MECHANISTIC STUDIES OF THE COPOLYMERIZATION OF EPOXIDES WITH CARBON DIOXIDE AND RING-OPENING POLYMERIZATION OF CYCLIC

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Zhiping Zhou, B.S.

*****

The Ohio State University 2004

Dissertation Committee: Approved by Professor Malcolm H. Chisholm, Advisor

Professor Philip J. Grandinetti ______Professor T. V. RajanBabu Advisor Graduate Program in Chemistry

ABSTRACT

The development of new generation is guided by Green Chemistry and sustainability. Biodegradable polymers derived from readily renewable resources, e.g., plants and carbon dioxide, are attractive as environmentally friendly alternatives to polymers derived from fossil fuels, and will play an increasing role in the market place.

Polyoxygenates, such as and polycarbonates, are one important family of such polymers. In this work, the microstructure of poly(propylene carbonate) (PPC) has been studied by C-13 NMR spectroscopy. By analyzing samples and model compounds, we observed tetrad and triad sensitivity for the carbonate and propylene carbon atoms, respectively. The reactions between (TPP)AlX, where TPP = tetraphenylporphyrin and X = Cl, O(CH2)9CH3 and O2C(CH2)6CH3, and propylene oxide,

PO, have been studied and shown to give (TPP)AlOCHMeCH2X and

(TPP)AlOCH2CHMeX compounds. The relative rates of ring opening of PO follow the order Cl > OR > O2CR, but in the presence of added 4-(dimethylamino)pyridine (DMAP)

(1 equiv), the order is changed to O2CR > OR. From studies of kinetics, the ring opening of PO is shown to be first order in [Al]. Carbon dioxide inserts reversibly into the Al–OR bond to give the compound (TPP)AlO2COR, and this reaction is promoted by the addition of DMAP. On the basis of these results, the reaction pathway leading to ring

ii opening of PO can be traced to an interchange associative mechanism, wherein coordination of PO to the electrophilic aluminum atom occurs within the vicinity of the

Al–X bond. The gas-phase studies in an electrospray tandem mass spectrometer revealed that, for a series of related ions LM+, L = TPP or (R,R)- N,N’-bis(3,5-di-tert-butyl- salicylidene)-1,2-cyclohexenediamino) ((R,R)-salen), the Lewis acidity order follows M

= Cr(III) ~ Al(III) > Co(III), and notable differences were observed when PO molecules coordinated to those metal centers. The preparations of LAlOCH2CH(S)MeCl, where L =

(R,R) or (S,S)-salen, are reported together with the respective LAlOEt compounds, and their reactivities toward L- and rac-lactides in various solvents reveal the surprising complexity of the stereopreference for the ring-opening event.

iii

Dedicated to my wife Na Guan

and

my parents

iv

ACKNOWLEDGMENTS

I wish to thank my advisor, Prof. Malcolm H. Chisholm, for his guidance and support during my Ph.D research. His patience, enthausiasm, encouragement and friendship made the past five years an enjoyable experience in my life. To me, he is not only a true scientist, but a great educator. I will always be grateful to him for the opportunities that he provided me with, to grow as a chemist.

I would like to thank the Chisholm group for the good time we shared. I thank Dr.

Hongshi Zhen, the only one who could speak my native language when I joined this group, for teaching me how to use the Schlenk-line and dry box. I thank Drs. Ewan

Delbridge, Diana Navarro-Llobet and Khamphee Phomphrai for all their help in the lab,

Dr. Matthew Byrnes for teaching me DFT calculations, Dr. Nathan Patmore for X-ray single crystal measurements and Dr. Ramkrishna Ramnauth for his comments on this manuscript. Mrs. Karen Weimer is acknowledged for her excellent clerical work.

I would also like to thank the outstanding faculty and staff at OSU. I thank Prof.

Christopher Hadad for his direction on the DFT calculations, also Prof. Philip Grandinetti and Prof. T. V. RajanBabu for serving on my dissertation committee and making this document possible. Many thanks are given to Drs. Karl Vermillion and Charles Cottrell

v from OSU NMR facility, Dr. Judith Gallucci from OSU X-Ray Laboratory, and the staff from CCIC Mass Laboratory.

I would like to thank Dr. William Simonsick at Dupont and Dr. Edward

Dexheimer at BASF for their valuable discussions, as well as Prof. Peter Chen and Dr.

Xiangyang Zhang at ETH for the oversea collaboration.

The U.S. Department of Energy is acknowledged for financial support throughout my graduate years.

I am eternally grateful to my wife, Na Guan, whose love and support means the world to me. I feel incredibly fortunate to have her in my life, and could not imagine how

I could have been through these years in the U.S. without her.

Finally, I am truly grateful to my parents and sister for their constant support and motivation during my academic pursuits.

vi

VITA

July 14, 1974...... Born – Hubei, China

July 1, 1996...... B.S., Chemistry, Wuhan University, China

1996 – 1999...... Graduate Research Associate, Wuhan University, China

1999 – 2001...... Graduate Teaching Associate, The Ohio State University

2001 – 2004...... Graduate Reasearch Associate, The Ohio State University

PUBLICATIONS

1. Chisholm, M. H.; Zhou, Z. Concerning the mechanism of the ring opening of propylene oxide in the copolymerization of propylene oxide and carbon dioxide to give poly(propylene carbonate). J. Am. Chem. Soc., 2004, 126, 11030-11039.

2. Byrnes, M. J.; Chisholm, M. H.; Hadad, C. M.; Zhou, Z. Regioregular and regioirregular oligoether carbonates: a 13C{1H} NMR investigation. Macromolecules, 2004, 37, 4139-4145.

3. Chisholm, M. H.; Navarro-Llobet, D.; Zhou, Z. Poly(propylene carbonate). 1. more about poly(propylene carbonate) formed from the copolymerization of propylene oxide and carbon dioxide employing a zinc glutarate catalyst. Macromolecules, 2002, 35, 6494-6504.

vii 4. Zhou, Z.; Wang, Z. Y.; Wu, C. Y.; Zhan, W.; Xu, Y. Sol- method for the preparation of solid phase microextraction fibers. Analytical Letters, 1999, 32, 1675-1681.

5. Zhang, D.; Zhou, Z.; Tang, Y.; Wu, C.; Zhan, W.; Xu, Y. Analysis of organochlorine compounds in by solid phase microextraction and gas chromatography. Fenxi Huaxue (The Chinese Journal of Analytical Chemistry), 1999, 27, 768-772.

6. Zeng, Z.; Zhou, Z.; Wu, C.; Liu, M. Retention and thermodynamic properties of solutes on two types of β-cyclodextrin derivative stationary phases. Fenxi Huaxue, 1997, 25, 874-878.

FIELDS OF STUDY

Major Field: Chemistry

viii

TABLE OF CONTENTS

P a g e

Abstract...... ii

Dedication...... iv

Acknowledgments...... v

Vita ...... vii

List of Tables ...... xi

List of Figures ...... xiii

List of Schemes...... xviii

Chapters

1. Introduction...... 1

2. Poly(propylene carbonate) microstructural studies by using polymers prepared by a zinc glutarate catalyst and model oligoether carbonates ...... 30 2.1 Introduction...... 30 2.2 Results and discussion ...... 31 2.2.1 13C{1H} NMR spectra of PPC ...... 31 2.2.1.1 The carbonate carbon region ...... 32 2.2.1.2 The methine, methylene, and methyl carbon regions...... 36 2.2.2 Regioregular and regioirregular oligoether carbonates ...... 44 2.2.2.1 Synthesis ...... 44 2.2.2.2 NMR spectra and assignments ...... 45 2.2.2.3 Structural considerations...... 48 2.2.2.4 NMR calculations...... 55 2.2.3 PO/CO2 copolymerization catalyzed by a zinc glutarate catalyst...59

ix 2.2.3.1 PPC formation as a function of time ...... 59 2.2.3.2 Degradation of PPC...... 62 2.3 Conclusions...... 66 2.4 Experimental section...... 67 2.5 References...... 77

3. Concerning the mechanism of the ring opening of propylene oxide in the copolymerization of propylene oxide and carbon dioxide to give poly(propylene carbonate)...... 79 3.1 Introduction ...... 79 3.2 Results and discussion...... 84 3.2.1 Studies of the initial ring opening of propylene oxide...... 84 3.2.2 The carbon dioxide insertion reaction...... 92 3.2.3 Polymer microstructure of poly(propylene carbonate) ...... 94 3.2.4 A proposed mechanism for the (TPP)AlCl/DMAP copolymerization of PO and CO2 ...... 98 3.2.5 Substituent effects on the porphyrin ligand ...... 102 3.2.6 Gas-phase studies ...... 106 3.2.6.1 Binding of propylene oxide ...... 106 3.2.6.2 Ring opening of propylene oxide in the gas phase ...... 111 3.2.6.3 Polymerization reactions in the solution phase...... 112 3.2.6.4 Solid-state and molecular structures...... 113 3.3 Concluding remarks...... 117 3.4 Experimental section ...... 119 3.5 References ...... 132

4. Concerning the relative importance of enantioporphic site versus chain end control in the stereoselective polymerization of lactides: reactions of (R,R-salen) and (S,S-salen) aluminum alkoxides LAlOCH2R complexes (R = CH3 and S- CHMeCl) ...... 137 4.1 Introduction...... 137 4.2 Results and discussion ...... 139 4.2.1 Synthesis and structural consideration...... 139 4.2.2 Stereoselectivity in 1:1 reactions with rac-lactide...... 142 4.2.3 Stereoselectivity in ring-opening polymerization of rac-lactide.....145 4.3 Concluding remarks ...... 146 4.4 Experimental...... 147 4.5 References...... 152

Appendix A. Supplemental materials for Chapter 2 ...... 155

Appendix B. Supplemental materials for Chapter 3 ...... 174

Bibliography ...... 181

x

LIST OF TABLES

Table Page

1.1 Green Chemistry Challenge Awards recognizing CO2 work...... 8

1.2 Homogeneous catalysts for PO/CO2 copolymerization...... 16

2.1 Calculated populations for the isomers of MeOCH2CHMeOCO2CHMeCH2OMe in the gas phase, chloroform and benzene ...... 52

2.2 Calculated NMR chemical shifts for the isomers of MeOCH2CHMeOCO2CHMeCH2OMe in the gas phase and chloroform ...... 57

2.3 Calculated NMR chemical shifts for the isomers of MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me in the gas phase and chloroform ...... 58

2.4 Percentage of PPC, PC, and polyether linkage formed in the copolymerization of rac-PO and CO2 as a function of time ...... 60

3.1 Homopolymerization of PO by (TPP) metal complexes ...... 82

3.2 Copolymerization of PO/CO2 by (TPP) and (salen) metal complexes .83

3.3 Copolymerization of PO/CO2 by (porp)AlO(CH2)9CH3 complexes in the presence of 1 equiv of DMAP...... 104

3.4 Selected bond distances (Å) and angles (deg) for (R,R- salen)AlOC(S)HMeCH2Cl...... 114

3.5 Selected bond distances (Å) and angles (deg) for (R,R-salen)AlOOCMe ...... 116

xi 3.6 Crystallographic details for (salen)AlOCHMeCH2Cl·0.5PO and (salen)AlOOCMe·1.5py...... 131

4.1 Selected bond distances (Å) and angles (deg) for (R,R- Salen)AlOCH2C(S)HMeCl and (S,S-salen)AlOCH2C(S)HMeCl ...... 141

4.2 Stereoselectivity in 1:1 reactions of (Salen)Al complexes and rac-lactide ...... 144

4.3 Crystallographic details for (R,R-salen)AlOCH2C(S)HMeCl and (S,S- salen)AlOCH2C(S)HMeCl...... 151

xii

LIST OF FIGURES

Figure Page

1.1 Reaction coordinate diagram for PO/CO2 coupling by (salen)CrCl from reference 30 ...... 11

1.2 TPP metal catalysts for the PO/CO2 copolymerization ...... 13

1.3 Salen metal catalysts PO/CO2 copolymerization ...... 14

1.4 Homogeneous well-defined catalysts for CHO/CO2 copolymerization ...... 15

1.5 The structure of (BDI) zinc acetate effective in PO/CO2 copolymerization ...... 16

1.6 Three stereo-isomers of lactide ...... 18

1.7 Single-site catalysts for ring-opening polymerization of lactide...... 20

1.8 Salen and salan Al catalysts for selective polymerization of lactide. ...22

13 1 2.1 C{ H} (100 MHz, CDCl3) NMR spectrum of the C=O region of PPC reported in ref 3 ...... 31

2.2 13C{1H} (150 MHz) NMR spectra of C=O region of PPC made from (a) rac-PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst...... 34

2.3 13C{1H} (100MHz) NMR spectrum showing the methine and methylene carbons of PPO prepared with a zinc glutarate catalyst...... 35

2.4 13C{1H} (150 MHz) NMR spectra of CH region of PPC made from (a) rac-PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst...... 39

xiii

13 1 2.5 C{ H} (150 MHz) NMR spectra of CH2 region of PPC made from (a) rac-PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst...... 41

13 1 2.6 C{ H}(150 MHz) NMR spectra of CH3 region of PPC made from (a) rac-PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst...... 43

13 1 2.7 C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of PPC made from rac-PO employing the (TPP)AlCl/EtPh3PBr catalytic system...... 44

13 1 2.8 C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regioregular oligoether carbonate compounds (n = 1, 2) ...... 46

13 1 2.9 C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regioirregular oligoether carbonate compounds (n = 1, 2)...... 48

2.10 Six local minimum conformers in the gas–phase calculations for RR isomers of MeOCH2CHMeOCO2CHMeCH2OMe...... 50

2.11 Six local minimum conformers in the gas–phase calculations for SR isomers of MeOCH2CHMeOCO2CHMeCH2OMe...... 51

2.12 Six local minimum conformers in both the gas phase and chloroform calculations for RR isomers having C2 symmetry of MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me...... 53

2.13 Newman projections for the RR isomers of MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me, viewing down the bonds of CH(Me)–O, CH2–CH(Me), and O–CH2 ...... 54

2.14 Eight local minimum conformers in both the gas phase and chloroform calculations for SR isomers having Cs symmetry of MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me...... 55

2.15 MALDI–TOF mass spectrum of PPC prepared with a zinc glutarate catalyst (t = 2.5 h)...... 61

2.16 MALDI–TOF mass spectrum of PPO prepared with zinc glutarate.....62

2.17 ESI–QTOF mass spectrum of PPC degradation products by LiOtBu ..65

xiv 1 3.1 H NMR (500 MHz, CDCl3) spectrum of PO ring-opened product by (TPP)AlCl, showing two regioisomers...... 84

1 3.2 H NMR (500 MHz, CDCl3) spectra of the reaction between (TPP)AlO(CH2)9CH3 and PO, showing the disappearance of initiator and appearance of product with time...... 86

3.3 Plots of -ln(It/I0) versus reaction time for the ring-opening reaction of the first PO molecule by initiators, in the absence and in the presence of 1 equiv of DMAP ...... 88

3.4 Plots of -ln(It/I0) versus reaction time for the ring-opening reaction of the first PO molecule by (TPP)AlO2C(CH2)6CH3, in the presence of varying [DMAP]:[initiator] ratios...... 89

1 3.5 H NMR (500 MHz, CDCl3) spectrum of R-PO ring-opened products by (R,R-salen)AlCl, showing three isomers ...... 90

3.6 The percentages of alkyl carbonate and alkoxide versus the 13 [DMAP]:[(TPP)AlO(CH2)9CH3] ratio for the CO2 insertion with (TPP)AlO(CH2)9CH3 ...... 93

13 1 3.7 C{ H} (150 MHz, CDCl3) NMR spectra of the carbonate carbon region of PPC made from (A1) zinc glutarate and rac-PO, (B1) (TPP)AlCl/DMAP and rac-PO, and (C1) (R,R-salen)CrCl and rac-PO ...... 95

13 1 3.8 C{ H} (150 MHz, CDCl3) NMR spectra of the carbonate carbon region of PPCs made from (A2) zinc glutarate and S-PO, (B2) (TPP)AlCl/DMAP and S-PO, (C2) (R,R-salen)CrCl and S-PO and (C3) (R,R-salen)CrCl and R-PO...... 97

3.9 MALDI-TOF mass spectrum of PPC prepared with the (TPP)AlO(CH2)9CH3/DMAP catalyst system ...... 99

3.10 The structures of (TPFPP)AlO(CH2)9CH3 and (OEP)AlO(CH2)9CH3..102

3.11 The percentages of alkyl carbonates and alkoxides versus [DMAP]: 13 [(porp)AlO(CH2)9CH3] ratio for the CO2 insertion with (porp)AlO(CH2)9CH3 complexes ...... 103

13 1 3.12 C{ H} (150 MHz, CDCl3) NMR spectra of the carbonate carbon region of PPC made from (porp)AlO(CH2)9CH3/DMAP and rac-PO/CO2 ...... 105

xv 3.13 ESI/MS from the reactions between (TPP)/(salen) metal cations and propylene oxide in the gas phase ...... 108

3.14 The plots of LM(PO)+ abundance versus CID voltage from the dissociation reaction of LM(PO)+ with xenon in the CID chamber .....108

+ 3.15 The plots of LM(PO)2 abundance versus CID voltage from the + dissociation reactions of LM(PO)2 with xenon in the CID chamber ..110

3.16 The plots of LM(PO)+ abundance versus CID voltage from the + dissociation reactions of LM(PO)2 with xenon in the CID chamber ..110

+ 3.17 ESI/MS from the reaction of (TPP)Al(PO)2 with pyridine in the gas phase...... 111

3.18 ORTEP view of (R,R-salen)AlOC(S)HMeCH2Cl ...... 113

3.19 ORTEP view of S-propylene oxide...... 115

3.20 ORTEP view of (R,R-salen)AlOOCMe ...... 117

4.1 ORTEP views of (R,R-salen)AlOCH2C(S)HMeCl and (S,S- salen)AlOCH2C(S)HMeCl...... 140

1 4.2 Selected H (500MHz, CDCl3) NMR spectra showing CH region in the products in ring-opening reactions of 1 equiv of lactide by (salen)AlOCH2R...... 143

1 4.3 H (500MHz, CDCl3) NMR spectra of the homodecoupled CH resonance of poly(rac-lactide) prepared in toluene using: (A) (R,R- salen)AlOCH2CH3, (B) (R,R-salen)AlOCH2C(S)HMeCl, and (C) (S,S- salen)AlOCH2C(S)HMeCl...... 146

13 1 A.1 C{ H} NMR (150 MHz, CDCl3) spectrum of PPO obtained from PO polymerization with KOEt...... 165

13 1 A.2 C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regioregular oligoether carbonate compounds (n = 3, 4) ...... 169

13 1 A.3 C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regio-regular oligoether carbonate compounds (n = ~10, ~30)...... 173

B.1 Plot of Ln(kobs]) vs Ln([Al]) for ring opening of PO by varying concentrations of (TPP)AlO(CH2)9CH3 (5 to 15 mM)...... 174

xvi

1 B.2 H NMR (500MHz, CDCl3) spectra of (TPP)AlX/DMAP, X = o o O(CH2)9CH3 and O2C(CH2)6CH3, at 25 C and -50 C ...... 175

B.3 Plots of -Ln(It/I0) versus reaction time for the ring-opening reaction of the first PO molecule by initiators, in the presence of 1 equiv of DMAP ...... 176

1 B.4 H (600 MHz, CDCl3) NMR spectra of (A) Complex 1; (B) product of

the reaction between (R,R-salen)AlCl and R-PO; (C) after addition of

Complex 1 into sample (B); (D) product of the reaction between (R,R-

salen)AlCl and S-PO; (E) after addition of Complex 1 into sample (D)

...... 178

B.5 ORTEP view of (R,R-salen)AlOCHMe(S)CH2Cl grown in S-PO...... 179

B.6 ORTEP view of (R,R-salen)AlOOCMe grown in pyridine ...... 180

xvii

LIST OF SCHEMES

Scheme Page

1.1 The biosynthetic pathway for poly(3-hydroxybutyrate) (PHB)...... 6

1.2 CO2 industrial utilizations in urea and methanol production...... 7

1.3 Copolymerization reaction of epoxides and CO2 ...... 8

1.4 The synthetic route for polycarbonate in industry...... 9

1.5 Three possible pathways in the PO/CO2 reaction ...... 11

1.6 Ring-opening polymerization of lactide...... 18

1.7 Transesterification reactions in the ring-opening polymerization of lactide ...... 20

2.1 Three possible regiosequences of PPC at the diad level when looking at the central carbonate carbon ...... 32

2.2 Ring opening of PO by a backside attack mechanism...... 35

2.3 Eight possible regiosequences for carbonate carbons (*) in PPC with central HT junctions at the tetrad level...... 36

2.4 Eight possible regiosequences for methine carbons (*) in PPC at the triad level...... 37

2.5 Eight possible regiosequences for methylene carbons (*) in PPC at the triad level...... 40

2.6 Eight possible regiosequences for methyl carbons (*) in PPC at the triad level...... 42

2.7 The synthetic route for oligoether carbonates ...... 45

xviii

2.8 Two possible mechanisms of PPC degradation by LiOtBu...... 64

3.1 The monometallic mechanism proposed for the CHO/CO2 copolymerization reaction by Darensbourg ...... 80

3.2 The stereochemistry in the PO ring-opening event involving attack at methine carbons by (R,R-salen)AlCl...... 92

3.3 The chain growing pathway showing HT carbonate junction (*) formation involving double attack at methine carbons in the PO/CO2 copolymerization ...... 96

3.4 The preferential stereochemistry involved in the formation of HH carbonate junctions during the formation of PPC by the catalyst derived from (R,R-salen)CrCl...... 98

3.5 Proposed interchange associative pathway for PO ring opening by (TPP)AlX/DMAP ...... 100

3.6 Proposed mechanism for PO/CO2 copolymerization by (TPP)AlX systems, where Nu represents the added Lewis base promoter ...... 101

3.7 Schematic representation of the modified Finnigan MAT TSQ-700 ESI tandem mass spectrometer...... 107

4.1 Proposed reaction scheme for the ring-opening polymerization of lactides by a metal alkoxide...... 138

4.2 Ring opening of 1 equiv of lactide by (salen)AlOCH2R...... 143

xix

CHAPTER 1

INTRODUCTION

Sustainable development has become an internationally accepted concept in modern society. The 1987 United Nations Commission on Environment and

Development initiated the thinking of sustainability and defined sustainable development as, “development which meets the needs of the present without compromising the ability of future generations to meet their own needs”.1 This definition is essentially broad and covers all aspects of society. Particularly it has a significant impact on chemistry-based industries. Currently 98% of all organic chemical products are produced from - based feedstocks. It is believed that the fossil fuels, mainly , oil and natural gas, are being depleted, and there will be a rapid decline in supply before the middle of this century.2,3 Ironically, despite the fact that it brings enormous benefits to society, chemistry suffers from a negative image or even hostile reaction from the general public, largely due to the environmental pollution caused by chemical industry.4,5 The awareness of these issues has resulted in an explosion in the number of environmental regulation laws in the past two decades. Regulations can prevent disasters and reduce the risk to a minimum amount. However, they cannot eliminate the problems, and are not economic for the additional cost in the production.

1 In the early 1990s the US Environment Protection Agency (EPA) started the

Green Chemistry Movement in order to “promote innovative chemical technologies that reduce or eliminate the use or generation of hazardous substances in the design, manufacture and use of chemical products”. The EPA established the well-known twelve principles of Green Chemistry.6 These principles are recognized as a blueprint for designing safer chemical products and processes, which can be briefly introduced as:

1. Prevent waste

2. Design safer chemicals and products

3. Design less hazardous chemical syntheses

4. Use renewable feedstocks

5. Use catalysts, not stoichiometric reagents

6. Avoid chemical derivatives

7. Maximize atom economy

8. Use safer solvents and reaction conditions

9. Increase energy efficiency

10. Design chemicals and products to degrade after use

11. Analyze in real time to prevent pollution

12. Minimize the potential for accidents

A green chemistry technology may address several of these principles, and in many instances it is not perfect but represents an improvement over existing processes.

Green Chemistry advocates a different philosophical approach from environment regulations. By applying environmentally friendly technologies, chemical industry can contribute to sustainable development in an economically sound way. The Green

2 Chemistry Movement has been popularized all over the world, as both a culture and a methodology.

The synthetic polymer industry has revolutionized the quality of human life since its origin in the early of the twentieth century. Polymer products are widely used in a variety of applications, such as packaging materials, coatings, automobile parts, electronics and medical devices. New polymers and new applications of polymers are continuously being developed and bringing great benefits to our daily lives. As the environmental awareness grows, challenges rise to the traditional polymer industry. The market of polymer products is over $300 billion per year in the U.S. alone, yet little of this is sustainable. Most polymer products are produced in the processes utilizing petroleum-based feedstocks. The nonrenewability of petroleum will in the near future cause the earth to be “out of gas”.7 Although almost all petroleum is used as fuel with only about 3% as a petrochemical feedstock, the declining supply of petrochemicals will result in a crisis in the polymer industry, considering the rapid growing demand for polymer materials. For example, the worldwide production is projected to increase from 55 × 106 metric tons in 2002 to 87 × 106 metric tons in 2010, and production is projected to increase from 35 × 106 to 60 × 106 metric tons in the same period.8 On the other hand, the enormous amount of polymers produced every year creates serious environmental problems after being discarded at the end of their use. This is because at present most synthetic polymers, such as polyolefins, are persistent and nondegradable in nature, leading ultimately to the accumulation of materials in the environment and increases of greenhouse gases in the atmosphere when incinerated. Polymer waste is a growing urgency for environment pollution. Incineration 3 is not either an economic or an ecological way to clean polymer waste. Mechanical recycling has been practiced to recover materials for second uses. However, its success is limited, because reprocessing requires energy and usually causes a loss in polymer physical and mechanical properties. Environmental concerns are also made to the hazardous substances utilized or generated in numerous polymerization processes, such as phosgene used in the polycarbonate production.9 The handling of such substances requires the additional costs for special precautions to minimize the risk of chemical accidents which could cause serious problems on the environment.

A considerable amount of attention has been paid to developing green technologies in polymerization processes in both industry and academia. Designing safer polymers has been practiced in several ways on the basis of Green Chemistry: existing processes to renew parts; process redesigns based on existing feedstocks; innovative processes (new feedstocks and new process routes).10 Reaction processes are developed in the media of supercritical CO2, water, and solid-state to eliminate the use of hazardous organic solvents. Atom economic reactions, such as Diels-Alder reactions, are more and more used in synthetic routes to obtain the maximum efficiency incorporating in atoms into final products as a way of minimizing waste. New catalytic processes are designed to allow reactions to occur at low temperatures to minimize energy consumption for heating and cooling. Meanwhile, a catalyst may be reusable and provide enhanced selectivity to eliminate separation processes in production.

Ultimately sustainable development in the polymer industry will be achieved by using the strategy of producing biodegradable products from readily renewable resources.

The obvious potential alternatives to petrochemicals are natural, sustainable sources, such

4 as wood, grasses, and crops, namely , derived from sunlight, carbon dioxide and water through the process of photosynthesis. The U.S. Department of Energy has set a target to achieve 10% of basic chemical building blocks arising from plant-derived renewable resources by 2020, and a further increase to 50% by 2050.11 Cellulose is very biodegradable, while the biodegradability of cellulose-based polymers is significantly reduced because a high proportion of hydroxyl groups have been acetylated.12,13

Polyhydroxyalkanoates (PHAs) can be synthesized in bacterial fermentation from natural sugars and oils (Scheme 1.1).14,15 PHAs can be processed to have similar properties to polyethylene and polypropylene, yet are completely biodegradable. ICI(UK) first launched PHAs in 1990 under the trade name Biopol.16 However, it has so far failed because the technology was not economically feasible. For this reason, the research has been redirected to producing PHAs in plants by developing genetic engineering techniques.17 Another biodegradable polymer, poly(lactic acid), PLA, can be produced from corn . This renewable route involves the fermentation of corn starch producing lactic acid, or further distilled to the cyclic dimer lactide. PLA is produced in both condensation polymerization of lactic acid and ring-opening polymerization of lactide. In January 2002 Cargill-Dow LLC started up the first full scale PLA plant with an annual capacity of 140 000 metric tons, using up to 40 000 bushels of locally grown corn per day.18,19 The development of NatureWorks brand PLA by Cargill-Dow LLC was shortly recognized with one of the 2002 Presidential Green Chemistry Challenge

Awards.20

5 O O O CoA SH 2 H CSCoAC SCoA 3 ketothiolase

NADPH + H+ reductase

NADP+

OH O O SCoA n PHA synthase O

Scheme 1.1. The biosynthetic pathway for poly(3-hydroxybutyrate) (PHB).

The development of the next generation of polymers is guided by Green

Chemistry, sustainability, industrial ecology and eco-efficiency. There is immense opportunity in developing new sustainable products from renewable resources. An intensive amount of research has been done in both exploring new processes and improving the existing processes for the production of biodegradable polymers from renewable feedstocks. The catalyst development plays a crucial role in making those processes succeed in industry, as a cost-effective process requires highly efficient and highly selective catalysts.

Polycarbonate Synthesis via Carbon Dioxide and Epoxides.

Carbon dioxide is known as a major in the atmosphere, responsible for 50-55% of global warming, along with for 15-18%, nitrous oxide

21 for 6%, etc. The rapid rise in CO2 level over the last 100 years, from 290 to 380 ppm,

6 was mostly caused by fossil fuel burning. Human activity annually adds about 24 billion tons of CO2 to the atmosphere, with about 15 billion tons absorbed by plants, soil and the oceans. Various approaches are used to reduce CO2 emission, such as sequestration, energy conservation, development of high fuel-efficiency automotives and using alternative energy sources. On the other hand, utilization of CO2 is of great interest for its nontoxicity and abundance. As a matter of fact, carbon dioxide is widely used in enhanced oil recovery, refrigeration, beverage carbonation, as well as in the chemical

22 productions of urea and methanol (Scheme 1.2). Recently, supercritical CO2 has been

23 used as a solvent for various reactions. The utilization of CO2 is one of the central topics in Green Chemistry, as shown in Table 1.1 a considerable number of Presidential

Green Chemistry Challenge Awards recognized the work in this area.24

(1) CO2 + 2 NH3 NH2COONH4 CO(NH2)2 + H2O

(2) CO2 + 3 H2 CH3OH+ H2O

Scheme 1.2. CO2 industrial utilizations in urea and methanol production.

Despite its low reactivity, carbon dioxide has been a target for decades in polymerization reactions for generating high-value polymeric materials, since it is considered an ideal C1 resource for its obvious advantages in not only environmental concern but production cost. Among quite a few polymerization reactions explored so far, the copolymerization of epoxides and CO2 to form polycarbonates is the most studied

(Scheme 1.3).25,26

7 O catalyst R2 O + CO2 O O n R1 R2 R1

Scheme 1.3. Copolymerization reaction of epoxides and CO2.

Green Chemistry Challenge Awards Receipents

Benign tunable solvents coupling reaction and Charles A. Echert & Charles L. separation processes Liotta, Georgia Institute of Technology, 2004

Design of non-fluorous, highly CO2-soluble Eric J. Beckman, University of materials Pittsburg, 2002

SCORR-supercritical CO2 resist remover SC Fluids, Inc., 2002

Design and application of surfactants for CO2 Joseph M. DeSimone, University of North Carolina at Chapel Hill, 1997 The development and commercial The Dow Chemical Company, 1996

implementation of 100% CO2 as an environmentally friendly blowing agent for the polystyrene foam sheet packaging market

24 Table 1.1. Green Chemistry Challenge Awards recognizing CO2 work.

The coupling of CO2 and propylene oxide into a regular alternating , poly(propylene carbonate), was first reported by Inoue in 1969.27,28 This might provide a profitable utilization of CO2 and also an environmentally friendly alternative to the industrially used polycarbonate production process, that is, condensation polymerization involving large amounts of noxious phosgene (Scheme 1.4).9 Considerable activity has

8 been carried out in catalyst development for this process since its discovery. A great number of heterogeneous and ill-defined systems were found to be active, most of which were made from a mixture of diethyl zinc and multi-protic compounds, i.e., water, di- and trialcohols, phenols, amines, dicarboxylic acids.25,29 However, none of these systems were truly effective, and among them the best yield reported were 70 g of PPC per g of catalyst in 40 h by using a zinc glutarate system.29 Meanwhile, the polymerization reactions could not be well controlled by those catalysts, and usually the polymers were produced with inconsistent yields, unpredictable molecular weights and broad molecular weight distributions. Because of the difficulties in understanding the structures of heterogeneous catalysts, especially the circumstances of active sites, they were not much informative for catalyst design.

HO OH + NaOHNaO ONa

O phosgene Cl Cl

O

O O n

Scheme 1.4. The synthetic route for polycarbonate in industry.

9 Besides the alternating copolymerization of PO/CO2 to polycarbonate, there are other competing processes in this system, which lead to the formation of cyclic propylene carbonate, PC, and polyether linkages (Scheme 1.5).30 Ring opening of PO by metal alkyl carbonates and insertion of CO2 into M−OR bond are the two elementary reactions. PO ring opening by metal complexes can be brought about by either coordination-cationic or

-anionic pathways, in both of which PO activation by coordination to the metal centers and nucleophilic attack at methine/methylene carbons are the two key steps. In order to eliminate the formation of polyether linkage, the insertion of CO2 must be kinetically rapid, and the formed metal alkyl carbonate must exist at a sufficiently high thermodynamic equilibrium concentration, such that it opens PO at a higher rate than the metal alkoxide does. The cyclic PC can be formed in either direct coupling of PO/CO2 or degradation of PPC by the metal alkoxide in the system. It has been shown the latter was responsible for PC formation in this copolymerization system, and also that at high temperatures PC production was favored over PPC formation.31-33 Thus, PPC is the kinetic rather than the thermodynamic product. Darensbourg has nicely detailed the activation parameters for PC vs PPC formation with a salen chromium (III) catalyst system, where salen = N,N’-bis(3,5-di-tert-butyl-salicylidene)-1,2-cyclohexenediamine, as shown in Figure 1.1.30

10 P O

O O O M

A P O O CO2 P O O M M C O O O B

O O

O P O M

Scheme 1.5. Three possible pathways in the PO/CO2 reaction.

Ea(PC) Ea(PPC) Ea(PPC) = 67.6 kJ/mol

O Ea(PC) = 100.5 kJ/mol

y rH = -54 kJ/mol g

r + e CO n 2

E O

H O O O r O O n

Reaction Coordinate

Figure 1.1. Reaction coordinate diagram for PO/CO2 coupling by (salen)CrCl from reference 30.

11 Recently the concept of single-site , originally used in olefin polymerization,34 was introduced to the field of polyoxygenate production, such as polyethers, polyesters and polycarbonates.35 The general formula for a single-site catalyst is LnMX, where L = auxiliary ligands, M = a metal center and X = the initiator group

(active site). The well-defined single-site catalysts can achieve well-controlled polymerizations, allowing for precise molecular weight control and producing polymers with narrow molecular weight distributions in a living manner. (TPP)AlX, where TPP = tetraphenylporphyrin, X = Cl, OR, etc., was the first example of single-site catalyst for ring-opening polymerization of cyclic esters and epoxides (Figure 1.2), although the term of “single-site” was not used for these catalysts historically.36-39 It could also catalyze the

40 copolymerization of PO/CO2 with a low CO2 incorporation. However, when a Lewis base, such as Et3NCl and EtPh3PBr, is present in the system, CO2 incorporation can be improved significantly.40,41 Some other porphyrin metal complexes, M = Cr(III) and

Co(III), have been reported in the PO/CO2 coupling reaction to give cyclic PC at elevated temperatures.42,43 Although porphyrin metal catalysts have been known for twenty years in this process, their low in common organic solvents limited their applications.

12 N N M N N X

M = Al, Cr, Co X = Cl, OR, OOCR, etc.

Figure 1.2. TPP metal catalysts for the PO/CO2 copolymerization.

The Schiff base type compounds, (salen)MX, have been extensively studied as catalysts in the hydrolytic kinetic resolution of terminal epoxides.44 Recently a number of

(salen)MX type complexes were found to be active in the copolymerization of PO/CO2

(Figure 1.3). Darensbourg and coworkers first used (salen)CrX compounds as catalysts

45 for epoxide/CO2 copolymerization process in 2002. Since then Darensbourg’s group has extensively studied the polymerization kinetics,30,45 cocatalyst (Lewis bases) effects and substituent effects on the salen backbone for this type of catalysts. They found that the copolymerization proceeded in a monometallic pathway, in contrast to the bimetallic mechanism found in the asymmetric ring opening of epoxides.44,46 The catalyst activity can be improved significantly if the optimum amount of an appropriate cocatalyst is added. For example, phosphines are more effective cocatalysts than N-methylimidazole,

N-MeIm.47,48 Coates and coworkers have reported discrete (salen)CoOAc catalysts for

PO/CO2 copolymerization, which showed relatively good activity, and a high selectivity 13 of PPC over PC.49 More recently, Lu and coworkers found when a cocatalyst, such as

Bu4NCl, was employed with (salen)CoX catalysts, the turnover frequency (TOF) of PPC could be greatly increased, and they have reported the highest TOF of 371 mol·mol-1·h-1

50 for PO/CO2 coplymerization reaction.

N N M tBu O O tBu M = Cr, Co X X = Cl, N OR, OOCR, etc. tBu tBu 3

Figure 1.3. Salen metal catalysts PO/CO2 copolymerization.

Zinc-based catalysts were the first and are the most widely used in PO/CO2 copolymerization. Although most zinc catalysts are heterogeneous, efforts have been continuously made to prepare well-defined homogeneous zinc catalysts for this process to gain detailed information about the active site and achieve control polymerization. Zinc pyrogallol reported by Kuran represents the first example. It was formed in the reaction of diethyl zinc and pyrogallol, and soluble in 1,4-dioxane and THF. However, it exists as aggregates in solution with MW > 3 000 Da, and thus the nature of the active site was still unkown.51-53 Researchers continued to make well-defined zinc catalysts for epoxide/CO2 copolymerization process. Many of the catalysts can copolymerize cyclohexene oxide, CHO, with CO2 to give poly(cyclohexene carbonate), PCHC, but for

PO/CO2 PC was the major product, and usually a small amount of polymer could only be obtained at low temperatures. A variety of homogeneous zinc catalysts for epoxide/CO2

14 copolymerization are presented in Figure 1.4. One of the key features of some zinc catalysts is that they do copolymerize PO/CO2 to PPC efficiently, but at the same time they are highly active in degrading PPC to PC through a backbiting mechanism. Despite these difficulties, Coates and coworkers successfully used (BDI)ZnOAc, when BDI ligand was deliberately modified as shown in Figure 1.5, to copolymerize PO/CO2 to

PPC with high efficiency, TOF of 138 mol·mol-1· h-1, and reasonably good selectivity,

PPC : PC = 93:7.54 A subtle change on the substituents on the BDI ligand can result in a large change in activity and selectivity. This work sheds light in this area and shows the possibility of preparing highly efficient well-defined zinc catalysts for PO/CO2 copolymerization. Table 1.2 summarizes the turnover frequencies and selectivity (PPC vs

PC) achieved by a number of homogeneous catalytic systems in PO/CO2 copolymerization.

Ph Ph F F F N O N O O O F Et H Zn Zn OR Zn Zn Zn H Et O N F O O O N F F Ph F Ph

Coates Darensbourg Nozaki

55-57 Figure 1.4. Homogeneous well-defined catalysts for CHO/CO2 copolymerization.

15

N Zn OAc N F3C

Figure 1.5. The structure of (BDI) zinc acetate effective in PO/CO2 copolymerization.

Catalysts Temp CO2 TOF Selectivity Carbonate PDI Ref (oC) (psi) (h-1) (% PPC) linkage (%) (salcy)CrCl/DMAP58 25 700 3 93 97 1.35 58 (salph)CrCl/DMAP 75 200 160 71 98 1.38 48

(salph)CrN3/PPNCl 60 500 192 - 97 - 47 (salcy)CoOAc 25 800 59 99 99 1.57 49

(salcy)CoOAr/Bu4NCl 40 600 371 99 99 1.37 50 (TPP)AlCl 25 800 1 92 22 1.10 40

(TPP)AlCl/Et4NBr 25 800 1 78 75 1.10 41 (BDI)ZnOAc 25 300 212 87 99 1.11 54 Zinc glutarate 60 700 3 96 98 9.10 33

Table 1.2. Homogeneous catalysts for PO/CO2 copolymerization.

(salcy = N,N’-bis(3,5-di-tert-butyl-salicylidene)-1,2-cyclohexenediamino; salph = N,N’- bis(3,5-di-tert-butyl-salicylidene)-1,2-phenylenediamino)

16

It should be noted that the polycarbonates studied so far in the epoxides/CO2 copolymerization processes are of inferior physical properties to the commercial poly(bisphenol-A carbonate). Practically, the glass transition temperatures (Tg) of those polycarbonates, which fall in the range of 0 to 120 oC, need to be improved to be comparable to that of poly(bisphenol-A carbonate), 150 oC, in order to make this process more industrially viable. The general strategies are searching for suitable from biorenewable resources and introducing cross-linking into the formation of polymer.

Nevertheless, breakthroughs have been seen in the past few years, and efforts are continuing to bring this process into industrial reality.

Ring-Opening Polymerization of Lactide

The cyclic lactide, 3,6-dimethyl-1,4-dioxane-2,5-dione (Figure 1.6), used to be derived from petrochemicals, can now be derived from plants, such as corn, by a fermentation process.59 This makes lactide an inexpensive material and more importantly, an annually renewable resource for the chemical industry. Ring-opening polymerization of lactide forms polylactide, PLA (Scheme 1.6), which can be biodegraded to lactic acid and absorbed as a metabolite in the human body. Thus, PLA is an ideal and very attractive for its biodegradability and biocompatibility. Indeed, PLA have found enormous applications in medical fields, such as sutures in surgery, control release agents in and scaffolds in .60-64 The full-scale plant for

PLA production built by Cargill-Dow in 2002 brought this process onto the stage of industrial commercialization, and its market was certainly boosted as PLA can be used as

17 environmentally friendly commodity , such as packaging materials, food containers.65

O O O

O O O O O O

O O O L-lactide D-lactide meso-lactide

rac-lactide

Figure 1.6. Three stereo-isomers of lactide.

O O O catalyst O n O O n O O

Scheme 1.6. Ring-opening polymerization of lactide.

New challeneges and opportunities still emerge in this industrially successful process. Lactide is usually ring-opening polymerized in melts at temperatures over 120oC in industry with catalysts, such as tin(II) octonoate (i.e. stannous bis(2-ethylhexanoate)), aluminum alkoxides, zinc metal, etc.66-68 Although these catalysts are efficient and robust, the polymerization is not carried out in a well-controlled manner. For instance, transesterification (Scheme 1.7) is a common side reaction (especially at elevated

18 temperatures).69 Moreover, lactide has a relative high equilibrium concentration at high temperatures,70,71 and little selectivity can be achieved when a racemic mixture of lactide is used. Even a very fundamental aspect, the initiation group of some catalysts, has not been well understood. For tin(II) octonoate, Sn–OOCC7H15 is not usually considered as the initiation group, and Sn–OR groups formed in and alcoholysis with impurities or added are likely to be responsible for the initiation.66 A great amount of research has been done to synthesize well-defined, highly efficient catalysts for ring-opening polymerization of lactide. Metals with low , such as Zn(II),

Mg(II), Ca(II), Sn(IV), Al(III), rare earth and others, with a variety of ligands, have been exploited and resulted in notable advances (Figure 1.7).72,73 In 1996, Chisholm documented the first study of well-defined single-site catalysts for ROP of lactide, which were magnesium and zinc alkoxides with η3-trispyrazolyl or η3-trisindazolyl-borate ligands.74,75 Coates pioneered the work of using β-diiminate (BDI) metal complexes for lactide polymerization,76 followed by intensive work on BDI type catalysts by several other groups.77-80 Tolman and Hillmeyer have recently developed a bimetallic zinc catalyst that mimics an enzymatic system and is kinetically very rapid, yielding 96% conversion in 5 min at 650 equiv of lactide.81,82

O O O O O P O P' O O ML ML O O O O P' LM O LM P O O O

Scheme 1.7. Transesterification reactions in the ring-opening polymerization of lactide. 19 t t Bu tBu Bu O N N But O t HB N N MOR N O Zn Zn O Bu N N M OR N N N N N tBu

M = Mg, Zn, Ca Tolman & Hillmeyer t M = Mg, Zn R = Et, Bu, Ph, SiMe3 OR = O iPr, OAc, etc Chisholm Coates

Figure 1.7. Single-site catalysts for ring-opening polymerization of lactide.

Along with the activity, the selectivity is also a critical criterion for catalyst development. PLAs with various tacticities show marked differences in physical properties. Stereoselective polymerization leads to enhanced crystallinity and higher Tm

o 83 values. For poly(L-lactide), Tm ~175 C, while for the stereocomplex polymer which consists of a 1:1 mixture of poly(L-lactide) and poly(D-lactide), the Tm value is even higher, 230oC.84,85 Isotactic PLA has excellent mechanical properties and is subject to a slower degradation than amorphous atactic PLA. Isotactic PLA can be produced by using pure L- or D-lactides, but the optically pure monomers are expensive. Thus, it is of great interest to produce controlled tacticity from less expensive racemic mixture of lactide.

Although versatile tacticities have been obtained in ROP of lactides (rac- and meso-) by various catalysts, the selective mechanism is far from clear. Catalytic site control mechanism and chain end control mechanism are thought to be responsible for the selectivity. The Schiff-base type catalysts, (salen)AlOR, are the most studied for

20 understanding these mechanisms. In 1996 Spassky and coworkers reported that by using a chiral and enantiomerically pure (salen)AlOR (R-I in Figure 1.8) predominantely (Pi =

0.96) isotactic PLA was obtained from rac-LA,86 while the achiral salen Al catalyst (II) produced PLA with slightly enhanced isotactic junctions from rac-LA.87 In 2000 Baker and Smith reported the preparation of isotactic PLA from rac-LA by using rac-I catalyst.88 Subsequently, Coates and coworkers found the polymers prepared in that system were actually stereoblocks.89,90 Highly isotactic PLA could also be synthesized by using achiral salen Al catalysts (III, IV) as reported independently by Nomura and

Chen.91,92 Chiral salen Al catalysts V were first used by Feijen in polymerizing rac-LA to

PLA with dominant isotactic junctions (Pi > 0.92). They also found (R,R)-V selectively polymerized L-LA over D-LA (14:1),93,94 in contrast S-I prefers L-LA (28:1).95 Other than salen catalysts, Coates and coworkers synthesized heterotactic PLA (Pi = 0.94) from

i 76 rac-LA and syndiotactic PLA (Pi = 0.76) from meso-LA by [(BDI)ZnO Pr]2. Gibson and coworkers recently developed a class of (salan)AlX catalysts (VI). They observed a dramatic switch in tacticity, isotactic to heterotactic component ratio, upon small changes to the substituents on the salan ligand. A remarkable range of PLA microstructure (Pi =

0.21 to 0.96) were obtained in the polymerization of rac-LA.96

21

tBu

t NO Bu Al OR N O N O N O Al OMe Al OiPr

N O N O tBu I II t Spassky Bu Baker & Smith Spassky III Nomura Coates

t tBu Bu R2

t tBu Bu R2 N O N O R1 N O Al OiPr Al R Al Me N O N O N O t tBu Bu R1 R2

t tBu Bu R2 IV V VI Chen Feijen Gibson

Figure 1.8. Salen and salan Al catalysts for selective polymerization of lactide.

22 This thesis consists of two parts, poly(propylene carbonate) (Chapter 2 and 3) and polylactide (Chapter 4). Chapter 2 mainly reports on our studies of the PPC microstructure, Chapter 3 provides the insights into the PO/CO2 copolymerization mechanism by using single-site catalysts. In Chapter 4 we present our studies on the stereocontrol mechanism in ring-opening polymerization of lactide by using (salen)AlOR catalysts.

References:

(1) Bruntland, G.; Editor Our Common Future: The World Commission on

Environment and Development; Oxford University Press, Oxford, 1987.

(2) Charpentier, R. R. Science 2002, 295, 1470.

(3) Young, S. Nature 2001, 414, 487-488.

(4) Clark, J. H.; Editor Chemistry of Waste Minimization; Blackie Academic,

Glasgow, UK, 1995.

(5) MORI. The Public Image of the Chemical Industry; Research study conducted for

the Chemical Industries Association. MORI, London, 1999.

(6) Anastas, P.; Warner, J. Green Chemistry: Theory and Practice; Oxford University

Press, Oxford, UK, 1998.

(7) Goodstein, D. Out of Gas: The End of the Age of Oil; W. W. Norton & Company,

New York, 2004.

(8) Tullo, A. Chem. Eng. News 2003, 81, 26-27.

(9) Wilks, E. S. Industial Polymers Handbook; WILEY-VCH, Germany, 2001.

23 (10) Anstas, P. T.; Bickart, P. H.; Kirchhoff, M. M. Designing Safer Polymers;

WILEY-INTERSCIENCE, New York, NY, 2001.

(11) "A Vision to Enhance U.S. Economic Security through Renewable Plant/Crop-

Based Resource Use," Plant/Crop-Based Renewble Resources 2020, http://www.oit.doe.gov/agriculture/pdfs/ag_vision.pdf.

(12) Perez, J.; Munoz-Dorado, J.; de la Rubia, T.; Martinez, J. Int. Microbiol. 2002, 5,

53-63.

(13) Mohanty, A. K.; Misra, M.; Hinrichsen, G. Macromol. Mater. Eng. 2000,

276/277, 1-24.

(14) Poirier, Y. Nat. Biotechnol. 1999, 17, 960-961.

(15) Lee, S. Y. Trends Biotechnol. 1996, 14, 431-438.

(16) Liddell, J. M. Special Publication - Royal Society of Chemistry 1992, 103, 10-25.

(17) Taguchi, S.; Doi, Y. Macromol. Biosci. 2004, 4, 145-156.

(18) Dartee, M.; Lunt, J.; Shafer, A. Chem. Fibers Int. 2000, 50, 546,548,551.

(19) Gruber, P.; O'Brien, M. Biopolymers 2002, 4, 235-250.

(20) Ritter, S. K. Chem. Eng. News 2002, 80, 26-30.

(21) Holder, J. V. In Handbook of Green Chemistry and Technology; Blackwell,,

2002, pp 28-55.

(22) Xiaoding, X.; Moulijn, J. A. Energy Fuels 1996, 10, 305-325.

(23) Beckman, E. J. J. Supercrit. Fluids 2004, 28, 121-191.

(24) http://www.epa.gov/greenchemistry/presgcc.html.

(25) Rokicki, A.; Kuran, W. J. Macromol. Sci., Rev. Macromol. Chem. 1981, C21,

135-186.

24 (26) Super, M. S.; Beckman, E. J. Trends Polym. Sci. 1997, 5, 236-240.

(27) Inoue, S.; Koinuma, H.; Tsuruta, T. Makromol. Chem. 1969, 130, 210-220.

(28) Inoue, S.; Koinuma, H.; Tsuruta, T. J. Polym. Sci., Pol. Lett. 1969, 7, 287-292.

(29) Ree, M.; Bae, J. Y.; Jung, J. H.; Shin, T. J. Korea Polym. J. 1999, 7, 333-347.

(30) Darensbourg, D. J.; Yarbrough, J. C.; Ortiz, C.; Fang, C. C. J. Am. Chem. Soc.

2003, 125, 7586-7591.

(31) Kuran, W.; Gorecki, P. Makromol. Chem. 1983, 184, 907-912.

(32) Kuran, W.; Listos, T. Macromol. Chem. Phys. 1994, 195, 1011-1015.

(33) Chisholm, M. H.; Navarro-Llobet, D.; Zhou, Z. Macromolecules 2002, 35, 6494-

6504.

(34) Coates, G. W. Chem. Rev. 2000, 100, 1223-1252.

(35) Coates, G. W. J. Chem. Soc., Dalton Trans. 2002, 467-475.

(36) Aida, T.; Inoue, S. Acc. Chem. Res. 1996, 29, 39-48.

(37) Aida, T.; Inoue, S. Porphyrin Handbook 2000, 6, 133-156.

(38) Inoue, S. J. Macromol. Sci., Chem. 1988, A25, 571-582.

(39) Inoue, S. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 2861-2871.

(40) Aida, T.; Ishikawa, M.; Inoue, S. Macromolecules 1986, 19, 8-13.

(41) Jung, J. H.; Ree, M.; Chang, T. J. Polym. Sci., Part A: Polym. Chem. 1999, 37,

3329-3336.

(42) Kruper, W. J.; Dellar, D. D. J. Org. Chem. 1995, 60, 725-727.

(43) Paddock, R. L.; Hiyama, Y.; McKay, J. M.; Nguyen, S. T. Tetrahedron Letters

2004, 45, 2023-2026.

(44) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421-431.

25 (45) Darensbourg, D. J.; Yarbrough, J. C. J. Am. Chem. Soc. 2002, 124, 6335-6342.

(46) Hansen, K. B.; Leighton, J. L.; Jacobsen, E. N. J. Am. Chem. Soc. 1996, 118,

10924-10925.

(47) Darensbourg, D. J.; Mackiewicz, R. M.; Rodgers, J. L.; Phelps, A. L. Inorg.

Chem. 2004, 43, 1831-1833.

(48) Eberhardt, R.; Allmendinger, M.; Rieger, B. Macromol. Rapid Commun. 2003,

24, 194-196.

(49) Qin, Z.; Thomas, C. M.; Lee, S.; Coates, G. W. Angew. Chem., Int. Ed. Engl.

2003, 42, 5484-5487.

(50) Lu, X.; Wang, Y. Angew. Chem. Int. Ed. 2004, 43, 3574-3577.

(51) Kuran, W.; Rokicki, A.; Pasynkiewicz, S. J. Organomet. Chem. 1978, 157, 135-

143.

(52) Kuran, W.; Pasynkiewicz, S.; Skupinska, J. Makromol. Chem. 1977, 178, 47-54.

(53) Kuran, W.; Pasynkiewicz, S.; Skupinska, J.; Rokicki, A. Makromol. Chem. 1976,

177, 11-20.

(54) Allen, S. D.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc.

2002, 124, 14284-14285.

(55) Cheng, M.; Moore, D. R.; Reczek, J. J.; Chamberlain, B. M.; Lobkovsky, E. B.;

Coates, G. W. J. Am. Chem. Soc. 2001, 123, 8738-8749.

(56) Darensbourg, D. J.; Wildeson, J. R.; Yarbrough, J. C.; Reibenspies, J. H. J. Am.

Chem. Soc. 2000, 122, 12487-12496.

(57) Nakano, K.; Nozaki, K.; Hiyama, T. J. Am. Chem. Soc. 2003, 125, 5501-5510.

(58) Chisholm, M. H.; Zhou, Z., Unpublished results.

26 (59) Datta, R.; Tsai, S.-P. ACS Symposium Series 1997, 666, 224-236.

(60) Benicewicz, B. C.; Hopper, P. K. J. Bioactive & Comp. Polym. 1990, 5, 453-472.

(61) Benicewicz, B. C.; Hopper, P. K. J. Bioactive & Comp. Polym. 1991, 6, 64-94.

(62) Slager, J.; Domb, A. J. Adv. Drug Delivery Rev. 2003, 55, 549-583.

(63) Huh, K. M.; Cho, Y. W.; Park, K. Drug Delivery Technology 2003, 3, 42, 44-49.

(64) Gunatillake, P. A.; Adhikari, R. Eur. Cells Mater. 2003, 5, 1-16.

(65) Drumright, R. E.; Gruber, P. R.; Henton, D. E. Adv. Mater. 2000, 12, 1841-1846.

(66) Penczek, S.; Duda, A.; Kowalski, A.; Libiszowski, J.; Majerska, K.; Biela, T.

Macromol. Symp. 2000, 157, 61-70.

(67) Schwach, G.; Coudane, J.; Engel, R.; Vert, M. Polym. Int. 1998, 46, 177-182.

(68) Kasperczyk, J. Polymer 1996, 37, 201-203.

(69) Baran, J.; Duda, A.; Kowalski, A.; Szymanski, R.; Penczek, S. Macromol. Symp.

1997, 123, 93-101.

(70) Duda, A.; Penczek, S. Macromolecules 1990, 23, 1636-1639.

(71) Witzke, D. R.; Kolstad, J. J.; Narayan, R. Macromolecules 1997, 30, 7075-7085.

(72) O'Keefe, B. J.; Hillmyer, M. A.; Tolman, W. B. J. Chem. Soc., Dalton Trans.

2001, 2215-2224.

(73) Nakano, K.; Kosaka, N.; Hiyama, T.; Nozaki, K. Dalton Trans. 2003, 4039-4050.

(74) Chisholm, M. H.; Eilerts, N. W. Chem. Commun. 1996, 853-854.

(75) Chisholm, M. H.; Eilerts, N. W.; Huffman, J. C.; Iyer, S. S.; Pacold, M.;

Phomphrai, K. J. Am. Chem. Soc. 2000, 122, 11845-11854.

(76) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.; Lobkovsky, E. B.;

Coates, G. W. J. Am. Chem. Soc. 2001, 123, 3229-3238.

27 (77) Chisholm, M. H.; Huffman, J. C.; Phomphrai, K. J. Chem. Soc., Dalton Trans.

2001, 222-224.

(78) Gibson, V. C.; Marshall, E. L.; Navarro-Llobet, D.; White, A. J. P.; Williams, D.

J. J. Chem. Soc., Dalton Trans. 2002, 4321-4322.

(79) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.; Williams, D. J.

Dalton Trans. 2004, 570-578.

(80) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.; Williams, D. J.

Chem. Commun. 2001, 283-284.

(81) Williams, C. K.; Breyfogle, L. E.; Choi, S. K.; Nam, W.; Young, V. G., Jr.;

Hillmyer, M. A.; Tolman, W. B. J. Am. Chem. Soc. 2003, 125, 11350-11359.

(82) Williams, C. K.; Brooks, N. R.; Hillmyer, M. A.; Tolman, W. B. Chem. Commun.

2002, 2132-2133.

(83) Spinu, M.; Jackson, C.; Keating, M. Y.; Gardner, K. H. J. Macromol. Sci., Pure

Appl. Chem. 1996, A33, 1497-1530.

(84) Yui, N.; Dijkstra, P. J.; Feijen, J. Makromol. Chem. 1990, 191, 481-488.

(85) Tsuji, H.; Ikada, Y. Polymer 1999, 40, 6699-6708.

(86) Spassky, N.; Wisniewski, M.; Pluta, C.; Le Borgne, A. Macromol. Chem. Phys.

1996, 197, 2627-2637.

(87) Wisniewski, M.; Le Borgne, A.; Spassky, N. Macromol. Chem. Phys. 1997, 198,

1227-1238.

(88) Radano, C. P.; Baker, G. L.; Smith, M. R., III J. Am. Chem. Soc. 2000, 122, 1552-

1553.

28 (89) Ovitt, T. M.; Coates, G. W. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 4686-

4692.

(90) Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc. 2002, 124, 1316-1326.

(91) Nomura, N.; Ishii, R.; Akakura, M.; Aoi, K. J. Am. Chem. Soc. 2002, 124, 5938-

5939.

(92) Tang, Z.; Chen, X.; Pang, X.; Yang, Y.; Zhang, X.; Jing, X. Biomacromolecules

2004, 5, 965-970.

(93) Zhong, Z.; Dijkstra, P. J.; Feijen, J. Angew. Chem., Int. Ed. Engl. 2002, 41, 4510-

4513.

(94) Zhong, Z.; Dijkstra, P. J.; Feijen, J. J. Am. Chem. Soc. 2003, 125, 11291-11298.

(95) Majerska, K.; Duda, A. J. Am. Chem. Soc. 2004, 126, 1026-1027.

(96) Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; White, A. J. P.; Williams, D. J. J.

Am. Chem. Soc. 2004, 126, 2688-2689.

29

CHAPTER 2

POLY(PROPYLENE CARBONATE) MICROSTRUCTURAL STUDIES BY USING

POLYMERS PREPARED BY A ZINC GLUTARATE CATALYST AND MODEL

OLIGOETHER CARBONATES

2.1 Introduction

It is well known that the physical properties of polymers are directly related to their microstructures, including both regiochemical and stereochemical sequences.

Meanwhile the microstructure of a polymer bears a memory of the reaction pathway leading to its formation. Among various of analytical techniques, NMR spectroscopy is particularly useful in the microstructural analysis of polymers.1,2 The interpretation of

NMR spectra of polymers requires knowledge of accurate assignments of the NMR signals. We were struck by the fact that we had little or no basis upon which to investigate the regio- and stereochemistry of PPC. Prompted by these considerations, we decided to start our research in this field from investigating the microstructure of PPC by using NMR spectroscopy. We report here our studies on the regio- and stereochemistry assignments in PPC’s 13C{1H} NMR Spectrum by analyzing real PPC samples prepared

30 from a zinc glutarate catalyst, as well as using oligoether carbonates as model compounds.

2.2 Results and Discussion

2.2.1. 13C{1H} NMR Spectra of PPC

Only one previous literature report has dealt with the microstructure of PPC and the focus of that work was on the carbonate carbon.3 Three regions in the 13C{1H} NMR spectrum of the carbonate carbon were assigned as HT, TT and HH as shown in Figure

2.1. These junctions are shown in Scheme 2.1 and, given that the PPC chain was formed from rac-PO, each regiosequence can be further broken down to i or s stereosequences.

As is evident from an inspection of Figure 2.1, there are more than two resonances in the

HT and HH regions, and this implies that the carbonate carbon must be sensitive to more than its adjacent PO units.

13 1 Figure 2.1. C{ H} (100 MHz, CDCl3) NMR spectrum of the C=O region of PPC reported in ref 3.

31 We have no reason to challenge this original assignment made at the diad level, and we build on this in the studies reported herein. We have examined the PPC prepared from a zinc glutarate catalyst at 13C frequencies of 150 MHz. Three different samples of

PPC were employed, namely those derived from rac-PO, 50:50 rac + S-PO and 99% S-

PO, having Mn of 17 000 to 67 000 Daltons and PDIs of 7.0 to 14.0.

O O

O * O O O O O HT O O O

O * O O O O O HH O

O O

O * O O O O O TT O

Scheme 2.1. Three possible regiosequences of PPC at the diad level when looking at the central carbonate carbon.

2.2.1.1 The carbonate carbon region. The 13C{1H} NMR spectra of the carbonate carbon signals are shown in Figure 2.2 for the three different samples. Details pertaining to concentration and spectra acquisition are given in the Experimental Section.

The ratio of the TT:HT:HH junctions is on the order 1:3:1. The HT junctions are thus favored over their statistical occurrence, and this indicates that the ring opening of PO occurs with some regiospecificity in the formation of PPC. This is not altogether surprising inasmuch as the zinc glutarate catalyst ring opens PO (in the absence of CO2)

32 to give regioregular PPO (HTHTHT) as shown in Figure 2.3. In the case of PPO formation, we anticipate a transition state for ring opening involving backside attack on a zinc bound PO molecule by a neighboring alkoxide group which in turn is bound to a zinc (II) center (Scheme 2.2). In this way the regioregular (HTHTHT) PPO is formed, and in coordinate catalysis this has been seen to favor the formation of isotactic junctions when the PO monomer is racemic.4 In the case of ring opening of PO in PPC formation, a carbonate is the nucleophile. This is less sterically demanding and apparently less discriminating in its attack on a PO carbon. From the observed TT:HT:HH junctions, we can estimate that the relative attack on a tail, T, the CH2 carbon, vs the head, H, the CH carbon, to be in the ratio of 3:1 (Appendix A).

The simplest of the spectra shown in Figure 2.2 arises from the copolymerization of S-PO and CO2. Each region (HH, HT and TT) shows more than one resonance providing information concerning the possible regiosequences that occur at the tetrad level. As shown in the drawings in Scheme 2.3, there are four regiosequences with central H–T junctions that are distinguishable by NMR for the carbonate carbons. The relative abundances of these regiosequences can be calculated on the basis of the 3:1 preference for HT over HH or TT junction formation. In this way we can assign the predominant carbonate signal of PPC derived from S-PO to the iii tetrad of the

(HT)(HT).(HT)(HT) sequence, which must, by accidental degeneracy, occur with one of the other less abundant regiosequences (HT)(HT).(HT)(TH) or (TH)(HT).(HT)(HT). The minor signal is assigned to one of the latter and the least favored (TH)(HT).(HT)(TH) sequence. Polymerization of the 50:50 mixture of S-PO + rac-PO in the formation of PPC will favor the eight stereosequences of the (HT)(HT).(HT)(HT) tetrad as follows: iii

33 (0.320), iis (0.117), sii (0.117), iss (0.117), ssi (0.117), isi (0.070), sis (0.070), and sss

(0.070). See Appendix I. However, because other regiosequences are in combination roughly equally probable to the major (HT)(HT).(HT)(HT) tetrad, we are unable to make any further unequivocal assignments.

a. HT

TT

HH

ppm b. HT

TT

HH

ppm c.

HT

TT HH

ppm

Figure 2.2. 13C{1H} (150 MHz) NMR spectra of C=O region of PPC made from (a) rac-

PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst.

34

ii i

is/si ss s

76.0 75.5 75.0 74.5 74.0 73.5 73.0 72.5 72.0 71.5 ppm

Figure 2.3. 13C{1H} (100MHz) NMR spectrum showing the methine and methylene carbons of PPO prepared with a zinc glutarate catalyst.

Zn O Zn O OR

Zn OR

Scheme 2.2. Ring opening of PO by a backside attack mechanism.

The other carbonate carbon signals arise from TT and HH junctions at the diad level. Again, assuming backside attack, we can assign the TT region as s at the diad level and the HH region similarly. This assignment is confirmed by analyzing model carbonate compounds, vida infra. Inspection of the HH region clearly shows evidence of several 35 partially overlapping resonances and we can make tentative assignments at the tetrad level based on knowledge of the preference for TH/HT formation.

Diad level Tetrad level

(HT)(HT).(HT)(HT) 1 (HT)(HT).(HT)(TH) 2 (HT).(HT) (TH)(HT).(HT)(HT) 3

(TH)(HT).(HT)(TH) 4

(HT)(TH).(TH)(HT) 5 = 4

(HT)(TH).(TH)(TH) 6 = 2 (TH).(TH) (TH)(TH).(TH)(HT) 7 = 3

(TH)(TH).(TH)(TH) 8 = 1

Scheme 2.3. Eight possible regiosequences for carbonate carbons (*) in PPC with central

HT junctions at the tetrad level. (4, 5), (2, 6), (3, 7) and (1, 8) are four distinguishable sequences by NMR spectroscopy

2.2.1.2 The methine, methylene, and methyl carbon regions. The assignments of the carbon signals derived from PO are complementary to those of the carbonate carbon. However, because of the structure of the polymer their sensitivity is examined at the triad level. There are eight possible regiosequences for CH, CH2 and CH3 carbons at

36 the triad level as the polymer is inherently asymmetric. Only four regiosequencies are distinguishable by NMR spectroscopy. These are shown in Scheme 2.4–2.6.

O O O 51'3 1 O 3' O 5' 4 O 2 O * 2' 4' 6' O O O (HT)(H.T)(HT) (5, 5') 1

(HT)(H.T)(TH) (5, 6') 2 (H.T) (TH)(H.T)(HT) (4, 5') 3

(TH)(H.T)(TH) (4, 6') 4

O O 6' 2 O 4' 2' * O O 4 5' O 3' O 1' 1 3 5 O O O (HT)(T.H)(HT) (6', 4) 5 = 4

(HT)(T.H)(TH) (6', 5) 6 = 2 (T.H) (TH)(T.H)(HT) (5', 4) 7 = 3

(TH)(T.H)(TH) (5', 5) 8 = 1

Scheme 2.4. Eight possible regiosequences for methine carbons (*) in PPC at the triad level. (4, 5), (2, 6), (3, 7) and (1, 8) are four distinguishable sequences by NMR spectroscopy.a (a The numbers in the brackets indicate the positions of the neighboring methine groups for that regiosequence.)

37 The 13C{1H} spectrum derived from S-PO reveals four regiosequences as shown in Figure 2.4. The most abundant signal at δ 72.34 ppm can be assigned to the most abundant regiosequence (HT)(H.T)(HT) (Scheme 2.4) which, assuming backside attack must be ii. Of the three other signals we notice that two are upfield at ca. δ 72.1 ppm while the other is very close to the signal of the regioregular sequence. All are roughly of the same integral intensity so an unequivocal assignment cannot be made. We are, however, inclined toward the view that the two upfield signals arise from sequences that have TH junction (Scheme 2.4).

In going from S-PO to rac-PO the overlap of the resonances is very pronounced.

Nevertheless, we feel we can, on the basis of deconvolution, assign the ii, ss and is/si signals of the most abundant triad sequence (HT)(H.T)(HT).

The methylene region of PPC derived from S-PO shows just two resonances of integral intensity 37% and 63% (Figure 2.5). This can only arise from accidental degeneracy of the major (HT)(H.T)(HT) regiosequence signal with one of the minor. On the basis of considerations of bond distances, we propose that the regiosequence with a

HH junction is coincident with the regioregular sequence. The upfield methylene resonance is thus assigned to the two regiosequences with TT junctions. On the basis the statistics of HT ring opening being favored in the ratio 3:1, as deduced from the carbonate carbon signals, we predict the ratio of 37:63, which fits exactly that observed.

Once again in going from polymerizations involving S-PO to rac-PO, new resonances appear which must correspond to s sequences. It is, however, not possible to correlate these reliably with the regiosequences.

38 a.

72.50 72.40 72.30 72.20 72.10 72.00 71.90 71.80 ppm b.

72.50 72.40 72.30 72.20 72.10 72.00 71.90 71.80 ppm c.

72.50 72.40 72.30 72.20 72.10 72.00 71.90 71.80 ppm

Figure 2.4. 13C{1H} (150 MHz) NMR spectra of CH region of PPC made from (a) rac-

PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst.

39

O O 2 O 6' 4' 2' * O O 4 5' O 3' O 1' 1 3 5 O O O (HT)(H.T)(HT) (6', 1', 4) 1

(TH)(H.T)(HT) (5', 1', 4) 2 (H.T) (HT)(H.T)(TH) (6', 1', 5) 3

(TH)(H.T)(TH) (5', 1', 5) 4

O O O 5 3 1 1' O 3' O 5' 4 O 2 O * 2' 4' 6' O O O

(HT)(T.H)(HT) (5, 1', 5') 5 = 4

(HT)(T.H)(TH) (5, 1', 6') 6 = 3 (T.H) (TH)(T.H)(HT) (4, 1', 5') 7 = 2

(TH)(T.H)(TH) (4, 1', 6') 8 = 1

Scheme 2.5. Eight possible regiosequences for methylene carbons (*) in PPC at the triad

level. (4, 5), (3, 6), (2, 7) and (1, 8) are four distinguishable sequences by NMR

spectroscopy.a (a The numbers in the brackets indicate the positions of the neighboring methine groups for that regiosequence.)

40 a.

69.60 69.40 69.20 69.00 68.80 68.60 68.40 ppm b.

69.60 69.40 69.20 69.00 68.80 68.60 68.40 ppm c.

69.60 69.40 69.20 69.00 68.80 68.60 68.40 ppm

13 1 Figure 2.5. C{ H} (150 MHz) NMR spectra of CH2 region of PPC made from (a) rac-

PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst.

41 The methyl carbon region obtained from PPC prepared with S-PO shows only three signals in the integral ratio 17:62:21 in Figure 2.6. The major signal can once again be assigned to the ii triad of the major regiosequence (HT)(H.T)(HT) with a coincidence with one of the minor regioirregular triads. This would have a calculated combined integral intensity of 62.5%. In changing from S-PO to rac-PO, only the downfield signals increase in intensity. While it is again clear that there is considerable overlap (accidental degeneracy), it would seem that the is/si and ss stereosequences of the regioregular

(HT)(H.T)(HT) triad must be embodied within this large and broad upfield signal.

O * O O 6 4 2 2' O 4' O 6' 5 O 3 O 1 3' 5' 7' O O O (HT)(H.T)(HT) (6, 6') 1 0.75 x 0.75 x 0.75 = 0.422

(HT)(H.T)(TH) (6, 7') 2 0.75 x 0.75 x 0.25 = 0.141 (H.T) (TH)(H.T)(HT) (5, 6') 3 0.25 x 0.75 x 0.75 = 0.141

(TH)(H.T)(TH) (5, 7') 4 0.25 x 0.75 x 0.25 = 0.047

O O O 7' 5' 3' 1 O 3 O 5 6' O 4' O 2' 2 4 6 OO * O (HT)(T.H)(HT) (7', 5) 5 = 4 0.75 x 0.25 x 0.75 = 0.141

(HT)(T.H)(TH) (7', 6) 6 = 2 0.75 x 0.25 x 0.25 = 0.047 (T.H) (TH)(T.H)(HT) (6', 5) 7 = 3 0.25 x 0.25 x 0.75 = 0.047

(TH)(T.H)(TH) (6', 6) 8 = 1 0.25 x 0.25 x 0.25 = 0.016 Scheme 2.6. Eight possible regiosequences for methyl carbons (*) in PPC at the triad level. (4, 5), (2, 6), (3, 7) and (1, 8) are four distinguishable sequences by NMR spectroscopy.a (a The numbers in the brackets indicate the positions of the neighboring methine groups for that regiosequence.)

42

a.

16.60 16.50 16.40 16.30 16.20 16.10 16.00 15.90 15.80 15.70 ppm b.

16.60 16.50 16.40 16.30 16.20 16.10 16.00 15.90 15.80 15.70 ppm c.

16.60 16.50 16.40 16.30 16.20 16.10 16.00 15.90 15.80 15.70 ppm

13 1 Figure 2.6. C{ H}(150 MHz) NMR spectra of CH3 region of PPC made from (a) rac-

PO, (b) 50% rac-PO and 50% S-PO, (c) S-PO, employing a zinc glutarate catalyst.

43 2.2.2. Regioregular and Regioirregular Oligoether Carbonates

A common impurity in certain samples of PPC is the occurrence of -rich segments. As shown in Figure 2.7, the 13C{1H} signals in the carbonate region reveal the existence of carbonate carbons flanked by more than one ether unit.5 In order to make further assignments on those PPC’s 13C{1H} carbonate signals and obtain helpful information on the polymer formation mechanism, we synthesized some model compounds and report herein the findings of these studies.

HT

HT’

TT HH’ TT’ i s HH

155.20 155.00 154.80 154.60 154.40 154.20 154.00 153.80 153.60 153.40 ppm

13 1 Figure 2.7. C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of PPC made from rac-PO employing the (TPP)AlCl/EtPh3PBr catalytic system. HH’, HT’ and

TT’ are ether-rich carbonate signals. The i and s assignment in the HH’ region is proposed on the basis of the work reported herein.

2.2.2.1. Synthesis. The general synthetic route to these ether carbonates is shown

4,6,7 in Scheme 2.7. The etherols Et(OCH2CHMe)nOH, where n = 1, 2, 3, and 4, were separated by spinning band column distillation. For n = 2, 3, and 4, the distillations were

44 carried out under reduced pressure. The ring opening of propylene oxide (PO) in Scheme

2.7 involves backside attack, and by selection of rac-PO and S-PO, we can determine the stereochemical sequences in the etherols. Then in the reaction with triphosgene by combinations of etherols derived from S-PO and rac-PO, we obtained carbonates containing statistical (i + s) and enriched (i) stereosequences.

O 1) KOC2H5 C H OH 3 2 5 O 2) HCl n n=1, 2, 3, 4, ... Propylene oxide separation

O C pyridine Cl3CO OCCl3

O O C H O O C H 2 5 n C n 2 5 O

Scheme 2.7. The synthetic route for oligoether carbonates.

2.2.2.2. NMR spectra and assignments. The 13C{1H} signals of the carbonate carbon for n = 1 and 2 are shown in Figure 2.8. For n = 2, it is clear that the 13C carbonate signal shows tetrad sensitivity, and furthermore, since this molecule has a regiosequence

(TH)(TH)·(HT)(HT) (H = methine group; T = methylene group), the signal for iis and sii cannot be distinguished. Similarly, the ssi and iss sequence is the same.

45

O C OC2H5 C2H5 O O O 1 1 a. i s

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

O C OCH C2H5 O O 2 5 2O 2 b. iis/sii iss/ssi

iii isi sis sss

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

c. iii

iis/sii iss/ssi isi

sis sss

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

13 1 Figure 2.8. C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regioregular oligoether carbonate compounds (n = 1, 2). Carbonate compounds in (b) were made from etherol derived from rac-PO, while the ones in (c) were prepared from etherol mixtures (1:1) derived separately from rac-PO and S-PO.

46

The spectra for n = 3 and 4 are shown in Figure A.2 in Appendix A. For n = 3, there are 20 possible stereosequences, but no more than 12 signals are partially resolved.

Thus, for n ≥ 4 we can only assign the signal at the diad level, i or s. The spectra obtained for polyether carbonates Et(PO)nOCO2(PO)nEt where n = ~10 and ~30, are shown in

Figure A.3 in Appendix A. These spectra further demonstrate that only diad sensitivity is seen.

From the use of commercial sources of regioirregular etherols,

CH3(PO)n(OCH2CHMe)OH where n = 1 and 2, we have made ether carbonates that have

HH junctions at the carbonate carbon but are otherwise regioirregular. The 13C{1H} spectra of the carbonate carbon are shown in Figure 2.9. For n = 1, we observed tetrad sensitivity and can identify the stereosequences associated with the regioregular component (see Appendix A for details). Once again, however, as the chain length increases on each side of the carbonate, the sensitivity is reduced so that only i and s regions can be unequivocally identified. Finally, we should note that the chemical shift difference between the unique resonances is quite small, typically less than 0.1 ppm.

47

O O C O O O H3CO OCH3 1 1 a.

iss /ssi sss iii isi

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

O O C O O O H3CO OCH3 2 2 b.

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

13 1 Figure 2.9. C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regioirregular oligoether carbonate compounds (n = 1, 2).

2.2.2.3. Structural considerations. As an aid to the interpretation of the NMR data, we have employed calculations using density functional theory (DFT) and the

Gaussian 98 suite of programs to examine the energies of various conformers.8,9 As a starting point, we examined dimethyl carbonate MeOCO2Me, which can exist in three local minima, namely cis–cis, trans–trans, and cis–trans conformers as shown in A, B, and C below, respectively. Experimentally, it is known that cis–cis is thermodynamically

48 preferred, and a related carbonate molecule has been structurally characterized by single- crystal crystallography in this cis–cis geometry.10 Previous DFT calculations, analogous to the ones reported here, also have revealed the preference for the cis–cis geometry, and our results closely parallel these earlier calculations.11 The cis–trans isomer is calculated to be higher in energy by 3.0 kcal/mol while the trans–trans is higher by 21.1 kcal/mol.

On the basis of these energy differences, the cis–cis isomer will constitute more than 99% of the mixture at room temperature, 298 K. Consequently, we started our calculations of the compounds MeOCH2CHMeOCO2CHMeCH2OMe with the cis–cis geometry.

O O O H C CH C H3C C 3 C 3 O O O O O O

H3C CH3 CH3 C A B

The results of the gas-phase calculations for the RR isomer, the molecule with the isotactic carbonate HH junction, are summarized in Figure 2.10. The conformations depicted in Figure 2.10 are viewed down the C–O bond of the carbonate, and all of these conformers represent local minima as confirmed by vibrational frequency analyses. The nomenclature for the calculated six minima indicates whether or not the molecule has C2 or C1 symmetry. Thus, structures [1,1], [2,2], and [3,3] have C2 symmetry whereas the combinations [2,3], [1,3], and [1,2] contain two different halves on either side of the molecule and have C1 symmetry. In the latter cases, only the gross geometry of the 1, 2, or 3 fragment is present in the respective [1,1], [2,2], or [3,3] isomer. The bond angles and bond lengths are fully optimized for each structure. On the basis of these gas-phase structures, we can estimate that [1,1] and [1,3] isomers constitute >95% of the species

49 present at 298 K, and the [3,3] and [1,2] isomers are less than 2% each. The structures depicted in Figure 2.10 all have an anti conformation of the ether oxygen with respect to the carbonate O–CHMe bond as shown. The gauche conformer is higher in energy by

0.94 kcal/mol in the case of the [1,1] isomer.

O MeOC OMe O O R R

O CI

l) 6 Me X X = - CH2OMe 5 H O CI 4 X H 3 O CI Me

nergy (kcal/mo 2 H Me

E 1 X 0 0123456[1,1] [2,2] [3,3] [1,2] [1,3] [2,3] Local minimum conformers Figure 2.10. Six local minimum conformers in the gas–phase calculations for RR isomers of MeOCH2CHMeOCO2CHMeCH2OMe.

The SR isomers are compared in Figure 2.11 for the cis–cis geometry about the carbonate. Again we show the calculated local minima, and here for the conformers [1,1],

[2,2], and [3,3], there is Cs symmetry but for all others only C1. In the Newman projections, we show only the S half of the molecule. Once again the MeO groups are anti to the carbonate O–CHMe bond. For the SR case, the gas-phase structures of the

[1,1], [1,3], and [3,3] conformers are very close in energy, differing by less than 0.15 kcal/mol, indicating that these would be the dominant species in equilibrium at 298 K.

50

O MeO C OMe O O S R O C 6 I X = - CH OMe l) X Me 2 5 H 4 3 O O C CI I H X nergy (kcal/mo 2 Me H E 1 X Me 0 0123456[1,1] [2,2] [3,3] [1,2] [1,3] [2,3] Local minimum conformers

Figure 2.11. Six local minimum conformers in the gas–phase calculations for SR isomers of MeOCH2CHMeOCO2CHMeCH2OMe.

We also investigated the solvent effects on the preferred conformations of the methoxy-terminated using the polarizable continuum model (PCM) in our calculations,12 and their respective energies in chloroform and benzene differed little from those calculated for the gas phase. These energies are compared in Table 2.1. This suggests that the dominant species present in the gas phase also exist in chloroform and benzene predominantly.

The molecules of MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me, considered as a small segment of the PPC chain, were also investigated in our calculations. The results of the calculations for RR isomers in both the gas phase and chloroform are summarized in

Figure 2.12. Because of the huge number of all possible conformations, we simplified the calculations by only dealing with those having C2 symmetry derived from [1,1] in Figure

51 2.10, and keeping all three carbonate groups in the cis–cis conformation. The Newman projections depicted in Figure 2.13 are viewed down the bonds of CH(Me)–O (1), CH2–

CH(Me) (2), and O–CH2 (3). The nomenclature for the calculated six minima in Figure

2.12 indicates a combination of those three views. For example, [1,3’,2”] represents that a half of this isomer has conformation 1 for the CH(Me)–O bond, 3’ for the CH2–CH(Me) and 2” for the O–CH2, and the entire isomer has a C2 symmetry. The bond angles and lengths are fully optimized. The relative energies of the six isomers are in a range 0–1.1 kcal/mol as shown in Figure 2.12, which indicates that all of them have contributions more or less to the Boltzmann average at 298 K. The relative energies for some isomers in chloroform differ significantly from the gas phase, in contrast to the molecules of

MeOCH2CHMeOCO2CHMeCH2OMe, which suggests that solvent effects will play a more important role on the conformations of this molecule.

conformers gas phase chloroform benzene ∆E population ∆E population ∆E population (kcal/mol) ratio (kcal/mol) ratio (kcal/mol) ratio RR [1,1] 0 1 0 1 0 1 [1,3] 0.050 0.918 0.077 0.879 0.056 0.909 SR [1,1] 0.010 1 0.004 1 0.012 1 [3,3] 0.131 0.815 0.080 0.880 0.099 0.863 [1,3] 0.027 0.971 0.005 0.998 0.014 0.997

Table 2.1. Calculated populations for the isomers of

MeOCH2CHMeOCO2CHMeCH2OMe in the gas phase, chloroform and benzene.

52

O 1 2 MeO O C O OMe O C C R O R 3 O O 2 gas phase chloroform l) 1.5

1

(kcal/mo nergy 0.5 E

0 0123456 Local Minimum Conformers

Figure 2.12. Six local minimum conformers in both the gas phase and chloroform calculations for RR isomers having C2 symmetry of

MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me. The conformers’ structures are according to Figure 2.13: 1, [1,1’,1”]; 2, [1,2’,2”]; 3, [1,3’,1”]; 4, [1,1’,2”]; 5, [1,2’,2”]; 6, [1,3’,2”].

The SR isomers of MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me are compared in Figure 2.14. Again, we only show the results of symmetric conformers (Cs) derived from [1,1] and [3,3] in Figure 2.11. We began with many initial starting geometries, but some potential structures optimized to a reduced number of final geometries. Indeed, we

53 obtained eight minima in the relative energy rang of 0–1.7 kcal/mol. Similarly to RR isomers, the relative energies for some SR isomers in chloroform were much different from the gas phase, as shown in Figure 2.14.

O 1 2 MeO O C O OMe C O O 3 C R O R O

1. O O O C CI CI I H Me Me X X H

X H Me (1) (2) (3) 2. OCOOMe H H I H I I Me H I I Me H I Me H H H H OCOOMe I MeOOCO I I OCOO- OCOO- OCOO- (1') (2') (3') 3. CHMe- CHMe- I MeOOC I O I O I I H IH H H MeOOC (1'') (2'') X = - CH2OCOOMe

Figure 2.13. Newman projections for the RR isomers of

MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me, viewing down the bonds of CH(Me)–O,

CH2–CH(Me), and O–CH2.

54

O 1 2 MeO O C O OMe C O O C S R 3 O O

2 gas phase

l) choroform 1.5

1

nergy (kcal/mo nergy 0.5 E

0 012345678

Local Minimum Conformers Figure 2.14. Eight local minimum conformers in both the gas phase and chloroform calculations for SR isomers having Cs symmetry of

MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me. The structures of conformers are as follows: 1, [1,1’,1”]; 2, [1,2’,2”]; 3, [1,3’,1”]; 4, [1,1’,2”]; 5, [1,3’,2”]; 6, [3,1’,1”]; 7,

[3,3’,1”]; 8, [3,1’,2”].

2.2.2.4. NMR calculations. With an attempt to simulate the 13C NMR chemical shifts of central carbonate carbons of the model compounds, we carried out NMR calculations employing the gauge-independent atomic orbital (GIAO) method available

13 in Gaussian. For the molecules of MeOCH2CHMeOCO2CHMeCH2OMe, the results are tabulated in Table 2.2. For the RR isomer, two carbonate carbon signals are predicted at

55 162.345 and 162.675 ppm in chloroform for the predominant local minima. Given that at room temperature, rotations about single bonds will be rapid on the NMR time scale, this predicts a single carbonate carbon resonance at 162.499 ppm. For the SR isomer where three local minima are predicted to contribute to the average carbon resonance of the carbonate carbon in chloroform, we find 162.377, 163.062, and 162.662 ppm. Given the relative population of these conformers at 298 K (1:0.880:0.998), the predicted carbonate carbon signal is 162.685 ppm. The calculated chemical shift separation for the carbonate carbon signals of the two isomers representing i and s junctions is 0.186 ppm, which may be compared to 0.042 ppm in CDCl3 solution. Moreover, the calculations predict that the

RR isomer should have a chemical shift upfield relative to the SR isomers, which is contrary to our experimental assignment in the model compounds (with the ethoxy-end groups) for i and s. It is probably unreasonable to expect our current level of computation of NMR chemical shifts to simulate with accuracy such small experimental variations.

For instance, the concentration of the solution (0.2 M CDCl3 solution used in NMR experiments), not simulated in the calculations, may have significant effects on the intermolecular chain interactions, which changes the relative energies and populations of the conformations and results in deviations in the calculated NMR chemical shifts from the experimental ones. However, the calculations do indicate that the conformations of molecules and the configurations of stereocenters can make differences in chemical shifts. Consistently, it is experimentally evident from Figure 2.8 that the carbonate carbon is sensitive to the stereocenters of the ring-opened PO units in a chain of the type

R(PO)nOCO2(PO)nR, where n = 2. For n > 2, the rapidly increasing number of stereoisomers leads to overlapping resonances, and only diad sensitivity can be claimed

56 by 13C NMR spectroscopy. In the case of longer chain molecules, such as

R(PO)10OCO2(PO)10R and R(PO)30OCO2(PO)30R, only single broad resonances assignable to i and s are observed (Appendix A, Figure A.3). For the isotactic prepared from S-PO, the half–width of the signal for i is less than 0.01 ppm for n ~ 10.

This is not significantly broader than the half width seen for the iii tetrad signal where n =

2, which indicates that the time averaging of various conformers in the long chain oligomer remains fast on the NMR time scale.

conformers gas phase chloroform (ppm) (ppm) RR [1,1] 162.529 162.345

[1,3] 162.820 162.675 a ave 162.668 162.499 SR [1,1] 162.429 162.377 [3,3] 163.352 163.062

[1,3] 162.903 162.662

ave 162.864 162.685 RR–SRb -0.196 -0.186

Table 2.2. Calculated NMR chemical shifts for the isomers of

MeOCH2CHMeOCO2CHMeCH2OMe in the gas phase and chloroform. a Statistical average chemical shift. b Chemical shift difference between RR and SR.

The NMR chemical shifts for MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me were also calculated, and the results are tabulated in Table 2.3. In the gas phase, for RR isomers, the average central carbonate carbon signal is predicted at 162.761 ppm, while

57 for the SR isomers, it is 162.766 ppm. In chloroform, the predicted chemical shifts for the central carbonate carbon are 162.741 ppm for RR and 162.167 ppm for SR.

conformers gas phase chloroform (ppm) (ppm) RR [1,1’,1”] 162.307 162.127 [1,2’,1”] 163.052 162.879 [1,3’,1”] 163.257 162.968 [1,1’,2”] 161.967 160.488 [1,2’,2”] 163.436 163.665 [1,3’,2”] 163.291 162.695 avea 162.761 162.741 SR [1,1’,1”] 162.468 161.976 [1,2’,1”] 162.995 161.951 [1,3’,1”] 163.233 162.645 [1,1’2”] 161.941 161.510 [1,3’,2”] 163.167 162.336 [3,1’,1”] 163.093 162.664 [3,3’,1”] 163.508 162.341 [3,1’,2”] 163.252 - ave 162.766 162.167 RR–SRb -0.005 0.574

Table 2.3. Calculated NMR chemical shifts for the isomers of

MeOCO2CH2CHMeOCO2CHMeCH2OCO2Me in the gas phase and chloroform. a Statistical average chemical shift. b Chemical shift difference between RR and SR.

58 2.2.3 PO/CO2 Copolymerization Catalyzed by a Zinc Glutarate Catalyst

2.2.3.1 PPC formation as a function of time. As noted in Chapter 1, one of the major problems in the production of PPC is that its formation is very slow and inefficient with respect to monomer to catalyst ratio, i.e., 70 g of PPC per 1 g of catalyst in 40 h represents the highest reported efficiency for heterogeneous catalysts. We have examined the formation of PPC in the zinc glutarate catalyzed reaction as a function of time. With a heterogeneous catalyst we always have to be concerned that the number of active sites will vary with catalyst preparation, i.e., with surface area and impurities. To minimize these effects in our studies, we have used data obtained from one specific catalyst preparation and we have stored the catalyst in a dry box with less than 1 ppm of O2/H2O.

Each catalyst run employed the same PO to zinc glutarate concentration (2 mL of PO and

o 100 mg of zinc glutarate catalyst), a CO2 pressure of 50 bar and temperature of 60 C.

The reactor was loaded in the dry box, transferred to the benchtop, purged with CO2 and then pressurized. Each reaction was stirred with the aid of a magnetic follower. The reactions were monitored as a function of time in the following way. At time t h, the reactor was cooled to room temperature and the pressure was slowly released. The reaction mixture was then examined by 1H NMR spectroscopy to determine or gain a reasonable estimate of the percent conversion of PPC along with the other impurities, namely PC and polyether linkages that have the appearance of PPO by 1H NMR spectroscopy. The data are summarized in Table 2.4.

A number of points are noteworthy. (1) The percent of PPC conversion with time increases to 90% at 40 h, but after 5–10 h 70% conversion is already achieved. (2) The percent of PC formation increases slowly with time whereas the percent of polyether

59 linkages remains essentially constant. (3) At high % PPC formation, the reactor contains the product as solid foam. (4) In the absence of CO2, PO is polymerized to regioregular

PPO at a rate comparable to PPC production (in the presence of CO2) during the initial 5–

10 h. The PDI of the PPO is also comparable to that of the PPC, ~7, suggesting many different reactive sites.

Reaction time PPC PC Polyether linkagea PO monomerb (h) (%) (%) (%) (%) 2.5 16.5 0.6 2.23 80.6

5 71.1 2.4 2.22 24.3

5c - - 72.4 27.6 10 68.8 3.8 2.17 25.2 20 77.4 4.4 2.17 16.0

40 90.5 5.7 2.15 1.6

Table 2.4. Percentage of PPC, PC, and polyether linkage formed in the copolymerization of rac-PO and CO2 as a function of time. a refers to two or more units of PO linked together. b refers to PO monomer left after the reactions, (c) refers to homopolymerization of PO in the absence of CO2.

We worked up the PPC by dissolving the foam (or viscous solution) in benzene and removing the zinc glutarate by centrifugation and filtration. We have examined the

PPC by GPC and, for the 2.5 h preparation, by MALDI–TOF/MS. Even at 16.5% conversion some of the PPC has a very high molecular weight ca. 200 000 Da (calibrated against polystyrene). In the mass spectrum of the low MW polymers present at 16.5%

60 conversion, we can identify the end groups as –OH and –H. Specifically, we can observe series in HO–[(PO)n-alt-(CO2)m]–H, where m = n–1, n–2, to n–5. These are all sodiated

+ ions HO–[(PO)n-alt-(CO2)m]–H·Na . See Figure 2.15.

Figure 2.15. MALDI–TOF mass spectrum of PPC prepared with a zinc glutarate catalyst

+ (t = 2.5 h). An = HO–[(C3H6O)n-alt-(CO2)n-1]–H·Na , n = 11, 12; Bn = HO–[(C3H6O)n-

+ + alt-(CO2)n-2]–H·Na , n = 12, 13; Cn = HO–[(C3H6O)n-alt-(CO2)n-3]–H·Na , n = 12, 13;

+ Dn = HO–[(C3H6O)n-alt-(CO2)n-4]–H·Na , n = 13, 14; En = HO–[(C3H6O)n-alt-(CO2)n-5]–

H·Na+, n = 13, 14.

On the basis of the data obtained from Table 2.4, we propose that the conversion of PO and CO2 to PPC is primarily limited by the number of active sites. Some sites must be very active since high MW polymer chains are formed early in the reaction. With time the average molecular weight does not increase significantly. Also after 20 h with less

61 than 16% PO remaining, the mixture is very viscous and conversion to PPC may be limited by mass transport.

o In the absence of CO2, the zinc glutarate polymerizes PO to PPO at 60 C in a regioregular manner (HTHTHT) and shows some preference in the polymerization of rac-PO for the formation of ii triads (Figure 2.3). By MALDI–TOF/MS we observe that

+ the PPO is –OH and –H terminated, HO–[(PO)n]–H·Na , with just a trace amount of

+ cycles [(PO)n]·Na (Figure 2.16).

Figure 2.16. MALDI–TOF mass spectrum of PPO prepared with zinc glutarate. An =

+ + HO–[(C3H6O)n]–H·Na , n = 28, 29; Bn = [(C3H6O)n]·Na , n = 28, 29.

2.2.3.2. Degradation of PPC. Since a heterogeneous catalyst may have many different active sites, it is important to determine whether or not the minor products detected by NMR, such as PC and polyethers, are formed separately or are formed competitively during PPC production. For example, by NMR spectroscopy, it is not easy 62 to distinguish between a small amount of PPO being present or PPC having some deficiency in carbonate such that there are sections in the polymer that have (PO)n units bounded by carbonates. To investigate the latter, we have degraded the PPC by reaction with LiOtBu in THF in the presence of tBuOH. Within 3 h at room temperature PPC was degraded to PC. The mechanism of the reaction presumably involves alkoxide ion attack on the carbonate carbon followed by backbiting. This effectively unzips the linear polymer producing the cyclic carbonate. A similar result will occur if a hydroxyl end group reacts with LiOtBu to give tBuOH and LiOP (Scheme 2.8). A polyether link, however, is not degraded by this mechanism, and PPO is inert to LiOtBu under the conditions described. The degraded polymer product has been examined by NMR, GPC and ESI–QTOF/MS. The major product ca. 98% was PC, and no long chain polyether was detected. However, by mass spectrometry we did detect some small oligomers of the form HO–[(PO)n(CO2)]–H, HO–[(PO)n-alt-CO2]2]–H and cycles [(PO)n(CO2)], where n =

3, 4, 5, 6 and trace amounts where n = 7 and 8. In addition, we can detect HO–[(PO)2]–

+ + H·Na and [(PO)3]·H as ions lacking in CO2 (Figure 2.17). This implies that the NMR evidence for polyether formation arises from (PO)n units (n ≥ 2) in the PPC chain. This conclusion is also supported by the observed mass spectrum of low molecular weight

PPC which shows the PO-rich chains (Figure 2.15).

63

O O O O O P P OH P O O O O LiOtBu + LiOtBu O P - tBuO O O O P OLi + tBuOH O O O P OLi O LiO

O O P LiO O O O P - LiOP

O O - LiOP O O O

O O

Scheme 2.8. Two possible mechanisms of PPC degradation by LiOtBu.

Since the work of Kuran previously established that certain organozinc compounds could be catalysts for both the production and degradation of PPC,14,15 We questioned whether or not zinc glutarate could degrade PPC leading with time to the appearance of small quantities of PC. From the mass spectral data, we infer that a hydroxyl group is the initiator, and this Zn–OH group could behave similarly to LiOtBu.

To test this notion we allowed a commercial sample of PPC of MW ~ 50 000 Da to react

o with the zinc glutarate catalyst in CH2Cl2 at 60 C for 40 h. After this time we examined 64 the products by GPC and NMR spectroscopy. Some PC was found to have been formed ca. 2% based on PPC. More significantly, however, the molecular weight of the polymer was decreased, and there was evidence of smaller chains with MW ~ 4 000 Da

(polystyrene equivalent). However, after 1 week under similar conditions the PPC polymer showed little degradation. We can conclude that the surface active sites responsible for the production of PPC from PO and CO2 are not very active in PPC degradation, although we have found some evidence of PC formation and chain breaking.

Figure 2.17. ESI–QTOF mass spectrum of PPC degradation products by LiOtBu. Peak at

+ + m/z = 157 for HO–[(PO)2]–H·Na , m/z = 175 for [(PO)2]·H , m/z = 205 for [(PO)2-alt-

+ + (CO2)2]·H , m/z = 259 for HO–[(PO)3(CO2)1]–H·Na , m/z = 303 for HO–[(PO)3-alt-

+ + (CO2)2]–H·Na , m/z = 317 for HO–[(PO)4(CO2)1]–H·Na , m/z = 361 for HO–[(PO)4-alt-

+ + (CO2)2]–H·Na , m/z = 419 for HO–[(PO)5-alt-(CO2)2]–H·Na , m/z = 477, HO–[(PO)6-alt-

+ + (CO2)2]–H·Na , m/z = 547 for [(PO)6-alt-(CO2)4]·Na , m/z = 605 for [(PO)7-alt-

+ (CO2)4]·Na .

65

2.3. Conclusions

The zinc glutarate catalyst system is active in the polymerization of PO and in the copolymerization of PO and CO2. The polymerization of PO gives regioregular PPO which is enhanced in isotactic triads. In the formation of PPC we have found preference for the regioregular HT carbonate junctions, but the polymer contains roughly 30% regioirregular HH and TT junctions. By use of 13C{1H} NMR spectroscopy, we can observe tetrad and triad sensitivity for the carbonate and propylene carbon atoms, respectively, and by use of S-PO certain stereosequences have been assigned.

The model compounds, R(PO)nOCO2(PO)nR, where R = Me, Et, or H, and n = 1,

2, 3, 4, ~10, and ~30, were synthesized for PPC microstructural assignments in NMR studies. The 13C{1H} NMR investigations of those compounds showed that the carbonate carbon signals have both regio- and stereosensitivity at the diad and the tetrad levels to its adjacent PO ring-opened units. The calculations on the model compounds indicated that the carbonate groups exist predominantly at cis–cis geometries with more than one stable conformation for each molecule. In addition, the 13C chemical shifts predicted in the calculations were sensitive to the conformations of the molecules and the configurations of the stereocenters in PO ring-opened units.

The zinc glutarate catalyst has been found to contain some very active sites, Zn–

OH groups on the surface, for PPC formation as judged by the formation of high molecular weight PPC at short reaction times and low percent conversion. The catalyst is relatively inert to the degradation of PPC. The challenge thus remains to design a molecular system that mimics the active site and does not endear itself to backbitting and

66 PC formation, which appears to be the common mode of reaction of single-site metal alkoxide catalysts.

2.4. Experimental Section

All syntheses and solvent manipulations were carried out under an inert atmosphere using standard Schlenk-line and drybox techniques. Solvents were dried in the standard procedures. Racemic propylene oxide (Fisher) and S-propylene oxide (Alfa

Aesar) were distilled from calcium hydride. Zinc glutarate was synthesized according to

16 the literature from zinc oxide (Fisher) and glutaric acid (Fisher). CO2 (99.99%) was obtained from BOC Gases. Potassium ethoxide (Aldrich), triphosgene (Aldrich), and anhydrous 1,4-dioxane (Aldrich) were used as received. Di- and tri(propylene glycol) methyl (Aldrich), poly(propylene glycol) (typical Mn = 425) (Aldrich), and deuterated solvents were stored over 4 Å molecular sieves for 24 h prior to use.

NMR Experiments. 1H and 13C{1H} NMR experiments were carried out with a

Bruker DRX-500 (5 mm broad band probe) and a Bruker DRX-600 (5 mm broad band probe) spectrometers, operating at proton Larmor frequencies of 500 and 600 MHz, respectively. The parameters used in 13C{1H} NMR experiments on a Bruker DRX-600 spectrometer were as follows: number of data point, TD = 65536; sweep width, SWH =

1502 Hz; relaxation time, D1 = 2 s; and chemical shift range 0–200 or 150–160 ppm.

Typically 0.2 M sample solutions in chloroform-d were used in the analyses. Their peak frequencies were referenced against the solvent, chloroform-d at 7.24 ppm for 1H and

77.0 ppm for 13C{1H} NMR. The deconvolution of 13C NMR spectra was done using an

Xwin-NMR Version 2.6 software.

67 Mass Spectrometries. Matrix assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI–TOF/MS) analysis was performed on a Bruker Reflex III

(Bruker, Breman, Germany) mass spectrometer operated in linear, positive ion mode with a N2 laser. Laser power was used at the threshold level required to generate signal. The accelerating voltage was set to 28 kV. The instrument was calibrated with standards bracketing the molecular weights of the samples. The 2,5-dihydroxybenzoic acid (DHBA) matrix was prepared as saturated solutions in THF. Allotments of 10 mL of matrix and 2 mL of a solution of the sample were thoroughly mixed together; 0.5 ml of this was spotted on the target plate and allowed to dry.

Small molecular weight di-, tri-, and tetra(propylene glycol) ethyl ethers and the carbonates produced from them were analyzed on a Bruker Esquire ion trap mass spectrometer (Bremen, Germany) equipped with an orthogonal electrospray source operated in positive ion mode. Samples were prepared in a solution containing methanol/acetic acid (99:1) and infused into the electrospray source at a rate of 5–10 mL/min. Optimal ESI conditions were as follows: capillary voltage, 3500 V; source temperature, 250 oC; the ESI drying gas, nitrogen. The ion trap was set to pass ions from m/z 50 to 2000 amu. Data were acquired in continuum mode until acceptable averaged data were obtained.

Chromatography. Gel permeation chromatographic (GPC) analysis was performed at 35 oC on a Breeze system equipped with a Waters 410 refractive index detector and a set of two columns, Waters Styragel HR-2 and HR-4 (Milford, MA).

THF (HPLC grade) was used as the mobile phase at 1.0 mL/min. The sample concentration was 0.1%, and the injection volume was 100 µL. The samples were

68 centrifuged and filtered before analysis. The calibration curve was made with nine polystyrene standards covering the molecular weight range from 580 to 3 150 000 Da.

High performance liquid chromatographic (HPLC) analysis was performed on the same instrument as GPC. Instead, a Symmetry C18 (4.6 × 150 mm) column (Milford, MA) and methanol/water (50:50) mobile phase at 1.0 mlL/min were used.

Calculations. The calculations were performed with density functional theory and the Gaussian 98 suite of programs. The geometry optimizations and vibrational frequency calculations were carried out at the B3LYP/6-31G(d) level.17 The calculations for NMR chemical shift prediction were done with the GIAO method and at the

B3LYP/6-311+G(2d,p) level in both the gas phase and with the PCM method for chloroform, using the optimized geometries of each local minimum.

DSC Measurements. The DSC experiments were performed on a Perkin Elmer

Pyris 1 differential scanning calorimeter with a Perkin Elmer Intracooler 1P cooling accessory (Norwalk, CT). The measurements were made in aluminum crimped pans with a heating rate of 10 oC/min, under nitrogen at a flow rate of 20 mL/min.

Homopolymerization of PO. The catalyst (0.1g, 0.5 mmol) was allowed to react with rac-PO (2mL, 29 mmol) at 60 oC for 5 h. The residue of the catalyst was removed by dissolving the product in benzene. The excess monomer, benzene and any other volatiles species were evaporated under vacuum, to give a white polymer, yield 72.4%.

The polymer was analyzed by 1H NMR, 13C{1H} NMR, GPC and MALDI–TOF/MS. 1H

NMR (CDCl3, δ, ppm): 0.98–1.04 (m, CH3 of PPO), 3.15–3.65 (m, CH2 and CH of PPO).

13 1 C{ H} NMR (CDCl3, δ, ppm): 17.28, 17.39 (m, CH3 of PPO), 72.77, 72.83, 72.87,

72.92, 73.32, 73.35, 73.37 (m, CH2 of PPO), 75.08, 75.28, 75.32, 75.48 (m, CH of PPO). 69 GPC data: Mn = 12 443, Mw = 485 752, and PDI = 39.0 (PDI = 7.4 after precipitated in methanol). MALDI/MS data: Two series are found: the major one is HO–[(C3H6O)n]–

+ + H·Na ions, and the other is [(C3H6O)n]Na , with no end groups. Peak at m/z = 1648 for

+ + [(C3H6O)28]·Na , m/z = 1666 for HO–[(C3H6O)28]–H·Na , m/z = 1706 for

+ + [(C3H6O)29]Na , m/z = 1724 for HO–[(C3H6O)29]–H·Na .

Copolymerization Reactions for PPC Microstructural Studies. Typical copolymerization reactions of PO and CO2 were done in an autoclave. 0.1 g (0.5mmol) of zinc glutarate was added into 2 mL (29 mmol) of PO. CO2 was then added to this system.

o The system was sealed and heated at 60 C for 20 h, under a CO2 pressure of 50 bar. The product was dissolved in benzene, and precipitated in methanol. The final products were analyzed by NMR, GPC and DSC.

1 Rac-PO/CO2/Zinc Glutarate. H NMR (CDCl3, δ, ppm): 1.29, 1.31 (d, CH3 of

PPC), 4.05–4.23 (m, CH2), 4.90–5.02 (m, CH). DEPT NMR (CDCl3, δ, ppm): CH3 carbons: 16.61, 16.64, 16.69. CH2 carbons: 69.42, 69.52, 69.60, 69.69. CH carbons:

13 1 72.52, 72.57, 72.76, 72.80. C{ H} NMR (50 mg/mL, CDCl3, δ, ppm): 16.14, 16.19,

16.23 (m, CH3 of PPC), 68.96, 69.05, 69.08, 69.14, 69.24 (m, CH2), 72.03, 72.08, 72.32,

72.33, 72.34 (m, CH), 153.65, 153.67, 153.70, 153.78, 153.80, 153.83 (m, C=O, H–H region), 154.14, 154.17, 154.19, 154.21, 154.23, 154.25, 154.27 (m, C=O, H–T), 154.64,

154.66 (m, C=O, T–T). GPC data: Mn = 66 326, Mw = 458 616, and PDI = 6.9. DSC data:

o o glass transition temperature Tg = 26.8 C, decomposition temperature Td = 264.7 C.

1 S-PO/CO2/Zinc Glutarate. H NMR (CDCl3, δ, ppm): 1.32, 1.34 (d, CH3 of PPC),

4.12–4.30 (m, CH2), 4.94–5.05 (m, CH). DEPT NMR (CDCl3, δ, ppm): CH3 carbons:

16.22, 16.25, 16.30. CH2 carbons: 69.03, 69.20. CH carbons: 72.14, 72.18, 72.38, 72.41. 70 13 1 C{ H} NMR (50 mg/mL, CDCl3, δ, ppm): 16.15, 16.19, 16.23 (m, CH3 of PPC), 68.97,

69.14 (m, CH2), 72.07, 72.08, 72.11, 72.31, 72.34 (m, CH), 153.63, 153.64, 153.67 (m,

C=O, H–H region), 154.14, 154.16 (m, C=O, H–T), 154.63, 154.64 (m, C=O, T–T). GPC

o data: Mn = 17 479, Mw = 243 017, and PDI = 13.9. DSC data: Tg = 37.8 C, and Td = 219.4 oC.

13 1 50% rac-PO/50% S-PO/CO2/Zinc Glutarate. C{ H} NMR (50 mg/mL, CDCl3,

δ, ppm): 16.14, 16.18, 16.22 (m, CH3 of PPC), 68.93, 68.96, 69.05, 69.07, 69.13, 69.23

(m, CH2), 72.00, 72.03, 72.06, 72.07, 72.10, 72.31, 72.33 (m, CH), 153.62, 153.64,

153.66, 153.69, 153.77, 153.80, 153.82 (m, C=O, H–H region), 154.14, 154.16, 154.18,

154.20, 154.22, 154.24, 154.26 (m, C=O, H–T), 154.63, 154.65 (m, C=O, T–T). GPC

o data: Mn = 20 268, Mw = 249 464, and PDI = 12.3. DSC data: Tg = 40.7 C, and Td = 218.1 oC.

Synthesis of Oligo(propylene glycol) Ethyl Ethers. In a typical reaction, potassium ethoxide (2.0 g, 23.8 mmol) was allowed to react with rac-PO (or S-PO) (5 mL, 71.6 mmol) in 40 mL of 1,4-dioxane at 90 oC for 20 h. After cooling to room temperature, the product was allowed to react with excess HCl solution. The solvents and any other volatile species were evaporated under vacuum. The residue was extracted with hexane.

After hexane was distilled out, a yellow liquid was obtained, which was a mixture of oligo(propylene glycol) ethyl ethers. This mixture was separated using spinning band column distillation under vacuum at elevated temperatures to obtain pure (>98%) di-, tri-, and tetra(propylene glycol) ethyl ethers (colorless liquids). The final products were analyzed by HPLC, ESI/MS, and NMR.

71 Di(rac-propylene glycol) Ethyl Ether. ESI/MS: peak at m/z = 185 for

+ 1 Et(PO)2OH·Na . H NMR (CDCl3, δ, ppm): 0.97 (m, 2CH3–CH), 1.06 (t, CH3–CH2),

13 1 3.78 (m, CH–OH), 3.00–3.60 (m, CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm):

14.76, 14.78, 16.22, 16.94, 18.00, 18.26 (CH3), 66.42, 66.44, 74.16, 74.26, 75.53 (CH2),

65.44, 66.88, 74.19, 75.94 (CH).

Tri(rac-propylene glycol) Ethyl Ether. ESI/MS: peak at m/z = 243 for

+ 1 Et(PO)3OH·Na . H NMR (CDCl3, δ, ppm): 0.90–1.10 (m, 4CH3), 3.77 (m, CH–OH),

13 1 3.00–3.60 (m, CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm): 14.84, 14.87, 14.88,

14.91, 16.68, 16.70, 16.78, 16.81, 16.86, 16.88, 16.94, 16.99, 18.00, 18.03, 18.30 (CH3),

66.42, 66.43, 72.92, 73.04, 73.06, 73.13, 74.14, 74.23, 74.31, 74.34, 75.67 (CH2), 65.31,

65.38, 66.87, 66.91, 74.47, 74.63, 74.90, 74.91, 74.98, 75.00, 76.24, 76.50 (CH).

Tetra(rac-propylene glycol) Ethyl Ether. ESI/MS: peak at m/z = 301 for

+ 1 Et(PO)4OH·Na . H NMR (CDCl3, δ, ppm): 0.90–1.10 (m, 4CH3), 3.77 (m, CH–OH),

13 1 3.00–3.60 (m, CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm): 14.78, 14.81, 14.82,

14.85, 16.62, 16.64, 16.67, 16.69, 16.72, 16.74, 16.79, 16.81, 16.84, 16.87, 17.97, 17.99,

18.00, 18.02 (CH3), 66.32, 66.36, 72.63, 72.67, 72.83, 72.85, 72.87, 72.92, 72.96, 72.99,

73.05, 73.19, 73.21, 74.08, 74.12, 74.14, 74.16, 74.18, 74.25, 74.28, 75.54, 75.59, 75.61

(CH2), 65.26, 65.31, 65.34, 66.77, 66.82, 74.41, 74.45, 74.57, 74.60, 74.75, 74.77, 74.83,

74.84, 74.89, 74.91, 74.93, 74.97, 75.03, 75.10, 75.16, 75.18, 76.14, 76.24, 76.26, 76.29,

76.30, 76.39 (CH).

Di(S-propylene glycol) Ethyl Ether. ESI/MS: peak at m/z = 185 for

+ 1 Et(PO)2OH·Na . H NMR (CDCl3, δ, ppm): 0.97 (m, 2CH3–CH), 1.06 (t, CH3–CH2),

72 13 1 3.78 (m, CH–OH), 3.00–3.60 (m, CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm):

14.89, 16.76, 18.29 (CH3), 66.53, 74.31, 74.38 (CH2), 65.50, 74.29 (CH).

Tri(S-propylene glycol) Ethyl Ether. ESI/MS: peak at m/z = 243 for

+ 1 Et(PO)3OH·Na . H NMR (CDCl3, δ, ppm): 0.90–1.10 (m, 4CH3), 3.77 (m, CH-OH),

13 1 3.00–3.60 (m, CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm): 15.01, 16.79, 17.01,

18.32 (CH3), 66.57, 73.17, 74.31, 74.44 (CH2), 65.48, 74.55, 75.12 (CH).

C2H5–[OCH2CH(CH3)]n–OH (Average n = 10). Potassium ethoxide (0.24 g, 2.86 mmol) was reacted with rac-PO (or S-PO) (2 mL, 28.6 mmol) in 20 mL of 1,4-dioxane at

90 oC for 20 h. After the addition of HCl solution, removal of solvents and extraction with hexane, a yellow liquid was obtained (90% yield). Using rac-PO: GPC, Mn = 740,

1 PDI = 1.28. H NMR (CDCl3, δ, ppm): 0.90–1.10 (m, CH3), 3.83 (m, CH–OH), 3.00–

13 1 3.60 (m, CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm): 15.02–18.63 (m, CH3), 67.01,

73.30–74.90, 76.24, 76.32 (CH2), 65.95, 65.98, 67.51, 67.57, 75.08–75.90, 77.46 (CH).

1 Using S-PO: GPC, Mn = 780, PDI = 1.25. H NMR (CDCl3, δ, ppm): 0.90–1.10 (m,

13 1 CH3), 3.83 (m, CH–OH), 3.00–3.60 (m, CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm):

15.46–19.20 (m, CH3), 73.00, 73.08, 73.13, 74.17, 74.18 (CH2), 65.27, 66.30, 74.38,

74.84, 75.13, 75.16, 75.20 (CH).

C2H5–[OCH2CH(CH3)]n–OH (Average n = 50). This large MW polymer sample was synthesized to investigate the regio- and stereoselectivity of PO ring opening in the above reactions (Appendix A). Potassium ethoxide (24 mg, 0.29 mmol) was allowed to react with rac-PO (2 mL, 28.6 mmol) in the same condition and following the same procedure as above. A yellow viscous liquid was finally yielded. GPC: Mn = 5040, PDI =

73 1 1.32. H NMR (CDCl3, δ, ppm): 0.90–1.10 (m, CH3), 3.83 (m, CH–OH), 3.00–3.60 (m,

13 1 CH2, CH, OH). C { H} NMR (CDCl3, δ, ppm): 17.71, 17.83 (CH3), 72.73, 72.78,

72.86, 72.99, 73.23, 73.26 (CH2), 74.99, 75.18, 75.22, 75.37 (CH).

Synthesis of Model Carbonate Compounds. R–(PO)n–OCOO–(PO)n–R, R = CH3,

C2H5, or H, were synthesized as following. R–(PO)n–OH (0.5 mL) was reacted with 1/6 molar equivalent of triphosgene in 2 mL of benzene, stirring.7 After 3 h, 2 mL of pyridine was added into the flask, and the temperature was increased to 60 oC for another 12 h.

After the removal of solvent and volatile species, the product was extracted with hexane.

The final products were analyzed with MS and NMR.

CH3(PO)2OH/Triphosgene. ESI/MS: peak at m/z = 373 for

+ 1 Me(PO)2OCO2(PO)2Me·Na . H NMR: 1.00–1.30 (m, CH3–CH), 3.10–3.60 (m, CH3–O,

13 1 CH2, CH), 4.86 (m, CH–OCO2). C { H} NMR: 16.83–17.43 (CH3–CH), 57.18, 59.52

(CH3–O), 71.67, 71.69, 71.75, 72.05, 72.10, 73.95, 74.01, 74.06, 75.46, 75.49, 77.00,

77.13 (CH2), 73.51-73.90, 74.09, 74.15, 75.66, 75.68, 76.28 (CH), 154.10–154.25 (C=O).

CH3(PO)3OH/Triphosgene. ESI/MS: peak at m/z = 489 for

+ 1 Me(PO)3OCO2(PO)3Me·Na . H NMR: 1.05–1.30 (m, CH3–CH), 3.20–3.70 (m, CH3–O,

13 1 CH2, CH), 4.87 (m, CH–OCO2). C { H} NMR: 16.48–18.30 (CH3–CH), 59.46 (CH3–

O), 71.50–73.85, 74.74, 74.83, 76.37–77.08 (CH2), 73.87, 74.05, 74.15, 74.93–76.31

(CH), 154.10–154.25 (C=O).

CH3CH2(rac-PO)1OH/Triphosgene. ESI/MS: peak at m/z = 257 for

+ 1 Et(PO)1OCO2(PO)1Et·Na . H NMR: 1.08 (t, CH3–CH2), 1.19 (m, CH3–CH), 3.00–4.00

13 1 (m, CH2, CH), 4.82 (m, CH–OCO2). C { H} NMR: 14.88, 16.49, 16.51 (CH3), 66.53,

66.54, 72.54, 72.60 (CH2), 73.13, 73.18 (CH), 154.13–154.18 (C=O). 74 CH3CH2(rac-PO)2OH/Triphosgene. ESI/MS: peak at m/z = 373 for

+ 1 Et(PO)2OCO2(PO)2Et·Na . H NMR: 1.00–1.30 (m, CH3), 3.00–3.60 (m, CH2, CH), 4.80

13 1 (m, CH–OCO2). C { H} NMR: 15.00–17.20 (CH3), 66.54 (CH3–CH2-O), 71.20, 71.26,

71.59, 71.62, 74.33, 74.37, 74.50 (CH2), 73.34, 73.38, 73.51, 73.59, 75.10, 75.12, 75.34,

75.35 (CH), 154.09–154.20 (C=O).

CH3CH2(rac-PO)3OH/Triphosgene. ESI/MS: peak at m/z = 489 for

+ 1 Et(PO)3OCO2(PO)3Et·Na . H NMR: 1.00–1.30 (m, CH3), 3.00–3.60 (m, CH2, CH), 4.78

13 1 (m, CH–OCO2). C { H} NMR: 15.00–18.30 (CH3), 66.49 (CH3–CH2–O), 71.10–73.30,

74.20–74.40, 75.83 (CH2), 73.31, 73.50, 73.58, 73.60, 74.42–75.62, 76.43 (CH), 154.10–

154.21 (C=O).

CH3CH2(rac-PO)4OH/Triphosgene. ESI/MS: peak at m/z = 605 for

+ 1 Et(PO)4OCO2(PO)4Et·Na . H NMR: 1.00–1.30 (m, CH3), 3.10–3.60 (m, CH2, CH), 4.78

13 1 (m, CH–OCO2). C { H} NMR: 15.00–17.30 (CH3), 66.48 (CH3–CH2-O), 71.09–71.70,

72.80–73.25, 73.34, 73.38, 74.30–74.77 (CH2), 73.27, 73.31, 73.49, 73.57, 73.59, 74.86–

75.64 (CH), 154.08–154.20 (C=O).

50% CH3CH2(rac-PO)2OH/50% CH3CH2(S-PO)2OH/Triphosgene. ESI/MS: peak

+ at m/z = 373 for Et(PO)2OCO2(PO)2Et·Na . It has similar chemical shifts for the signals

1 13 1 in H and C { H} NMR spectra to the product of CH3CH2(rac-PO)2OH/triphosgene, except that the relative intensities are different.

50% CH3CH2(rac-PO)3OH/50% CH3CH2(S-PO)3OH/Triphosgene. ESI/MS: peak

+ at m/z = 489 for Et(PO)3OCO2(PO)3Et·Na . It has similar chemical shifts for the signals

1 13 1 in H and C { H} NMR spectra to the product of CH3CH2(rac-PO)3OH/triphosgene, except that the relative intensities are different.

75 CH3CH2(rac-PO)nOH (Average n = 10) /Triphosgene. MALDI/MS: major series,

+ 1 Et(PO)nOCO2(PO)mEt·Na . H NMR: 1.00–1.30 (m, CH3), 3.00–3.60 (m, CH2, CH), 4.80

13 1 (m, CH–OCO2). C { H} NMR: 16.60–18.40 (CH3), 66.50 (CH3–CH2–O), 70.20–76.00

(CH2, CH), 154.08–154.20 (C=O).

CH3CH2(S-PO)nOH (Average n = 10) /Triphosgene. MALDI/MS: major series,

+ 1 Et(PO)nOCO2(PO)mEt·Na . H NMR: 1.00–1.30 (m, CH3), 3.10–3.60 (m, CH2, CH), 4.78

13 1 (m, CH–OCO2). C { H} NMR: 16.40–17.40 (CH3), 66.45 (CH3–CH2–O), 71.50–75.80

(CH2, CH), 154.21 (C=O).

Poly(propylene glycol) (Mn = 2000) /Triphosgene. MALDI/MS: two major series,

+ + 1 HO(PO)nOCO2(PO)mOH·Na , and HO(PO)nOCO2(PO)mOCO2(PO)lOH·Na . H NMR:

13 1 1.00–1.30 (m, CH3), 3.10–3.60 (m, CH2, CH), 4.78 (m, CH–OCO2). C { H} NMR:

17.00–17.50 (CH3), 72.50–73.40 (CH2), 74.70-75.50 (CH), 154.10–154.22 (C=O).

Degradation studies of PPC. Degradation reaction of 0.5 g of PPC was done in

THF at room temperature for 3 h, under an inert atmosphere, using 10 mg (0.12mmol) of

LiOtBu (Aldrich) and 0.02 mL (0.3mmol) of tBuOH (Aldrich). The product was analyzed

1 1 by H NMR, GPC and ESI/MS. H NMR (CDCl3, δ, ppm): 1.39, 1.40 (d, CH3 of PC),

3.10–3.70 (m, trace, CH2 and CH of PPO), 3.93, 3.95, 3.97, 4.46, 4.48, 4.50 (m, CH2 of

PC), 4.74, 4.76, 4.78, 4.80, 4.81, 4.83 (m, CH of PC). GPC data: Mn = 542, Mw = 630,

+ and PDI = 1.16. ESI/MS data: Peak at m/z = 157 for HO–[(PO)2]–H·Na , m/z = 175 for

+ + [(PO)2]·H , m/z = 205 for [(PO)2-alt-(CO2)2]·H , m/z = 259 for HO–[(PO)3(CO2)1]–

+ + H·Na , m/z = 303 for HO–[(PO)3-alt-(CO2)2]–H·Na , m/z = 317 for HO–[(PO)4(CO2)1]–

+ + H·Na , m/z = 361 for HO–[(PO)4-alt-(CO2)2]–H·Na , m/z = 419 for HO–[(PO)5-alt-

76 + + (CO2)2]–H·Na , m/z = 477, HO–[(PO)6-alt-(CO2)2]–H·Na , m/z = 547 for [(PO)6-alt-

+ + CO2]4]·Na , m/z = 605 for [(PO)7-alt-(CO2)4]·Na .

0.75 g of PPC obtained from Aldrich was allowed to react with 50 mg (0.26

o mmol) of zinc glutarate in 10 mL of CH2Cl2 at 60 C for 40 h under an inert atomosphere.

The product was analyzed by 1H NMR and GPC. The ratio of PC/PPC increased from

2.51% to 4.39%.

2.5 References

(1) Tonelli, A. E. Polym. Spectrosc. 1996, 55-95.

(2) Tonelli, A. E. Annu. Rep. NMR Spectrosc. 1997, 34, 185-229.

(3) Lednor, P. W.; Rol, N. C. J. Chem. Soc., Chem. Commun. 1985, 598-599.

(4) Chisholm, M. H.; Navarro-Llobet, D. Macromolecules 2002, 35, 2389-2392.

(5) Aida, T.; Ishikawa, M.; Inoue, S. Macromolecules 1986, 19, 8-13.

(6) Steiner, E. C.; Pelletier, R. R.; Trucks, R. O. J. Am. Chem. Soc. 1964, 86, 4678-

4686.

(7) Eckert, H.; Forster, B. Angew. Chem., Int. Ed. Engl. 1987, 26, 894-895.

(8) Parr, R. G.; Yang, W. Density-functional Theory of Atoms and Molecules, 1989.

(9) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;

Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R. E.; Burant, J.

C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.;

Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;

Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;

Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, 77 J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;

Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;

Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M.

W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98; revision A9 ed.; Gaussian, Inc.: Pittsburgh, PA, 1998.

(10) Mazhar ul, H.; Caughlan, C. N.; Smith, G. D.; Ramirez, F.; Glaser, S. L. J. Org.

Chem. 1976, 41, 1152-1154.

(11) Bohets, H.; van der Veken, B. J. Phys. Chem. Chem. Phys. 1999, 1, 1817-1826.

(12) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027-2094.

(13) Wiberg, K. B. J. Comput. Chem. 1999, 20, 1299-1303.

(14) Kuran, W.; Gorecki, P. Makromol. Chem. 1983, 184, 907-912.

(15) Gorecki, P.; Kuran, W. J. Polym. Sci., Pol. Lett. 1985, 23, 299-304.

(16) Ree, M.; Bae, J. Y.; Jung, J. H.; Shin, T. J. Korea Polym. J. 1999, 7, 333-347.

(17) Becke, A. D. Phys. Rev. A 1988, 38, 3098-3100.

78

CHAPTER 3

CONCERNING THE MECHANISM OF THE RING OPENING OF PROPYLENE

OXIDE IN THE COPOLYMERIZATION OF PROPYLENE OXIDE AND CARBON

DIOXIDE TO GIVE POLY(PROPYLENE CARBONATE)

3.1 Introduction

Catalytic reactions proceed via a series of elementary reactions, each of which is in principle reversible and collectively these determine the turnover limiting frequency that replaces what is commonly called a rate limiting step for a reaction. In the copolymerization of epoxides (oxiranes) and carbon dioxide to give poly(alkene carbonates), the two elementary reactions are (i) insertion of CO2 into a metal-alkoxide bond to form an alkyl carbonate group: M–OR + CO2 → M–O2COR and (ii) ring opening of epoxides with addition of the alkyl carbonate to regenerate the metal-alkoxide bond, as in the case of propylene oxide: M–O2COR + PO → M–OCHMeCH2OR’ or M–

OCH2CHMeOR’. Furthermore, in the ring-opening reaction of an epoxide, the coordination of the epoxide to the metal is one of the key steps.1-7 Coordination to the

Lewis acidic metal center facilitates ring opening by both an SN2 and an SN1 reaction pathway. 8,9

79 Detailed mechanistic studies of reactions of model compounds and the kinetics of the CHO/CO2 copolymerization reaction by the Darensbourg and Coates groups have provided considerable insight into these reactions.10-15 It is, however, the step involving the ring opening of the epoxide (the enchainment step) that is the least well understood.

In the case of dimeric zinc complexes, Coates has provided kinetic evidence that implicates a bimetallic pathway,13 while in the case of salen chromium (III) complexes

Darensbourg has provided evidence for a monometallic single-site pathway (Scheme

3.1).10

Polymer O Polymer Polymer C O O O O C C O O O O O O O Cr Cr Cr

N N N

N N N

Scheme 3.1. The monometallic mechanism proposed for the CHO/CO2 copolymerization reaction by Darensbourg.10

Certain catalysts, be they homogeneous or heterogeneous, are capable of both homopolymerizing PO and copolymerizing PO/CO2. Some, however, are selective in one over the other as shown in Table 3.1 and 3.2. In the homopolymerization of rac-PO, these catalysts yield highly regioregular polymers (HTHT)n with significant stereoselectivity in the preferential formation of isotactic junctions.16-19 In contrast, the copolymerization of

80 PO/CO2 yields regioirregular PPC with a ratio of HT:HH+TT junctions that is catalyst dependent.12,20-22 The addition of Lewis bases, such as Br-, N-methylimidazole or 4-

(dimethylamino)pyridine (DMAP), greatly enhance PPC production at the expense of

PPO.11,23-25 Moreover, there are notable and quite remarkable differences between the activity of the metals Al(III), Cr(III) and Co(III), all of which have a very similar ionic radius, and between the porphyrin and salen ligands. With porphyrin and salen complexes of metal (III) this begs the question of how the alkyl carbonate and PO react within the same face of the metal center.

In an attempt to clarify this matter, we have examined the stereochemistry and kinetics of the initial ring-opening event of PO by various aluminum porphyrin and salen initiators, as well as the influence of the promoter DMAP on this event and on the insertion of carbon dioxide into the Al–OR bond. We also studied the binding of propylene oxide to the porphyrin and salen metal (III) cations, LM+, in the gas phase in an electrospray ionization tandem mass spectrometer (ESI/MS/MS). We describe herein these findings.

81

Entry Catalytic systems Reaction PPOb Mnc PDIc times (h) (%) (×103) 1 (TPP)CrCl 120 76 61 1.53 2 (TPP)CrCl/DMAPd 120 59 44 1.38

3 (TPP)AlCl 168 50 28 1.32 4 (TPP)AlCl/DMAP 168 2 - - e 5 (TPP)CoCl 120 - - - 6 (TPP)CoCl/DMAP 120 - - -

7f Zinc glutarate 5 72 12 7.42 8g Union Carbide

Table 3.1. Homopolymerization of PO by (TPP) metal complexes.a

(a) The polymerization reactions were carried out in neat PO (2 mL), 1:1500 molar ratio used for (TPP) metal catalyst systems (25oC, 1 equiv of DMAP when needed), and 0.1 g of catalyst was used for zinc glutarate (60oC). No PPO was formed by using (TPP)GaCl and (salen)AlCl, and PPO could be formed by (salen)CrCl only at high [cat] and elevated temperatures. (b) Determined by 1H NMR. (c) Mn and PDI determined by gel permeation chromatography. (d) DMAP, 4-(dimethylamino)pyridine. (e) No or little formation. (f) 1 g of catalyst in 2 mL of neat PO at 60 oC.21 (g) Union Carbide system is known to be active in PO ring-opening polymerization.17,26

82

b c d e Entry Catalytic systems Reaction PPC PPO PC Mn PDI times (h) (%) (%) (%) × 3 ( 10 ) 1 (TPP)CrCl 48 8 92 -f 85 1.49

2 (TPP)CrCl/DMAP 48 49 26 - 23 1.41 3 (TPP)AlCl 144 4 21 - 12 1.19

4 (TPP)AlCl/DMAP 144 57 7 1 31 1.24 5 (TPP)CoCl 48 - - - - -

6 (TPP)CoCl/DMAP 48 - - 3 - - 7 (salen)CrCl 144 13 1 1 9 1.21

8 (salen)CrCl/DMAP 144 25 1 2 14 1.35 9 (salen)AlCl 144 - - - - -

10 (salen)AlCl/DMAP 144 - - - - - 11g (salen)CoOAc 3 35 - - 8 1.57

12 Zinc glutarate 5 71 2 2 27 9.16 13 Union Carbide - - - - -

a Table 3.2. Copolymerization of PO/CO2 by (TPP) and (salen) metal complexes.

(a) The polymerization reactions were carried out in neat PO (2 mL), 1:1500 molar ratio used for (TPP) and (salen) metal catalyst systems (25oC, 1 equiv of DMAP when needed), and 0.1 g of catalyst was used for zinc glutarate (60oC). (b) Carbonate linkages in the polymers obtained in the reaction mixtures determined by 1H NMR. (c) Ether linkages in the resulting polymers determined by 1H NMR. (d) Cyclic propylene

1 carbonate in the products determined by H NMR. (e) Mn and PDI determined by gel permeation chromatography. (f) No or little formation. (g) see reference (22), [cat]:[PO]

= 1:200.22

83 3.2 Results and Discussion

3.2.1 Studies of the Initial Ring Opening of Propylene Oxide.

The reactions between (TPP)AlX initiators, where TPP = tetraphenylporphyrin and X = Cl, O(CH2)9CH3 and O2C(CH2)6CH3, and PO give the ring-opened products

(TPP)AlOCHMeCH2X as the major products at short reaction times. Because of the large magnetic anisotropy of the porphyrin ring,27-29 the proton resonances of signals of CH,

CH2, and Me groups within ca. 5 Å of the Al center appear notably shifted upfield relative to Me4Si. Consequently, it is relatively easy to monitor these reactions in solution by 1H NMR spectroscopy. Furthermore, one can unequivocally distinguish between the regioisomers of ring opening, (TPP)AlOCH2CHMeX versus (TPP)AlOCHMeCH2X.

1b (1) (TPP)AlOCH(CH32 )CH Cl 1a 1b 1c/1c’

(2) (TPP)AlOCH23 CH(CH )Cl 2c 2a 2b

1c’ 1c 2b 1a 2a 2c

0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2 -1.4 -1.6 -1.8 -2.0 -2.2 -2.4 -2.6 ppm

1 Figure 3.1. H NMR (500 MHz, CDCl3) spectrum of PO ring-opened product by

(TPP)AlCl, showing two regioisomers.

A careful examination of the spectrum in this region reveals that both regioisomers are formed but that the major regioisomer (TPP)AlOCHMeCH2X

84 predominates by ca. 9:1, when X = Cl. The 1H NMR spectrum revealing the presence of the (TPP)AlOCHMeCH2Cl and (TPP)AlOCH2CHMeCl isomers is shown in Figure 3.1.

The minor regioisomer was prepared independently from the reaction between

(TPP)AlCH3 and the chloro S-ClCHMeCH2OH. The ring opening of PO by

(TPP)AlCl is extremely rapid, but the very low solubility of (TPP)AlCl in CDCl3 and

C6D6 precludes the study of the kinetics of this reaction by NMR methods.

The introduction of the long alkyl chains when X = O(CH2)9CH3 and

O2C(CH2)6CH3 rendered these derivatives soluble in CDCl3 and to some extent in C6D6.

The choice of the n-octanoate ligand is intended to model for the ring opening of PO by an alkyl carbonate ligand, vide infra.

PO (TPP)AlOCH222 CH CH (CH 263 ) CH > (TPP)AlOCH(CH32 )CH O(CH 293 ) CH a b c d e f/f’ e f d 5560

2800 1600 870

580 Time (min) 360 180

80 c a b 10 -0.4 -0.6 -0.8 -1.0 -1.2 -1.4 -1.6 -1.8 -2.0 -2.2 -2.4 -2.6 ppm

1 Figure 3.2. H NMR (500 MHz, CDCl3) spectra of the reaction between

(TPP)AlO(CH2)9CH3 and PO, showing the disappearance of initiator and appearance of product with time. 85

The ring opening of PO by these two compounds was followed by 1H NMR spectroscopy under pseudo first order conditions, where [PO]:[Al] was on the order

o 100:1. The reactions were followed at 25 C in CDCl3 by the disappearance of the Al–X proton signals (Figure 3.2). For the reaction involving X = O(CH2)9CH3, the product of ring opening was almost exclusively the (TPP)AlOCHMeCH2O(CH2)9CH3 isomer with the other regioisomer being barely detectable. However, for the n-octanoate initiator, both regioisomers were formed in an approximate ratio of 4:1, where the major isomer once again represents the formation of the secondary alkoxide by ring opening of PO at the methylene carbon. However, with time this ratio changes as the primary alkoxide, the product of ring opening of PO at the methine carbon, reacts faster with PO than does the n-octanoate. The secondary alkoxide also reacts with PO but at a significantly slower rate than the primary. Thus, we can state that in the initiation step the reactivity order for

(TPP)AlX compounds is X = Cl > O(CH2)9CH3 > O2C(CH2)6CH3 (Figure 3.3). We also determined, by varying the concentration of (TPP)AlO(CH2)9CH3 from 5 to 15 mM in four different experiments, that the rate of ring opening of PO was first order in

(TPP)AlO(CH2)9CH3 (Figure B.1 in Appendix B). The straight-line plots shown in Figure

3.3 also confirm this.

The influence of added 4-(dimethylamino)pyridine (DMAP) was examined in the following manner. In a NMR experiment, 1 equiv of DMAP was added to a CDCl3 solution of (TPP)AlX, where X = O(CH2)9CH3 or O2C(CH2)6CH3. In each case the signals associated with the aromatic protons of DMAP were shifted upfield and were significantly broadened relative to signals of DMAP in neat CDCl3 (Figure B.2 in

86 Appendix B). At 25 oC, DMAP is reversibly coordinating to the aluminum porphyrin center and there is a significant concentration of the 1:1 adduct. Upon cooling to –50 oC, this equilibrium becomes frozen out on the NMR time scale for X = O2C(CH2)6CH3 and

O2CO(CH2)9CH3 (see later). The 2,6-protons of the aromatic ring in DMAP appear at δ

0.56 ppm upon complexation to the Al center, which contrasts with δ 8.2 ppm for the free molecule in CDCl3. The 3,5-aromatic protons shift similarly from δ 6.5 ppm to 4.0 ppm on complexation, and the NMe resonances shift from δ 3.0 ppm to 2.0 ppm. In the case of

X = O(CH2)9CH3, the equilibrium could not be frozen out on the NMR time scale because the equilibrium constant for binding of DMAP was much smaller. At ambient temperature, the 1H signals of DMAP represent a time average between [free] and

o [complexed], which allows for the estimation of Keq for binding. At 25 C in CDCl3, we

-3 -1 -1 estimate Keq to be 5.8 × 10 mM for X = O(CH2)9CH3, 1.7 mM for X =

-1 O2C(CH2)6CH3, and 0.51 mM for X = O2CO(CH2)9CH3, which reveals that DMAP binds in the order O2CR > O2COR >> OR. This order follows roughly the inverse order of the basicity of the oxygen donor ligands.

The kinetics of the ring opening of PO by the (TPP)Al–X compounds in the presence of 1 equiv of DMAP was also determined by NMR studies at 25 oC by monitoring the disappearance of the α-CH2 protons of the alkoxide or carboxylate group signal. As shown in Figure 3.3, the influence of added DMAP was very dramatic in enhancing the rate. Now, the ring-opening event is much faster by the carboxylate than the alkoxide, although even the alkoxide is accelerated by an order of magnitude. We also noticed that the kinetic profiles of these two initiators in the presence of DMAP were polynomial rather than linear (Figure B.3 in Appendix B). This is because the secondary 87 alkoxide products are less strongly coordinated by DMAP than the initiators, and the effective [DMAP]:[initiator] ratio increases with time. Thus, the apparent rate constants increase with time in these two reaction systems.

4 (TPP)AlOR y = 0.0006x + 0.1603 R2 = 0.9967 3.5 (TPP)AlOOCR y = 0.0002x + 0.151 R2 = 0.9936 3 2 (TPP)AlOOCR/DMAP y = 6E-06x + 0.0004x + 0.0322 2.5 R2 = 0.9984 ) 0 2 /I t y = 0.0002x - 0.0004x + 0.0349 2 (TPP)AlOR/DMAP R2 = 0.9974 Ln(I 1.5

1

0.5

0 0 1000 2000 3000 4000 5000 6000 7000 Reaction Time (Minutes)

Figure 3.3. Plots of –ln(It/I0) versus reaction time for the ring-opening reaction of the first PO molecule by initiators, in the absence and in the presence of 1 equiv of DMAP. I0 is the initial initiator concentration, and It is the initiator concentration at time t.

(TPP)AlOR = (TPP)AlO(CH2)9CH3, (TPP)AlOOCR = (TPP)AlO2C(CH2)6CH3. In all of the reactions, [Al] = 15 mM, [PO] = 720 mM.

88 We have also studied the kinetics of reactions at varying [DMAP]:[initiator] ratios. As shown in Figure 3.4, when the addition of DMAP is over 2 equiv, the reaction rates become slower. One possible reason is that the open metal binding site could be saturated by too much DMAP, thus it inhibits the PO activation by binding to the metal center.

3 1 equiv of DMAP 2 equiv of DMAP 2.5 4 equiv of DMAP 2 6 equiv of DMAP 8 equiv of DMAP 1.5

1 -Ln(It/Io)

0.5

0 0 20406080100120140160 Reaction Time (min)

Figure 3.4. Plots of -ln(It/I0) versus reaction time for the ring-opening reaction of the first

PO molecule by (TPP)AlO2C(CH2)6CH3, in the presence of varying [DMAP]:[initiator] ratios.

Salen aluminum chloride is not capable of polymerizing PO under the conditions comparable to those of the porphyrin analogue, nor will it act in copolymerizing PO and

CO2. However, it will ring open PO to generate a chloroalkoxide. Because of this, we have studied the ring opening of S-PO and R-PO by the compound containing Jacobsen’s 89 chiral ligand bound to aluminum, (R,R)-N,N’-bis(3,5-di-tert-butyl-salicylidene)-1,2- cyclohexenediaminoaluminum chloride, (R,R-salen)AlCl. The ring opening of R-PO yields a 2.3:1 mixture of (R,R-salen)AlOC(R)HMeCH2Cl and (R,R-

30 salen)AlOCH2CHMeCl. The latter compound exists as a mixture of R- and S-

OCH2CHMeCl isomers in the ratio of 2:1, respectively. The related reaction between S-

PO and (R,R-salen)AlCl gave (R,R-salen)AlOC(S)HMeCH2Cl, and apparently only (R,R-

1 salen)AlOCH2C(R)HMeCl in the ratio of 5:1. The H NMR spectrum revealing the products derived from the ring opening of R-PO is shown in Figure 3.5, and other related spectra are given in Figure B.4 in Appendix B.

2a’ 2a 2b

1b 3a’ 3b 3a 1a’ 1a

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 ppm

1 Figure 3.5. H NMR (500 MHz, CDCl3) spectrum of R-PO ring-opened products by

(R,R-salen)AlCl, showing three isomers. (1) (R,R-salen)AlOCH2C(S)HMeCl; (2) (R,R- salen)AlOC(R)HMeCH2Cl; (3) (R,R-salen)AlOCH2C(R)HMeCl. (a and a’ represent CH2 signals, and b represents CH signals derived from PO units, respectively.)

90

The compound (R,R-salen)AlOC(S)HMeCH2Cl was crystallographically characterized (Figure 3.18, see later). The formation of (R,R-salen)AlOCH2C(S)HMeCl was confirmed by its independent synthesis from the reaction between (R,R-salen)AlMe and the chiral alcohol HOCH2C(S)HMeCl as described in the Experimental Section. This compound was also crystallographically characterized and will be shown in Figure 4.1 in

Chapter 4.

These studies of (R,R-salen)AlCl clearly parallel those of the porphyrin analogue in revealing the ring opening of PO occurs by competitive pathways where the methylene carbon is preferentially but not exclusively attacked in the ring-opening event. Moreover, the use of the chiral salen aluminum template shows that there is a preference in the ring opening of R-PO for the formation of the OCH2C(R)HMeCl alkoxide when attack occurs at the methine carbon. The ring opening of S-PO by attack at the methine carbon also preferentially forms the OCH2C(R)HMeCl ligand. The preferential formation of the β- chloroalkoxy ligand with the R- stereocenter is clearly a result of the R,R-salen template, but its formation in the reaction with R-PO occurs with retention of stereochemistry at the methine carbon whereas for S-PO it is formed by inversion (Scheme 3.2).

91

N O R Cl Al O N O n O tio R en ret N O Al Cl N O inv ers ion N O S Cl Al O N O

O S N O inversion N O R Cl Al Cl Al O N O N O

Scheme 3.2. The stereochemistry in the PO ring-opening event involving attack at methine carbons by (R,R-salen)AlCl.

3.2.2 The Carbon Dioxide Insertion Reaction.

Carbon dioxide reacts reversibly with (TPP)AlOR compounds to give

27,31 (TPP)AlO2COR compounds. This equilibrium is both temperature and CO2 pressure dependent, that is, dependent of [CO2] in solution, and when the CO2 atmosphere is removed only the alkoxide is recovered. In our studies, we have examined the 92 o 13 1 13 equilibrium shown in 1 in CDCl3 at 25 C under 4 atm of CO2 by H and C NMR spectroscopy. The concentration of aluminum complex was 11 mM.

k 13 1 13 (TPP)Al-O(CH2)9CH3 + CO2 (TPP)Al-O2 CO(CH2)9CH3 (1) k-1

At 25 oC, under these conditions the equilibrium favors the alkoxide. Upon the introduction of DMAP, however, the equilibrium shown in 1 is shifted markedly to the right, as shown in Figure 3.6. At a high concentration of DMAP (5 equiv) the alkoxide was completely converted to carbonate. When CO2 pressure was released under N2 atmosphere, ~95% of the Al complexes existed as carbonate in solution. However, if the solvent and CO2 pressure were removed and the products were dried under vacuum, the alkoxide was recovered. We note in this system that CO2 insertion/deinsertion is kinetically rapid with respect to the ring opening of PO.

100 90 80 % 70 60 alkylcarbonate 50 40 alkoxide 30 20 10 0 00.511.522.53 [DMAP] : [(TPP)AlOR] Ratio

Figure 3.6. The percentages of alkyl carbonate and alkoxide versus the

13 [DMAP]:[(TPP)AlO(CH2)9CH3] ratio for the CO2 insertion with (TPP)AlO(CH2)9CH3.

93

3.2.3 Polymer Microstructure of Poly(propylene carbonate).

We have previously examined the polymer microstructure of PPC formed in the

13 zinc glutarate catalyzed copolymerization of PO and CO2 by use of proton decoupled C

NMR (13C{1H}) spectroscopy as shown in Chapter 2. Specifically, we used rac-PO, S-

PO, and a 50:50 mixture of rac + S-PO, and we were thereby able to make assignments for the stereosequences i and s for the TT, HT, and HH carbonate junctions.

We have now examined the carbonate 13C signals of PPC prepared by

(TPP)AlCl/DMAP and (R,R-salen)CrCl catalyst systems. As shown in Figure 3.7, the regio- and stereo-sequences differ quite significantly with the catalyst system employed.

The PPC derived from (TPP)AlCl/DMAP has fewer TT and HH junctions, and Coates recently reported that (salen) cobalt (III) acetate produces approximately 80% HT regiosequences while (BDI) zinc acetate produces almost random TT:HT:HH ~ 1:2:1.12,22

We can reasonably assume that HT carbonate junctions normally retain the stereochemistry at the methine carbon as a consequence of the preferential ring opening at the methylene carbon. However, when ring opening at the methine carbon is relatively competitive, HT junctions may also be formed with either double inversion (backside attack) or with randomization of stereochemistry as shown in Scheme 3.3.

94 (A1) Zinc glutarate + rac -PO/CO2

HT i

TT s

i s HH

HT’ i s

ppm

(B1) (TPP)AlCl/DMAP + rac -PO/CO2

HT

TT HH i s HT’ i s

155.00 154.80 154.60 154.40 154.20 154.00 153.80 153.60 153.40 ppm

(C1) (R,R -salen)CrCl + rac -PO/CO2

HT

TT s i HH i s HT’

ppm

13 1 Figure 3.7. C{ H} (150 MHz, CDCl3) NMR spectra of the carbonate carbon region of

PPC made from (A1) zinc glutarate and rac-PO, (B1) (TPP)AlCl/DMAP and rac-PO, and

(C1) (R,R-salen)CrCl and rac-PO. (HH’ and HT’ are ether rich-carbonate signals.)

95

O O OP LnM O OP LnM O C C O O

CO2 O O PO C LnM O O C* O OP O

O C O * O LnM O C O OP O

Scheme 3.3. The chain growing pathway showing HT carbonate junction (*) formation involving double attack at methine carbons in the PO/CO2 copolymerization. (No mechanistic information is implied by this scheme which merely outlines the regiochemical consequence of attack on the methine carbon)

The maximum information concerning the stereochemistry of the ring-opening event can be gleaned from an examination of the HH carbonate 13C signals of PPC derived from enantiomerically pure PO. As shown in Figure 3.8, the (TPP)AlCl/DMAP system yields PPC derived from S-PO with both i and s HH (and TT) junctions, although the syndiotactic junctions are preferred by approximately a 2:1 ratio. Rather interestingly, the PPC derived from S-PO and (R,R-salen)CrCl has both i and s HH (and TT) carbonate junctions, but the ratio is now in the inverse order, i > s (Figure 3.8).

96

(C2) (R,R -salen)CrCl + S -PO/CO2 (A2) Zinc glutarate + S -PO/CO2

HT HT i

TT TT HH HH s s i s HT’ i s HH’ s

ppm 155.00 154.80 154.60 154.40 154.20 154.00 153.80 153.60 153.40 ppm

(C3) (R,R -salen)CrCl + R -PO/CO 2 (B2) (TPP)AlCl/DMAP + S -PO/CO2 HT i HT i TT

s HH TT HH i HH’ HT’ s HT’ s HH’ i i s i s

155.00 154.80 154.60 154.40 154.20 154.00 153.80 153.60 153.40 ppm ppm

13 1 Figure 3.8. C{ H} (150 MHz, CDCl3) NMR spectra of the carbonate carbon region of

PPCs made from (A2) zinc glutarate and S-PO, (B2) (TPP)AlCl/DMAP and S-PO, (C2)

(R,R-salen)CrCl and S-PO and (C3) (R,R-salen)CrCl and R-PO. (HH’ and HT’ are ether- rich carbonate signals.)

When R-PO is employed with the chiral salen CrCl catalyst, the ratio of i:s junctions again changes and is now in favor of s. From this, we can conclude that the salenCr catalyst prefers to open the PO molecule to generate the primary alkoxide of S stereochemistry by attack at the methine carbon (Scheme 3.4).

97 O R N O S OOR N O OP inversion Cr O C Cr O O R OP N O C N O O O

O S N O N O S OOS retention OP Cr O O S Cr O C OP N O C N O O O

Scheme 3.4. The preferential stereochemistry involved in the formation of HH carbonate junctions during the formation of PPC by the catalyst derived from (R,R-salen)CrCl.

3.2.4 A Proposed Mechanism for the (TPP)AlCl/DMAP Copolymerization of PO and CO2.

Our studies have shown that the ring opening of PO by (TPP)AlX compounds occurs by competitive attack on the methylene and methine carbons. When X =

O(CH2)9CH3, the reaction is first order in [Al], giving predominately the secondary alkoxide by attack at the methylene carbon. When X = O2C(CH2)6CH3, the regiochemistry of ring opening is less selective, and the ring opening is notably slower than that for X = O(CH2)9CH3. However, upon addition of DMAP, we note that the rate of ring opening of PO is accelerated greatly, and now the carboxylate group is much more active in ring opening PO relative to the alkoxide. This is particularly important as

(TPP)AlCl alone in the presence of PO and CO2 yields PPO in preference to PPC (Table 98 3.2). Furthermore, we note that CO2 insertion into the Al–OR bond is promoted by the addition of DMAP. Both of these factors assist in producing a regular alternating copolymer and serve to minimize the number of ether-rich linkages as denoted by HT’

13 1 and HH’ signals in the carbonate C{ H} spectra (Figure 3.7). The CO2 incorporation into the polymer formed in the PO/CO2 copolymerization with (TPP)AlO(CH2)9CH3 can be increased from 0.20 to 0.50 (molar fraction determined by 1H NMR) by the addition of

1 equiv of DMAP, yielding a perfectly alternating copolymer as shown by mass spectrometry (Figure 3.9).

Figure 3.9. MALDI-TOF mass spectrum of PPC prepared with the

(TPP)AlO(CH2)9CH3/DMAP catalyst system. An = H–[(C3H6O)n-alt-(CO2)n-1]–

+ + O(CH2)9CH3•Na ; Bn = H– [(C3H6O)n-alt-(CO2)n]–O(CH2)9CH3•Na ; Cn = H–

+ [(C3H6O)n-alt-(CO2)n-2]–O(CH2)9CH3•Na . Peaks (m/z): A38 (4014), A39 (4116), A40

(4218), B38 (4058), B39 (4160), B40 (4262), C39 (4072), C40 (4174), C41 (4276). 99

We propose that the role of DMAP is to labilize the ligand trans to it, either the alkoxide or the alkyl carbonate. The fact that HH and TT junctions can be formed with retention of stereochemistry clearly implies that an incipient carbonium ion is captured by a neighboring nucleophile. The PO and the alkyl carbonate must be cis to one another.

We pictorially represent this DMAP-promoted PO interchange associative process by the sequence shown in Scheme 3.5. The PO is activated by coordination to the aluminum center as the Al–O2COR bond is weakened by the trans-effect of the DMAP ligand.

Which oxygen of the carboxylate group is involved in attacking the PO carbon (Schemes

3.3 and 3.4) is unknown.

' + + O DMAP + PO Al X N N Al X N N Al X

Scheme 3.5. Proposed interchange associative pathway for PO ring opening by

(TPP)AlX/DMAP.

We can now envisage the (TPP)AlX-catalyzed reactions on the basis of the equilibria and cycles shown in Scheme 3.6. In the absence of DMAP the ring opening of

PO by alkoxide dominates even in the presence of CO2, such that very little carbonate is incorporated into what is essentially poly(propylene oxide) (Table 3.2). The addition of

DMAP serves to (1) increase the equilibrium concentration of the aluminum alkyl

100 carbonate at the expense of the aluminum alkoxide, and (2) vastly increase the rate of ring-opening PO by the alkyl carbonate ligand (relative to the alkoxide), such that alternating copolymerization of PO and CO2 to yield PPC now dominates. As shown in

Figures 3.7 and 3.8, there are few ether linkages in the PPC, and this can be contrasted with the carbonate incorporation into PPO in the absence of DMAP as shown in Table

3.2.

k Al OP 1 +AlOCO2 OP k-1 C O

O O O k B CO2 k > k 3 k2 A 2 3

O OP Al O C OP O Al O

(carbonate linkage) (ether linkage)

' k 1 Nu Al OP + CO Nu Al O OP 2 ' k -1 C O

O O ' O k 2 A ' ' ' k 2 < k 3 k 3 B CO2

OP Nu Al O O OP Nu Al O C O (ether linkage) (carbonate linkage)

Scheme 3.6. Proposed mechanism for PO/CO2 copolymerization by (TPP)AlX systems, where Nu represents the added Lewis base promoter. 101

3.2.5 Substituent Effects on the Porphyrin Ligand.

(TPFPP)AlO(CH2)9CH3 and (OEP)AlO(CH2)9CH3, where TPFPP = tetrapentafluorophenylporphyrin and OEP = octaethylporphyrin (Figure 3.10), were prepared for the study of substituent effects on porphyrin ligand. For the initial ring opening of PO, (TPFPP)AlO(CH2)9CH3 shows a much higher rate (25 times) than

(TPP)AlO(CH2)9CH3, while (OEP)AlO(CH2)9CH3 shows little reactivity at the condition described before. The Lewis acidity of Al center is significantly increased by introducing strong electron withdrawing (TPFPP) groups, which also increases the polarity of Al–OR bond. Thus, in the case of (TPFPP)AlO(CH2)9CH3, the PO ring opening is facilitated by stronger PO activation and more readily dissociation of Al–OR.

F F F

F F

F F F F N N N N F Al F Al N N N N F F X F F X

F F

F F X = O(CH ) CH F 2 9 3

Figure 3.10. The structures of (TPFPP)AlO(CH2)9CH3 and (OEP)AlO(CH2)9CH3.

The difference in Lewis acidity among these three porphyrin aluminum alkoxides can also be seen in the study of DMAP binding by the method described in 3.2.1. The Keq

102 -1 -3 -1 of DMAP binding is 0.50 mM for (TPFPP)AlO(CH2)9CH3 and 1.6 × 10 mM

-3 -1 (OEP)AlO(CH2)9CH3, in comparison with 5.8 × 10 mM for (TPP)AlO(CH2)9CH3.

These results clearly show that the Lewis acidity of these complexes is in the order of

(TPFPP) > (TPP) > (OEP).

120

100

80

60 %

40

20

0 00.511.522.53 [DMAP] : [(porp)AlOR] ratio

Figure 3.11. The percentages of alkyl carbonates and alkoxides versus

13 [DMAP]:[(porp)AlO(CH2)9CH3] ratio for the CO2 insertion with (porp)AlO(CH2)9CH3

13 13 complexes. ∆(OEP)AlOR, ▲ (OEP)AlO2 COR; □ (TPP)AlOR, ■ (TPP)AlO2 COR; ○

13 (TPFPP)AlOR, ● (TPFPP)AlO2 COR (precipitate formed in the reaction, and measure the species in solution).

We have examined the carbon dioxide insertion to (TPFPP)AlO(CH2)9CH3 and

(OEP)AlO(CH2)9CH3 as shown in Figure 3.11. Interestingly we noticed that the product

103 in the reaction of (TPFPP)AlO(CH2)9CH3 and CO2 was of relatively low solubility in

CDCl3, and part of the product precipitated out from the solution. In the absence of

DMAP, (OEP)AlO2CO(CH2)9CH3 exists at a slightly higher equilibrium concentration than (TPP)AlO2CO(CH2)9CH3. However, the addition of DMAP increases the equilibrium concentration of (TPP)AlO2CO(CH2)9CH3 more significantly than that of

(OEP)AlO2CO(CH2)9CH3.

b c d e Entry Catalytic systems Reaction PPC PPO PC Mn PDI times (h) (%) (%) (%) (×103)

1 (TPP)AlO(CH2)9CH3 144 20 - - 13 1.16

2 (TPFPP)AlO(CH2)9CH3 144 11 - - 19 1.08

3 (OEP)AlO(CH2)9CH3 144 34 - - 56 1.07

Table 3.3. Copolymerization of PO/CO2 by (porp)AlO(CH2)9CH3 complexes in the presence of 1 equiv of DMAP.a

(a) The polymerization reactions were carried out in neat PO (2 mL) at 25oC under 50 bar

CO2 pressure, [cat]:[PO] = 1:1500, 1 equiv of DMAP added. (b) Carbonate linkages in the polymers obtained in the reaction mixtures determined by 1H NMR. (c) Ether linkages in the resulting polymers determined by 1H NMR. (d) Cyclic propylene

1 carbonate in the products determined by H NMR. (e) Mn and PDI determined by gel permeation chromatography.

The copolymerization reactions of PO and CO2 catalyzed by the porphyrin aluminum alkoxides have been carried out in the presence of 1 equiv of DMAP as shown 104 in Table 3.3. The reactivity is in the order of (OEP) > (TPP) > (TPFPP), which is the reverse order of their Lewis acidity. We should note that Darensbourg found that the reactivity of (salen)CrX in PPC formation was also enhanced by introducing donor groups onto the salen ligand.25 Importantly, the polymer microstructural study reveals that PPCs were formed with essentially the same microstructure by using these three porphyrin aluminum catalysts (Figure 3.12). Thus we can reasonably believe that the geometry of the porphyrin Al center, rather than its Lewis acidity, determines the microstructure of PPC.

(TPFPP)AlOR/DMAP + rac- PO/CO 2

HT

TT HH i s HT’ i s

155.00 154.80 154.60 154.40 154.20 154.00 153.80 153.60 153.40 ppm

(OEP)AlOR/DMAP + rac- PO/CO2

HT

TT HH i s i HT’ s

155.00 154.80 154.60 154.40 154.20 154.00 153.80 153.60 153.40 ppm

13 1 Figure 3.12. C{ H} (150 MHz, CDCl3) NMR spectra of the carbonate carbon region of

PPC made from (porp)AlO(CH2)9CH3/DMAP and rac-PO/CO2. (HH’ are ether rich carbonate signals.) 105

3.2.6 Gas-Phase Studies

3.2.6.1 Binding of propylene oxide. In a very rudimentary study, the tetraphenylporphyrin metal chlorides (TPP)MCl were dissolved in methanol and examined by electrospray ionization mass spectrometry with varying ionization voltages in order to determine the relative ease of heterolytic cleavage of the metal-chlorine bond for M = Al, Cr and Co. From this study it was clear that formation of the (TPP)Co+ cation

(threshold voltage 700V) was much easier than formation of the aluminum (800V) and chromium (800V) analogues. The latter metals show similar ionization thresholds.

Subsequent to this, we studied the binding of PO to the (TPP)M+ cations by electrospray tandem mass spectrometry as depicted in Scheme 3.7.32 In the first reaction chamber the (TPP)M+ ions are allowed to react with PO (1×10-2 Torr), and the resultant

+ ions (TPP)M(PO)n , where n = 1 or 2, are detected and selected. As is shown in Figure

3.13 the cobalt cation has a very low affinity for PO and is present almost exclusively as

+ + (TPP)Co with just a detectable amount of (TPP)Co(PO)2 . This is a clear indication of the weaker Lewis acidity of (TPP)Co+ and parallels the ease of its formation noted

+ earlier. Aluminum forms both (TPP)Al(PO)n cations for n = 1 and 2 with a slight

+ preference for n = 1, while for chromium the (TPP)Cr(PO)2 cation is clearly much

+ preferred. Also shown in Figure 3.13 are ions due to (TPP)Ga(PO)n (n = 1, 2) and

+ related (salen)M(PO)n , where M = Al and Cr, and these will be discussed later. These

PO adducts formed in the gas phase are not ring-opening derivatives as will be shown by their collision induced displacement experiments, vida infra.

106

Scheme 3.7. Schematic representation of the modified Finnigan MAT TSQ-700 ESI tandem mass spectrometer.

In the second reaction chamber the relative binding affinities of the

+ (TPP)M(PO)n ions are examined by collision induced dissociation (CID) experiments. A

+ selected ion (TPP)M(PO)n (n = 1 or 2) at the same concentration is allowed to react with xenon (3×10-5 Torr) and the voltage is increased, and the relative abundance of the ions

(both the parent and daughter) are detected as shown in Scheme 3.7.

From the graphical data presented in Figure 3.14 we can conclude that for the ions containing one PO the ease of dissociation follows the order (TPP)Ga+ > (salen)Al+

> (TPP)Al+, which is the inverse order of their Lewis acidities. This makes good sense as gallium is notably softer than aluminum and element-element bond strengths decrease down a series in the main group. The difference in PO binding to the salen and porphyrin

Al+ cations is also particularly noteworthy and reflects the fact that salen is a significantly stronger donor ligand relative to TPP. The (TPP)Al+ cation is more Lewis acidic than its

(salen)Al+ counterpart. We could not examine (TPP)Cr(PO)+ ion as its concentration was relatively too low. However, a comparison is seen for the bis PO adducts. 107 Figure 3.13. ESI/MS from the reactions between (TPP)/(salen) metal cations and propylene oxide in the gas phase.

Figure 3.14. The plots of LM(PO)+ abundance versus CID voltage from the dissociation reaction of LM(PO)+ with xenon in the CID chamber. 108

+ Figure 3.15 and 3.16 compare the (TPP)M(PO)2 (M = Al, Cr) and

+ (salen)Cr(PO)2 ions with respect to their dissociation of PO. Specifically in Figure 3.15

+ + we can see the loss of the bis PO cations in the order (TPP)Al(PO)2 >> (TPP)Cr(PO)2 ~

+ (salen)Cr(PO)2 . This is a clear indication of the high affinity of the chromium(III) ion to retain six coordination. In Figure 3.16, we show the appearance of the mono PO cations.

Here we see the growth of (TPP)Al(PO)+ > (TPP)Cr(PO)+ > (salen)Cr(PO)+ with the latter being at a maximum in the order of only 5%. From this we can conclude that the binding of PO to the four metal ions is in the order Al > Cr and that for chromium the

TPP favors coordination relative to salen. For the chromium salen cations there is clear evidence of the cooperative binding of PO. As shown in the equation below, for

+ + (salen)Cr(PO)2 , k2 >> k1. Whereas for (TPP)Al(PO)2 the initial loss of PO forms

+ (TPP)Al(PO) predominantely, ie. k2 << k1.

k1 k2 LM(PO) + LM(PO)+ LM+ 2 -PO -PO

109

+ Figure 3.15. The plots of LM(PO)2 abundance versus CID voltage from the dissociation

+ reactions of LM(PO)2 with xenon in the CID chamber.

Figure 3.16. The plots of LM(PO)+ abundance versus CID voltage from the dissociation

+ reactions of LM(PO)2 with xenon in the CID chamber.

110

3.2.6.2 Ring opening of propylene oxide in the gas phase. Since the cationic mechanism for epoxide ring opening is well known for some metal complexes,33-37 we examined the possibility of a cationic pathway for PO ring-opening polymerization in the gas phase by the (TPP) and (salen) metal cations. As shown in Figure 3.13, no metal ions with PO adduct number > 2 were detected. Furthermore, we examined the possibility of

PO ring opening with the assist of neutral nucleophiles, such as diethylamine and

+ + pyridine. One cation, either LM(PO) or LM(PO)2 , was selected and reacted with the neucleophile (3×10-4 Torr) in the CID chamber. Interestingly, ligand displacements were observed rather than PO ring opening. A typical mass spectrum is shown in Figure 3.17.

We should notice that the species in the gas phase are in an extremely low concentration relative to the solution phase, and these cations are not among the ones with strong Lewis acidity. In fact, PO ring-opening polymerization was observed when a bimetallic cation

2+ 38 of [(salen)Cr]2 was used in the experiment.

+ Figure 3.17. ESI/MS from the reaction of (TPP)Al(PO)2 with pyridine in the gas phase.

111

3.2.6.3 Polymerization reactions in the solution phase. As shown in Table 1 only the (TPP) metal chlorides of aluminum and chromium are effective in polymerizing

PO at ambient temperature and of these chromium is the more active. The addition of the strong nucleophile (good Lewis base, 1 equiv) 4-(dimethylamino)pyridine, DMAP, suppresses the polymerization and for aluminum essentially renders the system inactive.

The related cobalt complex with or without DMAP is essentially inactive. The salen metal chlorides of chromium and aluminum are also inactive at 25oC with respect to homopolymerization of PO although both will ring open one molecule of PO.30,39

However, at 75 oC in neat PO in a bomb the salen chromium becomes active.38

In Table 3.2 we present data pertaining to the copolymerization of PO and CO2, where it is evident that the reactivity order is M = Cr > Al > Co, and that the added

DMAP favors the formation of PPC. Indeed in its absence only a trace of CO2 is incorporated into the polymer which is largely PPO. Also we can see that the activity order on the attendent ligands is TPP > salen. We should note that Darensbourg has extensively studied the salen chromium system with respect to both changes of the ligand itself (backbone and aromatic ring) and Lewis base additives.10,11,25 Also Coates has reported that salen cobalt acetate is active in the formation of PPC,22 and (salen)Co(III) compounds have been extensively studied as catalysts in the hydrolytic kinetic resolution of terminal epoxides.40-42 More recently, Lu and coworkers found when a cocatalyst, such as Bu4NCl, was employed with (salen)CoX catalysts, the catalyst efficiency can be

43 enhanced significantly for PO/CO2 coplymerization reaction.

112 3.2.6.4. Solid-state and molecular structures. The ring opening of S-PO by

(R,R-salen)AlCl yields two products: (R,R-salen)AlOC(S)HMeCH2Cl (1) as the major product and (R,R-salen)AlOCH2C(R)HMeCl as the minor product as shown in 3.2.1. The molecular structure of compound 1 was determined on a crystal grown in neat S-PO and is particularly pertinent to the present study. In the solid state there are two five coordinate aluminum (III) centers that are nestled together in a head-tail salen packing motif as shown in Figure B.5 in Appendix B. There is also one molecule of PO trapped in the lattice. The coordination about aluminum is best described as square based pyramid with the Al–OR (alkoxide) bond in the axial site. Figure 3.18 shows one of the two Al molecules in the unit cell. Selected bond distances are given in Table 3.4.

Figure 3.18. ORTEP view of (R,R-salen)AlOC(S)HMeCH2Cl.

113 Al(1a)─O(1a) 1.800(2) O(3a)─Al(1a)─O(2a) 112.2(1) Al(1a)─O(2a) 1.793(2) O(3a)─Al(1a)─O(1a) 108.1(1) Al(1a)─O(3a) 1.736 (2) O(2a)─Al(1a)─O(1a) 90.5(1) Al(1a)─N(1a) 2.027(2) O(3a)─Al(1a)─N(2a) 102.4(1) Al(1a)─N(2a) 2.007(2) O(2a)─Al(1a)─N(2a) 87.8(1) O(3a)─C(2a) 1.405(3) O(1a)─Al(1a)─N(2a) 147.6(1) C(1a)─C(2a) 1.540(5) O(3a)─Al(1a)─N(1a) 96.0(1) C(2a)─C(3a) 1.521(5) O(2a)─Al(1a)─N(1a) 150.8(1) Cl(1a)─C(3a) 1.806(3) O(1a)─Al(1a)─N(1a) 87.5(1) N(2a)─Al(1a)─N(1a) 78.7(1) C(2a)─O(3a)─Al(1a) 132.2(2) O(3a)─C(2a)─C(3a) 105.3(2) O(3a)─C(2a)─C(1a) 110.1(3) C(3a)─C(2a)─C(1a) 111.3(3) C(2a)─C(3a)─Cl(1a) 111.7(2)

Table 3.4. Selected bond distances (Å) and angles (deg) for (R,R- salen)AlOC(S)HMeCH2Cl.

The molecule of S-PO is shown in Figure 3.19. This is the first structural determination of the free (unligated) PO molecule.44-46 Pertinent bond distances and angles are given in the caption. What is particularly pertinent here is that the aluminum center remains five-coordinate even when PO is present as a solvent. For salen chromium azide Jacobsen reported the structure of the ring-opening product of cyclopentene oxide and found a solvent molecule of THF in the axial position.30,47

114

Figure 3.19. ORTEP view of S-propylene oxide. Selected bond distances (Å) and angles

(deg): O(1c)─C(1c) 1.428(4), O(1c)─C(2c) 1.430 (5), C(1c)─C(2c) 1.431(6),

C(2c)─C(3c) 1.454 (6); C(1c)─O(1c)─C(2c) 60.1(2), O(1c)─C(1c)─C(2c) 60.0(2),

O(1c)─C(2c)─C(1c) 59.9(2), O(1c)─C(2c)─C(3c) 117.6(3), C(1c)─C(2c)─C(3c) 123.5

(4).

An alkoxide ligand is known to have a higher trans-influence than that of a carboxylate or alkyl carbonate ligand.48 Consequently, we investigated the ability of

(salen)AlO2CMe to bind a stronger Lewis base, namely pyridine. The molecular structure of this compound determined from a crystal grown in neat pyridine as shown in Figure

3.20. The aluminum is again five-coordinate, and the acetate ligand is in the apical position of a square based pyramid. Significant note is the fact that the acetate is η1- coordinated. The long AlLO(acetate) distance is 3.45 Å, and the Al─O(acetate) bond of

1.77 Å is also longer than that in the related Al–O(alkoxide) bond (1.74 Å) (Tables 3.4 and 3.5). In the lattice there are two Al molecules and three noncoordinated pyridine molecules per unit cell (Figure B.6 in Appendix B)! This is a clear indication that the

Al(III) center is only weakly Lewis acidic. Of course, Al(III) six-coordinate ions are well 115 known and Atwood has shown than (salen)Al cations with weakly coordinating anions

- - (OTs , BPh4 , etc.) will readily pick up two molecules of O-donor solvents, such as

35,36,49 MeOH, THF or H2O.

Al(1a)─O(1a) 1.776(2) O(3a)─Al(1a)─O(2a) 100.6(1) Al(1a)─O(2a) 1.795(2) O(3a)─Al(1a)─O(1a) 104.3(1) Al(1a)─O(3a) 1.772 (2) O(2a)─Al(1a)─O(1a) 90.6(1) Al(1a)─N(1a) 1.998(3) O(3a)─Al(1a)─N(2a) 105.9(1) Al(1a)─N(2a) 2.002(3) O(2a)─Al(1a)─N(2a) 87.8(1) O(3a)─C(37a) 1.295(4) O(1a)─Al(1a)─N(2a) 149.5(1) O(4a)─C(37a) 1.210(4) O(3a)─Al(1a)─N(1a) 103.6(1) C(37a)─C(38a) 1.493(5) O(2a)─Al(1a)─N(1a) 154.9(1) O(1a)─Al(1a)─N(1a) 89.4(1) N(2a)─Al(1a)─N(1a) 79.7(1) C(37a)─O(3a)─Al(1a) 143.5(2) O(3a)─C(37a)─C(38a) 114.3(3) O(3a)─C(37a)─O(4a) 123.5(3) O(4a)─C(37a)─C(38a) 122.1(3)

Table 3.5. Selected bond distances (Å) and angles (deg) for (R,R-salen)AlOOCMe.

116

Figure 3.20. ORTEP view of (R,R-salen)AlOOCMe.

3.3 Concluding Remarks

The observation that the ring opening of PO by (TPP)AlO(CH2)9CH3 is first order in aluminum complex supports the earlier study on the kinetics of PPO formation employing (TPP)AlCl as initiator. These workers determined that the rate of PPO formation was first order in [Al] and first order in [PO].50-53 Others have shown that

(TPP)AlCl immobilized on a support will also ring open PO to give PPO.54 The mechanism is thus unequivocally mononuclear and different from that described by

30 Jacobsen for (R,R-salen)CrN3.

We have shown that added DMAP serves to accelerate two principal reactions.

(1) It promotes the insertion of CO2 into the Al-alkoxide, and (2) it labilizes the carboxylate ligand, and by inference the alkyl carbonate ligand, toward ring opening of

117 PO. In the latter reaction it is evident the DMAP binds more strongly to (TPP)AlO2CX species (X = R or OR) than it does to the parent alkoxide complex. Thus, whereas in the absence of added DMAP the (TPP)Al system is only effective in the homopolymerization of PO, when DMAP is added, alternating copolymerization of PO and CO2 is kinetically favored.

The stereochemical studies of the ring opening of PO in these systems reveal that attack occurs at both the methine and the methylene carbons, and furthermore when attack occurs at the methine carbon the ring-opening event can occur with retention or inversion of stereochemistry.8,9 Collectively, these findings serve to support a reaction pathway wherein PO and the growing chain are on the same side of the porphyrin ring.

The coordination of DMAP in the trans position serves to labilize the Al–OR or Al–

O2CX (X = R or OR) bonds toward dissociation and uptake of PO in an interchange substitution process. Upon coordination to the electrophilic aluminum center, the PO is activated toward carbonium ion-like behavior, and the attack by the nucleophilic growing chain leads to its enchainment in the polymerization reaction in a relatively nonregio- and nonstereoselective manner. We propose that the use of added ligands, such as 1- methylimidazole and Br- in related reactions, parallels the affects noted here for DMAP, but to a different degree.

For a series of related ions LM+, the Lewis acidity order follows M = Cr(III) ~

Al(III) > Co(III). For the formation of the mono PO adducts LM(PO)+, the metal–PO

+ bond strength follows the order Al > Cr. For the adducts LM(PO)2 , the loss of the first

PO is in the order Al > Cr. These notable differences in the metal porphyrin complexes can be traced to the dn configuration effects of the M(III) ions, d0 for Al, d3 for Cr and d6

118 for Co. The binding of PO to the (TPP)M+ ions was also noted to be stronger relative to the respective (salen)M+ ions which is taken to indicate the better donor properties of salen relative to TPP. Finally, the structural characterizations of the

(salen)AlOCHMeCH2Cl and (salen)AlO2CMe molecules in the presence of excess PO and pyridine, respectively, provide further testimony to the importance of five- coordination relative to six for the (salen)Al cation.

With respect to the homopolymerization of PO, the weaker Lewis acidity of the

Co(III) ion would seem to be responsible for its lack of activity. The difference between chromium and aluminum where the activity order is Cr > Al may well reflect the greater affinity for six-coordination which would allow for a bimolecular pathway in [Cr], as has

55 been shown by Jacobsen for (salen)CrN3 in asymmetric ring opening of epoxides.

However, different mechanisms may operate for different metal ions with different ligand sets. Clearly, more needs to be known about the details of these reaction profiles. An understanding of the factors controlling the reactions shown in Scheme 3.6 should allow for the development of highly active systems for the catalytic production of PPC and related polycarbonates.

3.4 Experimental Section

All syntheses and solvent manipulations were carried out under a nitrogen atmosphere using standard Schlenk-line and drybox techniques. Solvents were dried in the standard procedures. Propylene oxide (Alfa Aesar) was distilled from calcium hydride. 5,10,15,20-tetraphenylporphine (Fisher), diethylaluminum chloride (1.0 M solution in hexane, Aldrich), trimethylaluminum (2.0 M solution in hexane, Aldrich), 4-

119 dimethylaminopyridine (Aldrich), (R,R)-N,N’-bis(3,5-di-tert-butyl-salicylidene)-1,2- cyclohexenediamine (Aldrich), (R,R)-N,N’-bis(3,5-di-tert-butyl-salicylidene)-1,2- cyclohexenediaminoaluminum chloride (Aldrich) and (R,R)-N,N’-bis(3,5-di-tert-butyl- salicylidene)-1,2-cyclohexenediaminochromium chloride were used as received.

Chromium(II) chloride, cobalt(II) chloride and gallium(III) chloride were purchased from

Aldrich. n-Decanol (Fisher), n-octanoic acid (Fisher), (S) 2-chloro-1-propanol (Aldrich) and deuterated solvents were stored over 4 Å molecular sieves for 24 h prior to use.

NMR Experiments. 1H and 13C{1H} NMR experiments were carried out with a

Bruker DRX-400 (5 mm broad band probe), a Bruker DRX-500 (5 mm broad band probe) and a Bruker DRX-600 (5mm broad band probe) spectrometers, operating at proton Larmor frequencies of 400, 500 and 600 MHz, respectively. The parameters used in 13C{1H} NMR experiments on Bruker DRX-600 spectrometer were: number of data point, TD=65536, sweep width, SWH=1502 Hz, relaxation time, D1=2 sec, and chemical shift range 0 - 200 ppm. Their peak frequencies were referenced against the solvent, chloroform-d at 7.24 ppm for 1H and 77.0 ppm for 13C{1H} NMR .

Gel Permeation Chromatography. Gel permeation chromatographic (GPC) analysis was performed at 35 oC on a Waters Breeze system equipped with a Waters 410 refractive index detector and a set of two columns, Waters Styragel HR-2, and HR-4

(Milford, MA). THF (HPLC grade) was used as the mobile phase at 1.0 ml/min. The sample concentration was 0.1 %, and the injection volume was 100 µl. The samples were centrifuged and filtered before analysis. The calibration curve was made with six polystyrene standards covering the molecular weight range from 580 to 460, 000 Da.

120 ESI Tandem Mass Spectrometry/Gas Phase Studies. The complexes are

-5 typically diluted to 10 M with CH2Cl2 and electrosprayed on a modified Finnigan MAT

TSQ-700 tandem mass spectrometer (Scheme 3.7) at a flow rate of 7-15 ml min–1 and at a potential of 4-5 kV using nitrogen as sheath gas. The ions are then passed through a heated capillary (typically at 150°C) where they are declustered and the remaining solvent molecules are removed by high vacuum. The extent of desolvation and collisional activation can further be controlled by a tube lens potential, which typically from 50 V

(soft condition) to 120 V (hard condition). In this work even harder desolvation conditions was used to get strong signal, e.g. tube lens ~ 150 V. The transfer 24-pole

(O1) acts as an ion guide to separate the ions from neutral molecules which are pumped off by a turbo pump located underneath the octopole. O1 is fitted with an open cylindrical sheath around the rods into which, depending on the setup used, a collision gas can be bled for thermalization or reaction at pressures up to 100 mTorr. Modification of normal octopole by a 37cm radio-frequency (RF) 24-pole is based on requirements of the longer reaction time and bigger pressure of reagent gas. The ions then enter the actual mass spectrometer, which is at 10–6 Torr and 70°C manifold temperature during operation. The configuration is quadrupole/octopole/quadrupole (Q1/O2/Q2), with the two quadrupoles as mass selection stages and the second octopole operating as a collision-induced dissociation (CID) cell. Spectra can be recorded in different modes. In the normal ESMS mode, only one quadrupole is operated (either Q1 or Q2), and a mass spectrum of the electrosprayed ions is recorded. This mode serves primarily to characterize the ions produced by a given set of conditions. In the daughter-ion mode, Q1 is used to mass- select ions of a single mass-to-charge-ratio from among all of the ions produced in O1,

121 which are then collided or reacted with a target gas in O2 (CID cell), and finally mass- analyzed by Q2. This mode is used to obtain structural information (by analysis of the fragments) or the specific reactivity of a species of a given mass. Therefore, collision- induced dissociation (CID) could be done in either the radio-frequency (rf) 24-pole ion guide by collision with Xe (0.08 torr) or in a gas-filled (0.1 mtorr of either Xe or reagent gas) rf octopole ion guide. CID and ion-molecule reactions were all performed at low collision energy, i.e. –10 to –50 V in the laboratory frame.

X-Ray Crystallographic Studies. The data collection crystals for

(salen)AlOCHMeCH2Cl and (salen)AlOOCMe were both yellow chunks, which had been cut from a large crystal. Examination of the diffraction pattern on a Nonius Kappa

CCD diffractometer indicated a triclinic crystal system. All work was done at 200 K using an Oxford Cryosystems Cryostream Cooler. The data collection strategy was set up to measure a hemisphere of reciprocal space with a redundancy factor of 3.1 for

(salen)AlOCH2CHMeCl and 3.4 for (salen)AlOOCMe, which means that 90% of the reflections were measured at least 3.1 and 3.4 times, respectively. A combination of phi and omega scans with a frame width of 1.0 degree was used. Data integration was done with Denzo, and scaling and merging of the data was done with Scalepack. Merging the data (but not the Friedel pairs) resulted in a Rint value of 0.036.

The structure of (salen)AlOCHMeCH2Cl was solved in P1 by the direct methods in SHELXS-97. The asymmetric unit contains two molecules of the Al complex and one solvent molecule of propylene oxide. The Al complexes are labeled as molecules A and

B. One of the t-butyl groups on molecule B was rotationally disordered and was modeled as two isotropic sets of atoms.

122 The structure of (salen)AlOOCMe was solved in P1 by the direct methods in

SHELXS-97. The asymmetric unit contains two molecules of the Al complex and three molecules of pyridine. The Al complexes are labeled as molecules A and B. Some of the t-butyl groups were rotationally disordered and were modeled as two isotropic sets of atoms. For two of the three pyridine molecules (labeled as C and E), the location of the nitrogen atom was found by initially naming all of the atoms as carbon, and then seeing which atom had the smallest refined isotropic U value. This atom was then changed to a nitrogen atom. For the third pyridine ring (labeled as D), the location of the nitrogen atom was not obvious with this method, and so all of the atoms were designated as carbon atoms.

Experimental data relating to both structure determinations are displayed in Table

3.6.

Synthesis of (TPP) aluminum chloride. (TPP)AlCl was prepared according to the

16,27 literature. To a solution of (TPP)H2 (1.0 g, 1.6 mmol) in 30 mL of dichloromethane was added a solution of diethylaluminum chloride (Et2AlCl 1.0 M solution in hexane, 1.8 mL, 1.8 mmol). The addition was done slowly at room temperature. The resulting solution was left stirring 3 h after which the volatile fractions were removed under

1 vacuum to give a purple powder in 85% yield. H NMR (CDCl3, δ, ppm): 7.7, 8.2, 9.0

13 1 (aromatic). C{ H} NMR (CDCl3, δ, ppm): 120.73, 126.95, 128.06, 132.39, 134.23,

141.14, 148.51 (aromatic). ESI/MS data: m/z = 639, (TPP)Al+.

Synthesis of (TPP) aluminum methyl. (TPP)AlMe was prepared according to the

56 literature. To a solution of (TPP)H2 (1.0 g, 1.6 mmol) in 30 mL of dichloromethane was added a solution of trimethylaluminum (AlMe3 2.0 M solution in hexane, 0.96 mL, 1.9 123 mmol). The addition was done slowly at room temperature. The resulting solution was left stirring 3 h, after which the volatile fractions were removed under vacuum. After washing with hexane, the product was obtained as a purple powder in 83% yield. 1H

NMR (CDCl3, δ, ppm): -7.0 (s, CH3), 7.6, 8.1, 8.9 (aromatic).

Synthesis of (TPP) aluminum n-decoxide. To a solution of (TPP)AlMe (1.0 g, 1.5 mmol) in 30 mL of dichloromethane was added 0.34 mL of n-decanol (1.8 mmol), and the mixture was stirred overnight at 40 oC. After the reaction, the volatile fractions were removed at 120 oC under high vacuum. After washed twice with small amounts of

1 hexane, the product was obtained as a purple powder in 78% yield. H NMR (CDCl3, δ, ppm): -1.96, -1.42, -1.10, 0.06, 0.55, 0.85, 1.01, 1.09, 1.18 ((CH2)9), 0.82 (t, CH3), 7.75,

13 1 8.17, 9.02 (aromatic). C{ H} NMR (CDCl3, δ, ppm): 14.11 (CH3), 22.65, 23.98, 28.85,

29.25, 29.39, 31.13, 31.60, 31.84, 57.13 ((CH2)9), 120.55, 126.83, 127.85, 132.09,

134.24, 141.63, 148.79 (aromatic). Elemental analysis: Calcd.: C, 81.38; H, 6.20; N, 7.03.

Found: C, 80.09; H, 5.90; N, 7.01%.

Synthesis of (TPP) aluminum n-octanoate. To a solution of (TPP)AlMe (1.0 g,

1.5 mmol) in 30 mL of benzene was added 2.8 mL of n-octanoic acid (1.8 mmol), and the mixture was stirred overnight at room temperature. The volatile fractions were removed at 120 oC under high vacuum. After washed twice with small amounts of toluene, the

1 product was obtained as a purple powder in 74% yield. H NMR (CDCl3, δ, ppm): -1.20,

13 1 -0.74, -0.27, 0.43, 0.75, 0.99 ((CH2)6), 0.71 (t, CH3), 7.75, 8.16, 9.05 (aromatic). C{ H}

NMR (CDCl3, δ, ppm): 14.00 (CH3), 22.46, 23.68, 28.01, 28.44, 31.38, 33.86 ((CH2)6),

120.72, 126.90, 128.00, 132.32, 134.14, 141.23, 148.64 (aromatic), 170.23 (O-C=O).

124 Elemental analysis: Calcd.: C, 79.77; H, 5.54; N, 7.16. Found: C, 77.15; H, 5.44; N,

6.87%.

Synthesis of (TPP) aluminum 2-chloro-propoxide. To a solution of (TPP)AlMe

(1.0 g, 1.5 mmol) in 20 mL of dichloromethane was added 0.16 mL of (S) 2-chloro-1- propanol (1.8 mmol), and the mixture was stirred overnight at room temperature. After that, the volatile fractions were removed under vacuum. After washed twice with hexane, the product was obtained as a purple powder in yield of 83% and characterized by 1H

13 1 1 1 NMR, C NMR and COSY ( H- H correlation spectroscopy). H NMR (CDCl3, δ, ppm):

13 1 -1.36 (m, OCH2), -1.03 (d, CH3), -0.10 (m, CH-Cl), 7.75, 8.16, 9,04 (aromatic). C{ H}

NMR (CDCl3, δ, ppm): 18.96 (CH3), 63.64 (HC-Cl), 69.25 (H2C-O), 120.78, 126.93,

128.00, 132.30, 134.22, 141.38, 148.81 (aromatic).

Synthesis of (R,R-salen) aluminum 2-chloro-propoxide. To a solution of (R,R- salen)H (1.0 g, 1.8 mmol) in 20 mL of dichloromethane was added a solution of trimethylaluminum (AlMe3 2.0 M solution in hexane, 2.2 mL, 2.2 mmol), and the mixture was stirred for 3 h at room temperature. A yellow powder was obtained after the removal of volatile fractions under vacuum. The powder was redisolved in hexane, followed by the addition of 0.19 mL of (S) 2-chloro-1-propanol (2.2 mmol), stirred for 3 h. A yellow precipitate was obtained as the product in 75% yield. Crystals of this complex were obtained by slow evaporation of the solvent from a concentrated benzene

1 solution of the complex.. H NMR (CDCl3, δ, ppm): 1.17 (d, CH3-CHCl) 3.40, 3.50 (m,

OCH2), 3.68 (m, CHCl), 1.29 (d, C(CH3)), 1.51 (d, C(CH3)), 1.44, 2.07, 2.43, 2.58, 3.06,

3.90 (cyclohexyl), 7.00, 7.05, 7.48, 7.51 (aromatic), 8.17, 8.35 (HC=N). 13C{1H} NMR

(CDCl3, δ, ppm): 21.43 (CH3CH), 23.75, 24.27, 27.27, 28.80 (CH2 in cyclohexyl), 29.67, 125 29.94, 31.36, 31.43 ((CH3)3C), 33.96, 33.99, 35.56, 35.64 ((CH3)3C), 61.10, 62.64 (HC-

N), 65.86 (HC-Cl), 69.25(H2C-O), 118.12, 118.35, 127.37, 127.75, 129.84, 131.11,

137.82, 138.29, 140.59, 140.79, 162.27, 163.92 (phenyl), 163.02, 168.19 (HC=N).

Elemental analysis: Calcd.: C, 70.41; H, 8.79; N, 4.21. Found: C, 69.16; H, 8.44; N,

4.03%.

Synthesis of (TPFPP) aluminum n-decoxide. To a solution of (TPFPP)H2 (0.5 g,

0.5 mmol) in 10 mL of dichloromethane was added a solution of trimethylaluminum

(AlMe3 2.0 M solution in hexane, 0.3 mL, 0.6 mmol). The addition was done slowly at room temperature. The resulting solution was left stirring 3 h, after which the volatile fractions were removed under vacuum. After washing with hexane, the product was obtained as a purple powder, which was then dissolved in 30 mL of dichloromethane, and

0.11 mL of n-decanol (0.6 mmol) was added into the solution. The mixture was stirred overnight at 40 oC. After the reaction, the volatile fractions were removed at 120 oC under high vacuum. After washed twice with small amounts of hexane, the product was

1 obtained as a purple powder in 60% yield. H NMR (CDCl3, δ, ppm): -2.03, -1.46, -1.10,

-0.01, 0.47, 0.77, 0.97, 1.06, 1.17 ((CH2)9), 0.80 (t, CH3), 9.10 (aromatic). Elemental analysis: Calcd.: C, 56.07; H, 2.53; N, 4.84. Found: C, 55.15; H, 2.46; N, 4.68%.

Synthesis of (OEP) aluminum n-decoxide. To a solution of (OEP)H2 (0.5 g, 0.9 mmol) in 10 mL of dichloromethane was added a solution of trimethylaluminum (AlMe3

2.0 M solution in hexane, 0.6 mL, 1.2 mmol). The addition was done slowly at room temperature. The resulting solution was left stirring 3 h, after which the volatile fractions were removed under vacuum. After washing with hexane, the product was obtained as a purple powder, which was then dissolved in 30 mL of dichloromethane, and 0.24 mL of

126 n-decanol (1.2 mmol) was added into the solution. The mixture was stirred overnight at

40 oC. After the reaction, the volatile fractions were removed at 120 oC under high vacuum. After washed twice with small amounts of hexane, the product was obtained as a

1 purple powder in 65% yield. H NMR (CDCl3, δ, ppm): -2.39, -1.82, -1.30, -0.03, 0.51,

0.85, 1.03, 1.10, 1.20 ((CH2)9), 0.84 (t, CH3((CH2)9), 1.93 (t, CH3CH2), 4.14 (m,

CH3CH2), 10.25 (aromatic). Elemental analysis: Calcd.: C, 77.06; H, 9.14; N, 7.81.

Found: C, 76.76; H, 8.98; N, 7.75%.

Synthesis of (TPP) chromium chloride. (TPP)CrCl was prepared according to the

57,58 literature. (TPP)H2 (2.38 g, 3.9 mmol) was added to 200 mL of dimethylformamide

o (DMF). The mixture was stirred under reflux at 170 C. After 10 min the (TPP)H2 was dissolved and an excess of CrCl2 (0.75 g, 6.1 mmol) was added to the refluxing solution.

After 20 min, an aliquot was taken from the mixture for UV-vis. The spectrum showed free (TPP)H2 still remained in solution. A further addition of CrCl2 (0.5 g) was added to the solution, refluxing for another 20 min. No free (TPP)H2 was found in UV-vis measurement. The reaction mixture was allowed to cool to room temperature, and poured into 400 mL of ice-cold water. After filtered and washed three times with water, the solid was dried under vacuum at 100oC. The crude product was purified by column chromatography with CHCl3 as the eluent over aluminia column. After CHCl3 was removed, the product was dried overnight under vacuum at 100oC, yielding 1.95 g of product. UV-vis spectroscopy was used to confirm the formation of (TPP)CrCl.58

ESI/MS data: m/z = 664, (TPP)Cr+.

Synthesis of (TPP) cobalt chloride. (TPP)CoCl was prepared according to the

57,59 literature. (TPP)H2 (1.5 g, 2.4 mmol) was added to 200 mL of DMF. The mixture was

127 o stirred under reflux at 170 C. After 10 min the (TPP)H2 was dissolved and an excess of

CoCl2 (0.4 g, 3.1 mmol) was added to the refluxing solution. After 30 min, an aliquot was taken from the mixture for UV-vis. The spectrum showed free (TPP)H2 still remained in solution. A further addition of CoCl2 (0.4 g) was added to the solution, refluxing for another 30 min. No free (TPP)H2 was found in UV-vis measurement. The reaction mixture was allowed to cool to room temperature, and poured into 300 mL of concentrated HCl/MeOH solution overnight. After the removal of solvents and washing with a large amount of water, a purple solid was obtained and dried under vacuum at

o 100 C. The crude product was purified by column chromatography with CH2Cl2 (5%

MeOH) as the eluent over silica column. After solvents were removed, the product was dried overnight under vacuum at 100oC, yielding 1.25 g of product. UV-vis spectroscopy

59 1 was used to confirm the formation of (TPP)CoCl. H NMR (CDCl3, δ, ppm): 7.70, 8.19,

8.70 (aromatic). ESI/MS data: m/z = 671, (TPP)Co+.

Synthesis of (TPP) gallium chloride. (TPP)GaCl was prepared according to the

60,61 literature. (TPP)H2 (0.25g, 0.4mmol), anhydrous sodium acetate (0.8g, 9.6mmol) and

GaCl3 (0.2g, 1.1mmol) were added to 100 mL of acetic acid. The mixture was stirred and refluxed overnight at 120oC. The reaction mixture was allowed to cool to room temperature. After the removal of acetic acid, the product was extracted with CHCl3. The crude product was purified by column chromatography with CHCl3 as the eluent over aluminia column. After CHCl3 was removed, the product was dried overnight under vacuum at 100oC, yielding 0.15 g of product. UV-vis spectroscopy was used to confirm

61 1 the formation of (TPP)GaCl. H NMR (CDCl3, δ, ppm): 7.76, 8.20, 9.07 (aromatic).

ESI/MS data: m/z = 681, (TPP)Ga+. 128 Synthesis of (R,R-salen) aluminum acetate. To a solution of (R,R-salen)H2 (2.0 g,

3.7 mmol) in 20 mL of dichloromethane was added a solution of trimethylaluminum

(AlMe3 2.0 M solution in hexane, 2.0 mL, 4.0 mmol), and the mixture was stirred for 3 h at room temperature. A yellow powder was obtained after the removal of volatile fractions under vacuum. The powder was redissolved in hexane, followed by the addition of 0.25 mL of acetic acid (4.4 mmol), stirred for overnight. A yellow precipitate was obtained as the product (1.7 g). Crystals of this complex for x-ray measurement were obtained by slow evaporation of the solvent from a concentrated pyridine solution of the

1 complex. H NMR (CDCl3, δ, ppm): 1.79 (s, CH3-COO), 1.29 (d, C(CH3)), 1.50 (s,

C(CH3)), 1.44, 2.05, 2.41, 2.56, 3.09, 4.01 (cyclohexyl), 7.02, 7.09, 7.51 (aromatic), 8.18,

8.36(HC=N). Elemental analysis: Calcd.: C, 72.35; H, 8.79; N, 4.44. Found: C, 72.21; H,

8.66; N, 4.35%.

Short-time reaction between (TPP)AlCl and rac-PO. 1.0 mg of (TPP)AlCl

(0.0015mmol) was dissolved in 1 mL of chloroform, and 50 µL rac-PO (0.72 mmol) was added into the solution. After 15 min, the volatile fractions were removed from the mixture under vacuum. The obtained products were characterized by 1H NMR. 1H NMR

(CDCl3, δ, ppm): -2.75 (m, OCH), -1.91 (d, CH3CHO), -0.36, 0.31 (m, CH2Cl), -1.36 (m,

OCH2), -1.02 (d, CH3CHCl), -0.10 (m, CHCl), 7.75, 8.16, 9.02, 9.04 (aromatic) (Figure

3.1).

Reactions between (R,R-salen)AlCl and POs. In a typical reaction, 5 mg (0.008 mmol) of (salen)AlCl was allowed to react in 2 mL (28.6 mmol) of neat PO (S-, R- or rac-) for 2h at room temperature. A yellow powder was obtained after the removal of the excess of PO under vacuum. The products of all regio- and stereo-isomers were analyzed 129 by 1H NMR. The products were recrystalized in toluene to obtain the major regioisomers.

1 H NMR spectra are shown Figure S5. Crystals of (R,R-salen)AlOCH(S)MeCH2Cl were obtained by slow evaporation of the solvent from a concentrated S-PO solution of the complex.

Studies of Kinetics. The stock solutions of the initiators and DMAP in CDCl3 were prepared for the reactions. 50 µL of PO (0.72 mmol) was added to a certain amount of initiator solution (and DMAP solution if needed) to make a total volume of 1 mL, and a typical initiator concentration was 15 mM. The reactions were monitored at 25 oC by 1H

NMR, measuring the disappearance of initiator signals in the negative chemical shift region.

Carbon dioxide insertion reactions with (TPP)AlO(CH2)9CH3. In a typical

13 reaction, 4 atm of CO2 pressure was introduced into a J-Young NMR tube containing

o 11 mM (TPP)AlO(CH2)9CH3 in CDCl3 at 25 C, in the presence of a certain equivalent of DMAP. The reaction was monitored by 1H and 13C {1H} NMR. The signals from the

13 1 CO2 inserted complex are, H NMR (CDCl3, δ, ppm): -0.19, -0,05, 0.71, 0.87, 0.92,

13 1 1.04, 1.15, 1.26,1.35 (O2CO-(CH2)9-), 0.85 (t, CH3), 7.70, 8.16, 8.90 (aromatic); C { H}

13 NMR (CDCl3, δ, ppm): 149 (O2 CO).

Homopolymerization reactions of PO. In a typical reaction, 0.02 mmol of catalyst was allowed to react in 2 mL of neat PO (28.6 mmol) in a flask at room temperature, in addition of 2.4 mg of DMAP (0.02 mmol) if needed. After a certain time, the resulting mixture was analyzed by 1H NMR. Then the reaction was quenched by adding methanol/HCl solution, and the obtained polymer was analyzed by gel permeation chromatography. 130 Copolymerization reactions of PO/CO2. In a typical reaction, 0.02 mmol of catalyst was allowed to react in 2 mL of neat PO (28.6 mmol) in a stainless steel reaction vessel (Parr) under 50 bar CO2 pressure at room temperature, in addition of 2.4 mg of

DMAP (0.02 mmol) if needed. After a certain time, the resulting mixture was analyzed by 1H NMR. Then the reaction was quenched by adding methanol/HCl solution, and the obtained polymer was analyzed by gel permeation chromatography.

(salen)AlOCHMeCH2Cl (salen)AlOOCMe·1.5py ·0.5PO

Empirical formula C39H58AlClN2O3 + C38H55AlN2O3 +

1/2(C3H6O) 3/2(C5H5N) Formular weight 694.34 749.47 Temperature (K) 200 200 Wavelenghth (Å) 0.71073 0.71073 Crystal system triclinic triclinic Space group P1 P1 Unit cell dimensions a (Å) 12.776(1) 12.554(1) b (Å) 13.096(1) 12.887(1) c (Å) 13.855(1) 14.614(1) ß (o) 63.234(3) 85.652(3) V (Å3) 1990.7(3) 2151.1(3) Z 2 2 -3 Dcalc (mg m ) 1.158 1.157 Absorption coefficient (mm-1) 0.157 0.092 F(000) 752 810 Crystal size (mm-3) 0.27x0.31x0.42 0.27x0.35x0.35 Theta range for data collection 1.71-27.46 2.07-25.04 o 131 (o) Index ranges -16≤h≤16, -16≤k≤16, -14≤h≤14, -15≤k≤15, -17≤l≤17 -17≤l≤17 Reflections collected 54924 49416

Independent reflections 17967[Rint=0.036] 14712[Rint=0.035] Refinement method Full-matrix least-squares Full-matrix least-squares on F2 on F2 Data/restraints/parameters 17967/3/894 14712/66/985 Flack parameter 0.01(6) 0.03(15) Goodness-of-fit on F2 1.029 1.041

Final R indices [I>2σ(I)] R1=0.0572, wR2=0.1451 R1=0.0526, wR2=0.1324

R indices (all data) R1=0.0768, wR2=0.1569 R1=0.0671, wR2=0.1420 Largest difference peak and hole 0.588 and -0.592 0.493 and -0.312 (e/Å3)

Table 3.6. Crystallographic details for (salen)AlOCHMeCH2Cl·0.5PO and

(salen)AlOOCMe·1.5py.

3.5 References

(1) Darensbourg, D. J.; Holtcamp, M. W. Coord. Chem. Rev. 1996, 153, 155-174.

(2) Super, M. S.; Beckman, E. J. Trends Polym. Sci. 1997, 5, 236-240.

(3) Kuran, W. Prog. Polym. Sci. 1998, 23, 919-992.

(4) Nozaki, K. J. Polym. Sci., Part A: Polym. Chem. 2003, 42, 215-221.

(5) Rokicki, A.; Kuran, W. J. Macromol. Sci., Rev. Macromol. Chem. 1981, C21,

135-186.

(6) Nakano, K.; Kosaka, N.; Hiyama, T.; Nozaki, K. Dalton Trans. 2003, 4039-4050.

132 (7) Inoue, S. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 2861-2871.

(8) March, J. Advanced Organic Chemistry, 4th Ed.; Wiley: New York, 1992.

(9) Parker, R. E.; Isaacs, N. S. Chem. Rev. 1959, 59, 737-799.

(10) Darensbourg, D. J.; Yarbrough, J. C. J. Am. Chem. Soc. 2002, 124, 6335-6342.

(11) Darensbourg, D. J.; Yarbrough, J. C.; Ortiz, C.; Fang, C. C. J. Am. Chem. Soc.

2003, 125, 7586-7591.

(12) Allen, S. D.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc.

2002, 124, 14284-14285.

(13) Moore, D. R.; Cheng, M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc.

2003, 125, 11911-11924.

(14) Moore, D. R.; Cheng, M.; Lobkovsky, E. B.; Coates, G. W. Angew. Chem., Int.

Ed. Engl. 2002, 41, 2599-2602.

(15) Cheng, M.; Moore, D. R.; Reczek, J. J.; Chamberlain, B. M.; Lobkovsky, E. B.;

Coates, G. W. J. Am. Chem. Soc. 2001, 123, 8738-8749.

(16) Chisholm, M. H.; Navarro-Llobet, D. Macromolecules 2002, 35, 2389-2392.

(17) Antelmann, B.; Chisholm, M. H.; Iyer, S. S.; Huffman, J. C.; Navarro-Llobet, D.;

Pagel, M.; Simonsick, W. J.; Zhong, W. Macromolecules 2001, 34, 3159-3175.

(18) Le Borgne, A.; Spassky, N.; Jun, C. L.; Momtaz, A. Makromol. Chem. 1988, 189,

637-650.

(19) Schilling, F. C.; Tonelli, A. E. Macromolecules 1986, 19, 1337-1343.

(20) Lednor, P. W.; Rol, N. C. J. Chem. Soc., Chem. Commun. 1985, 598-599.

(21) Chisholm, M. H.; Navarro-Llobet, D.; Zhou, Z. Macromolecules 2002, 35, 6494-

6504.

133 (22) Qin, Z.; Thomas, C. M.; Lee, S.; Coates, G. W. Angew. Chem., Int. Ed. Engl.

2003, 42, 5484-5487.

(23) Aida, T.; Ishikawa, M.; Inoue, S. Macromolecules 1986, 19, 8-13.

(24) Eberhardt, R.; Allmendinger, M.; Rieger, B. Macromol. Rapid Commun. 2003,

24, 194-196.

(25) Darensbourg, D. J.; Mackiewicz, R. M.; Rodgers, J. L.; Phelps, A. L. Inorg.

Chem. 2004, 43, 1831-1833.

(26) Goeke, G. L.; Karol, F. J. U.S. Patent 1980, 4193892.

(27) Aida, T.; Inoue, S. J. Am. Chem. Soc. 1983, 105, 1304-1309.

(28) Aida, T.; Maekawa, Y.; Asano, S.; Inoue, S. Macromolecules 1988, 21, 1195-

1202.

(29) Asano, S.; Aida, T.; Inoue, S. Macromolecules 1985, 18, 2057-2061.

(30) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421-431.

(31) Takeda, N.; Inoue, S. Bull. Chem. Soc. Jpn. 1978, 51, 3564-3567.

(32) Chen, P. Angew. Chem., Int. Ed. Engl. 2003, 42, 2832-2847.

(33) Bochmann, M.; Lancaster, S. J.; Hannant, M. D.; Rodriguez, A.; Schormann, M.;

Walker, D. A.; Woodman, T. J. Pure Appl. Chem. 2003, 75, 1183-1195.

(34) Entelis, S. G.; Korovina, G. Makromol. Chem. 1974, 175, 1253-1280.

(35) Munoz-Hernandez, M.-A.; McKee, M. L.; Keizer, T. S.; Yearwood, B. C.;

Atwood, D. A. J. Chem. Soc., Dalton Trans. 2002, 410-414.

(36) Jegier, J. A.; Munoz-Hernandez, M.-A.; Atwood, D. A. J. Chem. Soc., Dalton

Trans. 1999, 2583-2588.

134 (37) Inoue, M.; Sugita, T.; Kiso, Y.; Ichikawa, K. Bull. Chem. Soc. Jpn. 1976, 49,

1063-1071.

(38) Shoen, E.; Zhang, X.; Zhou, Z.; Chisholm, M. H.; Chen, P., Manuscript in

preparation.

(39) Paddock, R. L.; Nguyen, S. T. J. Am. Chem. Soc. 2001, 123, 11498-11499.

(40) Tohma, H. Organometallic News 1995, 137.

(41) Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997, 277,

936-938.

(42) Nielsen, L. P. C.; Stevenson, C. P.; Blackmond, D. G.; Jacobsen, E. N. J. Am.

Chem. Soc. 2004, 126, 1360-1362.

(43) Lu, X.; Wang, Y. Angew. Chem. Int. Ed. 2004, 43, 3574-3577.

(44) Darensbourg, D. J.; Holtcamp, M. W.; Khandelwal, B.; Klausmeyer, K. K.;

Reibenspies, J. H. J. Am. Chem. Soc. 1995, 117, 538-539.

(45) Beckwith, J. D.; Tschinkl, M.; Picot, A.; Tsunoda, M.; Bachman, R.; Gabbaie, F.

P. Organometallics 2001, 20, 3169-3174.

(46) Harder, S.; Boersma, J.; Brandsma, L.; Kanters, J. A.; Duisenberg, A. J. M.; Van

Lenthe, J. H. Organometallics 1990, 9, 511-516.

(47) Hansen, K. B.; Leighton, J. L.; Jacobsen, E. N. J. Am. Chem. Soc. 1996, 118,

10924-10925.

(48) Coe, B. J.; Glenwright, S. J. Coord. Chem. Rev. 2000, 203, 5-80.

(49) Sahasrabudhe, S.; Yearwood, B. C.; Atwood, D. A.; Scott, M. J. Inorg. Synth.

2004, 34, 14-20.

(50) Yoo, Y.; McGrath, J. E. Pollimo 1994, 18, 892-894.

135 (51) Yoo, Y.; McGrath, J. E. Pollimo 1994, 18, 1063-1070.

(52) Nuytens, F.; Lopitaux, G.; Faven, C.; Coqueret, X. Macromol. Symp. 2000, 153,

17-25.

(53) Lopitaux, G.; Ricart, G.; Faven, C.; Coqueret, X. Macromol. Chem. Phys. 2002,

203, 2560-2569.

(54) Uno, H.; Takata, K.; Mizutani, Y. React. Polym. 1991, 15, 121-129.

(55) Kettle, S. F. A.; Editor Physical Inorganic Chemistry: Coordination Chemistry

Approach, 1996, p31-38.

(56) Aida, T.; Sugimoto, H.; Kuroki, M.; Inoue, S. J. Phys. Org. Chem. 1995, 8, 249-

257.

(57) Adler, A. D.; Longo, F. R.; Kampas, F.; Kim, J. J. Inorg. Nucl. Chem. 1970, 32,

2443-2445.

(58) Summerville, D. A.; Jones, R. D.; Hoffman, B. M.; Basolo, F. J. Am. Chem. Soc.

1977, 99, 8195-8202.

(59) Sakurai, T.; Yamamoto, K.; Naito, H.; Nakamoto, N. Bull. Chem. Soc. Jpn. 1976,

49, 3042-3046.

(60) Witowska-Jarosz, J.; Gorski, L.; Malinowska, E.; Jarosz, M. J. Mass Spectrom.

2002, 37, 1236-1241.

(61) Kadish, K. M.; Cornillon, J. L.; Coutsolelos, A.; Guilard, R. Inorg. Chem. 1987,

26, 4167-4173.

136

CHAPTER 4

CONCERNING THE RELATIVE IMPORTANCE OF ENANTIOMORPHIC SITE

VERSUS CHAIN END CONTROL IN THE STEREOSELECTIVE

POLYMERIZATION OF LACTIDES: REACTIONS OF (R,R-SALEN) AND (S,S-

SALEN) ALUMINUM ALKOXIDES LAlOCH2R COMPLEXES (R = CH3 AND S-

CHMeCl)

4.1 Introduction

The control of polymer microstructure is one of the most important goals in the development of single-site catalysis. One recent success story in this field has been in the stereoselective polymerization of propylene by metallocene based catalysis leading to the controlled formation of isotactic, syndiotactic, heterotactic and block polypropylene.1,2

There is considerable current interest in developing new classes of polymers derived from renewable resources, and the family of polylactides derived from the ring-opening polymerization of lactides, L-, rac- or meso-LA, represents a prime example.3,4

There are now several reports documenting the formation of both heterotactic

((SSRR)n = isi/sis) and isotactic stereoblocky ((RR)n(SS)n = (i)ns(i)n)) polymers formed by the ring-opening polymerization of rac-LA.5-17 However, although the mechanism of the

137 ring-opening of LA is generally recognized to proceed via the reaction pathway shown in

Scheme 4.1 involving the attack of an alkoxide group on the ketonic group of a coordinated LA molecule,18 the way in which stereoselectivity is achieved is not well understood. In the case of sterically demanding achiral β-diketonates employed with zinc and magnesium, the formation of heterotactic PLA must arise from chain end control.9,19

A similar situation would seem to pertain for bulky trispyrazolyl borate derivatives of calcium in the presence of donor solvents such as THF.13 However, the situation when the metal contains chiral chelating ligands is far from clear as the influence of the chiral chain end and the chiral ligand may be either constructive or destructive with respect to the overall stereoselectivity. We describe here a study of the ring opening of one equivalent of lactide (L and rac) by chiral–salen aluminum alkoxides in a variety of common solvents, where salen is (R,R) or (S,S)-N,N’-bis(3,5-di-tert-butyl-salicylidene)-

1,2-cyclohexenediamino, and the alkoxide initiator is OCH2CH3 or OCH2C(S) HMeCl.

These studies are particularly relevant to recent reports of the ring-opening polymerization of rac-LA by chiral and achiral salen and salan aluminum alkoxides.10,12,14,17

LM X L M O O X LM O O O O X O O O O O O

Scheme 4.1. Proposed reaction scheme for the ring-opening polymerization of lactides by a metal alkoxide pathway. 138

4.2 Results and Discussion

4.2.1 Synthesis and Structural Consideration

The chiral-salen aluminum alkoxides were prepared from the reactions shown in

Eq. 1 and 2.

CH2Cl2 (1) AlMe + (salen*)H (salen*)AlMe + 2CH4 3 2 r.t.

hexane (2) (salen*)AlMe + ROH (salen*)AlOR + CH r.t. 4

R = CH2CH3 or CH2CH(S)MeCl

The molecular structures of the (salen*)AlOCH2C(S)HMeCl are compared in

Figure 4.1. Here we use (R,R)-S and (R,R) to represent (R,R-salen)AlOCH2C(S)HMeCl and (R,R-salen)AlOCH2CH3, respectively. Similarly, (S,S)-S is for (S,S- salen)AlOCH2C(S)HMeCl and (S,S) is for (S,S-salen)AlOCH2CH3. In both structures, aluminum is five coordinated and the local AlN2O3 geometry can reasonably be described as square based pyramidal with the Al–OR group in the apical position. It is not unreasonable to believe that a similar five coordinate geometry is favored in hydrocarbon solvents. The mutual influence of the chiral salen and chiral alkoxide are seen in the conformations of the salen and OR groups as shown in Table 4.1. Little mechanistic information can been gleened from the ground state structures as the reactive intermediate leading to the C–O bond formation surely involves a six-coordinate Al(III) center with the Al–OR and Al–O(LA) groups in a cis position of a pseudo octahedral geometry with the salen forming either λ or δ chirality, as shown below.20

139

O O N LA N LA Al Al N OR NOR O O

I II

Figure 4.1. ORTEP views of (R,R-salen)AlOCH2C(S)HMeCl and (S,S- salen)AlOCH2C(S)HMeCl. 140 (R,R)-S (S,S)-S Al1─O1 1.797(3) Al2─O5 1.798(3) Al1─O2 1.804(3) Al2─O6 1.820(3) Al1─O3 1.746(3) Al2─O4 1.748(3) Al1─N1 2.021(3) Al2─N3 2.004(3) Al1─N2 1.999(3) Al2─N4 2.010(3) O3─C38 1.377(6) O4─C76 1.401(5) C38─C39 1.501(7) C76─C77 1.488(6) C39─C40 1.497(7) C77─C78 1.504(7) Cl1─C39 1.809(5) Cl2─C77 1.811(4)

O3─Al1─O1 109.7(1) O4─Al2─O5 111.0(1) O3─Al1─O2 107.2(1) O4─Al2─O6 106.5(1) O3─Al1─N1 96.5(1) O4─Al2─N3 97.0(1) O3─Al1─N2 104.3(1) O4─Al2─N4 103.2(1) O1─Al1─O2 89.8(1) O5─Al2─O6 89.8(1) O1─Al1─N1 88.8(1) O5─Al2─N3 88.8(1) O1─Al1─N2 144.9(2) O5─Al2─N4 144.9(1) O2─Al1─N2 88.2(1) O6─Al2─N4 88.2(1) O2─Al1─N1 155.2(1) O6─Al2─N3 155.3(1) N1─Al1─N2 78.9(1) N3─Al2─N4 79.0(1) C38─O3─Al1 127.3(3) C76─O4─Al2 124.4(3) O3─C38─C39 113.6(4) O4─C76─C77 113.0(4) C38─C39─C40 112.6(5) C76─C77─C78 112.0(4) C38─C39─Cl1 110.2(3) C76─C77─Cl2 111.0(3) C40─C39─Cl1 111.5(4) C78─C77─Cl2 111.4(3)

Table 4.1. Selected bond distances (Å) and angles (deg) for (R,R- salen)AlOCH2C(S)HMeCl and (S,S-salen)AlOCH2C(S)HMeCl.

141

Thus contributing to the stereoselective C–O bond forming transition state are (i) the chirality of the N―N backbone, (ii) the helicity of the η4-chelate, λ or δ, and (iii) the chirality of the alkoxide ligand, OR. A further complication arises from the solvent which may or may not hydrogen bond to the substrate (LA) and alkoxide oxygen atom or may coordinate to the [Al] center within the primary or secondary coordination sphere.

4.2.2 Stereoselectivity in 1:1 Reactions with rac-Lactide

In our study we have employed 1H NMR spectroscopy to evaluate the course of the reaction. Recognizing the significant difference in the rates of ring opening of lactide by primary and secondary alkoxide (salen)AlOR complexes, we carried out the reactions at 25oC to obtain the 1:1 adduct as shown in Scheme 4.2. The chain propagation can only proceed slowly at T ≥ 70 oC. As shown in Figure 4.2 the ring-opened moiety [Al]–

OCHMeC(O)OCHMeC(O)OR form a well defined group of resonances in CH(b) region

(as defined in Scheme 4.2) that can be reasonably assigned from the ring-opening of L-

LA.14 In the case of rac-LA, the ring opened L-LA to D-LA ratio may thus be determined. Based on the relative intensities of the methine protons we obtain a measure of the stereoselectivity in the ring-opening step based on the diastereomer excess, de%.

The results are collected in Table 4.2.

142

tBu tBu O

O tBu tBu O O N O N O O O b a Al OCH2R Al O OCH2R N O 1:20, r.t. N O O tBu tBu

tBu tBu

Scheme 4.2. Ring opening of 1 equiv of lactide by (salen)AlOCH2R.

1 Figure 4.2. Selected H (500MHz, CDCl3) NMR spectra showing CH region in the products in ring-opening reactions of 1 equiv of lactide by (salen)AlOCH2R. LA, unreacted lactide; LA*, LA satellite signals; L, L-LA ring opened product; D, D-LA ring opened product.

143 The data presented in Table 4.2 are based upon 1H NMR signal integration of methine proton signals of the –OCH(b)Me group. Each entry is an average of two independent reactions and a reasonable error bar of ±5% can be claimed. Thus for the entries with the OEt initiators we could expect the entries in the columns (R,R) and (S,S) to be equal in magnitude and opposite in sign. From the data presented it can be seen that

(R,R-salen)AlOEt shows a modest preference for reaction with L-LA while the (S,S- salen) complex prefers D-LA. The influence of the chiral donor solvent S-PO, R-PO has little effect but the influence of the chlorinated solvents CH2Cl2 and CHCl3 is more marked. Indeed, for CHCl3, a solvent capable of CHLO bonding, the stereoselectivity is inverted.

Solvents L – D (de%)a (R,R) (S,S) (R,R)–S (S,S)–S

C6H6 20 -18 -31 -40 Toluene 12 -17 -33 -37

CHCl3 -16 14 -30 -3

CH2Cl2 6 -6 -22 -4

THF 22 -20 -33 -13 Pyridine 20 -17 -39 -4

S-PO 17 -13 -32 -12 R-PO 14

rac-PO 17

Table 4.2. Stereoselectivity in 1:1 reactions of (salen)Al complexes and rac-lactide.

(a. diastereoner excess of L- and D-lactide ring-opened products)

144 Rather interestingly in reactions with the chiral initiator [Al]OCH2C(S)HMeCl, the (R,R-salen) ligand leads to the greatest stereoselectivity and now favors reaction with

D-LA. For (S,S-salen)AlOCH2C(S)HMeCl in the solvents benzene and toluene this preference is even more pronounced with de ~40% for D-LA, but in other solvents, most notably CH2Cl2 and CHCl3 this preference is considerably diminished.

4.2.3 Stereoselectivity in Ring-Opening Polymerization of rac-Lactide

The polymerization reactions of rac-LA were carried out in the presence of

o (salen)AlOCH2R initiators in toluene at 80 C for 10 days, yielding polylactide with ~40% conversion. The microstructure of obtained polymers was investigated by using 1H homodecoupled NMR and 13C proton decoupled NMR spectroscopies.21 As shown in

Figure 4.3, the PLAs were produced with dominantly isotactic junctions (>90%). No significant differences in i:s junction ratios were observed among the PLAs formed using these initiators, although (R,R-salen)Al polymerizes L-LA selectively and (S,S-salen)Al prefers D-LA as reported by Feijen.12,14 Also Feijen reported that (salen)AlOiPr produced

PLA in the ring-opening polymerization of rac-LA with a similar microstructure to the present work.

145

iii

iis sii (A) (R,R ) sis isi

(B) (R,R )- S

(C) (S,S )- S

5.35 5.30 5.25 5.20 5.15 5.10 5.05 ppm

1 Figure 4.3. H (500MHz, CDCl3) NMR spectra of the homodecoupled CH resonance of poly(rac-lactide) prepared in toluene using: (A) (R,R-salen)AlOCH2CH3, (B) (R,R- salen)AlOCH2C(S)HMeCl, and (C) (S,S-salen)AlOCH2C(S)HMeCl.

4.3 Concluding Remarks

From the results presented it is clear that the manner in which the chirality of the ligand bound to the metal, the chirality of the end group of the growing chain and the solvent all play a complex and rather unpredictable role in the preference for the ring opening of L- or D-LA in a racemic mixture. In the polymerization of rac-LA, the influence of the chiral end group would surely be greater than for OCH2C(S)HMeCl as the stereocenter will be closer to in the C–O bond forming step. It is then perhaps not surprising to find that stereoselective polymerizations of rac-LA have been observed with

146 up to 90% de. However, to ascribe this to chain-end control or enantiomorphic site control is quite problematic. As Gibson has recently found,15 we can expect subtle changes in the backbond of a chelating ligand which can adopt λ or δ stereoisomers in response to a chiral end group to greatly influence the outcome of stereoselectivity.

4.4 Experimental

All syntheses and solvent manipulations were carried out under a nitrogen atmosphere using standard Schlenk-line and drybox techniques. Solvents were dried in the standard procedures. Trimethylaluminum (2.0 M solution in hexane, Aldrich), 4- dimethylaminopyridine (Aldrich), (R,R)-N,N’-bis(3,5-di-tert-butyl-salicylidene)-1,2- cyclohexenediamine (Aldrich), (S,S)-N,N’-bis(3,5-di-tert-butyl-salicylidene)-1,2- cyclohexenediamine (Aldrich), and (S) 2-chloro-1-propanol (Aldrich) were used as received. Lactides (L and rac, Aldrich) were sublimed three times under reduced pressure prior to use. Deuterated solvents were stored over 4 Å molecular sieves for 24 h prior to use.

NMR Experiments. As described in Chapter 3.

Gel Permeation Chromatography. As described in Chapter 3.

X-Ray Crystallographic Studies. The data collection crystals for (R,R- salen)AlOCH2C(S)HMeCl and (S,S-salen)AlOCH2C(S)HMeCl were both yellow chunks, which had been cut from a large crystal. Examination of the diffraction pattern on a

Nonius Kappa CCD diffractometer indicated a monoclinic crystal system. All work was done at 200 K using an Oxford Cryosystems Cryostream Cooler. The data collection strategy was set up to measure a hemisphere of reciprocal space with a redundancy factor

147 of 3.5 for (R,R-salen)AlOCH2C(S)HMeCl and 4.2 for (S,S-salen)AlOCH2C(S)HMeCl, which means that 90% of the reflections were measured at least 3.5 and 4.2 times, respectively. A combination of phi and omega scans with a frame width of 1.0 degree was used. Data integration was done with Denzo, and scaling and merging of the data was done with Scalepack.22

The structure of (R,R-salen)AlOCH2C(S)HMeCl was solved in P21 by the direct methods in SHELXS-97.23 The asymmetric unit contains two molecules of the Al complex. Two of the tBu groups, one on each molecule, are disordered over two positions, and the corresponding carbon atoms were refined with isotropic displacement parameters. All non-hydrogen atoms were refined anistropically.

The structure of (S,S-salen)AlOCH2C(S)HMeCl was solved in P21 by the direct methods in SHELXS-97. The asymmetric unit contains two molecules of the Al complex.

Two of the tBu groups, one on each molecule, are disordered over two positions, and the corresponding carbon atoms were refined with isotropic displacement parameters. All non-hydrogen atoms were refined anistropically.

Experimental data relating to both structure determinations are displayed in Table

4.3.

Synthesis of (R,R-salen) aluminum ethoxide. To a solution of (R,R-salen)H2 (1.0 g, 1.8 mmol) in 20 mL of dichloromethane was added a solution of trimethylaluminum

(AlMe3 2.0 M solution in hexane, 1.1 mL, 2.2 mmol), and the mixture was stirred for 3 h at room temperature. A yellow powder was obtained after the removal of volatile fractions under vacuum. The powder was redissolved in hexane, followed by the addition of 0.2 mL of ethanol (3.4 mmol), stirred for 3 h. A yellow precipitate was obtained as the

148 1 product in 75% yield. H NMR (CDCl3, δ, ppm): 0.85 (t, CH3-CH2), 3.45 (m, CH3CH2),

1.29 (d, C(CH3)), 1.53 (d, C(CH3)), 1.44, 2.06, 2.42, 2.58, 3.05, 3.83 (cyclohexyl), 6.99,

13 1 7.04, 7.47, 7.50 (aromatic), 8.15, 8.35 (HC=N). C{ H} NMR (CDCl3, δ, ppm): 20.80

(CH3CH2), 23.76, 24.29, 27.28, 28.85 (CH2 in cyclohexyl), 29.71, 29.93, 31.39, 31.45

((CH3)3C), 33.95, 33.98, 35.58, 35.66 ((CH3)3C), 57.66(H2C-O), 62.63, 65.78 (HC-N),

118.15, 118.40, 127.29, 127.70, 129.66, 130.90, 137.53, 138.05, 140.66, 140.86, 162.42,

164.23 (phenyl), 162.86, 167.95 (HC=N). Elemental analysis: Calcd.: C, 73.99; H, 9.31;

N, 4.54. Found: C, 73.48; H, 9.33; N, 4.48%.

Synthesis of (S,S-salen) aluminum ethoxide. To a solution of (S,S-salen)H2 (1.0 g,

1.8 mmol) in 20 mL of dichloromethane was added a solution of trimethylaluminum

(AlMe3 2.0 M solution in hexane, 1.1 mL, 2.2 mmol), and the mixture was stirred for 3 h at room temperature. A yellow powder was obtained after the removal of volatile fractions under vacuum. The powder was redisolved in hexane, followed by the addition of 0.2 mL of ethanol (3.4 mmol), stirred for 3 h. A yellow precipitate was obtained as the

1 product in 73% yield. H NMR (CDCl3, δ, ppm): 0.85 (t, CH3-CH2), 3.44 (m, CH3CH2),

1.29 (d, C(CH3)), 1.53 (d, C(CH3)), 1.44, 2.06, 2.43, 2.58, 3.05, 3.83 (cyclohexyl), 6.99,

13 1 7.04, 7.48, 7.50 (aromatic), 8.15, 8.35 (HC=N). C{ H} NMR (CDCl3, δ, ppm): 20.80

(CH3CH2), 23.76, 24.29, 27.28, 28.85 (CH2 in cyclohexyl), 29.71, 29.93, 31.39, 31.45

((CH3)3C), 33.95, 33.98, 35.58, 35.66 ((CH3)3C), 57.66(H2C-O), 62.63, 65.78 (HC-N),

118.15, 118.40, 127.29, 127.70, 129.66, 130.90, 137.53, 138.05, 140.65, 140.86, 162.42,

164.23 (phenyl), 162.86, 167.95 (HC=N). Elemental analysis: Calcd.: C, 73.99; H, 9.31;

N, 4.54. Found: C, 73.25; H, 9.24; N, 4.42%.

Synthesis of (R,R-salen)AlOCH2C(S)HMeCl. As described in Chapter 3. 149 Synthesis of (S,S-salen)AlOCH2C(S)HMeCl. To a solution of (S,S-salen)H2 (1.0 g,

1.8 mmol) in 20 mL of dichloromethane was added a solution of trimethylaluminum

(AlMe3 2.0 M solution in hexane, 2.2 mL, 2.2 mmol), and the mixture was stirred for 3 h at room temperature. A yellow powder was obtained after the removal of volatile fractions under vacuum. The powder was redisolved in hexane, followed by the addition of 0.19 mL of (S) 2-chloro-1-propanol (2.2 mmol), stirred for 3 h. A yellow precipitate was obtained as the product in 70% yield. Crystals of this complex were obtained by slowly evaporation of the solvent from a concentrated benzene solution of the complex.

1 H NMR (CDCl3, δ, ppm): 1.22 (d, CH3-CHCl), 3.46 (m, OCH2), 3.69 (m, CHCl), 1.29

(d, C(CH3)), 1.52 (d, C(CH3)), 1.44, 2.06, 2.43, 2.58, 3.06, 3.93 (cyclohexyl), 7.00, 7.05,

13 1 7.48, 7.51 (aromatic), 8.17, 8.36 (HC=N). C{ H} NMR (CDCl3, δ, ppm): 21.40

(CH3CH), 23.78, 24.30, 27.28, 28.74 (CH2 in cyclohexyl), 29.67, 29.97, 31.38, 31.43

((CH3)3C), 33.97, 34.00, 35.57, 35.65 ((CH3)3C), 61.52 (HC-Cl), 62.55, 65.87(HC-N),

69.56(H2C-O), 118.18, 118.33, 127.41, 127.79, 129.84, 131.05, 137.75, 138.33, 140.54,

140.83, 162.19, 164.00 (phenyl), 163.00, 168.28 (HC=N). Elemental analysis: Calcd.: C,

70.41; H, 8.79; N, 4.21. Found: C, 69.64; H, 8.73; N, 4.02%.

Ring-opening reactions of 1 equiv of lactides. 20 mg of lactide (L or rac, 0.14 mmol) and 5 mg of (salen)AlOCH2R (~0.008 mmol) were dissolved in 1 mL of solvent in a J-Young NMR tube and allowed to react at room temperature for 12 h in the cases of

(salen)AlOCH2CH3, or 3 d in the cases of (salen)AlOCH2CHMeCl. The solvent was removed under vacuum, and the resulting mixture was redissolved in 1 mL of CDCl3 for

1HNMR measurement. The products in the reactions of L-lactide and all those four

150 1 (salen)AlOCH2R complexes were analyzed by H NMR as references for identifying products in the reactions involving rac-lactide.

Ring-opening polymerization reactions of rac-lactide. Typically

(Salen)AlOCH2R (20 mg, 0.03 mmol) and rac-lactide (0.45 g, 3.1 mmol) were allowed to react in 15 mL of toluene at 80oC for 10 d, yielding polylactide with 40% conversion.

The obtained polymers were analyzed by NMR and GPC. GPC: Mn = 9 000 – 12 000,

PDI = 1.13 –1.16.

(R,R)-S (S,S)-S

Empirical formula C39H58AlClN2O3 C39H58AlClN2O3 Formular weight 665.30 665.30 Temperature (K) 200 200 Wavelenghth (Å) 0.71073 0.71073 Crystal system monoclinic monoclinic

Space group P21 P21 Unit cell dimensions a (Å) 14.923(1) 14.927(1) b (Å) 11.021(1) 10.941(1) c (Å) 23.440(1) 23.393(1) ß (o) 93.397(1) 93.69(1) V (Å3) 3848.6(7) 3812.5(7) Z 4 4 -3 Dcalc (mg m ) 1.148 1.159 Absorption 0.159 0.160 coefficient (mm-1) F(000) 1440 1440 Crystal size (mm-3) 0.38x0.19x0.15 0.38x0.31x0.27

151 Theta range for data 1.58-22.99 1.57-25.03 collection (o) Index ranges -16≤h≤16, -12≤k≤12, -17≤h≤17, -13≤k≤12, -25≤l≤25 -27≤l≤27 Reflections collected 60183 60079

Independent 10692[Rint=0.073] 13387[Rint=0.058] reflections Refinement method Full-matrix least-squares Full-matrix least-squares on F2 on F2 Data/restraints/param 10692/1/886 13387/1/826 eters Flack parameter 0.00(9) 0.00(8) Goodness-of-fit on F2 1.021 1.020

Final R indices R1=0.0508, wR2=0.1237 R1=0.0555, wR2=0.1321 [I>2σ(I)]

R indices (all data) R1=0.0749, wR2=0.1364 R1=0.0852, wR2=0.1467 Largest difference 0.412 and -0.506 0.565 and -0.371 peak and hole (e/Å3)

Table 4.3. Crystallographic details for (R,R-salen)AlOCH2C(S)HMeCl and (S,S- salen)AlOCH2C(S)HMeCl.

4.5 References

(1) Coates, G. W. Chem. Rev. 2000, 100, 1223-1252.

(2) Coates, G. W.; Hustad, P. D.; Reinartz, S. Angew. Chem., Int. Ed. Engl.

2002, 41, 2236-2257.

(3) Gruber, P.; O'Brien, M. Biopolymers 2002, 4, 235-250.

152 (4) Kricheldorf, H. R. Chemosphere 2001, 43, 49-54.

(5) Spassky, N.; Wisniewski, M.; Pluta, C.; Le Borgne, A. Macromol. Chem.

Phys. 1996, 197, 2627-2637.

(6) Wisniewski, M.; Le Borgne, A.; Spassky, N. Macromol. Chem. Phys.

1997, 198, 1227-1238.

(7) Spassky, N.; Pluta, C.; Simic, V.; Thiam, M.; Wisniewski, M. Macromol.

Symp. 1998, 128, 39-51.

(8) Ovitt, T. M.; Coates, G. W. J. Polym. Sci., Part A: Polym. Chem. 2000,

38, 4686-4692.

(9) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.; Lobkovsky,

E. B.; Coates, G. W. J. Am. Chem. Soc. 2001, 123, 3229-3238.

(10) Nomura, N.; Ishii, R.; Akakura, M.; Aoi, K. J. Am. Chem. Soc. 2002, 124,

5938-5939.

(11) Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc. 2002, 124, 1316-1326.

(12) Zhong, Z.; Dijkstra, P. J.; Feijen, J. Angew. Chem., Int. Ed. Engl. 2002, 41,

4510-4513.

(13) Chisholm, M. H.; Gallucci, J.; Phomphrai, K. Chem. Commun. 2003, 48-

49.

(14) Zhong, Z.; Dijkstra, P. J.; Feijen, J. J. Am. Chem. Soc. 2003, 125, 11291-

11298.

(15) Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; White, A. J. P.; Williams,

D. J. J. Am. Chem. Soc. 2004, 126, 2688-2689.

153 (16) Radano, C. P.; Baker, G. L.; Smith, M. R., III J. Am. Chem. Soc. 2000,

122, 1552-1553.

(17) Tang, Z.; Chen, X.; Pang, X.; Yang, Y.; Zhang, X.; Jing, X.

Biomacromolecules 2004, 5, 965-970.

(18) Chisholm, M. H.; Delbridge, E. E. New J. Chem. 2003, 27, 1167-1176.

(19) Chisholm, M. H.; Phomphrai, K. Inorg. Chim. Acta 2003, 350, 121-125.

(20) Kettle, S. F. A.; Editor Physical Inorganic Chemistry: Coordination

Chemistry Approach, 1996, p432-434.

(21) Zell, M. T.; Padden, B. E.; Paterick, A. J.; Thakur, K. A. M.; Kean, R. T.;

Hillmyer, M. A.; Munson, E. J. Macromolecules 2002, 35, 7700-7707.

(22) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307-326.

(23) Sheldrick, G. M.; Schneider, T. R. Methods Enzymol. 1997, 277, 319-343.

154

APPENDIX A

SUPPLEMENTAL MATERIALS FOR CHAPTER 2

Deconvolution data. Deconvolution data for the 13C{1H} NMR from the polymer obtained between rac-PO and CO2 with the zinc glutarate. The data for carbonate, methine, methylene and methyl carbons are shown.

Carbonate carbons Region δ (ppm) Area % Area TT 154.66 26.7 18.2 154.64 22.2 HT 154.27 7.7 63.4 154.25 25.1 154.23 21.8 154.21 28.9 154.19 23.8 154.17 29.2 154.14 34.5 HH 153.83 6.4 18.4 153.80 6.6 153.78 7.6 153.70 8.2 153.67 9.6 153.65 11.0

Methine carbons δ (ppm) Area 72.34 81.4 72.33 168.2 72.32 149.0 155 72.08 97.6 72.03 71.0

Methylene carbons δ (ppm) Area 69.24 39.9 69.14 101.4 69.08 24.1 69.05 30.4 68.96 163.8

Methyl carbons δ (ppm) Area 16.23 28.5 16.18 120.9 16.14 142.9

Deconvolution data for the 13C{1H} NMR from the polymer obtained between S-

PO and CO2 with the zinc glutarate. The data for carbonate, methine, methylene and methyl carbons are shown.

Carbonate carbons Region δ (ppm) Area % Area TT 154.64 17.6 19.3 154.63 8.3 HT 154.16 21.8 61.0 154.14 60.4 HH 153.67 6.8 19.7 153.64 12.7 153.63 7.0

Methine carbons δ (ppm) Area 156 72.34 138.3 72.31 51.2 72.11 50.6 72.08 37.0 72.07 31.8

Methylene carbons δ (ppm) Area 69.14 94.0 68.97 160.7

Methyl carbons δ (ppm) Area 16.23 30.3 16.19 112.1 16.15 38.6

Deconvolution data for the 13C{1H} NMR from the polymer obtained between

50% rac-PO + 50% S-PO and CO2 with the zinc glutarate. The data for carbonate, methine, methylene and methyl carbons are shown.

Carbonate carbons Region δ (ppm) Area % Area TT 154.65 13.0 15.9 154.63 13.5 HT 154.26 8.8 61.0 154.24 11.1 154.22 10.0 154.20 11.8 154.18 11.0 154.16 16.7 154.14 32.5 HH 153.82 3.8 23.2 153.80 4.9 157 153.77 5.2 153.69 7.5 153.66 6.3 153.64 7.1 153.62 4.0

Methine carbons δ (ppm) Area 72.33 93.5 72.31 81.3 72.10 25.9 72.07 47.4 72.06 41.8 72.03 43.3 72.00 16.4

Methylene carbons δ (ppm) Area 69.23 21.9 69.13 61.9 69.07 10.9 69.05 17.1 68.96 102.3 68.93 26.9

Methyl carbons δ (ppm) Area 16.22 18.7 16.18 70.0 16.14 72.8

158

Statistical calculations

Statistical calculations for the probability of PO ring-opening site in the formation of PPC based on the deconvolution data of carbonate carbons.

Assume the probability of attacking the methylene carbon is x, which produces a HT PO unit,

Zn O Zn O OR Zn OR and the probability of attacking the methine carbon is y, which produces a TH PO unit with inversion of configuration at the chiral carbon.

Zn O Zn O OR Zn OR

Looking at the carbonate carbons at the diad level, HT junction can be formed in

(HT).(HT) with a probability of x2, and (TH).(TH) with a probability of y2; HH junction can only be formed in (TH).(HT) with a probability of yx; TT can only be formed in

(HT).(TH) with a probability of xy. Thus, for PPC made from rac-PO, based on the deconvolution data, we can get the equations

x2 + y2 = 63.4%

yx + xy = 18.4% + 18.2%

x + y = 1

By solving these, we get x=75.9%, and y=24.1%.

159 Similarly, we can calculate that, for PPC made from S-PO, x=73.5%, y=26.5%; and for PPC made from 50% rac-PO + 50% S-PO, x=73.5%, y= 26.5%.

In summary, x : y is about 0.75 : 0.25, i.e. 3 : 1.

Statistical calculations for regiosequences of carbonate carbons in PPC at the tetrad level derived from H-T diad junction, based on the selectivity of PO ring-opening site of 3 : 1.

Diad level Tetrad level

(HT)(HT).(HT)(HT) 1 0.75 x 0.75 x 0.75 x 0.75 = 0.316

(HT)(HT).(HT)(TH) 2 0.75 x 0.75 x 0.75 x 0.25 = 0.105 (HT).(HT) (TH)(HT).(HT)(HT) 3 0.25 x 0.75 x 0.75 x 0.75 = 0.105

(TH)(HT).(HT)(TH) 4 0.25 x 0.75 x 0.75 x 0.25 = 0.035

(HT)(TH).(TH)(HT) 5=4 0.75 x 0.25 x 0.25 x 0.75 = 0.035

(HT)(TH).(TH)(TH) 6=2 0.75 x 0.25 x 0.25 x 0.25 = 0.012 (TH).(TH) (TH)(TH).(TH)(HT) 7=3 0.25 x 0.25 x 0.25 x 0.75 = 0.012

(TH)(TH).(TH)(TH) 8=1 0.25 x 0.25 x 0.25 x 0.25 = 0.004

In summary, (1, 8) = 0.320, (2, 6) = 0.117, (3, 7) = 0.117, and (4, 5) = 0.070.

Statistical calculations for the stereosequences for the (HT)(HT).(HT)(HT) regiosequence of cabonate carbons in PPC made from 50% rac-PO + 50% S-PO.

160 O O OOR OOR iii 0.25 x 0.25 x 0.25 x 0.25 = 0.04 O R O O R OO O O O S O S OO O O iii 0.75 x 0.75 x 0.75 x 0.75 = 0.316 O S O O S OO O O O O OOR OOS iis 0.25 x 0.25 x 0.25 x 0.75 = 0.012 O R O O R OO O O O O S R OO OO iis 0.75 x 0.75 x 0.75 x 0.25 = 0.105 O S O O S OO O O O O R OOR OO sii 0.75 x 0.25 x 0.25 x 0.25 =0.012 O S O O R O O O O O S O S O O OO sii 0.25 x 0.75 x 0.75 x 0.75 = 0.105 O R O O S OO O O O O OOR OOS sis 0.75 x 0.25 x 0.25 x 0.75 = 0.035 O S O O R OO O O O O OOS OOR sis O R O O S OO 0.25 x 0.75 x 0.75 x 0.25 = 0.035 O O

O O R S OO O O isi 0.25 x 0.25 x 0.75 x 0.75 = 0.035 O R O O S O O O O O O S R OO O O isi 0.75 x 0.75 x 0.25 x 0.25 = 0.035 O S O O R OO O O O O OOR O O R iss 0.25 x 0.25 x 0.75 x 0.25 = 0.012 O R O O S O O O O O O S S O O O O iss 0.75 x 0.75 x 0.25 x 0.75 = 0.105 R O O S O O O O O O O S OOR OO ssi 0.75 x 0.25 x 0.75 x 0.75 = 0.105 O S O O S O O O O O O S R O O OO ssi 0.25 x 0.75 x 0.25 x 0.25 = 0.012 O R O O R O O O O O O R O O R O O sss 0.75 x 0.25 x 0.75 x 0.25 = 0.035 O S O O S OO O O

O S O OOS O O sss 0.25 x 0.75 x 0.25 x 0.75 = 0.035 O R O O R O O O O

Totally, iii = 0.320, iis = 0.117, sii = 0.117, sis = 0.070, isi = 0.070, iss =0.117, ssi

= 0.117 and sss = 0.070.

161

Statistical calculations for the regiosequences of methine carbons in PPC at the triad level, based on the selectivity of PO ring-opening site of 3 : 1.

O O O 51'3 1 O 3' O 5' 4 O 2 O * 2' 4' 6' O O O

(HT)(H.T)(HT) (5, 5') 1 0.75 x 0.75 x 0.75 = 0.422

(HT)(H.T)(TH) (5, 6') 2 0.75 x 0.75 x 0.25 = 0.141 (H.T) (TH)(H.T)(HT) (4, 5') 3 0.25 x 0.75 x 0.75 = 0.141

(TH)(H.T)(TH) (4, 6') 4 0.25 x 0.75 x 0.25 = 0.047

O O 2 O 6' 4' 2' * O O 4 5' O 3' O 1' 1 3 5 O O O (HT)(T.H)(HT) (6', 4) 5=4 0.75 x 0.25 x 0.75 = 0.141

(HT)(T.H)(TH) (6', 5) 6=2 0.75 x 0.25 x 0.25 = 0.047 (T.H) (TH)(T.H)(HT) (5', 4) 7=3 0.25 x 0.25 x 0.75 = 0.047

(TH)(T.H)(TH) (5', 5) 8=1 0.25 x 0.25 x 0.25 = 0.016

In summary, (1, 8) = 0.438, (2, 6) = 0.188, (3, 7) = 0.188 and (4, 5) = 0.188.

162 Statistical calculations for the regiosequences of methylene carbons in PPC at the triad level, based on the selectivity of PO ring-opening site of 3 : 1.

O O 2 O 6' 4' 2' * O O 4 5' O 3' O 1' 1 3 5 O O O

(HT)(H.T)(HT) (6', 1', 4) 1 0.75 x 0.75 x 0.75 = 0.422

(TH)(H.T)(HT) (5', 1', 4) 2 0.25 x 0.75 x 0.75 = 0.141 (H.T) (HT)(H.T)(TH) (6', 1', 5) 3 0.75 x 0.75 x 0.25 = 0.141

(TH)(H.T)(TH) (5', 1', 5) 4 0.25 x 0.75 x 0.25 = 0.047

O O O 5 3 1 1' O 3' O 5' 4 O 2 O * 2' 4' 6' O O O

(HT)(T.H)(HT) (5, 1', 5') 5 = 4 0.75 x 0.25 x 0.75 = 0.141

(HT)(T.H)(TH) (5, 1', 6') 6 = 3 0.75 x 0.25 x 0.25 = 0.047 (T.H) (TH)(T.H)(HT) (4, 1', 5') 7 = 2 0.25 x 0.25 x 0.75 = 0.047

(TH)(T.H)(TH) (4, 1', 6') 8 = 1 0.25 x 0.25 x 0.25 = 0.016

In summary, (1, 8) = 0.438, (2, 7) = 0.188, (3, 6) = 0.188 and (4, 5) = 0.188.

163 Statistical calculations for the regiosequences of methyl carbons in PPC at the triad level, based on the selectivity of PO ring-opening site of 3 : 1.

O * O O 6 4 2 2' O 4' O 6' 5 O 3 O 1 3' 5' 7' O O O (HT)(H.T)(HT) (6, 6') 1 0.75 x 0.75 x 0.75 = 0.422

(HT)(H.T)(TH) (6, 7') 2 0.75 x 0.75 x 0.25 = 0.141 (H.T) (TH)(H.T)(HT) (5, 6') 3 0.25 x 0.75 x 0.75 = 0.141

(TH)(H.T)(TH) (5, 7') 4 0.25 x 0.75 x 0.25 = 0.047

O O O 7' 5' 3' 1 O 3 O 5 6' O 4' O 2' 2 4 6 OO * O (HT)(T.H)(HT) (7', 5) 5 = 4 0.75 x 0.25 x 0.75 = 0.141

(HT)(T.H)(TH) (7', 6) 6 = 2 0.75 x 0.25 x 0.25 = 0.047 (T.H) (TH)(T.H)(HT) (6', 5) 7 = 3 0.25 x 0.25 x 0.75 = 0.047

(TH)(T.H)(TH) (6', 6) 8 = 1 0.25 x 0.25 x 0.25 = 0.016

In summary, (1, 8) = 0.438, (2, 6) = 0.188, (3, 7) = 0.188 and (4, 5) = 0.188.

164 Regio- and stereoselectivity in propylene oxide ring-opening polymerization with

KOEt.

Regio- and stereoselectivity in PO ring-opening polymerization by base catalysis, such as KOH, has been reported in the literature.14 PO ring-opening occurs >95% at methylene carbon to form exclusively head-tail (HT) junctions in the PPO chain with no significant preference on isotactic (i) or syndiotactic (s) junctions (~50% : 50%). PO ring- opening polymerization by KOEt (50 : 1) showed the similar selectivity as in Figure

A1.1. This result is used in the statistical calculations below. For details about the regio- and stereoselectivity study in the PPO formation by 13C{1H} NMR, see the following references: (S1) Schilling, F. C.; Tonelli, A. E. Macromolecules 1986, 19, 1337-1343;

(S2) Tonelli, A. E. Annu. Rep. NMR Spectrosc. 1997, 34, 185-229; (S3) Antelmann, B.;

Chisholm, M. H.; Iyer, S. S.; Huffman, J. C.; Navarro-Llobet, D.; Pagel, M.; Simonsick,

W. J.; Zhong, W. Macromolecules 2001, 34, 3159-3175.

CH carbons ii is/si ss CH2 carbons

iis/sii

iii sis iss/ssi sss isi

75.8 75.6 75.4 75.2 75.0 74.8 74.6 74.4 74.2 74.0 73.8 73.6 73.4 73.2 73.0 72.8 72.6 72.4 ppm 13 1 Figure A.1. C{ H} NMR (150 MHz, CDCl3) spectrum of PPO obtained from PO polymerization with KOEt.

165 Statistical calculations for stereosequences of carbonate carbons in regioregular

Et(PO)2OCO2(PO)2Et made from etherol derived from rac-PO.

O R O R O C (RR)(RR) R O O R OEt EtO iii 0.5 x 0.5 x 0.5 x 0.5 = 0.0625 O S O C O S (SS)(SS) OEt EtO S O O S iii 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O R O C O S OEt (RR)(RS) EtO R O O R iis 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O S O R O C (SS)(SR) S O O S OEt EtO iis 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O O R S O C (SR)(RR) R R OEt EtO O O sii 0.5 x 0.5 x 0.5 x 0.5 = 0.0625 O R O C O S (RS)(SS) S OEt EtO S O O sii 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O O S S O C (SR)(RS) R OEt EtO R O O sis 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O R C O R O (RS)(SR) EtO S O O S OEt sis 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

166 O R C O S O (RR)(SS) EtO R O O S OEt isi 0.5 x 0.5 x 0.5 x 0.5 = 0.0625 O S C O R O (SS)(RR) EtO S O O R OEt isi 0.5 x 0.5 x 0.5 x 0.5 = 0.0625 O R O C O R (RR)(SR) OEt EtO R O O S iss 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O S O C O S OEt (SS)(RS) EtO S O O R iss 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O S O C O S (SR)(SS) EtO R O O S OEt ssi 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O R O C O R OEt (RS)(RR) EtO S O O R ssi 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O O R S O C (SR)(SR) S OEt EtO R O O sss 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

O R O S O C (RS)(RS) S O O R OEt EtO sss 0.5 x 0.5 x 0.5 x 0.5 = 0.0625

In summary, the probability ratio of stereosequences is (iii) : (iis/sii) : (sis): (isi) :

(iss/ssi) : (sss) = 1 : 2 : 1 : 1 : 2 : 1.

167 Statistical calculations for stereosequences of carbonate carbons in regioregular

Et(PO)2OCO2(PO)2Et made from etherol derived from etherol mixtures (1 : 1) derived separately from rac-PO and S-PO.

Stereosequence probability of

O EtO OH

(SS) = 0.5 + 0.5 x 0.5 x 0.5 = 0.625

(SR) = 0.5 x 0.5 x 0.5 = 0.125

(RS) = 0.5 x 0.5 x 0.5 = 0.125

(RR) = 0.5 x 0.5 x 0.5 = 0.125

Stereosequence probability of

O O C O EtO O O OEt

iii: (RR)(RR) = 0.125 x 0.125, (SS)(SS) = 0.625 x 0.625 iis/sii: (RR)(RS) = 0.125 x 0.125, (SS)(SR) = 0.625 x 0.125,

(SR)(RR) = 0.125 x 0.125, (RS)(SS) = 0.125 x 0.625 sis: (SR)(RS) = 0.125 x 0.125, (RS)(SR) = 0.125 x 0.125 isi: (RR)(SS) = 0.125 x 0.625, (SS)(RR) = 0.625 x 0.125 iss/ssi: (RR)(SR) = 0.125 x 0.125, (SS)(RS) = 0.625 x 0.125,

(SR)(SS) = 0.125 x 0.625, (RS)(RR) = 0.125 x 0.125 sss: (SR)(SR) = 0.125 x 0.125, (RS)(RS) = 0.125 x 0.125

In summary, the probability ratio of stereosequences is (iii) : (iis/sii) : (sis): (isi) :

(iss/ssi) : (sss) = 13 : 6 : 1 : 5 : 6 : 1.

168 O C C H O O C H 2 5 O O 2 5 3 3

a. i s

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

b.

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

O C C H O O C H 2 5 O O 2 5 4 4

c.

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

13 1 Figure A.2. C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regioregular oligoether carbonate compounds (n = 3, 4). Carbonate compounds in (a)

169 were made from etherol derived from rac-PO, while the ones in (b) were prepared from etherol mixture (1:1) derived separately from rac-PO and S-PO.

Statistical calculations for stereosequences of carbonate carbons in regioregular

Et(PO)3OCO2(PO)3Et made from etherol derived from rac-PO, as well as from etherol mixtures (1 : 1) derived separately from rac-PO and S-PO.

The results are summarized in the table below, using the same method as described above for the statistical calculations.

Central diad Overall Probabilty × 128 stereosequences Stereosequences rac-etherols rac/S-etherols iiiii 4 41 iiiis/siiii 8 10 siiis 4 1 iiisi/isiii 8 10 iiiss/ssiii 8 10 i siisi/isiis 8 2 siiss/ssiis 8 2 isisi 4 2 isiss/ssisi 8 4 ssiss 4 2 iisii 4 9 iisis/sisii 8 10 sisis 4 1 iissi/issii 8 10 iisss/sssii 8 10 s sissi/issis 8 2

170 sisss/sssis 8 2 isssi 4 1 issss/ssssi 8 2 sssss 4 1

Statistical calculations for regio- and stereosequences of carbonate carbons in regioirregular Me(PO)2OCO2(PO)2Me made from regioirregular di-(propylene glycol) methyl ethers.

Di-(propylene glycol) methyl ethers were purchased from Aldrich and analyzed by 1H NMR, 13C{1H} NMR, DEPT, COSY and HMQC techniques. The results showed it contained both MeOCH2CHMeOCH2CHMeOH ((TH)(TH), 37%) and

MeOCHMeCH2OCH2CHMeOH ((HT)(TH), 63%) with no significant preference on stereochemistry.

1. Probability of regiosequences

171 O O C O MeO O O OMe

(TH)(TH)(HT)(HT) 37% x 37% = 13.7 %

O O C O MeO O O OMe

(TH)(TH)(HT)(TH) 37% x 63% = 23.3 % and (HT)(TH)(HT)(HT) 63% x 37% = 23.3 %

O O C O MeO O O OMe

(HT)(TH)(HT)(TH) 63% x 63% = 39.7 %

2. Probability of stereosequences

(1) (TH)(TH)(HT)(HT) (13.7%)

(iii) : (iis/sii) : (sis): (isi) : (iss/ssi) : (sss) = 1 : 2 : 1 : 1 : 2 : 1.

(2) (TH)(TH)(HT)(TH)/ (HT)(TH)(HT)(HT) (46.6%)

(iii) : (iis) : (sii) : (sis): (isi) : (iss) : (ssi) : (sss) = 1 : 1 : 1 : 1 : 1 : 1 : 1 : 1.

(3) (HT)(TH)(HT)(TH) (39.7%)

(iii) : (iis/sii) : (sis): (isi) : (iss/ssi) : (sss) = 1 : 2 : 1 : 1 : 2 : 1.

In summary, 20 carbonate signals from this sample are possibly distinguishable

13 1 by C{ H} NMR (in CDCl3), and the ratio is 13.7 : 27.4: 13.7: 13.7 : 27.4 : 13.7: 46.6:

46.6: 46.6 : 46.6 : 46.6 : 46.6 : 46.6 : 46.6 : 39.7 : 79.4 : 39.7 : 39.7 : 79.4 : 39.7.

172

O C O O C H C2H5 O O 2 5 10 10

a. s i

154.30 154.25 154.20 154.15 154.10 154.05 154.00ppm

O C C H O O C H 2 5 O O 2 5 10 10

b. i

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

O O C O O O 30 30

c.

154.30 154.25 154.20 154.15 154.10 154.05 154.00 ppm

13 1 Figure A.3. C{ H} (150 MHz, CDCl3) NMR spectra of carbonate carbon region of regio-regular oligoether carbonate compounds (n = ~10, ~30).

173

APPENDIX B

SUPPLEMENTAL MATERIALS FOR CHAPTER 3

Figure B.1. Plot of Ln(kobs]) vs Ln([Al]) for ring opening of PO by varying concentrations of (TPP)AlO(CH2)9CH3 (5 to 15 mM) .

174

c (TPP)AlOOCR/DMAP, 25 OC DMAP (free) N b

a N

c c

a b b a 9 8 7 6 5 4 3 2 1 0 -1 -2 9 8 7 6 5 4 3 2 1 0 -1 -2

(TPP)AlOOCR/DMAP, -50 OC (TPP)AlOR/DMAP, 25O C

c

c b a a b

9 8 7 6 5 4 3 2 1 0 -1 -2 9 8 7 6 5 4 3 2 1 0 -1 -2

1 Figure B.2. H NMR (500MHz, CDCl3) spectra of (TPP)AlX/DMAP, X = O(CH2)9CH3

o o and O2C(CH2)6CH3, at 25 C and –50 C. a, b, and c are the signals from DMAP.

175

3 2.5 2 1.5

-Ln(It/Io) 1 0.5 0 0 200 400 600 Reaction Time (min) (TPP)AlOOCR/DMAP (TPP)AlOR/DMAP

Figure B.3. Plots of -Ln(It/I0) versus reaction time for the ring opening reaction of the first PO molecule by initiators, in the presence of 1 equiv of DMAP. I0 is the initial initiator concentration, It is the initiator concentration at time t. (TPP)AlOR =

(TPP)AlO(CH2)9CH3, (TPP)AlOOCR = (TPP)AlO2C(CH2)6CH3. In all the reactions, [Al]

= 15 mM, [PO] = 720 mM.

176 Regioselctivity and stereoselectivity in the ring opening of propylene oxides by (R,R- salen)AlCl

* Complex 1 was synthesized independently from the reaction of (R,R-salen)AlMe and

HOCH2C(S)HMeCl. Compounds 1 and 4 were characterized by single crystal X-ray determinations (Figure S6).

1a,1a’ 1b Complex 1* (R,R-salen)AlOCH2C(S)HMeCl

2b 2a,2a’ Complex 2 (R,R-salen)AlOC(R)HMeCH2Cl

3a,3a’ 3b Complex 3 (R,R-salen)AlOCH2C(R)HMeCl

4b 4a,4a’ Complex 4* (R,R-salen)AlOC(S)HMeCH2Cl

(A) Complex 1

1b 1a’ 1a

4.20 4.00 3.80 3.60 3.40 3.20 3.00 2.80

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 ppm

(B) (R,R -salen)AlCl/ R -PO

2a’ 2a 2b

1b 3b 3a’ 3a 1a’ 1a

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 ppm

177 (C) (R,R -salen)AlCl/ R -PO, add Complex 1 1b 3b 1a’ 1a 2a’ 2a 2b

3a’ 3a

4.20 4.00 3.80 3.60 3.40 3.20 3.00 2.80

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 ppm

(D) (R,R -salen)AlCl/ S -PO

4a 4b 4a’

3b

3a 3a’

4.20 4.00 3.80 3.60 3.40 3.20 3.00 2.80

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 ppm

(E) (R,R -salen)AlCl/ S -PO, add Complex 1

4a

4b 4a’ 1b 3b

3a’ 3a 1a’ 1a

4.20 4.00 3.80 3.60 3.40 3.20 3.00 2.80

pp m 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

1 Figure B.4. H (600 MHz, CDCl3) NMR spectra of (A) Complex 1; (B) product of the reaction between (R,R-salen)AlCl and R-PO; (C) after addition of Complex 1 into sample

(B); (D) product of the reaction between (R,R-salen)AlCl and S-PO; (E) after addition of

Complex 1 into sample (D).

178

Figure B.5. ORTEP view of (R,R-salen)AlOCHMe(S)CH2Cl grown in S-PO.

179

Figure B.6. ORTEP view of (R,R-salen)AlOOCMe grown in pyridine.

180

BIBLIOGRAPHY

(1) Bruntland, G.; Editor Our Common Future: The World Commission on

Environment and Development; Oxford University Press, Oxford, 1987.

(2) Charpentier, R. R. Science 2002, 295, 1470.

(3) Young, S. Nature 2001, 414, 487-488.

(4) Clark, J. H.; Editor Chemistry of Waste Minimization; Blackie Academic,

Glasgow, UK, 1995.

(5) MORI. The Public Image of the Chemical Industry; Research study

conducted for the Chemical Industries Association. MORI, London, 1999.

(6) Anastas, P.; Warner, J. Green Chemistry: Theory and Practice; Oxford

University Press, Oxford, UK, 1998.

(7) Goodstein, D. Out of Gas: The End of the Age of Oil; W. W. Norton &

Company, New York, 2004.

(8) Tullo, A. Chem. Eng. News 2003, 81, 26-27.

(9) Wilks, E. S. Industial Polymers Handbook; WILEY-VCH, Germany,

2001.

(10) Anstas, P. T.; Bickart, P. H.; Kirchhoff, M. M. Designing Safer Polymers;

WILEY-INTERSCIENCE, New York, NY, 2001.

181 (11) "A Vision to Enhance U.S. Economic Security through Renewable

Plant/Crop-Based Resource Use," Plant/Crop-Based Renewble Resources 2020,

http://www.oit.doe.gov/agriculture/pdfs/ag_vision.pdf.

(12) Perez, J.; Munoz-Dorado, J.; de la Rubia, T.; Martinez, J. Int. Microbiol.

2002, 5, 53-63.

(13) Mohanty, A. K.; Misra, M.; Hinrichsen, G. Macromol. Mater. Eng. 2000,

276/277, 1-24.

(14) Poirier, Y. Nat. Biotechnol. 1999, 17, 960-961.

(15) Lee, S. Y. Trends Biotechnol. 1996, 14, 431-438.

(16) Liddell, J. M. Special Publication - Royal Society of Chemistry 1992, 103,

10-25.

(17) Taguchi, S.; Doi, Y. Macromol. Biosci. 2004, 4, 145-156.

(18) Dartee, M.; Lunt, J.; Shafer, A. Chem. Fibers Int. 2000, 50, 546,548,551.

(19) Gruber, P.; O'Brien, M. Biopolymers 2002, 4, 235-250.

(20) Ritter, S. K. Chem. Eng. News 2002, 80, 26-30.

(21) Holder, J. V. In Handbook of Green Chemistry and Technology;

Blackwell,, 2002, pp 28-55.

(22) Xiaoding, X.; Moulijn, J. A. Energy Fuels 1996, 10, 305-325.

(23) Beckman, E. J. J. Supercrit. Fluids 2004, 28, 121-191.

(24) http://www.epa.gov/greenchemistry/presgcc.html.

(25) Rokicki, A.; Kuran, W. J. Macromol. Sci., Rev. Macromol. Chem. 1981,

C21, 135-186.

(26) Super, M. S.; Beckman, E. J. Trends Polym. Sci. 1997, 5, 236-240.

182 (27) Inoue, S.; Koinuma, H.; Tsuruta, T. Makromol. Chem. 1969, 130, 210-

220.

(28) Inoue, S.; Koinuma, H.; Tsuruta, T. J. Polym. Sci., Pol. Lett. 1969, 7, 287-

292.

(29) Ree, M.; Bae, J. Y.; Jung, J. H.; Shin, T. J. Korea Polym. J. 1999, 7, 333-

347.

(30) Darensbourg, D. J.; Yarbrough, J. C.; Ortiz, C.; Fang, C. C. J. Am. Chem.

Soc. 2003, 125, 7586-7591.

(31) Kuran, W.; Gorecki, P. Makromol. Chem. 1983, 184, 907-912.

(32) Kuran, W.; Listos, T. Macromol. Chem. Phys. 1994, 195, 1011-1015.

(33) Chisholm, M. H.; Navarro-Llobet, D.; Zhou, Z. Macromolecules 2002, 35,

6494-6504.

(34) Coates, G. W. Chem. Rev. 2000, 100, 1223-1252.

(35) Coates, G. W. J. Chem. Soc., Dalton Trans. 2002, 467-475.

(36) Aida, T.; Inoue, S. Acc. Chem. Res. 1996, 29, 39-48.

(37) Aida, T.; Inoue, S. Porphyrin Handbook 2000, 6, 133-156.

(38) Inoue, S. J. Macromol. Sci., Chem. 1988, A25, 571-582.

(39) Inoue, S. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 2861-2871.

(40) Aida, T.; Ishikawa, M.; Inoue, S. Macromolecules 1986, 19, 8-13.

(41) Jung, J. H.; Ree, M.; Chang, T. J. Polym. Sci., Part A: Polym. Chem.

1999, 37, 3329-3336.

(42) Kruper, W. J.; Dellar, D. D. J. Org. Chem. 1995, 60, 725-727.

183 (43) Paddock, R. L.; Hiyama, Y.; McKay, J. M.; Nguyen, S. T. Tetrahedron

Letters 2004, 45, 2023-2026.

(44) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421-431.

(45) Darensbourg, D. J.; Yarbrough, J. C. J. Am. Chem. Soc. 2002, 124, 6335-

6342.

(46) Hansen, K. B.; Leighton, J. L.; Jacobsen, E. N. J. Am. Chem. Soc. 1996,

118, 10924-10925.

(47) Darensbourg, D. J.; Mackiewicz, R. M.; Rodgers, J. L.; Phelps, A. L.

Inorg. Chem. 2004, 43, 1831-1833.

(48) Eberhardt, R.; Allmendinger, M.; Rieger, B. Macromol. Rapid Commun.

2003, 24, 194-196.

(49) Qin, Z.; Thomas, C. M.; Lee, S.; Coates, G. W. Angew. Chem., Int. Ed.

Engl. 2003, 42, 5484-5487.

(50) Lu, X.; Wang, Y. Angew. Chem. Int. Ed. 2004, 43, 3574-3577.

(51) Kuran, W.; Rokicki, A.; Pasynkiewicz, S. J. Organomet. Chem. 1978,

157, 135-143.

(52) Kuran, W.; Pasynkiewicz, S.; Skupinska, J. Makromol. Chem. 1977, 178,

47-54.

(53) Kuran, W.; Pasynkiewicz, S.; Skupinska, J.; Rokicki, A. Makromol.

Chem. 1976, 177, 11-20.

(54) Allen, S. D.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem.

Soc. 2002, 124, 14284-14285.

184 (55) Cheng, M.; Moore, D. R.; Reczek, J. J.; Chamberlain, B. M.; Lobkovsky,

E. B.; Coates, G. W. J. Am. Chem. Soc. 2001, 123, 8738-8749.

(56) Darensbourg, D. J.; Wildeson, J. R.; Yarbrough, J. C.; Reibenspies, J. H.

J. Am. Chem. Soc. 2000, 122, 12487-12496.

(57) Nakano, K.; Nozaki, K.; Hiyama, T. J. Am. Chem. Soc. 2003, 125, 5501-

5510.

(58) Chisholm, M. H.; Zhou, Z., Unpublished results.

(59) Datta, R.; Tsai, S.-P. ACS Symposium Series 1997, 666, 224-236.

(60) Benicewicz, B. C.; Hopper, P. K. J. Bioactive & Comp. Polym. 1990, 5,

453-472.

(61) Benicewicz, B. C.; Hopper, P. K. J. Bioactive & Comp. Polym. 1991, 6,

64-94.

(62) Slager, J.; Domb, A. J. Adv. Drug Delivery Rev. 2003, 55, 549-583.

(63) Huh, K. M.; Cho, Y. W.; Park, K. Drug Delivery Technology 2003, 3, 42,

44-49.

(64) Gunatillake, P. A.; Adhikari, R. Eur. Cells Mater. 2003, 5, 1-16.

(65) Drumright, R. E.; Gruber, P. R.; Henton, D. E. Adv. Mater. 2000, 12,

1841-1846.

(66) Penczek, S.; Duda, A.; Kowalski, A.; Libiszowski, J.; Majerska, K.; Biela,

T. Macromol. Symp. 2000, 157, 61-70.

(67) Schwach, G.; Coudane, J.; Engel, R.; Vert, M. Polym. Int. 1998, 46, 177-

182.

(68) Kasperczyk, J. Polymer 1996, 37, 201-203.

185 (69) Baran, J.; Duda, A.; Kowalski, A.; Szymanski, R.; Penczek, S. Macromol.

Symp. 1997, 123, 93-101.

(70) Duda, A.; Penczek, S. Macromolecules 1990, 23, 1636-1639.

(71) Witzke, D. R.; Kolstad, J. J.; Narayan, R. Macromolecules 1997, 30,

7075-7085.

(72) O'Keefe, B. J.; Hillmyer, M. A.; Tolman, W. B. J. Chem. Soc., Dalton

Trans. 2001, 2215-2224.

(73) Nakano, K.; Kosaka, N.; Hiyama, T.; Nozaki, K. Dalton Trans. 2003,

4039-4050.

(74) Chisholm, M. H.; Eilerts, N. W. Chem. Commun. 1996, 853-854.

(75) Chisholm, M. H.; Eilerts, N. W.; Huffman, J. C.; Iyer, S. S.; Pacold, M.;

Phomphrai, K. J. Am. Chem. Soc. 2000, 122, 11845-11854.

(76) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.; Lobkovsky,

E. B.; Coates, G. W. J. Am. Chem. Soc. 2001, 123, 3229-3238.

(77) Chisholm, M. H.; Huffman, J. C.; Phomphrai, K. J. Chem. Soc., Dalton

Trans. 2001, 222-224.

(78) Gibson, V. C.; Marshall, E. L.; Navarro-Llobet, D.; White, A. J. P.;

Williams, D. J. J. Chem. Soc., Dalton Trans. 2002, 4321-4322.

(79) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.; Williams, D.

J. Dalton Trans. 2004, 570-578.

(80) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.; Williams, D.

J. Chem. Commun. 2001, 283-284.

186 (81) Williams, C. K.; Breyfogle, L. E.; Choi, S. K.; Nam, W.; Young, V. G.,

Jr.; Hillmyer, M. A.; Tolman, W. B. J. Am. Chem. Soc. 2003, 125, 11350-11359.

(82) Williams, C. K.; Brooks, N. R.; Hillmyer, M. A.; Tolman, W. B. Chem.

Commun. 2002, 2132-2133.

(83) Spinu, M.; Jackson, C.; Keating, M. Y.; Gardner, K. H. J. Macromol. Sci.,

Pure Appl. Chem. 1996, A33, 1497-1530.

(84) Yui, N.; Dijkstra, P. J.; Feijen, J. Makromol. Chem. 1990, 191, 481-488.

(85) Tsuji, H.; Ikada, Y. Polymer 1999, 40, 6699-6708.

(86) Spassky, N.; Wisniewski, M.; Pluta, C.; Le Borgne, A. Macromol. Chem.

Phys. 1996, 197, 2627-2637.

(87) Wisniewski, M.; Le Borgne, A.; Spassky, N. Macromol. Chem. Phys.

1997, 198, 1227-1238.

(88) Radano, C. P.; Baker, G. L.; Smith, M. R., III J. Am. Chem. Soc. 2000,

122, 1552-1553.

(89) Ovitt, T. M.; Coates, G. W. J. Polym. Sci., Part A: Polym. Chem. 2000,

38, 4686-4692.

(90) Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc. 2002, 124, 1316-1326.

(91) Nomura, N.; Ishii, R.; Akakura, M.; Aoi, K. J. Am. Chem. Soc. 2002, 124,

5938-5939.

(92) Tang, Z.; Chen, X.; Pang, X.; Yang, Y.; Zhang, X.; Jing, X.

Biomacromolecules 2004, 5, 965-970.

(93) Zhong, Z.; Dijkstra, P. J.; Feijen, J. Angew. Chem., Int. Ed. Engl. 2002, 41,

4510-4513.

187 (94) Zhong, Z.; Dijkstra, P. J.; Feijen, J. J. Am. Chem. Soc. 2003, 125, 11291-

11298.

(95) Majerska, K.; Duda, A. J. Am. Chem. Soc. 2004, 126, 1026-1027.

(96) Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; White, A. J. P.; Williams,

D. J. J. Am. Chem. Soc. 2004, 126, 2688-2689.

(97) Tonelli, A. E. Polym. Spectrosc. 1996, 55-95.

(98) Tonelli, A. E. Annu. Rep. NMR Spectrosc. 1997, 34, 185-229.

(99) Lednor, P. W.; Rol, N. C. J. Chem. Soc., Chem. Commun. 1985, 598-599.

(100) Chisholm, M. H.; Navarro-Llobet, D. Macromolecules 2002, 35, 2389-

2392.

(101) Steiner, E. C.; Pelletier, R. R.; Trucks, R. O. J. Am. Chem. Soc. 1964, 86,

4678-4686.

(102) Eckert, H.; Forster, B. Angew. Chem. 1987, 99, 922-923.

(103) Parr, R. G.; Yang, W. Density-functional Theory of Atoms and Molecules,

1989.

(104) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.

A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R. E.;

Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain,

M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;

Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.;

Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;

Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko,

A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.;

188 Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.;

Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-

Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98; revision A9 ed.; Gaussian,

Inc.: Pittsburgh, PA, 1998.

(105) Mazhar ul, H.; Caughlan, C. N.; Smith, G. D.; Ramirez, F.; Glaser, S. L. J.

Org. Chem. 1976, 41, 1152-1154.

(106) Bohets, H.; van der Veken, B. J. Physical Chemistry Chemical Physics

1999, 1, 1817-1826.

(107) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027-2094.

(108) Wiberg, K. B. J. Comput. Chem. 1999, 20, 1299-1303.

(109) Gorecki, P.; Kuran, W. J. Polym. Sci., Pol. Lett. 1985, 23, 299-304.

(110) Becke, A. D. Physical Review A: Atomic, Molecular, and Optical Physics

1988, 38, 3098-3100.

(111) Darensbourg, D. J.; Holtcamp, M. W. Coord. Chem. Rev. 1996, 153, 155-

174.

(112) Kuran, W. Prog. Polym. Sci. 1998, 23, 919-992.

(113) Nozaki, K. J. Polym. Sci., Part A: Polym. Chem. 2003, 42, 215-221.

(114) March, J. Advanced Organic Chemistry, 4th Ed.; Wiley: New York, 1992.

(115) Parker, R. E.; Isaacs, N. S. Chem. Rev. 1959, 59, 737-799.

(116) Moore, D. R.; Cheng, M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem.

Soc. 2003, 125, 11911-11924.

(117) Moore, D. R.; Cheng, M.; Lobkovsky, E. B.; Coates, G. W. Angew.

Chem., Int. Ed. Engl. 2002, 41, 2599-2602.

189 (118) Antelmann, B.; Chisholm, M. H.; Iyer, S. S.; Huffman, J. C.; Navarro-

Llobet, D.; Pagel, M.; Simonsick, W. J.; Zhong, W. Macromolecules 2001, 34,

3159-3175.

(119) Schilling, F. C.; Tonelli, A. E. Macromolecules 1986, 19, 1337-1343.

(120) Goeke, G. L.; Karol, F. J. In U.S. Patent; 4193892, 1980.

(121) Aida, T.; Inoue, S. J. Am. Chem. Soc. 1983, 105, 1304-1309.

(122) Aida, T.; Maekawa, Y.; Asano, S.; Inoue, S. Macromolecules 1988, 21,

1195-1202.

(123) Asano, S.; Aida, T.; Inoue, S. Macromolecules 1985, 18, 2057-2061.

(124) Takeda, N.; Inoue, S. Bull. Chem. Soc. Jpn. 1978, 51, 3564-3567.

(125) Chen, P. Angew. Chem., Int. Ed. Engl. 2003, 42, 2832-2847.

(126) Bochmann, M.; Lancaster, S. J.; Hannant, M. D.; Rodriguez, A.;

Schormann, M.; Walker, D. A.; Woodman, T. J. Pure Appl. Chem. 2003, 75,

1183-1195.

(127) Entelis, S. G.; Korovina, G. Makromol. Chem. 1974, 175, 1253-1280.

(128) Munoz-Hernandez, M.-A.; McKee, M. L.; Keizer, T. S.; Yearwood, B. C.;

Atwood, D. A. J. Chem. Soc., Dalton Trans. 2002, 410-414.

(129) Jegier, J. A.; Munoz-Hernandez, M.-A.; Atwood, D. A. J. Chem. Soc.,

Dalton Trans. 1999, 2583-2588.

(130) Inoue, M.; Sugita, T.; Kiso, Y.; Ichikawa, K. Bull. Chem. Soc. Jpn. 1976,

49, 1063-1071.

(131) Shoen, E.; Zhang, X.; Zhou, Z.; Chisholm, M. H.; Chen, P., Manuscript in

preparation.

190 (132) Paddock, R. L.; Nguyen, S. T. J. Am. Chem. Soc. 2001, 123, 11498-11499.

(133) Tohma, H. Organometallic News 1995, 137.

(134) Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997,

277, 936-938.

(135) Nielsen, L. P. C.; Stevenson, C. P.; Blackmond, D. G.; Jacobsen, E. N. J.

Am. Chem. Soc. 2004, 126, 1360-1362.

(136) Darensbourg, D. J.; Holtcamp, M. W.; Khandelwal, B.; Klausmeyer, K.

K.; Reibenspies, J. H. J. Am. Chem. Soc. 1995, 117, 538-539.

(137) Beckwith, J. D.; Tschinkl, M.; Picot, A.; Tsunoda, M.; Bachman, R.;

Gabbaie, F. P. Organometallics 2001, 20, 3169-3174.

(138) Harder, S.; Boersma, J.; Brandsma, L.; Kanters, J. A.; Duisenberg, A. J.

M.; Van Lenthe, J. H. Organometallics 1990, 9, 511-516.

(139) Coe, B. J.; Glenwright, S. J. Coord. Chem. Rev. 2000, 203, 5-80.

(140) Sahasrabudhe, S.; Yearwood, B. C.; Atwood, D. A.; Scott, M. J. Inorg.

Synth. 2004, 34, 14-20.

(141) Yoo, Y.; McGrath, J. E. Pollimo 1994, 18, 892-894.

(142) Yoo, Y.; McGrath, J. E. Pollimo 1994, 18, 1063-1070.

(143) Nuytens, F.; Lopitaux, G.; Faven, C.; Coqueret, X. Macromol. Symp.

2000, 153, 17-25.

(144) Lopitaux, G.; Ricart, G.; Faven, C.; Coqueret, X. Macromol. Chem. Phys.

2002, 203, 2560-2569.

(145) Uno, H.; Takata, K.; Mizutani, Y. React. Polym. 1991, 15, 121-129.

191 (146) Kettle, S. F. A.; Editor Physical Inorganic Chemistry: Coordination

Chemistry Approach, 1996, p31-38.

(147) Aida, T.; Sugimoto, H.; Kuroki, M.; Inoue, S. J. Phys. Org. Chem. 1995,

8, 249-257.

(148) Adler, A. D.; Longo, F. R.; Kampas, F.; Kim, J. J. Inorg. Nucl. Chem.

1970, 32, 2443-2445.

(149) Summerville, D. A.; Jones, R. D.; Hoffman, B. M.; Basolo, F. J. Am.

Chem. Soc. 1977, 99, 8195-8202.

(150) Sakurai, T.; Yamamoto, K.; Naito, H.; Nakamoto, N. Bull. Chem. Soc.

Jpn. 1976, 49, 3042-3046.

(151) Witowska-Jarosz, J.; Gorski, L.; Malinowska, E.; Jarosz, M. Journal of

Mass Spectrometry 2002, 37, 1236-1241.

(152) Kadish, K. M.; Cornillon, J. L.; Coutsolelos, A.; Guilard, R. Inorg. Chem.

1987, 26, 4167-4173.

(153) Coates, G. W.; Hustad, P. D.; Reinartz, S. Angew. Chem., Int. Ed. Engl.

2002, 41, 2236-2257.

(154) Kricheldorf, H. R. Chemosphere 2001, 43, 49-54.

(155) Spassky, N.; Pluta, C.; Simic, V.; Thiam, M.; Wisniewski, M. Macromol.

Symp. 1998, 128, 39-51.

(156) Chisholm, M. H.; Gallucci, J.; Phomphrai, K. Chem. Commun. 2003, 48-

49.

(157) Chisholm, M. H.; Delbridge, E. E. New J. Chem. 2003, 27, 1167-1176.

(158) Chisholm, M. H.; Phomphrai, K. Inorg. Chim. Acta 2003, 350, 121-125.

192 (159) Zell, M. T.; Padden, B. E.; Paterick, A. J.; Thakur, K. A. M.; Kean, R. T.;

Hillmyer, M. A.; Munson, E. J. Macromolecules 2002, 35, 7700-7707.

(160) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307-326.

(161) Sheldrick, G. M.; Schneider, T. R. Methods Enzymol. 1997, 277, 319-343.

193