using Sn(Oct)2 as catalyst and TBBA as initiator (Table 5). The [M]/[I] ratio was kept at 100 while monomer feed ratios ranged from 25%, 50% to 75% lactide content. The percent conversions of L-LA were high (>93%) while the percent conversions of 58 were higher than 80%. Due to the overlap of peaks in the NMR spectra, the percent conversions of 59 were estimated to be 85% 106 from Table 4. As shown in Table 5, the calculated molecular weights closely match the experimental molecular weights. Table 5. Mn and Tg of copolymers (P58-co-L-LA, P59-co-L-LA) a Conv. (%) Polymer Feed ratio Polymer LA:M Mn b Mn c Tg Calc. Exp. -3 (x10 ) -3 (x10 ) (oC) LA a 95 86 1:3 21:50 15.2 12.2 30.8 97 86 1:1 38:39 14.8 12.6 36.6 93 80 3:1 52:19 13.7 11.0 46.1 85 d 1:3 32:53 19.1 18.0 19.9 96 85 d 1:1 47:41 17.7 17.1 26.4 96 P59-co-LLA LA:M 97 P58-co-LLA M 85 d 3:1 71:21 15.8 15.5 34.0 1 Determined by H NMR. conversion. c b g/mol, calculated from the [M]/[I] and corrected for 1 g/mol, calculated from H NMR using end group analysis. 1 d estimated from Table 4 due to the overlap in H NMR. The onset decomposition temperature of both copolymers was about 230-240 oC which is similar to their homopolymers (Figure 17 and Figure 18). The pendant alkyne groups along the polymer backbone act as a stabilizer and slows down the decomposition at 300 oC via cross-linking as discussed before. A higher alkyne group content leads to higher weight retention at 300 oC and thus a higher temperature for 100% decomposition. 107 100 1:3 (L-LA:58) 1:1 (L-LA:58) 3:1 (L-LA:58) Weight Retention / % 80 60 40 20 0 100 200 300 400 500 600 700 o Temperature / C Figure 17. TGA curves of P58-co-L-LA with different feed ratios of L-LA to 58. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. 100 1:3 (L-LA:59) 1:1 (L-LA:59) 3:1 (L-LA:59) Weight Retention / % 80 60 40 20 0 100 200 300 400 500 600 o Temperature / C 700 800 Figure 18. TGA curves of P59-co-L-LA with different feed ratios of L-LA to 59. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. 108 As shown in Figure 19 and Figure 20, DSC analysis shows that the Tg of both of the two series of copolymers increases with increasing lactide content (from 30 to 46 oC and 20 to 34 oC for P58-co-L-LA and P59-co-L-LA, respectively). No melting peaks were observed in any of the DSC scans up to 200 oC (not shown here), indicating that the copolymers are completely amorphous. The absence of any crystallinity, even at the highest level of lactide feeding (75%) reveals that the lactide units/blocks are of insufficient length to crystallize (usually at least 12 lactic acid units are required). 75,150 In addition, there was only one glass transition observed for all of the copolymers with different lactide contents, which indicates that the distribution of the propargyloxy lactides and L-LA is statistical in nature (more random than blocks). 108 Specifically, the alkyne groups are not concentrated in block segments, but instead are distributed along the polymer backbone. The Fox equation (4) can be applied to determine the glass transition temperature in polymer blends weight fractions homopolymers. of monomers, 151 and and statistical copolymers the Tg values of 152 based on the the respective 153,154 (W L-LA/Tg,PLLA) + (WM/Tg, PM) = 1/Tg, copolymer (4) WL-LA, W M and M denote the weight percentage of L-lactide and monomer 58 or 59 in our case. The inverse Tg values of the copolymers as a function of the weight percentage of 58 and 59 are shown in Figure 21. The theoretical lines were generated using the Fox equation (4) based on the measured Tg values of the homopolymers (9 109 o C for P58 and 25 oC for P59) and poly(L-lactide) (PLLA, 59 oC). The experimental values fit well with the theory suggesting that the Tg of these copolymers can be tuned and predicted by varying the molar ratios of L-LA and 58 and 59 in the feed during the polymerization, because the proximity of a given temperature to the glass transition temperature is one of the most important factors in determining physical properties of an amorphous polymer. 155 o 30.8 C o Heat flow 36.6 C o 46.1 C 1:3 (L-LA:58) 1:1 (L-LA:58) 3:1 (L-LA:58) -40 -20 0 20 40 o Temperature / C 60 Figure 19. DSC curves of P58-co-L-LA with different feed ratios of L-LA to 58. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. 110 o 19.9 C o Heat flow 26.4 C o 34.0 C 1:3 (L-LA:59) 1:1 (L-LA:59) 3:1 (L-LA:59) -40 -20 0 20 o Temperature / C 40 60 Figure 20. DSC curves of P59-co-L-LA with different feed ratios of L-LA to 59. Conditions: 130 oC, [M]:[Sn(Oct)2]:[TBBA] = 50:1:1. 0.0036 -1 1/Tg (K ) 0.0034 0.0032 0.0030 P58-co-L-LA P59-co-L-LA 0.0 0.2 0.4 0.6 0.8 1.0 Weight percentage of 58 or 59 Figure 21. Plot of 1/Tg versus weight percentage of 58 and 59 in their copolymers. (Solid lines were generated using Fox equation (4) based on the measured Tg values of homopolymers) 111 This statistical nature of the copolymers was also confirmed by comparing the chemical shift changes in the carbonyl region of the 13 C NMR spectra from homopolymers PLLA, P58, P59 and their copolymers P58-co-L-LA, P59-co-L-LA, as 77,156,157 shown in Figure 22 and Figure 23. The carbonyl signals from P58 and P59 and PLLA are well-separated, with resonances of P58 and P59 shifted ~3.0 ppm upfield relative to PLLA. For both P58-co-L-LA and P59-co-L-LA, the “lactide” carbonyl shows new broad resonances shifted upfield ~ 0.3 ppm (marked with “*”). The carbonyls of P58 or P59 exhibit either new broad resonances shifted downfield ~ 0.2 ppm or a more highly split resonance pattern (marked with “+”). These results indicate a random distribution of the incorporated propargyl groups along copolymer backbone. Although there are remaining carbonyl resonances in copolymers which might be from block PLLA segments and P58 or P59 segments, those block segments should be short enough to be omitted to account for the statistical nature of the copolymers due to the single glass transition observed in DSC analyses. 112 P58-co-L-LA Figure 22. 125 MHz 13 C NMR carbonyl region of homopolymers PLLA, P58 and copolymer P58-co-L-LA. Solvent: CDCl3. 113 P59-co-L-LA Figure 23. 125 MHz 13 C NMR carbonyl region of homopolymers PLLA, P59 and copolymer P59-co-L-LA. Solvent: CDCl3. 3.2.4 Thermo-responsive Properties of mPEG/Alkyl-Grafted Polymers from P59 via Click Chemistry Emrick et al. were the first to report the Cu-catalyzed click functionalization of 158 polyesters in the form of poly(α-propargyl-δ-valerolactone). clickable polyesters have been reported 159,160,161 Since then, a variety of which were readily functionalized with different groups to afford new materials with tunable smart properties. Thermoresponsive polymers that show a lower critical solution temperature (LCST) have widely 114 been investigated due to their potential applications such as in controlled drug delivery, biomimetic actuators and gene-transfection agents. At LCST, polymers undergo a solution-gel transition and expulse the solvating water molecules from the polymer. It has been reported that the LCST can be tuned easily over a broad range of temperatures by adjusting the hydrophilic/hydrophobic balance. 77,162,163,164 As shown in Scheme 42, the clickable polylactide P59 was reacted with 1-(2azidoethoxy)-2-(methoxyethoxy) ethane (m3PEGN3), 1-azidodecane (C10H21N3) or a mixtures thereof with different molar ratios (2:1 and 7:2) in the presence of CuSO 4 and sodium ascorbate using DMF as solvent for 12 h at room temperature after removal of oxygen. The 1 H NMR spectra of these polymers were recorded and are shown in Figure 24. The disappearance of the resonances at 2.5 ppm (-CH2-CCH) and 4.20 ppm (-CH2-CCH) and the appearance of a new peak at 7.6 ppm (H on the triazole ring) indicate the success of the click functionalization of P59. This is also confirmed by the appearance of characteristic proton resonances corresponding to m3PEG (3.4-3.7 1 ppm), methoxy (3.3 ppm) and alkyl groups (0.8-1.5 ppm) in the H NMR spectra of P59g-m3PEG (P60), P59-g-C10H21 (P61) and P59-g-C10H21/m3PEG (P62, 2 : 1 feed ratio of m3PEGN3 : C10H21N3). The click reaction of P59 and m3PEGN3 did not run to 1 completion as shown in H NMR and FT-IR spectra. The click functionalization was further tested by FT-IR which showed the disappearance of the triple bond stretching -1 -1 (2117 cm ) and the terminal alkyne C-H stretching (3282 cm ) accompanied by the 115 appearance of a triazole ring stretching vibration at 3150 cm -1 and 1460 cm -1 in fingerprint region (Figure 25). The incorporation of m3PEG and alky chains can account for the stronger absorbance at 2852-2970 and 1100 cm -1 compared to the parent polymer P59 after normalizing the FT-IR spectra. Scheme 42. Click functionalization of P59 with alkyl and PEG azides 116 H on triazole ring H on alkyl OCH2CH2O OCH3 CH3 -CCH ppm 1 Figure 24. H NMR spectra of P59 and its click-grafted polymers P60 (P59-g-m3PEG), P61 (P59-g-C10H21) and P62 (P59-g-C10H21/m3PEG, 2 : 1 feed ratio of m3PEGN3 : C10H21N3). Solvent: CDCl3. 117 Absorbance P61 P62 P60 P59 4000 3500 3000 2500 2000 1500 1000 500 -1 Wavenumber / cm Figure 25. FT-IR spectra of P59 and its click-grafted polymers P60 (P59-g-m3PEG), P61 (P59-g-C10H21) and P62 (P59-g-C10H21/m3PEG, feed ratio of m3PEGN3 and C10H21N3 is 2 : 1). By changing the feed ratios of m3PEGN3 and C10H21N3, one polymer P60 with only pendant hydrophilic side chains and two polymers P62 with different ratios of hydrophilic and hydrophobic pendant chains (2:1 and 7:2) were obtained and used to study their thermo-responsive properties and to measure LCST using the cloud point method. The 1 composition of the pendant chains of the mix-clicked polymers were determined by H NMR spectroscopy to be 58% m3PEG chains and 42% decyl chains from a 2:1 feed ratio, and 63% m3PEG chains and 37% decyl chains from a 7:2 feed ratio, which indicates that the 1-azidodecane reacts faster than m3PEGN3. This also supports the orthogonal nature of click functionalization via simple mixing the azido compounds to afford polymers with different functionalities. Aqueous solution s(~3 wt%) of these 118 polymers were used to measure the absorbance at 450 nm at different temperatures using a UV-visible spectrophotometer. As shown in Figure 26, two polymers P62 with different ratios of hydrophilic and hydrophobic pendant chains (2:1 and 7:2) exhibit very sharp solution-gel transitions at ~ 22 and 27 oC which were recorded as a rapid increase of absorbance due to the dramatic change in light scattering. P60 with only hydrophilic chains has a moderate phase transition through LCST at ~68 oC. This higher LCST resulting from incorporation of more hydrophilic side chains could be attributed to the increased overall hydrogen bonding which requires higher temperature to be disrupted. Notably, those samples remained unchanged after performing more than 8 heating-cooling cycles in order to take pictures of the clear phase transition by a camera in the lab, which indicates their relative high stability upon heating and cooling (data is not shown here). For medical applications, it would be more interesting to have the LCST around body temperature (37 oC) and this presumably could be controled by tuning the hydrophobic/hydrophilic balance. The relationship between LCST and the molar fraction of m3PEG in polymers P60 and P62 is plotted in Figure 27, which shows a nearly linear relationship. This should allow for tuning the LCST to the desired temperature required for different applications simply by adjusting the side chain composition during the click functionalization. The dramatic increase in light scattering for all samples is directly visualized in Figure 28 as the transition from a transparent solution to a cloudy mixture which can be re-dissolved upon cooling. 119 P62 (7:2) P62 (2:1) Absorbance P60 0 10 20 30 40 50 60 70 o Temperature / C 80 90 Figure 26. Plots of absorbance as a function of temperature for aqueous solutions (~3 mg/mL, in D2O) of P60 and P62. The solid lines are the non-linear curve fit to the data. 70 60 LCST 50 40 30 20 0.6 0.7 0.8 0.9 1.0 Real molar fraction of m3PEG in copolymers Figure 27. Plot of LCST as a function of mole fraction of m3PEG in polymers P60 and P62. The solid line is the linear fit to the data. 120 Figure 28. Visualization of thermo-responsive properties of P60 (P59-g-m3PEG), P62 (P59-g-C10H21/m3PEG, 2:1 and 7 : 2 feed ratio of m3PEGN3 : C10H21N3) at Size distribution by intensity / % temperatures below (left) and above (right) their LCST. Solvent: D2O. 60 o P60 25 C o P60 80 C o P62 (7:2) 15 C o P62 (7:2) 45 C o P62 (2:1) 21 C o P62 (2:1) 50 C 45 30 15 0 1 10 100 1000 Diameter / nm 10000 Figure 29. DLS of P60 (P59-g-m3PEG), P62 (P59-g-C10H21/m3PEG, 2:1 and 7 : 2 feed ratio of m3PEGN3 : C10H21N3) at temperatures below (filled symbol) and above (open symbol) their LCST. Solvent: D2O. 121 The dynamic light scattering (DLS) curves in Figure 29 show that the hydrodynamic sizes of all samples rapidly increased from ~10-150 nm to hundreds of nanometers when heating through the LCST, which suggests a transition from a hydrated state to an agglomerated insoluble state. In general, the PEG segments of the polymer can form hydrogen bonding with water to facilitate the dissolution and exhibit reversible dehydration and hydration process with variations in the temperature, which changes the local environment around the polymers and results in a shift of their NMR resonances. Indeed, when heating aqueous solutions of P60 and P62 through their LCST, the proton chemical shifts of the triazole ring (~7.6 ppm), of the mPEG unit (3.4-3.7 ppm), of the methoxy group (3.3 ppm) and of the alkyl groups (0.8-1.5 ppm) are all shifted downfield in a similar pattern as shown in 1 their H NMR spectra (Figure 30). Each of these proton resonances is broader above the LCST due to the phase transition from a soluble to an insoluble state. This clearly further confirms the solution-gel phase transition of the clicked polymers when heated through the LCST. 122 1 Figure 30. H NMR spectra of P60 (P59-g-m3PEG), P62 (P59-g-C10H21/m3PEG, 7 : 2 feed ratio of m3PEGN3 : C10H21N3) at temperatures below and above their LCST. Solvent: D2O. The broken line denotes the significant changes of chemical shift. 3.2.5 Formation of Core Cross-linked Biodegradable Micelles and the Modulation of Redox-Responsive Properties of PEG Block Copolymers of 58 and 59 via Click Chemistry The application of amphiphilic block copolymers as drug carriers has been an active research area, and among the materials surveyed, polylactide (PLA) based copolymers are the most widely studied due to their easy availability. Amphiphilic block copolymers 123 of PEG and PLA form micelles in aqueous solution via self-assembly arising from the large solubility differences between hydrophilic and hydrophobic moieties and these micelles have been investigated in a variety of delivery systems. Typically, the hydrophobic blocks of the copolymers form the core of the micelles through hydrophobic interactions, electrostatic interactions or stereocomplex formation while the hydrophilic blocks of the copolymers form the corona shell all leading to the stabilization of the micellar structure. 165 These structures consist of water-insoluble cores and water- soluble corona shells each of which can be used to modify the micelle and thus improve the delivery performance. The amphiphilic polymers could form micelles in water when their concentrations are above the critical micelle concentrations (CMCs); However, the micelles will dissemble back to free polymer chains when their concentrations are below the CMCs. Due to the amphiphilic nature and low application concentration of the formed micelles, core cross-linking approach has been used to maintain their shape and structure under a variety of conditions. The utilization of functional or reversible cross linkers can give cross-linked micelles stimuli responsive properties depending on the functionalities of the cross linkers. In order to synthesize amphiphilic copolymers, monomers 58 and 59 were polymerized at 130 oC using Sn(Oct)2 as catalyst and mPEG as initiators of different molecular weights (Mn = 350, 2k and 5k) to give block copolymers (P58B, P59B, P59C and P59D) (Scheme 41). As shown in Table 6, the polymerization using mPEG as 1 initiator achieved high conversion and the molecular weights calculated from H NMR analysis are close to the feed ratio of [M]/[I] after taking percent conversion into account. 124 Table 6. Block copolymers of mPEG (initiator) and monomers 58 and 59. entry P I [M]/[I] t Conv. (min) (%) b Xn c Xn a Mn d Tg -3 e o C (x 10 ) 1 mPEG350 50 15 96 48 47 9.0 16 2 P59B mPEG350 50 45 95 48 44 11.4 5 3 P59C mPEG2k 30 20 91 27 30 9.6 -22 4 a P58B P59D mPEG5k 30 20 88 26 29 12.3 -33 Sn(Oct)2 used as the catalyst, 130 oC. b Measured by Calculated from the [M]/[I] and corrected for conversion. spectra using end group analysis. e d d 1 H NMR spectra. c 1 Determined by H NMR g/mol, calculated from Xn . The critical micelle concentration (CMC), an effective indicator of the micellar stability, was determined with nile red (NR) as fluorescence hydrophobic probe. 166 All amphiphilic copolymers generated for the study summarized in Table 7 have relatively low CMC values and the CMC values decreased with decreasing hydrophilic PEG content. Water contact angle measurement is the one of the most common methods for determining a material's wettability and hydrophilicity. The materials with lower contact angle have greater tendency for water to wet the solid, and thus are more hydrophilic. The copolymers P59C and P59D showed similar water contact angles, but which were much lower than those of their homopolymer P59 and the copolymer P59B. This is consistent with the observed CMC values. Based on the observed higher CMC and smaller contact angle than those of P59 and P59B, homopolymer P58 and its copolymer P58B were more hydrophilic compared to other polymers due to the lack of one side chain on each monomer unit. 125 Table 7. CMC and contact angles of polymers P59 and P59B, P59C and P59D Substrate CMC -2 (x 10 mg/mL) Contact angle (o) P58 P58B P59 P59B P59C P59D -- -- 6.6 -- 5.7 11 15 32 86 72 95 82 25 23 The amphiphilic copolymers P58B, P59B, P59C and P59D were used to prepare uncross-linked micelles (UCMs) by nano-precipitation method. 167 , 168 Acetone solutions of copolymers P58B and P59B were added to deionized water dropwise under vigorous stirring to give UCMP58B and UCMP59B, respectively. All UCMs of copolymers P59C and P59D were prepared by adding water dropwise to their DMF solution under vigorous stirring to give UCMP59C and UCMP59D, respectively. The final UCM/copolymer concentration in water was set to 0.5 mg/mL for all samples which is above their CMC. 1 The formation of a core–shell structure of the UCMs was characterized by H NMR 1 spectra of polymeric micelles as shown in Figure 31. The H NMR spectra only show the proton signals of the mPEG component (3.2 ppm for terminal methoxy group and 3.6 ppm for methylene groups from the mPEG) in solvent D2O (PEG segment is dissolved). This indicates that the PEG segment is in an extended solvated state to form the hydrophilic corona shell to protect UCMs in water. Since D2O is a nonselective solvent of P59 segment and it cannot dissolve P59 segment, the protons in the P59 segment (refer to P59 in Figure 24 for proton resonances in CDCl3) have almost disappeared, implying that P59 segment forms the central hydrophobic cores of 126 micelles and P59 segment has restricted motion within the inner core. 169 The zeta potential, the charge that develops at the interface between a colloid particle surface and its liquid medium, is most often used as an indicator of dispersion stability. Surface functionalities of the colloid particles will change the zeta potential values of particles. The fact that PEG shell staying on the surface of UCMs was verified by the increase of UCMs’ zeta potential (ζ) from -50 mV to about -10 mV when PEG composition in copolymers increased as shown in Table 8. Compared to the zeta potential (-50 mV) 170 reported for PLA nanoparticles, the increase of the non-ionic and hydrophilic corona shell thickness shields the core better and thus reduces the ionization degree of carbonyl groups, resulting in a higher zeta potential. 127 UCMP59B UCMP59C UCMP59D 1 Figure 31. H NMR spectra of UCMs prepared from P59B, P59C and P59D in D2O. Table 8. Properties of UCMS and CMs UCM b c d e c D D UCMP58B 52 45 -50 UCMP59B 55 32 -30 UCMP59C 61 58 -11 CMP59C 51 UCMP59D 128 59 a b CM Cross-linker: 63 a ζ -7 d CM Cross-linker: 64 e D D -- -- -- -- -- -- -- -- c ζ d e D D ζ -- -- -- -- -- -- -- -- 67 27 -20 CMP59D 114 29 -14 CM*P59D 134 50 -10 UCMs were dialyzed against water. nm, measured by DLS in water. mV, measured by DLS in water. d b 28 -25 CM*P59C CMs were dialyzed against water. Diameter, nm, measured by TEM. 128 e c Diameter, Zeta potential, - To maintain the original micelles shape and structure, core cross-linking requires the micelles to be dispersed well in solution after addition of the chemicals and catalysts required for core cross-linking to the micelle solution. However, different dispersion stability of the UCMs was observed upon addition of aqueous CuSO4 (10% equiv. to alkyne groups). As shown in Figure 32, all UCM suspensions in water were clear. However, the suspensions of UCMP58B and UCMP59B immediately turned cloudy after adding CuSO4 and a precipitate formed upon standing overnight, which indicates their low dispersion stability. The suspensions of UCMP59C and UCMP59D remained unchanged after adding CuSO4. This phenomenon might be due to the different amounts of PEG composition in the copolymers with a lower degree of stabilization associated with a lower percentage of PEG in the copolymer. The addition of CuSO4 might reduce ionization degree of carbonyl groups on hydrophobic core segment and thus the increased hydrophobic interaction between micelles can be dominant and induce aggregation due to the much lower micelle surface coverage by PEG chains in P58B and P59B compared to P59C and P59D. 129 UCMP58B UCMP59B UCMP59C UCMP59D Figure 32. Visualization of UCMs’ integrity stability prepared from P58B, P59B, P59C and P59D in D2O without (a) and with (b) addition of aqueous CuSO4. Core cross-linking via click chemistry was performed only on those UCMs that were stable to the addition of CuSO4. The cross-linking of UCMP59C and UCMP59D was carried with 1,5-diazidopentane (63), bis(azidoethyl) disulfide (64) or a mixture (2:8) of the two in the presence of CuSO4 and sodium ascorbate to afford the core cross-linked micelles (CMP59C and CMP59D using 63 as cross-linker, CM*P59C and CM*P59D using 64 as cross-linker, CM**P59C and CM**P59D using a mixture (2:8) of 63 and 64 as cross-linker, Scheme 43). As shown in Table 8, after cross-linking, the diameters of all CMs as measured from DLS remained close to their precursor UCMs, indicating that the cross-linking was occurring in the core and did not substantially affect the micelle dispersion stability. The larger DLS diameters of UCMP59D and CMP59D were probably due to the more extended PEG brush chains and due to the higher viscosity of the DMF solution during the micelle preparation. 130 171 All micelles showed medium or high negative zeta potentials (ζ), which is indicative of their high stability in water (Table 8). Indeed, they were stable in water for up to 5 days during the test period, and their stability for longer periods was not tested. This indicates that aging did not induce aggregation of the micelles (Figure 33). Micelle diameters measured by TEM were smaller than those measured by DLS due to the fact that dried samples are required for the TEM tests, whereas the extended micelle structure in water were measured by DLS. 1 The cross linking only occurring in the core was also confirmed by the H NMR spectra of UCMP59D and CMP59D in DMSO-d6. This is because DMSO can dissolve both PEG and P59 segments of UCMP59D to give all proton signals for both of segments in 1 H NMR spectra, whereas DMSO can only dissolve PEG segment of CMP59D to give 1 PEG proton signals in H NMR spectra without dissolving the cross-linked P59 core segment. As shown in Figure 34, only hydrophilic PEG resonances (3.2 and 3.5 ppm) could be observed for CMP59D cross-linked by 63, whereas UCMP59D also showed P59 core segment (3.5, 3.9, 4.1 and 5.5 ppm), indicating the limited molecular motion of the P59 backbone in the core and also that the micelle structure is maintained in DMSO. 131 Scheme 43. Preparation of UCMs and CMs via click chemistry and their response to environment change. 132 135 Diameter / nm 120 UCMP59D CM4P59D 105 90 UCMP58B UCMP59B UCMP59C CM3P59C 75 60 0 24 48 72 Aging time / h 96 120 Figure 33. Hydrodynamic diameters of prepared UCMs and CMs versus aging time (Cross-linker: 63). 133 UCMP59D CMP59D 1 Figure 34. H NMR spectra of UCMP59D and CMP59D (cross-linker: 63) in DMSO-d6. To further demonstrate that the cross-linking of the micelles occurred in the hydrophobic core and to demonstrate their redox-responsiveness to reducing agent (DTT), the changes in the hydrodynamic diameters of both UCMs and CMs as a response to the environmental changes was evaluated by DLS in water or DMF in the presence or absence of DTT. The results of these studies will be described below. DMF is capable of dissolving both outspread PEG segment and buried hydrophobic core segment of the micelles. Therefore DMF would be expected to completely disrupt the UCMs by dissolution, whereas CMs would maintain the micelle structure in DMF, 134 although a larger diameter as measured by DLS would be expected due to the swelling effect of the cross-linked micelle core DMF. As shown in Figure 35, both UCM3 and UCM4 were disrupted by 10-fold dilution with DMF which is confirmed by the disappearance of DLS peaks (Figure 35, (a)). However, the micellar structure of UCMP59C and UCMP59D was preserved after core crossing-linking with 1,5 diazidopentane 63 as evidenced by the presence of DLS peaks for CMP59C and CMP59D after dilution with DMF (Figure 35, (b)). These with the fact that DLS signals are shifted to a larger diameter, indicates that the micelles underwent swelling in DMF. These swelled micelles were unchanged after 24h (data shown in Table 9, and stability was not tested over longer times). These results also indicate the high micelle structure stability of the core cross-linked micelles. Intensity / % (a) 15 UCMP59C in H2O 10 UCMP59C in DMF UCMP59D in H2O 5 UCMP59D in DMF 0 10 Intensity / % (b) 20 15 10 5 100 1000 CMP59C in H2O Cross-linker: 63 CMP59C in DMF CMP59D in H2O CMP59D in DMF 0 10 100 Diameter/ nm 1000 Figure 35. DLS size distribution of hydrodynamic diameters of (a) UCMs and (b) CMs in water before and after 10-fold dilution with DMF (Cross-linker: 63). 135 To track the release of nile red (NR, used as a drug substitute to study the cargo release behavior of the micelles in the present study) from UCMP59C, UCMP59D, CMP59C and CMP59D, the NR-loaded micelles were placed in a dialysis bag (molecular weight cutoff (MWCO) = 3.5 kD; concentration 0.25 mg/mL, 2 mL) and dialyzed with DI water. The fluorescence spectra of the NR-loaded micelles were measured by removing aliquots of the micelle solution from the dialysis bag, which allowed us to monitor the continual release of nile red over 70 hours. As shown in Figure 36, all micelles exhibited a burst release of NR in the first 2 h period. But the cross-linked micelles CMP59C and CMP59D only released ~45% of their original loading, which is much lower than ~ 65% release from the uncross-linked micelles UCMP59C and UCMP59D. After burst release, NR release was monitored with time and it was observed that NR release from the cross-linked micelles CMP59C and CMP59D was more controlled and slower compared to the uncross-linked micelles UCMP59C and UCMP59D which continued until almost all of the entrapped nile red was released. This clearly indicates that core cross-linking has no adverse effect on cargo release; instead, it can slow down but not ultimately inhibit cargo release. 172 The structural stability of micelles after nile red release was complete was also investigated using DLS to study the cross-linking/cargo release effect on particle size distribution. As shown in Figure 37 (a and b), both UCMP59C and UCMP59D showed a bimodal size distribution after NR release different from UCMP59C and UCMP59D without NR loading and release. However, the cross-linked micelles CMP59C and CMP59D showed almost no change compared to CMP59C and CMP59D without NR loading and release. These results again indicate that the micelles structural stability is greatly 136 Normalized fluorescence intensity enhanced by core cross-linking without affecting their cargo release behavior. 1.0 UCMP59C-Nile Red release CMP59C-Nile Red release UCMP59D-Nile Red release CMP59D-Nile Red release 0.8 0.6 0.4 0.2 0.0 0 12 24 36 48 Release time / h 60 72 Figure 36. Continual release profiles of Nile red from NR-loaded uncross-linked micelles and cross-linked micelles. (a) 30 24 18 12 6 0 UCMP59C in H2O before NR loading Intensity / % UCMP59C in H2O after NR release CMP59Cin H2O bedore NR loading CMP59C in H2O after NR release 1 Intensity / % (b) 40 32 10 100 UCMP59D in H2O before NR loading 1000 UCMP59D in H2O after NR release 24 CMP59D in H2O before NR loading 16 CMP59D in H2O after 8 NR release 0 1 10 100 Diameter/ nm 1000 Figure 37. DLS size distribution of both uncross-linked and cross-linked micelles before and after nile red release. 137 The thiol-disulfide exchange reaction is an important reaction in many biological processes and protects important cellular components from reactive oxygen species. Cross-linkers that respond to external triggers such as redox agents would provide an element of control in the design and disassembly of crosslinked micelles and of control cargo release (Scheme 43). So cross-linker 64, which should be responsive to the reducing agent (DTT) was used for cross linking. As shown in Figure 38 and Table 9, the resulted cross-linked CM*P59C and CM*P59D showed an obvious size increase as indicated by DLS peaks which shifted after a 10-fold dilution with DMF. Upon treatment with 10 nM DTT, the diameters of both CM*P59C and CM*P59D in DMF further increased. Due to the fact that the swelled micelles CM*P59C and CM*P59D using cross-linkers 63 or 64 remained the same in DMF without DTT after 24h, the further diameter increase of CM*P59C and CM*P59D can be attributed to cleavage of the disulfide bond of cross-linker 64 via thiol-disulfide exchange which partially and gradually breaks the core cross linking, resulting in further swelling of the cross-linked micelles CM*P59C and CM*P59D. That the cross-linked core was only partially and gradually broken down was confirmed by the further size increase after a DMF solution of CM*P59C and CM*P59D micelles that had been treated with DTT were allowed to stand for 24h. Theoretically, the micelles should be disrupted completely and show no obvious DLS signals; however, DLS profiles showed light scattering of particles with diameters and PDI as 571 nm, 0.70 and 518 nm, 0.49 for CM3 and CM4, respectively (Table 9). These diameters were even larger than the corresponding DLS peak values. Further extending the reaction time or adding more DTT did not change the DLS traces. The micelle quality report from the DLS measurement indicated that the samples were 138 too poor or polydisperse for proper distribution analysis. So it is reasonable to conclude that DTT breaks all of the disulfide cross-linkers and disrupts the core cross-linked micelles in DMF. The reason why DLS still shows light scattering after DTT treatment remains unclear at this point. In order to control the level of core cross-linking and thus the degree of disassembly upon treatment with DTT, a mixture of the inert cross-linker 63 and redox active crosslinker 64 (2:8) was used to give a mixed cross-linked core. As shown in Figure 39 and Table 9, both mix cross-linked CM**P59C and CM**P59D swelled upon dilution with 10fold DMF. After addition of DTT to the DMF solution of micelles, both the crosslinked CM**P59C and CM**P59D showed further hydrodynamic diameter increase which is attributed to disulfide bond cleavage of the cross-linker 64. This reduces the level of core cross linking, and induces further swelling and size increases in the micelles CM**P59C and CM**P59D. In contrast to the micelles with only the redox active crosslinker 64, the mix cross-linked CM**P59C and CM**P59D maintained micelle structure as evidenced by the DLS profiles after 24h in the presence of DTT at room temperature. This clearly demonstrates that the level of core cross-linking can be tuned by changing the feed ratios of the cross-linkers and thus can be used to control cargo release through tunable redox responsive properties of CMs. 139 Table 9. Hydrodynamic diameters and PDI of UCMs and CMs in different solvent a Crosslinker H2O b c DMF Solvent DMF d 24h D PDI DMF (DTT) D PDI e DMF (DTT) f 24h D PDI D UCMP5 9C UCMP5 9D CMP59 C CMP59 D CM**P5 9C CM**P5 9D CM*P59 C CM*P59 D a PDI D PDI -- 61 0.23 -- -- -- -- -- -- -- -- -- 128 0.14 -- -- -- -- -- -- -- -- 63 51 0.26 85 0.35 87 0.33 -- -- -- -- 63 114 0.13 166 0.06 162 0.09 -- -- -- -- 78 0.25 0.26 245 0.37 168 0.22 162 0.24 256 0.21 346 0.23 63:64 65 0.32 (2:8) 63:64 138 0.22 (2:8) 76 0.28 148 64 67 0.33 75 0.28 0.25 113 0.28 571 g 64 134 0.18 220 0.19 224 0.18 300 0.21 518 g Dispersion media of micelles. b-f 76 g 0.70 g 0.49 Diameter (D, nm) and particle polydispersity index (PDI) of micelles were measured by DLS. b Micelles were dialyzed against water. Micelles were diluted 10-fold with DMF and measured immediately by DLS. d c Micelles e were measured by DLS after dilution with 10-fold DMF and standing for 24 h in lab. Micelles were diluted with 10-fold DMF in the presence of 10 nM DTT and measured f immediately by DLS. Micelles were measured by DLS after 10-fold dilution with DMF in g the presence of 10 nM DTT and standing for 24 h in lab. DLS reports were noted as poor or too polydisperse for distribution analysis. These data were the reported diameter and PDI. 140 Cross-linker: 64 CM*P59C in H2O Intensity / % (a) 27 CM*P59Cin DMF CM*P59C in DMF(10 nM DTT) CM*P59C in DMF (10 nM DTT, 24h) 18 9 0 10 Intensity / % (b) 100 1000 CMP59D in H2O 27 CMP59D in DMF CMP59D in DMF (10 nM DTT) CMP59D in DMF (10 nM DTT, 24h) 18 9 0 10 100 Diameter/ nm 1000 Figure 38. DLS size distribution of (a) CM*P59C and (b) CM*P59D in water before and after 10-fold dilution in DMF with or without 10 nM DTT (Cross-linker:64). Cross-linker: 63:64 = 2:8 CM**P59C in H2O Intensity / % (a) 27 CM**P59C in DMF CM**P59C in DMF(10 nm DTT) CM**P59C in DMF (10 nm DTT, 24h) 18 9 0 1 Intensity / % (b)27 10 100 1000 CM**P59D in H2O CM**P59D in DMF CM**P59D in DMF (10 nM DTT) CM**P59D in DMF (10 nM DTT, 24h) 18 9 0 1 10 100 Diameter/ nm 1000 Figure 39. DLS size distribution of (a) CM**P59C and (b) CM**P59D in water before and after 10-fold dilution in DMF with or without 10 nM DTT (Cross-linker: 63:64 = 2:8). 141 3.3 Conclusion The facile synthesis of propargyloxylactic acid (12) provides a convenient approach to the preparation of clickable polylactides (P58 and P59) by the polymerization of hetero- and homo- monomers 58 and 59. The kinetics of bulk polymerization was studied and clickable polylactides with pendant propargyl groups were obtained for further post-modification. The random copolymerization of monomers 58 and 59 with Llactide was also successful to afford statistical copolymers as evidenced by DSC and 13 C NMR analysis. The click grafting of PEG chains, alkyl chains and a mixture of both to these polylactides provides side-chain grafted amphiphilic polylactides which exhibit thermo-responsive properties with LCST ranging from 22 to 65 oC depending on the ratio of incorporated PEG and alkyl chains that were grafted to the polymer. The greater the percentage of PEG chain that is incorporated onto the polylactide, the greater would be the increase in LCST due to the stronger hydrogen bonding with water. The amphiphilic block copolymers of PEG with 58 and 59 were also prepared to study their solution assembly behavior and the cross-linking of the core in the self-assembled 1 micelles. The cross-linking of the core was confirmed by DLS and H NMR spectra. The core cross-linked micelles have high dispersion stability in different solvents and provide controlled cargo release without affecting the micellar structure. Core cross-linked micelles with the disulfide cross-linker 64 conferred a redox-responsive property to the micelles which was triggered upon treatment with the reducing agent (DTT). These observations are suggestive potential application of this new class of cross-linked micelles in delivering pharmaceutical agents in a controlled release. 142 3.4 Experimental Section Materials. Sodium azide, tosyl chloride (TsCl), tosylic acid (TsOH), bromoacetyl bromide, 1,5-dibromopentane, 2,2’-dithiodiethanol, D,L-dithiothreitol (DTT), 4-tertbutylbenyl alcohol (TBBA), triethylene glycol monomethyl ether (m3PEG) and 1bromodecane were purchased from Sigma-Aldrich, Poly(ethylene glycol) monomethyl ether (Mn = 350, mPEG350; Mn = 2000, mPEG2k; Mn = 5000, mPEG5k) were purchased from Sigma-Aldrich and dried under high vacuum for 40h before use. Stannous octoate (Sn(Oct)2) was freshly distilled under vacuum and kept in dry box before use. THF and toluene were dried over sodium and distilled before use. Triethylamine (Et3N) was dried over calcium hydride and distilled before use The silica gel (60 Å porosity) for column chromatography was purchased from SiliCycle Inc. All other reagents and solvents were ACS grade and used as received unless specified. Characterization. Melting points were measured on Electrothermo® melting point 1 apparatus. H NMR and 13 C NMR were recorded on a Varian 300 MHz, VXR-500 MHz or 600 MHz instrument in CDCl3 unless otherwise noted. CDCl3 was used as the 1 internal standard for both H NMR ( = 7.24) and 13 C NMR ( = 77.0). Analytical thin- layer chromatography (TLC) was performed on silica gel plates with F-254 indicator. FTIR spectra were taken with Mattson Galaxy FT-IR 3000. HRMS were taken on Waters Xevo G2-S TOF UPLC/TOF MS. Thermogravimetric analysis (TGA) was recorded on Perkin-Elmer TGA 7 at a heating/cooling rate of 10 °C under air. Differential scanning calorimetry analysis of the polymers (DSC) was recorded on DSC-Q100 at a 143 heating/cooling rate of 10 °C under nitrogen and the second heating scan was used for analysis. Dynamic Light Scattering (DLS) was performed in a Malvern NanoZS ZetaSizer with a 178 degree backscattering detection (Malvern, UK) to give both size and zeta potential of the samples. TEM images were collected on a JOEL 2200FS electron transmission microscope. The micelle was deposited by putting one drop of the solution on the carbon-coated copper grid and then the solution was sucked away after 5 mins to keep the micelle integrity. Polymer molecular weights were determined by GPC-MALLS 35 °C using two PLgel 10µ mixed-B columns in series (manufacturer-stated linear molecular weight range of 6 500-1010 g/mol). The eluting solvent was THF at a flow rate of 1 mL/min. A Waters 2410 differential refractometer was used as the detector and an Optilab rEX (Wyatt Technology Co.) and a DAWN EOS 18-angle light scattering detector (Wyatt Technology Co.) with a laser wavelength of 684 nm were used to calculate absolute molecular weights. Uv-vis spectra were recorded with Evolution 600 BB Uv-visible spectrophotometer (Thermo Electron Corporation) with a polymer concentration at ~3 mg/mL. Fluorescence was recorded using E4500 fluorescence spectrophotometer from Hitachi and the fluorescence emission spectra were observed from 580-900 nm with excitation wavelength at 561 nm. Contact angle was measured on VCA2000 (video contact angle system) with Fostec controller. Synthesis of hetero-monomer (58). Propargyloxylactic acid (7.5 g, 52 mmol) and 200 mL of dry THF were added to a 500 mL 3-neck flask under nitrogen. After cooled to 144 0 °C, 2-bromoacetyl bromide (5.7 mL, 65 mmol) was added. Then a solution of Et3N (11 mL) in THF (50 mL) was added dropwise. After stirring for 5 h at 0 °C, the mixture was filtered to remove white precipitate and the filtrate was concentrated to remove THF and redissolved in ethyl acetate (200 mL). After washed with 1 M HCl (4 x 100 mL) and saline (8  100 mL) and dried over MgSO4, filtration and removal of the solvent in vacuo gave the crude linear acid ester as a viscous liquid (13.8 g). Then this ester was dissolved in a acetone (1.5 L) and refluxed for 24 h in the presence of NaHCO3. The solids were removed by filtration and the filtrate was concentrated in vacuo. The residue was dissolved in ethyl acetate (250 mL), washed with 1 M HCl (2 x 100 mL), saturated NaHCO3 (2 x 100 mL) and saturated NaCl (5  50 mL), and then dried over MgSO4. The product was purified by flash chromatography with CH2Cl2/EtOAc (19:1) and dried under vacuum to give viscous liquid which turned to wax solid upon standing in the dry 1 box, 1.86 g, 39%. Rf = 0.42. H NMR (500 MHz, CDCl3) δ 5.12 (d, 1H, J = 15 Hz), 5.10 (t, 1H, J = 5 Hz), 4.84 (d, 1H, J = 15 Hz), 4.24 (d, 1H, J = 5 Hz), 4.16 (dd, 1H, J = 10 and 5 Hz), 4.03 (dd, 1H, J = 10 and 5 Hz), 2.52 (t, 1H, J = 5 Hz); 13 C NMR (125 MHz, CDCl3) δ 164.22, 163.32, 77.46, 76.34, 76.23, 70.38, 65.39, 59.03; IR 3286, 2944, -1 2878, 2122, 1763, 1443, 1374, 1286, 1237, 1107, 1068, 1038, 848 cm ; HRMS(m/z) + calcd for C8H9O5 185.0451, [M+1] , found 185.0457. Synthesis of homo-monomer (59). Racemic propargyloxylactic acid (10.5 g, 73 mmol) and p-toluenesulfonic acid monohydrate (1 g) were added to a 2-L round bottom 145 flask charged with 1.5 L of toluene. The flask was heated gently overnight until all acids were dissolved in toluene to give a slight yellow solution (Note: fast heating could decompose the propargyloxylactic acid). Then the mixture was heated to reflux for 11 days to achieve a conversion of 69% while removing water azeotropically using a DeanStark trap. After cooled to room temperature, the toluene was removed and the residue was dissolved in 500 mL of CH2Cl2, washed with saturated aqueous NaHCO3 (3 x 150 mL) and dried over MgSO4. Filtration and removal of the CH2Cl2 gave the product as a light brown liquid as a mixture of trans-diasteremer (RS) and cis-diastereomers (RR, SS), 5.51 g, ~60% yield. This brown liquid was purified by flash chromatography with CH2Cl2/EtOAc (19:1) to give white solid as the pure trans-diastereomer which was dried 1 under vacuum and kept in dry box, 2.90 g, 32%. Rf = 0.60. m.p. 125-128 C. H NMR (500 MHz, CDCl3) δ 5.20 (t, 2H, J = 2.5 Hz), 4.23 (d, 4H, J = 3 Hz), 4.14 (dd, 2H, J = 10 and 2.5 Hz), 4.03 (dd, 2H, J = 10 and 2.5 Hz), 2.50 (t, 2H, J = 2.5 Hz); 13 C NMR (125 MHz, CDCl3) δ 164.09, 77.94, 76.43, 75.94, 69.52, 59.01; IR 3291, 2933, 2878, 2122, -1 1760, 1463, 1446, 1361, 1311, 1278, 1231, 1198, 1123, 1057, 922 cm ; HRMS(m/z) + calcd for C12H13O6 253.0712, [M+1] , found 253.0711. Synthesis of 1-(2-azidoethoxy)-2-(methoxyethoxy) ethane (m3PEGN3). A saturated aqueous solution of NaOH (40 g, 1.0 mol) was added to a mixture of triethylene glycol monomethyl ether (34.0 g, 210 mmol) and tosyl chloride (TsCl) (59.0 g, 300 mol) in 400 mL of THF dropwise at 0. The mixture was stirred overnight at room 146 temperature. The mixture was poured into ice water and the water layer was extracted with ether (3 x 150 mL). The combined organic layer was washed with saturated aqueous NaCl, dried over Na2SO4 and concentrated. The resulted tosylate was used without further purification (60 g, 91 %). The prepared tosylate (29 g, 91 mmol), NaI (1 g) and sodium azide (9.50 g, 137 mmol) were mixed in 150 mL of DMF and stirred at 60 C for 24 h. After cooled and added 100 mL of water, the product was extracted with ether (4 x 100 mL) and was washed with saturated aqueous NaCl (5x 100 ml) and dried over Na2SO4. After removal of ether, the product was purified via vacuum distillation to provide 1-(2-azidoethoxy)-2-(methoxyethoxy) ethane (m3PEGN3) as a colorless liquid, 1 b. p. 78 C, 38 mTorr, 10.6 g, 62 %. H NMR (500 MHz, CDCl3)  3.65-3.62 (m, 8H), 3.53-3.51 (m, 2H), 3.36-3.34 (m, 5H); 13 C NMR (125 MHz, CDCl3)  71.88, 70.66, -1 70.61, 70.57, 69.98, 58.89, 50.57; IR 2920, 2872, 2096, 1445, 1338 cm . Synthesis of 1-azidodecane (C10H21N3). 1-bromodecane (22.1 g, 100 mmol), NaI (1 g) and sodium azide (9.50 g, 137 mmol) in DMF (100 mL) were stirred at 60 °C overnight. After cooled, the mixture was filter and dissolved in CH2Cl2, washed with water (6x 100 mL) and dried over Na2SO4. After removal of the solvent, the product 1 was obtained as a colorless liquid, 16.2 g, 88 %. H NMR (500 MHz, CDCl3)  3.26 (t, 2H, J = 7.5 Hz), 1.61 (m, 2H), 1.22-1.42 (m, 14H), 0.89 (t, 2H, J = 7.5 Hz); 13 C NMR (125 MHz, CDCl3)  51.49, 31.86, 29.49, 29.47, 29.27, 29.14, 28.83, 26.71, 22.65, 147 -1 14.07; IR 2927, 2856, 2100, 1457, 1352, 1286, 892, 727 cm . Synthesis of 1,5-diazidopentane (63). 1,5-dibromopentane (11.4 g, 50.0 mmol), NaI (1 g) and sodium azide (14.2 g, 218 mmol) were dissolved in DMF (75 mL) and stirred overnight in 60 °C oil bath. Water (100 mL) was added and the solution was extracted with diethyl ether (2 x 75 mL). The organic phase was combined, washed with water (2 x 30 mL), and dried over MgSO4. After filtration and solvent evaporation, a 1 colorless oil was obtained, 5.0 g, 65 %. H NMR (500 MHz, CDCl3)  3.25 (t, 4H, J = 7.5 Hz), 1.61 (m, 4H), 1.44 (m, 2H); 13 C NMR (125 MHz, CDCl3)  51.07, 28.28, 23.80; -1 IR 2946, 2878, 2104, 1458, 1351, 1266, 900, 858 cm . Synthesis of bis-azidoethyl disulfide (64). 2,2’-dithiodiethanol (6.1 g, 40 mmol) and triethylamine (TEA, 11.1 g, 110 mmol) were dissolved in chloroform (65 mL) and cooled to 0 °C. TsCl (22.8 g, 120 mmol) in chloroform (30 mL) was added dropwise. After stirring at room temperature for 16 h, the reaction mixture was filtered, washed with saturated sodium carbonate solution (3 × 40 mL) and then with DI water (3× 50 mL). The organic phase was dried over MgSO4, filtered, and evaporated under vacuum to give the tosylate which was used directly, 11.4 g, 62 %. Sodium azide (9 g, 65 mmol) in DI water (30 mL) was added to the acetone solution (150 mL) containing the obtained tosylate. Then the mixture was stirred at 80 o C overnight followed by removal of acetone under reduced pressure. The product was extracted with dichloromethane (4 × 50 mL), and the combined organic layer was dried over MgSO4, filtered, and 1 concentrated under vacuum to afford a light yellow oil, 4.2 g, 51 %. H NMR (500 MHz, 148 CDCl3)  3.61 (t, 4H, J = 5 Hz), 2.88 (t, 4H, J = 7.5 Hz); 13 C NMR (125 MHz, CDCl3)  -1 49.94, 37.62; IR 2927, 2872, 2104, 1448, 1413, 1346, 1291, 1256, 1224, 903 cm . General procedure for bulk polymerizations. In a dry box, a small magnetic stir bar was put into a bulb prepared from a 3/8 in. diameter glass tubing with a 1-inch diameter bulb blown on one end. Then weighed monomer in 5-mL vial was dissolved in THF followed by the addition of predetermined toluene solution (0.04 M) of catalyst (Sn(Oct)2) and initiator (TBBA) separately. Then this solution was transferred to the bulb via pipette. After the bulb was connected to a T-shape vacuum adapter via cajon fitting, it was taken out of the dry box and connected to vacuum to remove the solvent carefully followed by holding under high vacuum for 10 h at room temperature. After the o bulb was filled back with nitrogen, it was immersed in 130 C oil bath for polymerization. At the desired period of time, the bulb was removed from the bath, quenched in ice 1 water and opened. A portion of the polymer was analyzed by H NMR for conversion. The remaining polymer was dissolved in CH2Cl2, precipitated from cold methanol/hexanes twice and dried under high vacuum at room temperature for 24 hours. Polymerization of Hetero-monomer (58). 58 (92 mg) was polymerized for 15 min 1 with [M]/[I] = 50. The conversion calculated from H NMR was 89%. Precipitation and 1 drying under vacuum gave 65 mg P58 as a pale red sticky solid (70.6%). H NMR: δ 5.37-5.55 (br, 1H), 4.72-4.97 (br, 2H), 4.20-4.27 (br, 2H), 3.95-4.10 (br, 2H), 2.44-2.58 (br, 1H); Mn, NMR = 8,300 g/mol; GPC (THF): Mn = 10,800 g/mol, PDI = 1.42. Other 149 ratios at 25, 100 and 200 were also prepared. Polymerization of Homo-monomer (59). 59 (126 mg) was polymerized for 40 min 1 with [M]/[I] = 50. The conversion calculated from H NMR was 87%. Precipitation and 1 drying under vacuum gave 95.0 mg P59 as a wax-like sticky solid (75.4%). H NMR: δ 5.37-5.60 (br, 1H), 4.17-4.27 (br, 2H), 3.90-4.12 (br, 2H), 2.45-2.58 (br, 1H); Mn, NMR = 13,000 g/mol; GPC (THF): Mn = 10,100 g/mol, PDI = 1.46. Other ratios at 25, 100 and 200 were also prepared. Copolymerization of 58 and L-lactide. A mixture of 58 (23.0 mg, 0.125 mmol) and L-lactide (18.0 mg, 0.125 mmol) was polymerized for 15 min with [M]/[I] = 100 Precipitation and drying under vacuum gave 30.0 mg of copolymer P58-co-LA as a light 1 yellow solid (73.2%). H NMR: δ 5.38-5.56 (br, 1H), 5.17-5.32 (br, 2H), 4.66-4.98 (br, 2H), 4.20-4.27 (br, 2H), 3.90-4.10 (br, 2H), 2.45-2.56 (br, 1H), 1.45-1.62 (br, 3H); Mn, NMR = 12,600 g/mol. Other feed ratios of 58 to lactide at 3:1 and 1:3 were also conducted. Copolymerization of 59 and L-lactide. A mixture of 59 (31.0 mg, 0.125 mmol) and L-lactide (18.0 mg, 0.125 mmol) was polymerized for 40 min with [M]/[I] = 100 Precipitation and drying under vacuum gave 30.0 mg of copolymer P59-co-LA as a light 1 yellow solid (77.6%). H NMR: δ 5.40-5.56 (br, 2H), 5.17-5.33 (br, 2H), 4.18-4.28 (br, 2H), 3.89-4.12 (br, 2H), 2.45-2.56 (br, 1H), 1.45-1.62 (br, 6H); Mn, NMR = 17,100 g/mol. Other feed ratios of 59 to lactide at 3:1 and 1:3 were also conducted. 150 Amphiphilic block copolymers of 58 and 59 with mPEGs as initiators. P59D was prepared from initiator mPEG5k (150 mg) and 59 (151 mg, 0.600 mmol) with [M]/[I] = 30. The conversion was 88%. The isolated polymer was as light yellow solid, 260 mg, 1 86.4%. H NMR: δ 5.36-5.55 (br, 1H), 4.17-4.26 (br, 2H), 3.89-4.12 (br, 2H), 3.45-3.78 (br, 420H), 3.35 (s, 3H), 2.42-2.54 (br, 1H); Mn, NMR = 12,300 g/mol. P58B, P59B and P59C were also prepared with high conversions (>90%) and good yield. General procedure for “click” functionalization. The desired amount of propargyl functionalized polymer, three equivalents of azide and 10 mol % of sodium ascorbate (10 mg/mL in water) were dissolved in DMF in a 20-mL vial, and the solution was deoxygenated by two freeze-pump-thaw cycles. Then 5 mol % of CuSO4 (10 mg/mL) was added to the frozen mixture under a nitrogen purge followed by evacuation. After backfilled with nitrogen, the solution was stirred at room temperature under nitrogen for 12 h. At the end of the reaction, DMF was blown away by gentle nitrogen flow and the viscous liquid was then washed with hexane and ether to remove the impurities and then dried under vacuum. 1-Azidodecane (C10H21N3)-grafted P59. P59 (([M]/[I] = 50, 9.2 mg) and 40 mg of 1-azidodecane were dissolved in 6 mL DMF for the click reaction. The decane-grafted 1 P59 was isolated as a sticky solid, conversion: 100%, 19 mg, 81.2%. H NMR: δ 7.5-7.9 (br, 1H), 5.2-5.6 (br, 1H), 4.5-4.8 (br, 2H), 4.2-4.4 (br, 2H), 3.7-4.15 (br, 2H), 1.8-2.0 (br, 2H), 1.1-1.4 (br, 14H), 0.8-0.9 (br, 3H); Mn, NMR = 20,900 g/mol. 1-(2-azidoethoxy)-2-(methoxyethoxy) ethane (m3PEGN3)-grafted P59. P59 151 (([M]/[I] = 50, 7.5 mg) and 34 mg of m3PEGN3 were dissolved in 5 mL DMF for the click reaction. The m3PEG-grafted P59 was isolated as a sticky solid, conversion: 69%, 11 1 mg, 73.3%. H NMR: δ 7.5-7.9 (br, 1H), 5.2-5.6 (br, 1H), 4.5-4.8 (br, 2H), 4.2-4.4 (br, 2H), 3.7-4.15 (br, 4H), 3.4-3.68 (br, 8H), 3.2-3.4 (BR, 3H), 2.5 (br, 0.4H, leftover terminal alkyne), 1.1-1.4 (br, 14H), 0.8-0.9 (br, 3H); Mn, NMR = 18,300 g/mol. m3PEGN3/C10H21N3 (2:1 and 7:2)-grafted P59. Two ratios of m3PEGN3/C10H21N3 were used to graft both PEG and alkyl chains on to P59. Conversions were quantitative for both of them. Measurement of LCST. The LCST of the grafted polymer samples was tested by measuring the absorbance of the solution at 450 nm while varying the temperature. The aqueous polymer solution (1 mL) was placed in cuvette and put into Uv-vis spectrophotometer with a programmed temperature control function. Then the absorbance at 450 nm was taken and plotted as a function of temperature. The midpoint of the solution-gel transition was taken as the LCST. Preparation of uncross-linked micelles (UCM) and measurement of critical micelle concentration (CMC). A. Acetone solution method: P58B or P59B (1.8 mg each) was dissolved in acetone (0.6 mL; 3 mg/mL polymer) and it was added to water (3.6 mL; 0.5 mg/mL UCM) slowly added under vigorous stirring in 20-mL vial. After vigorous stirring for another 10 mins at room temperature, the UCM micelles were purified by dialysis against DI water for 24 h to remove acetone (MWCO of 3.5k Da).The CMC of the micelle was tested using Nile Red as a fluorescence probe. Excess of Nile Red in chloroform (0.1 mg/mL, 10 μL) was 152 added to a glass vial and dried completely. Then a micelle solution (1 mL) with different concentration was added and shaken before settling down for 1 h. Then the fluorescence of the polymer solutions excited at 557 nm was taken and the peak fluorescence intensities were plotted versus concentration to give their CMC. B. DMF solution method: In a 20-mL vial, water (10 mL) was slowly added to the solution of P59C or P59D (5 mg each) in DMF (1 mL; 5 mg/mL polymer) dropwise under vigorous stirring. The final concentration of polymers was 0.5 mg/mL. After vigorous stirring for another 10 mins at room temperature, the UCM micelles were purified by dialysis against DI water for 24 h to remove DMF (MWCO of 3.5k Da). The CMC of the micelle was tested similar as discussed above. Preparation of cross-linked micelles (CM) of P59C and P59D. Water (10 mL) was slowly added to the solution of P59C or P59D (5 mg each) and 0.6 equiv. of crosslinker (63, 64 or a mixture 63 and 64 (2:8)) in DMF (1 mL; 5 mg/mL polymer) dropwise under vigorous stirring to afford UCM solution at 0.5 mg/mL. The vigorous stirring was kept for another 10 mins at room temperature before the UCM micelles were purified by dialysis against DI water. Then the UCM solution was deoxygenated by two freezepump-thaw cycles followed by the addition of 5 mol % of CuSO4 (10 mg/mL) and 10 mol % of sodium ascorbate (10 mg/mL in water) to the frozen mixture under nitrogen flow. After another freeze-pump-thaw cycle, the vial was filled with nitrogen and stirred at room temperature for 12 h. After the cross-linking, the CMs were purified by dialysis (MWCO of 3.5k Da) and subjected for further characterization. Nile Red release experiment. Nile Red loaded UCMs or CMs were placed into dialysis tubing and dialyzed with deionized water. Removing aliquots of the micelle 153 solution from the dialysis bag and measuring their fluorescence to monitor the continual release of Nile Red over 70 hours. After analysis, the micelle solution was returned to the dialysis bag and the surrounding water was replaced. And after Nile Red released, the hydrodynamic diameter of the micelles were measured by DLS. 154 CHAPTER 4 SYNTHESIS OF POLYSPIROLACTIDES 4.1 Introduction Thermal properties of the polymeric materials define their process and application temperature range. Glass transition, one important concept in polymer science, is a function of polymer chain flexibility and mobility when there is enough vibrational/thermal energy to create sufficient free-volume to permit the polymer chain movement. A polymer behaves in an increasingly rigid and brittle manner, termed as glassy state, when temperature is below its Tg. Above the Tg of one polymer, the polymer would be more tough and leathery as rubbery state. All polymers have some temperature range where their physical properties are rigid and glassy. In glassy state, the structural and mechanical behavior of polymers is relatively stable and do not change significantly with temperature due to the restricted chain diffusion. Several approaches can be used to tune the Tg of polymer, including molecular weight, tacticity, polymer architecture and the presence of plasticizer. In addition, the Tg of polymers can be raised by increasing the rotational barrier or interaction of polymer backbone. Increasing the polymer chain-chain interaction to make the polymer chains more “sticky” and thus raise the Tg, such as including polar groups or hydrogen bonding in the polymer structure. Incorporating rigid elements or bulky ring structures in or along polymer backbone, such as para-aromatic rings, would restrict the single bond rotation to raise the Tg. For PLA, replacing the methyl group on the lactide ring with bulky substituents, such as branched or cyclic groups can increase the bond rotation barrier 155 and thus the Tg. Baker and co-workers have prepared a series of lactide monomer derivatives with substituents ranging from linear, cyclic alkyl groups to phenyl groups to explore monomer structural effect on the thermal properties of PLA based materials, whose Tg increased to 100 and 98 respectively. 61, 173 o C for phenyl or cyclohexyl substituents, More recently, Hillmyer and coworkers have reported that spirolactide monomers can be prepared by Diels-Alder reaction to afford norbornane, cyclohexadiene and isoprene substituted lactides with high yield. 74,108 The polyspirolactides containing different cyclic structures on polymer backbones after polymerization exhibit much higher Tg than polylactide in the range of 76 to 119 oC. We are interested in the synthesis of polyspirolactides incorporated with different sized cyclic rings and their effect on thermal properties, such as Tg, of the corresponding polylactides due to the essentially rigid ring structure of cyclic rings. 4.2 Results and Discussion 4.2.1 Monomer Synthesis The spirolactide monomers (72a-c) were prepared using cyclic ketones (69a-c) as starting materials. The ketones were treated with KCN in the presence of Na 2S2O5 to give cyanohydrins after extraction which were directly hydrolyzed to their α-hydroxy acids (70a-c) at ~ 80% overall yield after continuous extraction and recrystallization. Then they were condensed with bromopropionyl bromide to form linear esters (71a-c) which were cyclized under mild basic condition in acetone to afford crude spirolactide 156 monomers (72a-c). The monomers were purified by recrystallization from ethyl acetate/hexane and flash column chromatography using dichloromethane as eluent to give white solid at ~55-62% yield. It is worth noting that sublimation after recrystallization cannot supply monomers pure enough for polymerization. Scheme 44. Synthesis of spirolactide monomers and polyspirolactides 4.2.2 Solution Polymerization The solution polymerization was conducted in dichloromethane using triazabicyclodecene (TBD) as catalyst and TBBA as initiator. Due to the basic and nucleophilic nature of the catalyst TBD and the slight acidic and electrophilic nature of the solvent dichloromethane, we conducted the stability experiments of TBD in 1 dichloromethane at different temperature by recording their H NMR spectra in CD2Cl2 157 in order to search for the best condition. As shown in Figure 40, TBD right after dissolved in CD2Cl2 showed only three sets of well splited proton resonances at 1.9, 3.1 and 3.2 ppm which correspond to three methylene groups at 25, 5 and -20 oC. After standing for 96 h at 25 and 5 oC, TBD gave completely different proton resonances in 1 H NMR spectra, which indicates the TBD lost its original structure and catalytic activity. This is confirmed by the very low yield and low molecular weight of the resulted polymers (data is not shown here). However, TBD displayed very high stability by showing exact same three sets of proton resonances as those at 0 h after standing for 96 h at -20 oC. The preliminary polymerization of 72b gave polymer at 43% yield. Apparently, TBD and solvent CD2Cl2 reacted with each other more likely through SN2 reaction to give one inactive catalyst. 174,175 158 TBD in CH2Cl2: TBD in CH2Cl2: TBD in CH2Cl2: TBD in CH2Cl2: TBD in CH2Cl2: TBD in CH2Cl2: 1 Figure 40. H NMR spectra of TBD in CD2Cl2 at different time and temperature. The kinetics was conducted in dichloromethane in order to optimize the polymerization conditions. Solution polymerizations were carried out using TBD as catalyst and TBA as initiator with fixed [M]/[I] ratio at 100:1 but slight different catalyst 159 feeding in order to obtain polymer products. The ratio of catalyst to initiator ([cat]/[I]) is 1:1 for 72a and 72b, and 1.5:1 for 72c. Due to the steric hindrance of spirolactide monomers, the solution polymerizations proceeded fairly slow compared to lactide which was polymerized too fast under same conditions (less than 1 min in our case) and kinetics data were failed to be taken. The solution polymerization of lactide catalyzed by 22 TBD at room temperature was reported to give 99% conversion after 20s (Figure 41). The polymerization of 72a and 72c achieved higher conversions (up to 95%) than 72b which only gave 83% after 100 h. And the polymerization rates depend on the ring size of the lactide ring substituents. Solution polymerization of the spirolactides should also follow first order kinetics as described by equation (2, Chapter 3), and plots of – ln([M]t/[M]0) versus time t should be linear for low monomer conversions as shown in Figure 42. The apparent rate constant values were extracted from the slope of the linear fitted plots and listed in Table 10. As we can see from Table 10, ring size of the substituents showed significant effect on their polymerization rates with 72a about 6x and 9x times faster than 72b and 72c, respectively. The much smaller size of cyclopentyl group compared to cyclohexyl and cycloheptyl groups enables 72a to polymerize faster; however, increasing ring size to cycloheptane did not slow down the polymerization too much compared to 72b, which is probably due to the more flexible ring structure in cycloheptane than cyclohexane, though the steric hindrance is bigger. 176 This effect was also reflected in reversed glass transition temperature trend which will be discussed later. 160 100 Conversion / % 80 60 40 72a 72b 72c 20 0 25 50 75 100 Time / min 125 150 Figure 41. Conversions for solution polymerization of 72a, 72b and 72c. Conditions: -20 o C in CH2Cl2, [M] = 1 M, [M]/[cat]/[I] = 100:1:1 for 72a and 72b; [M]/[cat]/[I] = 100:1.5:1 for 72c. Table 10. Solution polymerization rates of 72a, 72b and 72c. monomer Kp (L S -1 72a 9.28 72b 1.49 72c 1.08 161 -1 mol ) x 10 2 72a 72b 72c 2.5 -ln([M]t/[M]0) 2.0 1.5 1.0 0.5 0.0 0 5 10 15 Time / min 20 25 Figure 42. Solution polymerization kinetics of 72a, 72b and 72c. Conditions: -20 oC in CH2Cl2, [M] = 1 M, [M]/[cat]/[I] = 100:1:1 for 72a and 72b; [M]/[cat]/[I] = 100:1.5:1 for 72c. As shown in Figure 43, Mn of polyspirolactides increased with increasing the polymerization time until it reached high conversions. The PDI also remains narrow (~1.1) and then raised gradually when high conversions were achieved. These results indicate that ring-opening polymerization by TBD at low conversions is living “in nature” and intra- and intermolecular transesterification happen at high conversions which results in the drop of Mn and increase of PDI. The polymerization of these spirolactide monomers with different [M]/[I] ratios were carried out using the optimized condition for each monomer. The results are listed in Table 11, which showed relatively good agreement between feed ratio and Xn from NMR with higher deviation at higher [M]/[I] ratio. GPC results indicate the increase of Mn and narrow PDI for all the polymer 162 samples. GPC profiles of the polymers shift to shorter elution times with increasing [M]/[I] ratios as shown in Figure 44, Figure 45 and Figure 46, since large polymers elute out of GPC faster by traveling less in size exclusion chromatography (SEC). 177 These results indicate polymers with higher molecular weight were obtained by changing feed ratio of monomer to initiator. The PDI as determined from GPC is around 1.4-1.6 for all of the polymers and is not affected by the GPC calculation. P72a,Mn x10 -3 P72b,Mn x10 -3 P72c,Mn x10 2 P72a,PDI P72b,PDI P72c,PDI -3 12 Mn x 10 -3 8 PDI 16 1 4 0 25 50 75 100 Time / h 125 150 175 Figure 43. The evolution of molecular weight (Mn) from GPC and PDI with time for 72a, 72b and 72c. Conditions: -20 oC in CH2Cl2, [M] = 1 M, [M]/[cat]/[I] = 100:1:1 for 72a and 72b; [M]/[cat]/[I] = 100:1.5:1 for 72c. 163 Table 11. Solution polymerizations of 72a, 72b and 72c at -20 oC. entry M 1 2 3 4 5 6 7 8 9 a 72b 10 [M]/[I] LA t/h Conv. Xn b (%) c Xn Mn d PDI -3 e (x 10 ) a Tg e o C 100 200 300 100 200 300 100 200 300 72c 94 95 93 72 75 79 89 85 87 94 190 279 72 150 247 89 170 261 91 175 242 60 132 200 74 127 202 24.0 30.7 36.0 10.6 12.5 14.2 9.2 9.6 10.2 1.32 1.32 1.37 1.16 1.09 1.28 1.11 1.12 1.15 63 64 65 69 70 71 53 55 56 -- 72a 24 50 90 50 110 230 90 200 300 -- -- -- -- -- -- 55 f TBD as the catalyst and TBBA as initiator, -20 oC in CH2Cl2, [M] = 1 M, [cat]/[I] = 1:1 for 72a and 72b; [cat]/[I] = 1.5:1 for 72c. [M]/[I] and corrected for conversion. analysis. e d b 1 Measured by H NMR. 1 c Calculated from Calculated from H NMR using end group f g/mol, measured by GPC in THF using 0.03 as dn/dc for samples. ref. 57. P72a-C5LA P72a 100:1 200:1 300:1 13 14 15 Elution time / min 16 Figure 44. GPC profiles of P72a with different [M]/[I]. 164 P72b-C6LA P72b 100-1 200-1 300-1 15 16 17 Elution time / min 18 Figure 45. GPC profiles of P72b with different [M]/[I]. P72c-C7LA P72c 100:1 200:1 300:1 15 16 17 Elution time / min 18 Figure 46. GPC profiles of P72c with different [M]/[I]. The obtained polymer samples were analyzed by TGA under air and DSC under 165 nitrogen to measure their decomposition temperature and glass transition temperature. The TGA curves are shown in Figure 47, which showed the onset of decomposition temperatures of P72a, P72b and P72c was 290, 275 and 255 oC, respectively. The 100% decomposition temperatures were very close to each other at 350 oC. Both the onset of and 100% decomposition temperatures were higher than polylactide (235 oC for onset and 300 oC for 100%, prepared from [M]/[I] ratio at 100, data not shown here), which indicates slight higher thermal stability than PLA probably due to the more rigid polymer architecture and restricted polymer backbone rotation from the cyclic rings along backbone. This is consistent with previous report where polyspirolactides exhibited higher thermal stability than PLA. 108 Figure 48 showed the DSC traces of three polymers prepared with [M]/[I] at 300 and all other Tgs were listed in Table 11, which showed a significant effect of ring structure on Tg of polymers. The incorporation of bulky substituents along polymer backbone can increase the rotation barrier and thus increase the Tg. P72a introduced cyclopentyl group onto polylactide which increased Tg to 65 oC, and introducing more bulky cyclohexyl group onto P72b further increased its Tg to 71 oC. This clearly confirms the assumption of raising Tg by increasing backbone rotational barrier with bulky substituents. However, when further increase the ring size to cycloheptyl group, somewhat surprising was a Tg of 56 oC for P72c which fell into the same Tg range as PLA (Table 11, entry 10) which showed unexpected reverse trend based on the assumption. It has been known that the intermolecular dipole-dipole 178 interaction between chains of PLA determine its Tg. 166 When dipole−dipole interactions of polylactides become important, bulky substituents not only increase rotation barriers (higher Tg), but also interrupt dipole−dipole interactions (lower Tg) by shielding ester groups. One other factor which cannot be neglected is the higher flexibility of cycloheptyl group than cyclopentyl and cyclohexyl groups. Cycloheptyl group is more flexible than cyclopentyl and cyclohexyl groups, 61 and the rotational barrier increased by cycloheptyl group is lower than those by cyclopentyl and cyclohexyl groups, which also tends to decrease Tg. Apparently, the increase of rotational barrier of P72c to give higher Tg caused by the cycloheptyl groups might be compensated by the decrease of dipole−dipole interactions and the increase of ring flexibility, which leads to a similar Tg to PLA. Weight Retention / % 100 72a 72b 72c 75 50 25 0 200 300 400 500 600 o Temperature / C 700 800 Figure 47. TGA curves of P72a, P72b and P72c. Conditions: [M] = 1 M, [M]/[I] = 300:1 for 72a, 72b and 72c, -20 oC in CH2Cl2. 167 P72a Heat flow P72b P72c 25 50 75 100 125 Time / h 150 175 200 Figure 48. DSC scans of P72a, P72b and P72c. Conditions: [M] = 1 M, [M]/[I] = 300:1 for 72a, 72b and 72c, -20 oC in CH2Cl2. 4.3 Conclusion A series of spirolactide monomers (72a, 72b and 72c) were prepared from condensation of corresponding cyclic ring substituted α-hydroxy acid and bromopropionyl bromide followed by ring-closure with medium yield. The solution polymerization of spirolactides using TBD as catalyst and TBBA as initiator at -20 oC afforded polymers at medium yield with relatively good molecular weight control and narrow PDI. The polyspirolactides showed slightly higher decomposition temperature (≥ 255 oC) than polylactide. The glass transition temperature of polyspirolactides depends on the cyclic ring substituents along backbone. Cyclo- pentyl and hexyl groups substituted polyspirolactides exhibited Tg with values at 65 and 71 oC, respectively, which are higher than polylactide. This indicates the bulky and rigid cyclic structure 168 along polymer backbone can raise their Tg by increasing their rotation barrier. However, cycloheptyl group, a bigger ring substituent, did not raise the Tg. Instead, it resulted in polyspirolactide with similar Tg to polylactide probably due to the disruption of dipoledipole interactions between chains and more flexible ring structure. These results demonstrated the direct attachment of rigid and bulky substituents can be used to tune the Tg and thermal properties of polylactides. 4.4 Experimental Section Materials. Bromopropionyl bromide, 4-tert-butylbenyl alcohol (TBBA), triazabicyclodecene (TBD), cyclopentanone, cyclohexanone and cycloheptanone were purchased from Sigma-Aldrich and used directly. THF was dried over sodium/benzophenone and distilled before use. Dichloromethane and triethyl amine were dried over calcium hydride and distilled before use. The silica gel (60 Å porosity) for column chromatography was purchased from SiliCycle Inc. All other reagents and solvents were ACS grade and used as received unless specified. Characterization. Melting points were measured on Electrothermo® melting point 1 apparatus. H NMR and 13 C NMR were recorded on a VXR-500 MHz instrument in 1 CDCl3 unless otherwise noted. CDCl3 was used as the internal standard for both H NMR ( = 7.24) and 13 C NMR ( = 77.0). Analytical thin-layer chromatography (TLC) was performed on silica gel plates with F-254 indicator. FT-IR spectra were taken with Mattson Galaxy 3000 FT-IR. HRMS were taken on Waters Xevo G2-S TOF UPLC/TOF 169 MS. Thermogravimetric analysis (TGA) was recorded on Perkin-Elmer TGA 7 at a heating/cooling rate of 10 °C under air. Differential scanning calorimetry analyses of the polymers (DSC) was recorded on DSC-Q100 at a heating/cooling rate of 10 °C under nitrogen and the second heating scan was used for analysis. The molecular weights of polymers were determined by gel permeation chromatography (GPC) at 35 °C using two PLgel 10µ mixed-B columns in series with THF as the eluting solvent at a flow rate of 1 mL/min. A Waters 2410 differential refractometer was used as the detector, and monodisperse polystyrene standards were used to calibrate the molecular weights. Synthesis of α-hydroxy acid (70a, 70b and 70c). A mixture of 190 mL of water and cycloketones (69, 0.5 mol) was mixed with a aqueous solution of NaCN (49 g, 1 mol) in 75 mL of water and cooled down in ice bath. After the addition of sodium metabisulfite (Na2S2O5, 95 g, 1 mol) in 160 mL of water at 0 oC, the mixture was vigorously stirred overnight to give a light yellow suspension. After extracted with dichloromethane (4 x 100 mL), the organic layer was dried, filtered and concentrated. The resulted oily product was directly hydrolyzed in concentrated HCl under reflux for 12 h. Then the mixture was extracted continuously with dichloromethane and recrystallized with dichloromethane/hexne (1:10) to afford products as white solids. 1 1-Hydroxycyclopentanecarboxylic acid (70a), 71 g, 78%. m. p. 99-101 oC. H NMR (500 MHz, CDCl3) δ 2.10-2.22 (m, 2H), 1.75-1.95 (m, 6H); 13 C NMR (125 MHz, CDCl3) δ 182.30, 81.76, 39.88, 24.94; IR 3308, 3186, 2960, 2875, 2596, 1733, 1429, -1 12811198, 1099, 1066, 1030, 947, 886 cm . 1 Hydroxycyclohexanecarboxylic acid (70b), 81 g, 82%. m. p. 102-105 oC. H NMR 170 (500 MHz, CDCl3) δ 1.75-1.87 (m, 2H), 1.54-1.72 (m, 7H), 1.20-1.34 (m, 1H); 13 C NMR (125 MHz, CDCl3) δ 181.87, 73.69, 34.41, 25.01, 20.98; IR 3407, 2960, 2933, 2872, -1 2856, 2635, 1735, 1454, 1377, 1278, 1234, 1154, 1041, 994, 743, 674 cm . 1 1-Hydroxycycloheptanecarboxylic acid (70c), 80 g, 76%. m. p. 67-69 oC. H NMR (500 MHz, CDCl3) δ 1.97-2.07 (m, 2H), 1.75-1.85 (m, 2H), 1.47-1.75 (m, 8H); 13 C NMR (125 MHz, CDCl3) δ 182.94, 76.95, 38.87, 29.54, 22.56; IR 3247, 2922, 2856, 2600, -1 1700, 1463, 1418, 1267, 1253, 1179, 1063, 895cm . Synthesis of spirolactide monomers (72a, 72b and 72c). α-Hydroxy acids (70, 0.1 mol) and 300 mL of dry THF were added to a round bottom flask under nitrogen. 2Bromopropionyl bromide (11.6 mL, 0.11 mol) was added at 0 °C. Then a solution of Et3N (19 mL) in THF (75 mL) was added dropwise. After stirring overnight, the mixture was filtered to remove white precipitate and the filtrate was concentrated and dissolved in ethyl acetate (350 mL). After washed with 1 M HCl (4 x 100 mL), saline (2  100 mL) and dried over MgSO4, filtration and removal of the solvent in vacuo gave the crude linear acid ester as a white solid which was used directly. The ester was dissolved in acetone (1.8 L) and refluxed for 24 h in the presence of NaHCO3. After filtration, the filtrate was concentrated in vacuo. The residue was dissolved in ethyl acetate (400 mL), washed with 1 M HCl (3 x 100 mL), saturated NaHCO3 (2 x 150 mL) and saturated NaCl (5  100 mL), and then dried over MgSO4. 171 After recrystallization from ethyl acetate/hexane, the product was purified by flash chromatography with CH2Cl2 and dried under vacuum to give white solids and kept in dry box 1 72a, Rf = 0.80. 11.4 g, 62%. m. p. 68-71 oC. H NMR (500 MHz, CDCl3) δ 5.03 (q, 1H, J = 6.5 Hz), 2.58 (m, 1H), 2.25 (m, 1H), 2.14 (m, 2H), 1.82-2.00 (m, 4H), 1.69 (d, 3H, J = 6.5 Hz); 13 C NMR (125 MHz, CDCl3) δ169.35, 167.30, 89.30, 72.64, 38.51, 38.29, 25.03, 24031, 16.88; IR 2971, 2883, 1438, 1383, 1327, 1267, 1154, 1103, 1071, -1 + 1044, 1000, 925, 787cm ; HRMS(m/z) calcd for C9H13O4 185.0869, [M+H] , found 185.0864. 1 72b, Rf = 0.80. 10.9 g, 55%. m. p. 73-75 oC. H NMR (500 MHz, CDCl3) δ 5.06 (q, 1H, J = 6.5 Hz), 2.14 (m, 1H), 1.95-2.0.8 (m, 2H), 1.90 (m, 1H), 1.65-1.80 (m, 8H), 1.39(m, 1H); 13 C NMR (125 MHz, CDCl3) δ; IR 2966, 2938, 2856, 1769, 1738, 1449, -1 1374, 1339, 1319, 1261, 1275, 1132, 1071, 1046, 994, 964, 917, 724 cm ; HRMS(m/z) + calcd for C10H15O4 199.0957, [M+H] , found 199.0955. 1 72c, Rf = 0.80. 12.3 g, 58%. m. p. 62-64 oC. H NMR (500 MHz, CDCl3) δ 5.02 (q, 1H, J = 6.5 Hz), 2.40 (m, 1H), 2.20 (m, 1H), 2.09 (m, 2H), 1.56-1.82 (m, 11H); 13 C NMR (125 MHz, CDCl3) δ169.21, 166.87, 85.61, 72.68, 38.03, 36.73, 29.53, 29.41, 22.25, 21.91, 17.63; IR 3004, 2933, 2856, 1755, 1463, 1355, 1289, 1380, 1250, 1115, -1 + 1063, 1033, 986, 939, 727 cm ; HRMS(m/z) calcd for C11H17O4 213.1174, [M+1] , found 213.1178. 172 Solution polymerizations of spirolactides. Solution polymerizations were carried out in a glass bulb prepared from 3/8 in. diameter glass tubing which was sealed by connecting to another one-end closed 3/8 in. glass tubing via cajon fitting. The general procedure is as follows. In a dry box, a catalyst solution (0.04 M) was prepared by adding TBD (13.9 mg, 0.1 mmol) into 2.5 mL of dichloromethane and used freshly. The initiator solution (0.04 M) of TBBA was prepared by dissolving TBBA (131 mg, 0.8 mmol) in 20 mL of dichloromethane. The desired amount of spirolactides was added to one 5-ml vial, then predetermined catalyst solution, initiator solution and dichloromethane were transferred into the vial to make the monomer concentration at 1 M. Then the solution was transferred to polymerization apparatus and taken out of the dry box for polymerization in -20 oC cooler. At required polymerization time, the apparatus was taken out and quenched immediately by adding excess of benzoic acid. The polymers were collected by precipitating their dichloromethane solution to ice-clod methanol and dried under vacuum to give white solid. For polymerization kinetics, one polymerization mixture ([M]/[I] = 100) was prepared and separated to different polymerization apparatus for desired polymerization time before quenching for conversion and GPC measurement. 1 P72a: Yield 45%. H NMR (500 MHz, CDCl3) δ 4.97-5.17 (br, 1H), 2.35-2.55 (br, 1H), 2.16-2.35 (br, 1H), 1.92-2.15 (br, 2H), 1.65-1.82 (br, 4H), 1.45-1.55 (br, 3H). 1 P72a: Yield 51%. H NMR (500 MHz, CDCl3) δ 5.02-5.21 (br, 1H), 2.05-2.28 (br, 2H), 1.87-1.99 (br, 1H), 1.40-1.87 (br, 9H), 1.24-1.37 (br, 1H). 1 P72a: Yield 43%. H NMR (500 MHz, CDCl3) δ 4.97-5.17 (br, 1H), 2.23-2.45 (br, 173 1H), 1.95-2.23 (br, 3H), 1.40-1.83 (br, 11H). 174 CHAPTER 5 SURFACE FUNCTIONALIZATION OF SMALL SILICA NANOPARTICLES VIA CLICK CHEMISTRY 5.1 Introduction Nanoparticles are characterized by their size which is in the range of several nanometers to several hundred nanometers, well below the micrometer range. Because of their nanoscale dimensions and hence large specific surface area, nanoparticles exhibit remarkable physicochemical properties, which are usually not observed in their bulk materials. Among them, silica nanoparticles (SiNPs) have attracted much attention due to their chemical inertness, low or non toxicity and excellent thermal stability. Particularly, SiNPs have been widely studied in applications such as drug/gene delivery, 179 imaging agents, 180 immnunoassay agents, 181 because of the easy access of surface functionalization. The silanol surface of SiNPs has been modified by introducing functional groups such as amino, thiol, epoxy, carboxy, azido or alkyne groups. 182,183,184,185 The latter two allow for the efficient functionalization of SiNPs by copper-catalyzed 1,3-dipolar cycloaddition reaction (click chemistry) between an organic azido group and an alkyne group. 184,185 This click reaction is now widely used in biosciences because they proceed with high efficiency at room temperature, in aqueous solution over a wide range of pH and functionalities. Click chemistry has not only been used in the fields of homogenous reactions, but also moved into areas such as nanoparticle surface chemistry with promising applications in controlled drug delivery, 186 187 diagnostics, fluorescent labeling. 175 184 The properties of nanoparticles can be modified based one surface functionalization. For example, poly(ethylene glycol) (PEG), one of the best known surface coating agents, has been reported to reduce the non-specific protein binding and enable nanoparticles stealth characteristics to enhance permeation, retention and circulation time of nanoparticles. 188,189 Aggregation is a common and complex phenomenon for small particles and makes, which can be overcome by suitable surface coating. Surface coating can change the surface charge of the colloidal suspension of nanoparticles to stabilize the particles via electrostatic repulsion. The surface layer can also create a barrier between particles to stabilize nanoparticles, which is often referred to as steric stabilization. In many cases, particle stabilization is attributed to a combination of electrostatic repulsion and steric repulsion. We prepared a series of small and monodisperse silica nanoparticles, which were surface modified with different PEG chains and different functionalities via click chemistry. The surfaces of monodisperse silica nanoparticles were directly coated with alkynyl-substituted triethoxysilanes via condensation with silanol groups, which enabled click function of SiNPs surfaces. Since the click reaction is the final step in the synthesis, the particles are identical except for the surface chemistry, which allows the preparation of surface functionalized SiNPs in a reproducible fashion. 5.2 Results and Discussion 5.2.1 Synthesis of Clicking Reagents As shown in Scheme 45, silane coupling reagent (74) containing terminal alkyne group was synthesized from one-step addition reaction between difunctional 176 organosilane (73) and propargyl alcohol in 90% yield leaving the condensable site untouched for further SiNPs surface functionalization. 190 , 191 Linear azido clicking reagents with different functionalities (76, 78, 79, 83) were synthesized using the corresponding starting materials oligo- or polyethylene glycol monomethyl ether (75), tetraethylene glycol (77) or triethylene glycol monochlorohydrin (80) without column 192,193 purification as shown in Scheme 45. 1 All of these compounds were characterized 13 by H NMR, C NMR and FT-IR. 177 Scheme 45. Synthesis of linear clicking reagents. 5.2.2 SiNPs Surface Modification with mPEG Azides (76) Commercially available bare silica nanoparticles with different sizes were condensed with triethoxysilane 74 in basic EtOH-H2O mixture for three days, as shown in Scheme 46. After condensation, the alkyne modified silica nanoparticles were purified using dialysis tubing with different molecular weight cut off (MWCO) based on size of the 178 particles for 1.5 to 8 days. This thorough purification was confirmed by the absence of 1 organic chemicals in H NMR spectra of dialysate and particle suspension in dialysis tubing after removal of the solvent. The alkyne moiety was condensed on the surface successfully and also confirmed by TGA weight loss. Second, the clicking reagents (76a,b,c,d, 78, 79 and 83) were attached onto the silica particles surface by click reaction (copper catalyzed azide alkyne cycloaddition) to afford the designed monodisperse particles with specific surface functionalities and sizes. To avoid aggregation, the modified particles were purified by dialysis and kept in the corresponding solvent until further characterization or application. As shown in Figure 49, the surface chemistry of nanoparticles was tracked by normalized FTIR spectra of -1 silica nanoparticles using characteristic absorbance of silica at 806 cm , which allows semi-quantitative comparison of the nanoparticles during the syntheses. Alkynyl modified SiNPs (25 nm, SiNP25/74) show an increased absorption band at 2900-3300 cm -1 and two new bands at 1700cm -1 -1 and 1430 cm , which correspond to the alkyl group and the carbonyl group, confirming successful surface condensation. Although the characteristic of triple bond was not observed probably due to the low content of organic layer and small change of dipole moment of triple bond, the subsequent click modification with different mPEG azides (76) resulted in particles. The Ft-IR of the resulting particles clearly showed three sets of stronger absorption bands at 2930, 1460 -1 and 1630 cm (SiNP25/76a-d, Figure 49) due to more methylene groups from mPEG units and formed triazole rings, 194 indicating successful PEG functionalization on the particle surface. Those absorption bands corresponding to methylene groups from 179 mPEG units became stronger with increasing PEG chain length due to the higher surface mPEG grafting. The higher surface grafting for longer mPEG units was also confirmed by their TGA data in Table 12 and Figure 50, which will be discussed later. The zeta potential of the pegylated particles (-6 to -15 mV) was higher than bare SiNPs in buffer (-25 mV, 100 mM KCl and 10 mM HEPES), which indicates the surface 195 modification by mPEG units shields the silica core and reduced the surface charges. Because of the steric hindrance of grafting layers on surface, not all the hydroxyl groups on the particle surface reacted with alkynyl triethoxy silanes (74) and then clicked with azides, thus the zeta potentials were not zero. 196 Scheme 46. Surface modification of silica nanoparticles using click chemistry. 180 Absorbance SiNP25/76d SiNP25/76c SiNP25/76b SiNP25/76a SiNP25/74 SiNP25 3500 3000 2500 2000 1500 1000 -1 Wavenumber / cm Figure 49. Normalized FT-IR spectra of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different mPEG azides (76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000). The TGA data in Table 12 and Figure 50 confirm the step-by-step modification of silica particles with different surface functionalities. The grafting amount on the particle surface can be calculated by dividing the TGA weight losses by the molecular weight of the corresponding organic compound, which correspond to oxidation of the organic layer on the nanoparticles. Table 1 shows a 5% weight loss for alkyne-modified nanoparticles (SiNP25/74), which corresponds to a grafting amount of 0.36 mmol/g on silica. The pegylation step shows 7.6-25% weight loss for peglyated silica particles corresponding to a grafting amount of 0.03 - 0.33 mmol/g on silica, and this indicates the successful click surface modification and synthesis of water-soluble nanoparticles. However, the grafting amount dramatically decreased from 0.33 to 0.03 mmol/g with increasing the length of mPEG units from PEG750 to mPEG5000. This is a typical self181 limiting process of grafting process due to the steric hindrance which makes it more difficult for the longer polymer chains penetrate through existing polymer layer to reach the reactive sites on the surface. 197 Table 12. Surface grafting of silica nanoparticles (25 nm) Weight a Loss % -5 7.6 17 25 17 NP SiNP25 SiNP25/74 SiNP25/76a SiNP25/76b SiNP25/76c SiNP25/76d a Grafting a amount mmol/g -0.36 0.33 0.22 0.12 0.03 Click a Conv. % --91 61 34 10 Size nm b 30 c 37 44 63 56 66 Zeta d Potential mV -25 e --10 -15 -13 -6 Weight loss, grafting amount, and click conversion were calculated for each modification step, not for total modification. sonication for 4 h. c and 10 mM HEPES. b Size was measured in H2O by DLS after d Size was measured in EtOH. Measured in a buffer of 100 mM KCl e Data are not available. 182 100 SiNP25 weight Retention / % SiNP25/74 90 SiNP25/76a 80 SiNP25/76b SiNP25/76d SiNP25/76c 70 150 300 450 600 o Temperature / C 750 Figure 50. Normalized TGA curves of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and click-modified silica particles with different mPEG azides (76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000). The DLS measurements were conducted after each modification step of SiNP25, as shown in Table 12 and Figure 51. All surface modified nanoparticles showed monomodal dispersity by intensity in aqueous solution and the hydrodynamic diameters of the particles increased after each modification. The diameter of bare SiNP25 was 30 nm, and after alkyne modification, the diameter increased to 37 nm. After the pegylation step, the size of particles increased to 44, 63, 56, and 66 nm for different mPEG azides (76a: 4PEG; 76b: PEG750; 76a: PEG2000; 76b: PEG5000) with longer mPEG units giving slightly larger hydrodynamic diameters, which indicates the swelled surface layer increases the micro viscosity around SiNPs and then changes their hydrodynamic impermeability during the Brownian diffusion resulting in larger diameters from 198,199 DLS. The effect of thicker surface mPEG layer on DLS size should also not be 183 omitted due to the higher grafting amount and denser surface layer structure. However, the TEM results (data not shown here) measured from dry particles show that the particle sizes are almost the same as those without surface modification, confirming that the TEM sizes only provide a good measure of the individual nanoparticle cores after thin surface layer collapses during drying on TEM copper grid and have no relevance to solution behavior before and after surface modification. 196,199 The solution dispersion stability of mPEG modified SiNPs in water was studied by monitoring the DLS size distribution versus aging time. The hydrodynamic particle size distributions were narrow and confirmed the monodispersity. As shown in Figure 52, the hydrodynamic diameters of all modified SiNPs from DLS were almost similar during aging in the lab up to 4 days. This long term disperse stability might be due to the combination of electrostatic stabilization (negatively charged surface for all particles) and steric stabilization (surface layer coating) to increase repulsive forces between particles. Thus agglomeration is suppressed or kinetically slow. 200 The slight decrease of sizes might be due to the sediment of larger particles or dust in the solution. We have also investigated the structural stability of surface layer grafted to SiNPs surface when the modified particles were sonicated for 4 h at 60 oC. As shown in Figure 53, both alkynyl SiNP25/74 and SiNP25/76a showed no obvious loss of grafting amount of surface layer following ultrasound sonication. This is consistent with the results reported by Bielawski et al., which demonstrated that short polymers (up to 36 kDa) linked through 1,2,3-triazole moiety (which is often prepared using the click coupling of azides and alkynes) do not undergo cycloreversion 184 201 or any other bonds breaking. 25 SiNP25/76b SiNP25/76a Intensity / % 20 15 SiNP25/76c SiNP25/74 SiNP25/76d SiNP25 10 5 0 10 100 Diameter / nm Figure 51. DLS p ofiles (size distribution by intensity) of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different mPEG azides (76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000) after 4 h sonication at 60 oC. 72 Diameter / nm SiNP25/76d SiNP25/76b 64 SiNP25/76c 56 48 40 SiNP25/76a 0 20 40 60 Aging time / h 185 80 Surface Grafting Amount / (mmol/g) Figure 52. Dispersion stability of silica particles (SiNP25) click-modified with different mPEG azides (25 nm; 76a: 4PEG; 76b: PEG750; 76c: PEG2000; 76d: PEG5000) in aqueous solution. 0.40 0.36 SiNP25/74 0.32 0.28 SiNP25/76a 0.24 0 1 2 3 Sonication Time / h 4 Figure 53. Structural stability of click-grafted surface layer of silica particles modified with alkynyl silane (SiNP25/74) and mPEG azide (SiNP25/76a, 4PEG) during 4 h sonication at 60 oC. 5.2.3 SiNPs Surface Modification with Different Functionalities The surface chemistry of nanoparticles was tracked by normalized FT-IR (Figure 54) of silica nanoparticles (SiNP25) after modification with different organic azides (76a: 4PEG-OMe; 78: 4PEG-OH; 79: 4PEG-NH2; 83: 4PEG-COOH). After “click” modification of particles, two stronger bands at 2930 and 1640 cm -1 appeared corresponding to higher methylene groups content of PEG units and triazole ring. The TGA weight loss of clicked surface layers (7.2 – 8.7 %) and the calculated grafting amount on the surface in Table 13 were very close to the grafted alkyne layer before clicking. These results 186 indicate the successful click chemistry on the particle surface. FT-IR and TGA is not very useful for characterizing surface functionalities (-OMe, -OH, -NH2 and –COOH terminated particles), but the zeta potential of modified particles showed higher values than bare SiNPs, indicating the shielding effect of grafted surface layer on silanol group from bare silica surface, as shown in Table 13. -NH2 terminated particles (SiNP25/79) showed highest zeta potential, which indicates highly negative zeta potential of bare SiNPs was compensated by both PEG layer and -NH2 (+) surface charge. And -COOH terminated particles (SiNP25/83) in buffer gave the lowest zeta potential. The zeta potential was higher than that of bare SiNP25 due to both of PEG layer shielding and negative charged carboxylate group. The DLS showed sizes of modified particles increased slightly from 37 nm before clicking to ~50-60 nm after clicking due to the click grafting and thus microenvironment changes around particles. Only one particle population was observed in all DLS measurements, which suggested that the particles were mono-dispersed in aqueous solution (Figure 55). These results indicate click chemistry could introduce different surface functionalities and afford monodisperse silica nanoparticles. Table 13. Surface grafting of SiNPs (25 nm) with different functionalities. Weight NP SiNP25 Loss % -- SiNP25/74 a Click Grafting a amount mmol/g a Size nm b Zeta -- Conv. % -- 5 0.36 -- SiNP25/76a 7.6 0.33 91 44 -10 SiNP25/78 7.5 0.34 94 50 -14 187 30 37 c Potential mV -25 -- e d SiNP25/79 0.33 90 66 -7 SiNP25/83 a 7.2 8.7 0.37 100 51 -16 Weight loss, grafting amount, and click conversion were calculated for each modification step, not for total modification. sonication for 4 h. c Absorbance and 10 mM HEPES. b Size was measured in H2O by DLS after d Size was measured in EtOH. Measured in a buffer of 100 mM KCl e Data are not available. SiNP25/79 SiNP25/83 SiNP25/78 SiNP25/76a SiNP25/74 SiNP25 3500 3000 2500 2000 1500 1000 -1 Wavenumber / cm Figure 54. Normalized FT-IR spectra of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different surface functionalities (76a: 4PEGOMe; 78: 4PEG-OH; 79: 4PEG-NH2; 83: 4PEG-COOH). 188 20 SiNP25/78 Intensity / % 15 10 SiNP25/76a SiNP25/83 SiNP25/79 SiNP25/74 SiNP25 5 0 10 Diameter / nm 100 Figure 55. DLS profiles (size distribution by intensity) of bare SiNP (SiNP25, 25 nm), alkynyl SiNP (SiNP25/74) and SiNPs click-modified with different functional azides (76a: 4PEG-OMe; 78: 4PEG-OH; 79: 4PEG-NH2; 83: 4PEG-COOH). 5.2.4 Click Surface Modification of Different Sized SiNPs with Azido Carboxylic Acid (83) The surface functionalities of –COOH terminated silica nanoparticles with different sizes (silica core diameter declared by manufacturer: SiNP8, 8 nm; SiNP25, 25 nm; SiNP45, 45 nm; SiNP70, 70 nm.) were also tracked by FT-IR (Figure 56) normalized by using the IR band at 806 cm -1 for semi-quantitative comparison of the nanoparticles during the syntheses. All the pegylated particles showed IR absorption bands at 29003300 cm -1 -1 and 1700cm , which correspond to the alkyl group and carbonyl group, confirming successful modification on the particle surface. Since smaller silica nanoparticles have higher specific surface area, they have higher grafting amount of 189 organic layers on the surface per gram of particles. As shown in Figure 56, with decreasing nanoparticles size, the IR absorption bands at 2900-3300 cm 1 -1 and 1700cm - became stronger, indicating higher grafting amount for smaller silica nanoparticles. TGA results in Table 14 further confirmed this observation by showing higher weight loss and grafting amount for smaller particles. The click conversion was calculated from TGA data and it showed high conversion for all particles (>80%). The DLS traces and results after click modification of silica nanoparticles were shown in Figure 57 and Table 14. The average hydrodynamic diameters of bare silica particles were measured by DLS in water and they were bigger than the sizes claimed by manufacturers. And all pegylated nanoparticles showed monomodal dispersity (intensity distribution) in aqueous solution with their diameters closer to the corresponding bare SiNPs. This suggests the surface coating, highly negative charged surface and hydrophilicity of the surface modified particles facilitate the dispersion in solution. These results confirm the surface modification of silica nanoparticles with different sizes. Table 14. Surface grafting of different sized SiNPs with carboxylic acid (83). Weight a Loss Grafting a amount Click a Conv. % mmol/g % SiNP8/83 13.4 0.57 SiNP25/83 8.7 SiNP45/83 SiNP70/83 NP a Size b Size c Size d Zeta e Potential nm nm nm 97 8 20-30 36 -14 0.37 100 25 30 51 -16 2.8 0.12 84 45 80 82 -32 2.1 0.10 79 70 150 137 -37 mV Weight loss, grafting amount, and click conversion were calculated from click modification step only. b The average diameter of bare SiNPs declared by 190 manufacturer. c d The hydrodynamic diameter of bare SiNPs measured in H2O by DLS. Absorrbance The hydrodynamic diameter of SiNPs modified with 83 measured in H2O by DLS. Measured in a buffer of 100 mM KCl and 10 mM HEPES. SiNP8/83 SiNP25/83 SiNP45/83 SiNP70/83 3500 3000 2500 2000 1500 1000 -1 Wavenumber / cm Figure 56. Normalized FT-IR spectra of different sized SiNPs click-modified with 83 (4PEG-COOH after 4 h sonication. Core diameter: SiNP8, 8 nm; SiNP25, 25 nm; SiNP45, 45 nm; SiNP70, 70 nm. 191 e SiNP70/83 25 SiNP45/83 Intensity / % 20 SiNP25/83 15 SiNP8/83 10 5 0 10 100 Diameter / nm Figure 57. DLS profiles (size distribution by intensity) of different sized SiNPs clickmodified with 83 (4PEG-COOH) after 4 h sonication. Core diameter: SiNP8, 8 nm; SiNP25, 25 nm; SiNP45, 45 nm; SiNP70, nm. 5.2.5 In Vitro Colloidal Stability Studies of SiNPs Modified with Different Surface Functionalities Colloidal stability under physiological conditions is one of the most important factors relating to the biological applications of nanomaterial. 202,203 The colloidal stability of the silica nanoparticles in RPMI media 1640 with 2% bovine calf serum was investigated by measuring their hydrodynamic sizes by DLS for 3 days at room temperature. As shown in Figure 58, SiNP25/76a terminated with –OMe group was incubated in the media. DLS profiles of SiNPs only, media only and SiNPs in media (0.1 mg/mL) at different incubation time were recorded to show how the modified particles evolved. The media exhibited one smaller population and a bigger particle population in DLS profile. After incubation for 68 h, the intensity of the peak with larger diameter 192 grew. It is not clear what these two peaks are corresponding to. The shift of the peak may be due to the protein aggregation. SiNP25/76a in media showed a similar particle population during the 68 h incubation. Very similar DLS size distribution behavior was observed for SiNP25/78 terminated with –OH group (Figure 59). Although SiNP25/83 terminated with –COOH group exhibited very good solution stability in media when incubated for 42 h, it eventually showed gradual size increase in media by presenting a much bigger particle population close to that from media after incubated for 68 h as shown in Figure 60. However, SiNP25/79 terminated with –NH2 group showed immediate size increase from ~66 nm to ~100 nm after mixed with media and its size remained with little changes very close after incubation for 68 h (Figure 61). Therefore, the dispersion of surface-modified SiNPs in media crucially depends on their surface functionalities. However, as it is not clear why the two components in media changed with longer incubation time, it is difficult to draw a clear conclusion for the colloidal stability of surface-modified SiNPs. Further studies are needed to evaluate and understand the colloidal stability and the interactions between the media and the modified SiNPs, such as by measuring the change of media components before and after incubation with SiNPs. 193 0h Intensity / % Intensity / % 12 21 h NPs 12 NPs 8 Media+NPs Media 4 8 Media+NPs 0 0 12 1 10 42 h NPs 4 Media 100 1 68 h 12 10 100 NPs 8 Media+NPs Media 4 8 Media+NPs Media 4 0 0 1 10 100 Diameter / nm 1 10 100 Diameter / nm Figure 58. Size distribution of SiNPs (SiNP25/76a, -OMe terminated) when incubated in media at 0, 21, 42 and 68 h. 16 Intensity / % 16 0 h NPs 12 Media+NPs 8 Media Media+NPs 8 Media 4 4 0 0 16 Intensity / % 12 21 h NPs 12 10 42 h NPs 100 Media+NPs 8 Media 16 12 8 4 100 Media+NPs Media 4 0 10 68 h NPs 0 10 100 Diameter / nm 10 100 Diameter / nm Figure 59. Size distribution of SiNPs (SiNP25/78, -OH terminated) when incubated in media at 0, 21, 42 and 68 h. 194 16 20 h NPs Intensity / % 16 0 h NPs 12 Media+NPs 8 Media Media+NPs 8 Media 4 4 0 0 10 42 h NPs 16 Intensity / % 12 100 12 16 68 h 12 Media+NPs 8 Media 10 100 NPs 4 8 Media+NPs Media 4 0 0 10 100 Diameter / nm 10 100 Diameter / nm Figure 60. Size distribution of SiNPs (SiNP25/83, -COOH terminated) when incubated in media at 0, 20, 42 and 68 h. Intensity / % 16 16 0h NPs 12 12 NPs Media+NPs 8 Media Media+NPs 8 Media 4 4 0 0 16 Intensity / % 21 h 12 42 h 10 100 NPs Media+NPs 8 Media 16 68 h 12 8 4 4 0 10 100 NPs Media+NPs Media 0 10 100 Diameter / nm 10 100 Diameter / nm Figure 61. Size distribution of SiNPs (SiNP25/79, -NH2 terminated) when incubated in media at 0, 21, 42 and 68 h. 195 5.3 Conclusion We developed a facile route to synthesize stable mono-disperse SiNPs with specific sizes and surface chemistries. SiNPs were grafted with an alkynyl-substituted triethoxysilane (74) to afford alkynyl silica nanoparticles, which enable the modification of particle surfaces via click chemistry, the Cu-catalyzed azide-alkyne cycloaddition. Subsequent click reactions with different organic azides provide nanoparticles with a broad range of surface chemistries (carboxylic acids, amines, alcohols, and methoxyterminated polyethylene glycol chains) and narrow size distributions. The click modification on different sized silica nanoparticles was also investigated. The surface modification was characterized by FT-IR, TGA and DLS. The DLS of surface modified nanoparticles in water confirmed that the nanoparticles did not aggregate. The colloidal stability studies of modified nanoparticles in cell culture media showed the different dispersion behavior depending on the surface functionalities, but further study is needed to evaluate the interactions between the media and the modified SiNPs. This surface modification of silica nanoparticles provides one facile and general method to produce engineered nanoparticles with specific surface functionalities. 5.4 Experimental Section Materials. LUDOX TM-40 colloidal silica (25 nm, SiNP25, 40 wt% suspension in H2O), 3-(triethoxysilyl)propyl isocyanate, propargyl alcohol, tetraethylene glycol monomethyl ether, poly(ethylene glycol) monomethyl ether (Mn = 750, 2000, 5000),ethyl bromoacetate, tosyl chloride, sodium ascorbate and sodium azide were 196 purchased from Sigma-Aldrich and used as received. Colloidal silica SNOWTEX S (8 nm, SiNP8, 40 wt% suspension in H2O), SNOWTEX 20L (45 nm, SiNP45, 20 wt% suspension in H2O), and SNOWTEX ZL (70 nm, SiNP70, 40 wt% suspension in H2O) were received as gifts from Nissan Chemical Industries, Ltd. [2-[2-[2- chloroethoxy]ethoxy]ethanol (80) was purchased from TCI America. Other reagents and ® solvents were used as received. Dialysis membrane (Spectra/Por Biotech Cellulose ester, MWCO: 300K, 50K, 12-14K) was purchased from Spectrum Laboratories Inc. Ion exchange resin (Amberlite ® IRC-748) was purchased from Alfa Aesar. 1 Instruments and Characterization. H NMR and 13 C NMR spectra were recorded in CDCl3 on a VXR-500 MHz instrument. The CDCl3 resonance was used as the internal standard for 13 1 C NMR ( = 77.0 ppm) and residual CHCl3 for H NMR ( = 7.24 ppm). Fourier transform infrared (FT-IR) spectra were recorded on a Mattson Galaxy series FT-IR 3000. Thermogravimetric analyses (TGA) were obtained in air from a Perkin-Elmer TGA 7. Samples were held at 120 °C for 30 min to remove absorbed water from particle surfaces, and then heated to 850 °C at a rate of 10 °C/min. All samples for FT-IR and TGA were dried from water solution and dried under vacuum at room temperature for 24 h. Dynamic Light Scattering (DLS) was performed in a Malvern NanoZS ZetaSizer with a 178 degree backscattering detection. Intensity average diameters were calculated from the autocorrelation function using Malvern’s Zetasizer Software 6.12. Nanoparticles dispersion stability in media was measured by adding particles to fresh 197 media and vortex them, then the DLS was taken at certain time. Synthesis of ((2-propynylcarbamate)propyl)triethoxysilane (74). Propargyl alcohol (12.0 g, 210 mmol) and triethylamine (19.0 g, 190 mmol) were dissolved in 120 mL of dry CH2Cl2 and cooled in an ice bath. A solution of 3-(triethoxysilyl)propyl isocyanate (73, 46.0 g, 180 mmol) in 60 mL of dry CH2Cl2, was added dropwise under nitrogen. The reaction mixture was stirred for 24 h at room temperature and then filtered. CH2Cl2, excess propargyl alcohol, and NEt3 were evaporated under reduced pressure to afford 50 g of 74 as colorless oil. 1 H NMR (500 MHz, CDCl3):  4.55 (s, 2H), 3.71 (q, 6H, J = 7 Hz), 3.08 (m, 2H), 2.37 (s, 1H), 1.52 (m, 2H), 1.11 (t, 9H, J = 8 Hz), 0.52 (t, 2H, J = 8Hz); 13 C NMR (125 MHz, CDCl3) 155.4, 78.40, 74.40, 58.46, 52.26, 43.47,23.14, 18.25, 7.56. IR 3310, 2981, 2936, 2891, 1718, 1536, 1443, 1249, -1 1107, 1079, 957 cm . Synthesis of 1-azido-2-(2-(2-(2-methoxyethoxy)ethoxy)ethoxyethane (76a). A solution of NaOH (6.00 g, 150 mmol) in a water/THF mixture (6:4 , 30 mL) was added to tetraethylene glycol monomethyl ether (5.2 g, 25 mmol) in 50 mL of THF dropwise at 0 C. The mixture was stirred for 30 minute and then tosyl chloride (9.5 g, 50 mmol) in 50 mL of THF was added dropwise. The mixture was stirred at 0 C for 3 hours and then for 6 hours at room temperature. The mixture was poured into ice water and the water layer was extracted with ether (3 x 50 mL). The combined organic layer was washed with saturated aqueous NaCl, dried over Na2SO4 and concentrated. The resulted tosylate was used without further purification (8.0 g, 91 %). The prepared tosylate (8 g, 198 22 mmol), NaI (1 g) and sodium azide (4.8 g, 70 mmol) were dissolved in 50 ml of DMF and heated at 60 C for 24 h. The solution was cooled and added to 100 mL of water. The solution was extracted with ether (4 x 50 mL) and the combined organic layer was washed with saturated NaCl (2 x 100 ml) and dried over Na2SO4. The ether was 1 evaporated to give product (76a) as light yellow oil, 4.3 g, 85 %. H NMR (500 MHz, CDCl3):  3.69–3.64 (m, 12 H), 3.52 (m, 2 H), 3.36-3.34 (m, 5H); 13 C NMR (75 MHz, CDCl3):  72.87, 70.75, 70.72, 70.66, 70.63, 70.50, 70.02, 59.05, 50.7; IR 2928, 2874, -1 2094, 1441, 1329, 1245, 1130, 956, 852 cm . Synthesis of mPEG azides (76b-d). Polyethylene glycol monomethyl ether (25 mmol) and TsCl (14 g, 75 mmol) were dissolved in 150 ml of dry CH2Cl2. Then 28 mL of triethylamine was added dropwise. The mixture was stirred for 48 h at room temperature. After washed with NaHCO3 (100 mL), 1 M HCl (3 x 200 mL) and water (200 mL), CH2Cl2 layer was dried over Na2SO4 and concentrated. The resulted tosylate was used directly by mixed with NaI (1 g) and sodium azide (6.5 g, 100 mmol) in 150 mL of DMF and the suspension was heated at 90 C for 24 h. The solution was cooled and concentrated to remove DMF under vacuum. After dissolved in 100 mL of CH2Cl2 (caution: be careful for the possible explosive risk from CH2Cl2 and NaN3), the product was washed with water, dried and concentrated to give light yellow liquid or white solid. 1 76b: light yellow liquid, 17 g, 90%. H NMR (500 MHz, CDCl3):  3.56–3.75 (m, 70 H), 3.50-3.54 (m, 2 H), 3.34-3.38 (m, 5H); 13 C NMR (75 MHz, CDCl3):  72.44, 71.85, 199 70.61-70.36 (br), 70.13, 69.94, 61.53, 58.94, 50.61; IR 2872, 2111, 1460, 1352, 1300, -1 1250, 1118, 953, 853 cm . 1 76c: white solid, 37 g, 75%. H NMR (500 MHz, CDCl3):  3.45–3.77 (m, 200 H), 3.32-3.38 (m, 5H); 13 C NMR (75 MHz, CDCl3):  71.88, 70.60-70.45 (br), 69.97, 58.98, -1 50.67; IR 2889, 2106, 1468, 1361, 1278, 1146, 1115, 1063, 950, 842 cm . 1 76d: white solid, 87 g, 70%. H NMR (500 MHz, CDCl3):  3.48–3.77 (m, 520 H), 3.35-3.39 (m, 5H); 13 C NMR (75 MHz, CDCl3):  71.91, 70.68-70.30 (br), 70.00, 59.00, -1 50.67; IR 2889, 2100, 1471, 1358, 1344, 1278, 1148, 1112, 1066, 950, 845 cm . Synthesis of 2-[2-[2-[2-azidoethoxy]ethoxy]ethoxy]ethanol (78). A solution of TsCl (14 g, 71 mmol) in 250 mL of THF was added dropwise to a solution of tetraethylene glycol (34 g, 180 mmol), triethylamine (14 mL) and THF (250 mL). After stirring the resulting solution for 36 h at room temperature, THF was removed by rotary evaporation and the crude product was dissolved in 200 mL of CH2Cl2. The organic layer was washed with water (3 × 150 mL), aqueous NaHCO3 (200 mL), and brine (200 mL). After drying over Na2SO4, the solution was concentrated and dried under vacuum to give 18 g of a light yellow liquid comprised of a mixture of mono-tosylate (85%) and di-tosylate (15%), 58% yield. The product was used in the next step without further 1 purification. H NMR (500 MHz, CDCl3) δ 7.77 (m, 2H), 7.32 (m, 2H), 4.29 (m, 2H), 3.69-3.50 (m, 14H), 2.5 (br, 1H), 2.42 (s, 3H). The tosylates were dissolved in DMF, (210 mL), treated with NaN3 (10.4 g, 15 mmol), and then heated to 60 °C for 24 h. Then 200 the mixture was concentrated, re-dissolved in ethyl acetate, filtered to remove the solid, and concentrated under reduced pressure. The resulting oily product was dissolved in water and washed with hexane until ,-diazido tetraethylene glycol was not detected in the aqueous layer. Drying under vacuum gave the product as a light yellow liquid (8.0 g, 72%). 1 H NMR (500 MHz, CDCl3) δ 3.69 (m, 2H), 3.64 (m, 10H), 3.58 (m, 2H), 3.36 (t, 2H, J = 5.5 Hz), 2.8 (br, 1H); 13 C NMR (125 MHz, CDCl3) 72.41, 70.55, 70.51, 70.43, -1 70.18, 69.90, 61.56, 50.54; IR 3412, 2872, 2106, 1457, 1350, 1284, 1123, 944 cm . Synthesis of 2-[2-[2-[2-azidoethoxy]ethoxy]ethoxy]ethyl amine (79). A solution of TsCl (39.9 g, 210 mol) in THF (150 mL) was added dropwise into an ice-cold mixture of tetraethylene glycol (13.5 g, 70 mol), aqueous KOH (100 g, 47%, w/w), and THF (100 mL). After stirring for 36 h at room temperature, THF was removed by rotary evaporation, and then CH2Cl2 (200 mL) was added to the residue. The organic layer was washed with water (3 × 150 mL), aqueous NaHCO3 (200 mL), and brine (200 mL). After drying the organic layer over Na2SO4, the solution was concentrated and dried 1 under vacuum to give 32 g of a light yellow liquid (91%). H NMR (300 MHz, CDCl3) δ 7.81(m, 4H), 7.36 (m, 4H), 4.20 (m, 4H), 3.70-3.65 (m, 12H), 2.47 (s, 6H). The ditosylate of tetraethylene glycol was treated with NaN3 (11.7 g, 180 mmol) in DMF (300 mL) at 90 °C for 24 h. The mixture was concentrated, dissolved in water, and extracted with CH2Cl2 (4 × 100 mL). The solution was dried over anhydrous Na 2SO4 and concentrated under reduced pressure to give 12 g of a yellow liquid. Aqueous HCl (5%, 201 100 mL) was added to the yellow liquid, and under vigorous stirring at room temperature, a solution of PPh3 (11.4 g, 41.0 mmol) in 75 mL of ether was added dropwise. After stirring for 24 h, the ether was removed by rotary evaporation and the aqueous layer was extracted with CH2Cl2 until triphenylphosphine oxide was not detected in the aqueous layer. The aqueous layer was adjusted to pH=12, and amine (79) was extracted from the aqueous layer with CH2Cl2 (6 × 50 mL). Drying under 1 vacuum gave the product as a light yellow liquid (9.0 g, 68%). H NMR (500 MHz, CDCl3) δ 3.72-3.58 (m, 10H), 3.48 (t, 2H, J = 5 Hz), 3.35 (t, 2H, J = 5 Hz), 2.84 (t, 2H, J = 5Hz), 2.24 (s, 2H); 13 C NMR (125 MHz, CDCl3) 72.79, 70.50, 70.47, 70.43, 70.07, 69.85, 50.50, 41.44; IR 3350, 2872, 2111, 1589, 1460, 1347, 1297, 1123, 939, 859 cm 1 - . Synthesis of 2-[2-[2-[2-azidoethoxy]ethoxy]ethoxy]acetic acid (83). A solution of [2-[2-[2-chloroethoxy]ethoxy]ethanol (80, 16.8 g, 100 mmol) and NaN3 (13 g, 200 mmol) in 75 mL of water was heated at 75 °C with stirring for 24 h. Water was removed by rotary evaporation and the crude product was extracted from the salts with 200 mL of CH2Cl2. Concentration and drying under vacuum gave the product as a light yellow 1 liquid (17 g, 97%), which was used without further purification. H NMR (500 MHz, CDCl3) δ 3.72-3.62 (m, 8H), 3.58 (m, 2H), 3.37 (t, 2H, J = 5 Hz). The light yellow liquid product (14 g, 80 mmol) was added to NaH (100 mol) in 350 mL of dry THF cooled in an 202 ice bath. After stirring for 1 h at room temperature, ethyl bromoacetate (17.7 mL, 160 mmol) was added dropwise and then stirred for 48 h at room temperature. The THF was removed, and 150 mL of water was added. The aqueous layer was extracted with CH2Cl2 (3 × 50 mL), dried over anhydrous Na2SO4, and concentrated under reduced pressure at 55 °C to give an amber liquid (26 g), which was used without further purification. 1 H NMR (500 MHz, CDCl3) δ 4.24 (q, 2H, J = 10 Hz), 4.17 (s, 2H), 3.77- 3.65 (m, 10H), 3.41 (t, 2H, J = 5 Hz). The ester was hydrolyzed with 53 mL of 3M NaOH at room temperature for 18 h. After extracting the aqueous layer with CH 2Cl2 (10 × 30 mL) to remove mineral oil and other impurities, the aqueous layer was acidified to pH=1 and extracted with EtOAc (1 × 80 mL, 7 × 30 mL). After dred over anhydrous Na2SO4, 1 the residue was concentrated to a pale red liquid (14 g, 75%). H NMR (500 MHz, CDCl3) δ 4.18 (s, 2H), 3.78-3.67 (m, 10H), 3.41 (t, 2H, J = 5 Hz); 13 C NMR (125 MHz, CDCl3) 177.88, 76.71, 75.95, 75.69, 75.48, 75.31, 74.02, 55.9; IR 3120, 2927, 2853, -1 2536, 2111, 1741, 1443, 1354, 1289, 1123, 936, 853 cm . Surface modification of SNPs Synthesis of alkynyl SiNP. commercial available SiNPs (9.0 g of 40% or 20% suspension in water) were diluted with 37.5 mL of water followed by addition of 37.5 ml of ethanol and sonicated. The pH of the silica suspension was adjusted to ~9 by adding drops of aqueous ammonia followed by the addition of a solution of 74 (3.0 g, 10 mmol) in 10 mL of ethanol dropwise. After stirring for 3 days, the suspension was diluted with EtOH until it was homogeneous. The modified particles were purified by dialysis in 203 EtOH and water (9:1) using dialysis tubing (SiNP8, MWCO: 50 K; other SiNPs, MWCO: 300K). The surface coverage of the alkyne functional group was evaluated by TGA and FT-IR. The hydrodynamic diameters were measured by DLS. The particle concentration (mg/g) was determined by evaporating 2 - 4 mL of purified particle solutions. Click surface modification of SiNPs. Click chemistry between alkynyl SiNPs and clicking reagents is described using the synthesis of SiNP25/76a as example. A SiNP25 suspension in 13.8 g of EtOH/water (9:1) (100 mg of SiNP25, 0.036 mmol of alkyne functional groups) was diluted with 30 mL of DMF/water (3:1), then 76a (41.9 mg, 0.180 mmol, 5 equiv.) and sodium ascorbate (14.3 mg, 2 equiv.) were added to the mixture. After trace O2 was removed by three freeze-pump-thaw cycles, aqueous CuSO4 (72 μL, 0.10 M, 0.2 equiv.) was added to the solution followed by a final degassing process, and the solution changed to light brown-yellow. After stirring at room temperature for 36 h, 1 mL of aqueous EDTA (0.1M) was added and the solution was stirred for another 6 h. The resulting particles were dialyzed against (MWCO:12-14K) EtOH/water (9:1, 5 × 2 L) and against distilled water to afford purified surface-modified SiNPs with different sizes and surface chemistry. For the synthesis of -NH2 modified SiNP25/79, 1 mL of aqueous EDTA (0.1M) was not added after click chemistry, instead, it was dialyzed directly as mentioned above followed by sonication of the SiNP25/79 suspension to be homogeneous. Then ion exchange resin (2 g, Amberlite remove Cu 2+ ® IRC-748) was added to under gentle stirring, and this resin was removed via filtration through cotton after stirring for 8 hours. All modified SiNPs were kept in deionized water until further characterization. The surface coverage of surface functional groups after each 204 step modification was evaluated by TGA. DLS samples were prepared by diluting solutions to 0.1 or 1 mg/mL. The particle concentration (mg/g) was determined by evaporating 2-4 mL of purified particle solutions. . 205 APPENDIX 206 1 Figure A 1. H and 13 C NMR spectra of propargyl glycolic acid (1) in CDCl3. 207 1 Figure A 2. H and 13 C NMR spectra of propargyloxy lactic acid (12) in CDCl3. 208 1 Figure A 3. H and 13 C NMR spectra of poly(ethylene glycol) monopropargy lether lactic acid (23, n =1) in CDCl3. 209 1 Figure A 4. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (23, n =2) in CDCl3. 210 1 Figure A 5. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (23, n =3) in CDCl3. 211 1 Figure A 6. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (23, n =4) in CDCl3. 212 1 Figure A 7. H and 13 C NMR spectra of tetraethylene glycol monobenzyl ether (29) in CDCl3. 213 1 Figure A 8. H and 13 C NMR spectra of monotetrahydropyran-protected tetraethylene glycol monotosylate (35) in CDCl3. 214 1 Figure A 9. H and 13 C NMR spectra of octa(ethylene glycol) monobenzyl ether (36) in CDCl3. 215 1 Figure A 10. H and 13 C NMR spectra of 38 in CDCl3. 216 ppm 1 Figure A 11. H and 13 C NMR spectra of octa(ethylene glycol) monopropargyl ether lactic acid (40) in CDCl3. 217 1 Figure A 12. H and 13 C NMR spectra of poly(ethylene glycol) monopropargyl ether lactic acid (50) in CDCl3. 218 1 Figure A 13. H and 13 C NMR spectra of allyloxy lactic acid (54) in CDCl3. 219 1 Figure A 14. H and 13 C NMR spectra of azido lactic acid (56) in DMSO-d6. 220 ppm 1 Figure A 15. H and 13 C NMR spectra of hetero-monomer (58) in CDCl3. 221 1 Figure A 16. H and 13 C NMR spectra of homo-monomer (59) in CDCl3. 222 ppm 1 Figure A 17. H NMR spectra of polymer P58A in CDCl3. ppm 1 Figure A 18. H NMR spectra of polymer P59A in CDCl3. 223 ppm 1 Figure A 19. H NMR spectra of polymer P58B in CDCl3. ppm 1 Figure A 20. H NMR spectra of polymer P59B in CDCl3. 224 ppm 1 Figure A 21. H NMR spectra of polymer P59C in CDCl3. 225 ppm 1 Figure A 22. H NMR spectra of polymer P59D in CDCl3. 226 Figure A 23. Water contact angle of glass substrate (a); P58 film (b); P58B film (c); P59 film (d); P59B film (e); P59C film (f); P59D film (g); P60 film (h). 227 100 nm Figure A 24. TEM of UCMP58B. Scale bar: 100 nm. 50 nm Figure A 25. TEM of UCMP59B Scale bar: 50 nm. 228 100 nm Figure A 26. TEM of UCMP59C Scale bar: 100 nm. 50 nm Figure A 27. TEM of CMP59C from cross-linker 63. Scale bar: 50 nm. 229 50 nm Figure A 28. TEM of CM*P59C from cross-linker 64. Scale bar: 50 nm. 200 nm Figure A 29. TEM of UCMP59D. Scale bar: 200 nm. 230 100 nm Figure A 30. TEM of CMP59D from cross-linker 63. Scale bar: 100 nm. 200 nm Figure A 31. TEM of CM*P59D from cross-linker 64. Scale bar: 200 nm. 231 ppm ppm 1 Figure A 32. H and 13 C NMR spectra of monomer (72a) in CDCl3. 232 ppm ppm 1 Figure A 33. H and 13 C NMR spectra of monomer (72b) in CDCl3. 233 ppm ppm 1 Figure A 34. H and 13 C NMR spectra of monomer (72c) in CDCl3. 234 1 Figure A 35. H NMR spectra of polymer (P72a) in CDCl3. 1 Figure A 36. H NMR spectra of polymer (P72b) in CDCl3. 235 1 Figure A 37. H NMR spectra of polymer (P72c) in CDCl3. 236 Figure A 38. TEM images of SiNP25 (a); SiNP25/74 (b); SiNP25/76a (c); SiNP25/78 (d); SiNP25/83 (e); SiNP25/79 (f). 237 REFERENCES 238 REFERENCES 1. Datta, R.; Henry, M. J. Chem. Technol. Biotechnol. 2006, 81, 1119-1129. 2. Dechy-Cabaret, O.; Martin-Vaca, B.; Bourissou, D. Chem. Rev. 2004, 104, 6147– 6176. 3. Drumright, R. E.; Gruber, P. R.; Henton, D. E. Adv. Mater. 2000, 12, 1841–1846. 4. Sinclair, R.G. J. Macromol. Sci., Pure Appl. Chem. 1996, 33, 585–597. 5. Prokop, A.; Jubel, A.; Helling, H. J.; Eibach, T.; Peters, C.; Baldus, S. E.; Rehm, K. E. Biomaterials 2004, 25, 259–267. 6. Claes, L. E.; Ignatius, A. A.; Rehm, K. E.; Scholz, C. Biomaterials 1996, 17, 1621–1626. 7. Meyer, D. C.; Fucentese, S. F.; Ruffleux, K.; Jacob, H. A. C.; Gerber, C. Arthroscopy 2003, 19, 188-193. 8. Albertsson, A. C.; Varma, I. K. Biomacromolecules 2003, 4, 1466-1486. 9. Pan, Z.; Ding, J. Interface doi:10.1098/rsfs.2011.0123 2042-8901. 10. Vij, N.; Min, T.; Marasigan, R.; Belcher, C. N.; Mazur, S.; Ding, H.; Yong, K-T.; Roy, I. J. Nanobiotechnol. 2010, 8:22. 11. Cao, Z.; Yu, Q.; Xue, H.; Cheng, G.; Jiang, S. Angew. Chem. Int. Ed., 2010, 49, 3771–3776. 12. Acharya, S.; Sahoo, S. K. Adv. Drug Deliver. Rev. 2011, 63, 170-183. 13. Khanna, A.; Sudha, S. S.; Pillai, S.; Rath, S. S. J. Mol. Model. 2008, 16, 367-374. 14. Chen, G.; Kim, H.; Kim, E.; Yoon, J. Eur. Polym. J. 2006, 42, 468-472. 15. Moon, S. II; Lee, C. W.; Miyamoto, M.; Kimura, Y. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 1673-1679. 16. Marques, D. S.; Gil, M. H.; Baptista, C. M. S. G. J. Appl. Polym. Sci. 2012, 128, 2145–2151. 17. Ryner, M.; Stridsberg, K.; Albertsson, A.-C.; von Schenk, H.; Svensson, M. Macromolecules 2001, 34, 3877-3881. 239 Focus 2012, rsfs20110123; 18. Kowalski, A.; Duda, A.; Penczek, S. Macromol. Rapid Commun. 1998, 19, 567572. 19. Kricheldorf, H. R.; Kreiser-Saunders, I.; Stricker, A. Macromolecules 2000, 33, 702-709. 20. Nederberg, F.; Connor, E. F.; Moller, M.; Glauser, T.; Hedrick, J. L. Angew. Chem. Int. Ed. 2001, 40, 2712-2715. 21. Blakey, I.; Yu, G.; Howdle, S. M.; Whittaker, A. K.; Thurech, K. J. Green Chem. 2011, 13, 2032–2037. 22. Pratt, R. C.; Lohmeijer, B. G.G.; Long, D. A.; Waymouth, R. M.; Hedrick, J. L. J. Am. Chem. Soc. 2006, 128, 4556-4557. 23. Coady, D. J.; Engler, A. C.; Horn, H. W.; Bajjuri, K. M.; Fukushima, K.; Jones, G. O.; Nelson, A.; Rice, J. E.; Hedrick, J. L. ACS Macro Lett. 2012, 1, 19−22. 24. Nederberg, F.; Connor, E. F.; Glausser, T.; Hedrick, J. L. Chem. Commun. 2001, 2066-2067. 25. Grasa, G. A.; Kissling, R. M.; Nolan, S. P. Org. Lett. 2002, 4, 3583–3586. 26. Connor, E. F.; Nyce, G. W.; Myers, M.; Mock, A.; Hedrick, J. L. J. Am. Chem. Soc 2002, 124, 914-915. 27. Nyce, G. W.; Glauser, T.; Connor,. E. F.; Mock, A.; Waymouth, R. M.; Hedrick, J. L. J. Am. Chem. Soc. 2003, 125, 3046–3056. 28. Myers, M.; Connor, E. F.; Glauser, T.; Mock, A.; Nyce, G.; Hedrick, J. L. J. Polym. Sci., Part A: Polym. Chem. 2002, 40, 844-851. 29. Dove, A. P.; Pratt, R. C.; Lohmeijer, B. G.; Waymouth, R. M.; Hedrick, J. L. J. Am. Chem. Soc. 2005, 127, 13798 -13799. 30. Kricheldorf, H. R.; Dunsing, R. Makromol. Chem. 1986, 187, 1611-1625. 31. Kricheldorf, H. R.; Kreiser, I. Makromol. Chem. 1987,188, 1861-1873. 32. Bhaw-Luximon, A.; Jhurry, D.; Spassky, N.; Pensec, S.; Belleney, J. Polymer 2001, 42, 9651-9656. 33. Xie, W.; Chen, D.; Fan, X.; Li, J.; Wang, P. G.; Cheng, H. N.; Nickol, R. G. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 3486-3491. 34. Kricheldorf, H. R.; Lee, S.R. Polymer 1995, 36, 2995-3003. 35. Padovani, M.; Hilker, I.; Duxbury, c. J.; Heise, A. Macromolecules 2008, 41, 240 2439-2444. 36. Matsumura, S.; Mabuchi, K.; Toshima, K. Macromol. Rapid Commun. 1997, 18, 477-482. 37. Matsumura, S.; Tsukada, K.; Toshima, K. Int. J. Biol. Macromol. 1999, 25, 161167. 38. Kobayashi, S.; Uyama, H. Macromol. Symp. 1999, 144, 237-246. 39. Parrish,; B.; Breitenkamp, R. B.; Emrick, T. J. Am. Chem. Soc. 2005, 127, 74047410. 40. Li, S.; Vert, M. Macromolecules, 2003, 36, 8008–8014. 41. Yang, L.; El Ghzaoui, A.; Li, S. Int. J. Pharm. 2010, 96-103. 42. Ohya, Y.; Maruhashi, S.; Ouchi, T. Macromolecules 1998, 31, 4662-4665. 43. Ouchi, T.; Uchida, T.; Ohya, Y. Macromol. Biosci. 2001, 1, 371-375. 44. Stassen, S.; Archambeau, S.; Dobois, Ph.; Jerome, R.; Teyssie, Ph. J. Polym. Sci. Part A:  Polym. Chem. 1994, 32, 2443-2455. 45. Tessmar, J. K.; Mikos, A. G.; Göpferich, A. Biomacromolecules 2002, 3, 194-200. 46. Nguyen, C. A.; Allémann, A.; Schwach, G.; Doelker, E.; Gurny, R. Eur. J. Pharm. Sci. 2003, 20, 217–222. 47. Kang, E-H.; Lee, I-H.; Choi, T-L. ACS Macro Lett. 2012, 1, 1098-1102. 48. Liu, X.; Li, K.;Feng, X.; Cao, Y. J. Macromol. Sci., Pure Appl. Chem.2009, 46, 937-942. 49. Basko, M.; Bednarek, M. React. Funct. Polym. 2012, 72, 213-220. 50. Lav, T.; Lemechko, P.; Renard, E.; Amiel, C.; Langlois, V.; Volet, G. React. Funct. Polym. 2013, 73, 1001-1008. 51. Molla, M. R.; Ghosh, S. Macromolecules 2012, 45, 8561-8570. 52. Barker, I. A.; Hall, D. J.; Hansell, C. F.; Du Prez, F.; O’Reilly, R. K.; Dove, A. P. Macromol. Rapid Commu. 2011, 32, 1362–1366. 53. Hwang, J. J.; Iyer, S. N.; Li, L. S.; Claussen, R.; Harrington, D. A.; Stupp, S. I. PNAS 2002, 99, 9662-9667. 54. Klok, H.-A.; Hwang, J. J.; Hartgerink, J. D.; Stupp, S. I. Macromolecules 2002, 241 35, 6101-6111. 55. Stendahla, J. C.; Lia, L.; Claussen, R.; Stupp, S. I. Biomaterials 2004, 25, 5847– 5856. 56. Klok, H. A.; Hwang, J. J.; Iyer, S. N.; Stupp, S. I. Macromolecules 2002, 35, 746759. 57. Baker, G. L.; Vogel, E.B.; Smith III, M. R. Polym. Rev. 2008, 48, 64–84. 58. Yin, M.; Baker, G. L. Macromolecules 1999, 32, 7711–7718. 59. Simmons, T. L.; Baker, G. L. Biomacromolecules 2001, 2, 658–663. 60. Liu, T. Q.; Simmons, T. L.; Bohnsack, D. A.; Mackay, M. E.; Smith III, M. R.; Baker, G. L. Macromolecules 2007, 40, 6040–6047. 61. Jing, F.; Smith III, M. R.; Baker, G. L. Macromolecules 2007, 40, 9304-9312. 62. Gerhardt, W. E.; Noga, D. E.; Hardcastle, K. I.; García, A. J.; Collard, D. M.; Weck, M. Biomacromolecules 2006, 7, 1735-1742. 63 . Leemhuis, M.; van Nostrum, C. F.; Kruijtzer, J. A. W.; Zhong, Z. Y.; Breteler, M. R.; Dijkstra, P. J.; Feijen, J.; Hennink, W. E. Macromolecules 2006, 39, 35003508. 64. Noga, D. E.; Petrie, T. A.; Kumar, A.; Weck, M.; García, A. J.; Collard, D. M. Biomacromolecules 2008, 9, 2056–2062. 65. Ghassemi, A. H.; van Steenbergen, M. J.; Talsma, H.; van Nostrum, C. F.; Crommelin, D. J. A.; Hennink, W. E. Pharm. Res. 2010, 27, 2008-2017. 66. Lu, Y.; Yin, L.; Zhang, Y.; Zhang, Z.; Xu, Y.; Tong, R.; Cheng. J. ACS Macro Letters 2012, 1, 441-444. 67. Taguchi, K. Y., S.; Hiratani, K.; Norikio, M.; Okahata, Y. Macromolecules 1988, 21, 3338-3340. 68. Ouchi, T.; Fujino, A. Makromol. Chem. 1989, 190, 1523-1530. 69. Porjazoska, A.; Dimitrov, P.; Dimitrov, I.; Cvetkovska, M.; Tsvetanov, C. B. Macromol. Symp. 2004, 210, 427-436. 70. Lutz, J. F.; Hoth, A. Macromolecules 2006, 39, 893-896. 71. Jiang, X.; Smith III, M. R.; Baker, G. L. Macromolecules 2008, 41, 318-324. 72. Jiang, X.; Vogel, E. B.; Smith, M. R.; Baker, G. L. J. Polym. Sci. A Polym. 242 Chem.2007, 45, 5227–5236. 73. Castillo, J. A.; Borchmann, D. E.; Cheng, A. Y.; Wang, Y.; Hu, C.; García, A. J.; Weck, M. Macromolecules 2012, 45, 62-69. 74. Jing, F.; Hillmyer, M. A. J. Am. Chem. Soc. 2008, 130, 13826-13827. 75. Leemhuis, M.; Akeroyd, N.; Kruijtzer, J. A. W.; van Nostrum, C. F.; Crommelin, D. J. A.; Hennink, W. E. Eur. Polym. J. 2008, 44, 308-317. 76. Zou, J.; Hew, C. C.; Themistou, E.; Li, Y.; Chen, C.-K.; Alexandridis, P.; Cheng, C. Adv. Mater. 2011, 23, 4274–4277. 77. Jiang, X.; Vogel, E. B.; Smith III, M. R.; Baker, G. L. Macromolecules 2008, 41, 1937-1944. 78. Rubinshtein, M.; James, C. R.; Young, J. L.; Ma,Y. J.; Kobayashi, Y.; Gianneschi, N. C.; Yang, J. Org. Lett. 2010, 12, 3560–3563. 79. Gutiérrez-Villarreal, M. H.; Guzmán-Moreno, J. G.; J. Polym. Res. 2013, 20:149. 80. Carlisle, E. S.; Mariappan, M. R.; Nelson, K. D.; Thomes, B. E.; Timmons, R. B.; Constantinescu, A.; Eberhart, R. C.; Bankey, P. E. Tissue Eng. 2000, 6, 45-52. 81. Edlund, U.; Källrot, M.; Albertsson, A. C. J. Am. Chem. Soc. 2005, 127, 88658871. 82. Källrot, M.; Edlund, U.; Albertsson, A. C.Biomacromolecules 2007, 8, 2492-2496. 83. Pan, J.; Wang, Y. L.; Qin, S. H.; Zhang, B. B.; Luo, Y. F. J. Biomed. Mater. Res. Part B 2005, 74B, 476-480. 84. Porter, K.; Hossain, M.; Wang, M.; Radano, C. P.; Baker, G.; Smith, M. R.; McCabe, L. R. Mol. Biol. Rep. 2006, 33, 1-12. 85. du Boullay, O. T.; Bonduelle, C.; Martin-Vaca, B.; Bourissou, D. Chem. Commun. 2008, 1786-1788. 86. Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem. Int. Ed. 2001, 40, 2004–2021. 87. Ott, C.; Hoogenboom, R.; Schubert, U. S. Chem. Commun. 2008, 3516–3518. 88. Becer, C. R.; Babiuch, K.; Pilz, K.; Hornig, S.; Heinze, T.; Gottschaldt, M.; Schubert, U. S. Macromolecules 2009, 42, 2387–2394. 89. Killops, K. L.; Campos, L. M.; Hawker, C. J. J. Am. Chem. Soc. 2008, 130, 50625064. 243 90. Witczak, Z. J.; Lorchak, D.; Nguyen, N.; Carbohydr. Res. 2007, 342, 1929–1933. 91. Stanford,, M. J.; Dove, A. P.Macromolecules 2009, 42, 141–147. 92. Dag, A.; Durmaz, H.; Hizal, G.; Tunca, U. J. Polym. Sci. Part A Polym. Chem. 2008, 46, 302–313. 93. Williams, R. J.; Barker, I. A.; O’Reilly, R. K.; Dove, A. P. ACS Macro Lett. 2012, 1, 1285−1290. 94. Chen, C.-K.; Law, W.-C.; Aalinkeel, R.; Nair, B.; Kopwitthaya, A.; Mahajan, S. D.; Reynolds, J. L.; Zou, J.; Schwartz, S. A.; Prasad, P. N.; Cheng, C. Adv. Healthcare Mater. 2012, 1, 751-761. 95. Jones, C. H.; Chen, C. K.; Jiang, M.; Fang, L.; Cheng, C.; Pfeifer, B. A. Mol. Pharm. 2013, 4, 1138-1145. 96. Zhang, Z.; Yin, L.; Tu, C.; Song, Z.; Zhang, Y.; Xu, Y.; Tong, R.; Zhou, Q.; Ren, J.; Cheng, J. ACS Macro Lett. 2013, 2, 40–44. 97. Zhang, Z.; Yin, L.; Xu, Y.; Tong, R.; Lu, Y.; Ren, J.; Cheng, J. Biomacromolecules 2012, 13, 3456−3462. 98. Dimitrov, I.; Trzebicka, B.; M ller, A. H. E.; Dworak, A.; Tsvetanov, C. B. Prog. Polym. Sci. 2007, 32, 1275–1343. 99. Nalluri, S. K. M.; Voskuhl, J.; Bultema, J. B.; Boekema, E. J.; Ravoo, B. J. Angew. Chem. Int. Ed. 2011, 50, 1–6. 100. Akhoury, A.; Bromberg, L.; Hatton, T. L. ACS Appl. Mater. Interfaces 2011, 3, 1167–1174. 101. Jeong, B.; Bae, Y. H.; Lee, D. S.; Kim, S. W. Nature 1997, 388, 860–862. 102. Cunningham, A.; Oh, J. K. Macromol. Rapid Commun. 2013, 34, 163–168. 103. Fujishige, S.; Kubota, K.; Ando, I. J. Phys. Chem. 1989, 93, 3311–3313 104. Chen, G. H.; Hoffman, A. S. Nature 1995, 373, 49–52. 105. Chen, S.; Pieper, R.; Webster, D. C.; SinghInt. J. Int. J. Pharm. 2005, 288, 207– 218. 106. Feil, H.; Bae, Y. H.; Feijen, J.; Kim, S. W. Macromolecules 1993, 26, 2496– 2500. 107. Verma, A.; Simard, J. M.; Worrall, J. W. E.; Rotello, V. M. J. Am. Chem. Soc. 2004, 126, 13987-13991 244 108. Fiore, G. L.; Jing, F.; Young, Jr., V. G.; Cramer, C. J.; Hillmyer, M. A. Polym. Chem. 2010, 1, 870-877. 109. The routes shown in Scheme 30 are unpublished and will not be presented in this thesis. 110. Downs, A. J. Chemistry of Aluminium, Gallium, Indium and Thallium, 1st ed., 1993,pp. 372 Springer. 111. Murray, T. F.; Samsel, E. G.; Varma, V.; Norton, J. R. J. Am. Chem. Soc. 1981, 103, 7520–7528. 112. Buser, D.; Pauling, H.; Thum, A.; Bonrath, W. Molecules 2002, 7, 341-352. 113. Holmes, A. B.; Sporikou, C. N. Org. Synth.1987, 65, 61-65. 114. Suzuki, T.; Saimoto, H.; tomioka, H.; Oshima, K.; Nozaki, H. Tetrahedron Lett. 1982, 23, 3597-3600. 115. Roush, W. R.; Blizzard, T. A.; Basha, F. Tetrahedron Lett. 1982, 23, 2331-2334. 116. Akita, H.; Matsukura, H.; Oishi, T. Tetrahedron Lett. 1986, 27, 5397-5400. 117. Larcheveque, M.; Peit, Y. Bul.l Soc. Chim. Fr. 1986, 130-139. 118. Mole, T,; Surtee, J. R. Chem. Ind. 1963, 43, 1727-1728. 119. Srinivasan, R.; Uttamchandani, M.; Shao Q. Yao, S. Q. Org. Lett. 2006, 8, 713716. 120. Brea, R. J.; López-Deber, M. P.; Castedo, L.; Granja, J. R. J. Org. Chem. 2006, 71, 7870-7873. 121. Dragovich, P. S.; Prins, T.J.; Zhou, R. et al. J. Med. Chem. 2003, 46, 4572–4585. 122. Feifel, S. C.; Schmiederer, T.; Hornbogen, T.; Berg, H.; Sssmuth, R. D.; Zocher, R. ChemBioChem 2007, 8, 1767–1770. 123. Davison, E. C.; Fox, M. E.; Holmes, A. B. et al. J. Chem. Soc. Perkin Trans. 1 2002, 1494-1514. 124. Nishida, H.; Miyazaki, Y.; Mukaihira, T.; Saitoh, F.; Fukui, M.; Harada, K.; Itoh, M.; Muraoka, A.; Matsusue, T.; Okamoto, A.; Hosaka, Y.; Matsumoto, M.; Ohnishi, S.; Mochizuki, H. Chem. Pharm. Bull. 2002, 50, 1187-1194. 125. Liu, C. F.; Tam, J. P. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 6584-6588. 126. Vugts, D. J.; Aktas, H.; Al-Mafraji, K.;Orru, R. V. A. et al. Eur. J. Org. Chem. 245 2008, 1336-1339. 127. Compernolle, F.; Toppet, S. J. Heterocyclic Chem. 1986, 23, 541–544. 128. Bitan,G.; Muller, D.; Kasher, R.; Gluhov, E. V.; Gilon, C. J. Chem. Soc., Perkin Trans. 1 1997, 1501-1510. 129. Plank, C.; Mechtler, K.; Szoka, F. S. Jr.; Wanger, E. Hum. Gene Ther. 1996, 7, 1437-1446. 130. Pozzo, A. D.; Vigo, A.; Donzelli, G. Makromol. Chem. 1989, 190, 2457–2461. 131. Chen, Y.; Baker, G. L. J. Org. Chem. 1999, 64, 6870-6873. 132. Bhattacharya, B.; Dileep, P. V. Bioconjugate Chem. 2004, 15, 508-519. 133. Wang, C.; Zhang, H. Synth. Commun. 2002, 32, 3465-3468. 134. Ahmed, S. A.; Tanaka, M. J. Org. Chem. 2006, 71, 9884-9886. 135. Liang, C.; Feng, Y.; Vojkovsky, T. PCT Int. Appl. WO2006127961, 2006. 136. Krasovskiy, A.; Knochel, P. Synthesis 2006, 5, 890–891. 137. Torchilin, V. P. Pharm. Res. 2007, 24, 1–16. 138. O’Reilly, R. K.; Hawker, C. J.; Wooley, K. L. Chem. Soc. Rev. 2006, 35, 1068– 1083. 139. Weaver, J. V. M; Tang, Y. Q.; Liu, S. Y.; Iddon, P. D.; Grigg, R.; Billingham, N. C.; Armes, S. P.; Hunter, R.; Rannard, S. P. Angew. Chem. Int. Ed. 2004, 43, 1389–1392. 140. Jiang, X. Z.; Liu, S. Y.; Narain, R. Langmuir 2009, 25, 13344–13350. 141. Opsteen, J. A. and van Hest, J. C. M., Chem. Commun. 2005, 57-59. 142. Malkoch, M., Schleicher, K., Drockenmuller, E., Hawker, C. J., Russell, T. P., Wu, P., and Fokin, V. V., Macromolecules, 2005 38, 3663-3678. 143. Parrish, B., Breitenkamp, R. B., and Emrick, T., J. Am. Chem. Soc., 2005 127, 7404-7410. 144. Dubois, P.; Ropson, N.; Jerome, R.; Teyssie, P. Macromelecules, 1996, 29, 1965-1975. 145. Yin, M.; Baker, G. L. Macromolecules 1999, 32, 7711-7718. 246 146. Settle, F. A. Handbook of Instrucmental Techniques for Analytical Chemsitry; Prentice Hall: 1997, p 856, Chapter 4. 147. Khosravi, E.; Yagci, Y.; Savelyev, Y. New Smart Materials via Metal Mediated Macromolecular Engineering; Springer: 2009, p 157. 148. Bertini, F.; Audisio, G.; Kiji, J.; Fujita, M. J. Anal. Appl. Pyrolysis 2003, 68–69, 61–81. 149. Fournier, D.; Du Prez, F. Macromolecules 2008, 41, 4622-4630. 150. Leemhuis, M.; van Nostrum, C. F.; Kruijtzer, J. A. W.; Zhong, Z. Y.; Feijen, J.; Hennink, W. E. Macromolecules 2006, 39, 3500-3508. 151. Kim, J. H.; Jang, J.; Lee, D-Y.; Zin, W-C. Macromolecules, 2002, 35, 311–313. 152. Tsai, Y.; Fan, C.-H.; Hung, C.-Y.; Tsai, F.-J. J. Appl. Polym. Sci. 2008, 109, 2598–2604. 153. Hiemenz, P. C.; Lodge, T. P. Polymer Chemistry; Boca Raton, CRC Press, 2007, p 493. 154. Sperling, L.H. Glass-Rubber Transition Behavior, in Introduction to Physical Polymer Science, Fourth Edition, John Wiley & Sons, Inc., Hoboken, NJ, USA: 2005, p 401. 155. Kraus, G.; Childers, C. W.; Gruver, J. T. J. Appl. Polym. Sci. 1967, 11, 1581– 1591. 156. Guo, M.; Yu, T.; Xue, Z. Makromol. Chem., Rapid Commun. 1987, 8, 601–605. 157. Zhou, X.-M; Jiang, Z.-H. J. Polym. Sci. B Polym. Phys. 2005, 43, 1624–1630. 158. Parrish, B.; Breitenkamp, R. B.; Emrick, T. J. Am. Chem. Soc. 2005, 127, 74047410. 159. Silvers, A. L.; Chang, C.-C.; Emrick, T. J. Polym. Sci. A Polym. Chem. 2012, 50, 3517–3529. 160. Lu, C.; Shi, Q.; Chen, X.; Lu, T.; Xie, Z.; Hu, X.; Ma, J.; Jing, X. J. Polym. Sci. A Polym. Chem. 2007, 45, 3204-3217. 161. Khosravi, E. et al. Editor, New Smart Materials via Metal Mediated macromolecular Engineering, Springer Science: 2009, p 77. 162. Tachibana, Y.; Kurisawa, M.; Uyama, H.; Kobayashi, S. Biomacromolecules 2003, 4, 1132-1134. 247 163. Shimokuri, T.; Kaneko, T.; Akashi, M. J. Polym. Sci. A Polym. Chem. 2004, 42, 4492-4501. 164. Chen, C.; Wang, Z.; Li, Z. Biomacromolecules 2011, 12, 2859-2863. 165. Rijcken, C. J. F.; Soga, O.; Hennink, W. E.; van Nostrum, C. F. J. Control. Release 2007, 120, 131-148. 166. Behnke, T.; Würth, C.; Hoffmann, K.; Hübner, M.; Panne, U.; Resch-Genger, U. J Fluoresc. 2011, 21, 937-944. 167. Zhang, Y.; Zhang, Q.; Zha, L.; Yang, W.; Wang C.; Jiang, X.; Fu, S. Colloid Polym. Sci. 2004, 282, 1323-1328. 168. Ren, J.; Hong, H.; Song, J.; Ren, T. J. Appl. Polym. Sci. 2005, 98, 1884–1890. 169. Li, Y.; Qi, X. R.;Matani, Y.; Nagai, T. Nanotechnology 2009, 20, 055106. 170. Govender, T.; Stolnik, S.; Garnett, M. C.; Illum, L.; Davis, S. S. J. Control. Release 1999, 99, 171–185. 171. Gref, R.; Mück, M.; Quellec, P.; Marchand, M.; Dellacherie, E.; Harnisch, S.; Blunk, T.; Müller, R. H. Colloids Surf., B 2000, 18, 301-313. 172. Wu, Y.; Chen, W.; Meng, F.; Wang, Z.; Cheng, R.; Deng, C.; Liu, H.; Zhong, Z. J. Control. Release 2012, 164, 338-345. 173. Liu, T.; Simmons, T. L.; Bohnsack, D. A.; Mackay, M. E.; Smith, M. R.; Baker, G. L. Macromolecules 2007, 40, 6040-6047. 174. Rudine, A. B.; Walter, M. G.; Wamser, C. C. J. Org. Chem. 2010, 75, 42924295. 175. Mills, J. E.; Maryanoff, C. A.; McComsey, D. F.; Stanzione, R S.; Scott, S. J. Org. Chem. 1987, 52, 1857-1859. 176. McQuillin, F. J.; Baird, M. S. Acycli Chemistry, Second Edition, Cambridge University Press, 1983, p 10. 177. Settle, F. A. Handbook of Instrucmental Techniques for Analytical Chemsitry; Prentice Hall: 1997, p 856, Chapter 4. 178. Zhang, J. M.; Sato, H.; Tsuji, H.; Noda, I.; Ozaki, Y. Macromolecules 2005, 38, 1822−1828. 179. Ravi Kumar, M. N.; Sameti, M.; Mohapatra, S. S.; Kong, X.; Lockey, R. F.; Bakowsky, U.; Lindenblatt, G.; Schmidt, H.; Lehr, C. M. J. Nanosci. Nanotechnol. 248 2004, 4, 876-881. 180. Mahtab, F.; Lam, J. W. Y.; Yu, Y.; Liu, J.; Yuan, W.; Lu, P.; Tang, B. Z. Small, 2011, 7, 1448–1455. 181. Wang, T,; Jiang, H.; Zhao, Q.; Wang, S.; Zou, M.; Cheng, G. Int. J. Pharm. 2012, 436, 351-358. 182. Mader, H.; Li, X.; Saleh, S.; Link, M.; Kele, P.; Wolfbeis, O. S. Ann. N.Y. Acad. Sci. 2008, 1130, 218–223. 183. Bagwe, R. P.; Hilliard, L. R.; Tan, W. Langmuir 2006, 22, 4357–4362. 184. Achatz, D. E.; Heiligtag, F. J.; Li, X.; Link, M.; Wolfbeis, O. S. Sens. Actuators, B 2010, 150, 211-219. 185. Lu, X.; Sun, F.; Wang, J.; Zhong, J.; Dong, Q. Macromol. Rapid Commu. 2009, 30, 2116-2120. 186. Chen, C., Geng, J., Pu, F., Yang, X., Ren, J. and Qu, X. Angew. Chem. Int. Ed. 2011, 50, 882–886. 187. Kele, P., Mezö, G., Achatz, D. and Wolfbeis, Otto S. Angew. Chem. Int. Ed. 2009, 48, 344–347. 188. Howard, M. D.; Jay, M.; Dziublal, T. D.; Lu, X. L. J. Biomed. Nanotechnol. 2008, 4, 133-148. 189. Karakoti, A.; Das, S.; Thevuthasan, S.; Seal, S. Angew. Chem. Int. ed. 2011, 50, 2-17. 190. Choualeb, A.; Ros, J.; Braustein, P.; Welter, R. Organometallics 2003, 22, 26882693. 191. Lu, X.; Sun, F.; Wang, J.; Zhong, J.; Dong, Q. Macro. Rapid Commun. 2009, 30, 2116-2120. 192. Li, G.; Bhosale, Sh. V.; Wang, T.; Hackbarth, S.; Roeder, B.; Siggel, Fuhrhop, J-H J. Am. Chem. Soc. 2003, 125, 10693-10702. 193. James, M.; Ciampi, S.; Darwish, T. A.; Hanley, T. L.; Sylvester, S.O.; Gooding, J. J. Langumir 2011, 27, 10753-70762. 194. Pandey, P.; Farha, O.; Spokoyny, A.; Mirkin, C.; Kanatzidis, M.; Hupp, J.; Nguyen, S. J. Mater. Chem. 2011, 21, 1700-1703. 195. Yildirim, A.; Ozqur, E.; Bayindir, M. J. Mater. Chem. B 2013, 1, 1909-1920. 249 U.; 196. Wang, J.; Meng, G.; Tao, K.; Feng, M.; Zhao, X.; Li, Z.; Xu, H.; Xia, D.; Lu, J. R. PLoS ONE 2012, 7, e43478. 197. Zdyrko, B.; Luzinov, I. Macromol. Rapid Commun.2011, 32, 859–869. 198. Bevan, M. A.; Prieve, D. C. Langmuir 2000, 16, 9274-9281. 199. Ito, T.; Sun, L.; Bevan, M. A.; Crooks, R. M. Langmuir 2004, 20, 6940–6945. 200. Jiang, J.K.; Oberdorster, G.; Biswas, P. J. Nanopart. Res. 2009, 11, 77–89. 201. Brantley, J. N.; Wiggins, K. M.; Bielawski, C. W. Science 2011, 16, 1606-1609. 202. Kamart, M.; El-Boubbou, K.; Zhu, D.; Lansdell, T.; Lu, X.; Li, W.; Huang, X. Bioconjugate Chem. 2010, 21, 2128-2135. 203. Hu, F.; MacRenaris, K. W.; Waters, E. A.; Liang, T.; Schultz-Sikma, E. A.; Eckermann, A. L.; Meade, T. J. J. Phys. Chem. C 2009, 113, 20855–20860. 250