Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An unusually stable chiral ethyl zinc complex : reactivity and polymerization of lactide 2008

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_2008_fall_labourdette_guillaume.pdf [ 1.6MB ]
Metadata
JSON: 1.0061770.json
JSON-LD: 1.0061770+ld.json
RDF/XML (Pretty): 1.0061770.xml
RDF/JSON: 1.0061770+rdf.json
Turtle: 1.0061770+rdf-turtle.txt
N-Triples: 1.0061770+rdf-ntriples.txt
Citation
1.0061770.ris

Full Text

AN UNUSUALLY STABLE CHIRAL ETHYL ZINC COMPLEX: REACTIVITY AND POLYMERIZATION OF LACTIDE by  GUILLAUME LABOURDETTE  B.Sc., Bishop’s University, 2006      A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE   in   THE FACULTY OF GRADUATE STUDIES  (Chemistry)     THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)      August 2008   © Guillaume Labourdette, 2008 ii ABSTRACT    The racemic (±)-2,4-di-tert-butyl-6-(((2-(dimethylamino)cyclohexyl)(methyl) amino)methyl)phenol ((±)-(NNMeOtBu)H), (±)-2,4-di-tert-butyl-6-((2-(dimethylamino) cyclohexylamino)methyl)phenol ((±)-(NNHOtBu)H), and (±)-2-(((2-(dimethylamino) cyclohexyl)(methyl)amino) methyl)phenol ((±)-(NNMeOH)H) are chiral ancillary NNO proligands, which synthesis was adapted from a published procedure. Reaction of (±)-(NNMeOtBu)H ((±)-2), (±)-(NNMeOH)H  ((±)-3) and (±)-(NNHOtBu)H ((±)-1) with ZnEt2 successfully yielded the corresponding zinc ethyl complexes (±)-5, (±)-6 and (±)-7 respectively; the enantiomerically pure (R,R)-5 was synthesized from (R,R)-2. NMR spectroscopy experiments and X-ray crystallography allowed identification of two stereoisomers for (±)-5, which were observed in solution and in the solid state. The two stereoisomers, 5-α and 5-β, are in equilibrium in solution, with 5-β being thermodynamically favored. The zinc ethyl complexes were found to be unreactive towards weakly acidic alcohols (methanol, ethanol, isopropanol). However, the zinc chloride complex (±)-(NNMeOtBu)ZnCl ((±)-8) and the zinc phenoxide (NNMeOtBu)ZnOPh ((±)-9 and (R,R)-9) could be isolated and characterized. Comparison of the reactivity of both (±)-5 and the reported L1ZnEt (L1 = 2,4-di-tert-butyl-6- {[(2'-dimethylaminoethyl) methylamino]methyl}phenolate) in presence of pyridine led to the proposal of a dissociative mechanism explaining the fundamental difference between the two zinc ethyl species. Polymerization of rac-lactide catalyzed by 9 showed that the complex, in its racemic or enantiomerically pure version, has a slow activity and is not stereoselective.  iii TABLE OF CONTENTS    Abstract ................................................................................................................................ ii Table of Content..................................................................................................................iii List of Tables........................................................................................................................ v List of Figures ..................................................................................................................... vi List of Charts.....................................................................................................................viii List of Schemes................................................................................................................... ix List of Abbreviations............................................................................................................ x Acknowledgements............................................................................................................xii   Chapter 1. Literature Review  1.1 Poly(lactic acid): a biodegradable polymer ................................................................... 1 1.2 Synthesis of PLA ........................................................................................................... 5 1.3 Metal catalysts for ROP of lactide ................................................................................. 9 1.4 Single-site metal catalysts for lactide ROP.................................................................. 10 1.5 Single-site zinc catalysts .............................................................................................. 15 1.6 Organozinc compounds ............................................................................................... 22  Chapter 2. Synthesis and Reactivity of some New Zinc Complexes  2.1 L1ZnOEt, a highly active zinc catalyst for ROP of lactide .......................................... 30 2.2 Compounds synthesis................................................................................................... 33 2.2.1 Synthesis of the proligands ................................................................................... 33 2.2.2 Synthesis of the zinc complexes ........................................................................... 37 2.3 Reaction of zinc ethyl complexes with alcohols.......................................................... 52 2.3.1 Formation of alkoxides via alcoholysis ................................................................ 52 2.3.2 Mechanistic approach - coordination to pyridine ................................................. 59 2.4 Polymerization ............................................................................................................. 62 iv Conclusion ........................................................................................................................ 69 Experimental ...................................................................................................................... 71 Bibliography ...................................................................................................................... 82 Appendix............................................................................................................................ 90 v LIST OF TABLES    Table 1. 1. Summary of comparative data for the polymerization of lactide initiated by various zinc catalysts ......................................................................................................... 19  Table 2.1. Results for the polymerization of rac-lactide by L1ZnOEt............................... 32  Table 2.2. Selected interatomic distances (Ǻ) and bond angles for zinc ethyl complexes .......................................................................................................................... 39  Table 2.3. Selected interatomic distances (Ǻ) and bond angles (°) for compound (±)-6... 42  Table 2.4. Selected interatomic distances (Ǻ) and bond angles (°) for compound (±)-8... 54  Table 2.5.Selected interatomic distances (Ǻ) and bond angles (°) for compound (±)-9.... 58  Table 2.6. Probability of tetrad sequences in PLA according to the Bernoullian model ... 65  Table 2.7. Selected data for ring opening polymerization of rac-LA................................ 66  Table A.1. Crystal and refinement data for complexes (±)-5, (R,R)-5, (±)-6, (±)-8, and (±)-9•phenol ....................................................................................................................... 90     vi LIST OF FIGURES    Figure 1.1. [MeZn(DAD)][B(C6F5)4] ................................................................................ 25  Figure 1.2. Representation of the molecular structure of L1ZnEt, a discrete monoorganozinc compound bearing a Zn-O σ-bond, with selected nonhydrogen atoms labeled and hydrogen atoms omitted for clarity ................................................................ 26  Figure 1.3. Dimeric organozinc complex [L3ZnEt]2.......................................................... 27  Figure 1.4. (BDI-1)ZnEt .................................................................................................... 28  Figure 1.5. Tris(3-tert-butylpyrazolyl)hydroborato methyl zinc, [TptBu]ZnMe................. 28  Figure 2.1. Chiral tridentate ancillary proligands (±)-(NNHOtBu)H ((±)-1), (±)- (NNMeOtBu)H ((±)-2) and (±)- (NNMeOH)H ((±)-3).............................................................. 33  Figure 2.2. Molecular structure of two enantiomers of (±)-5 with ellipsoids at 50% probability level. Most hydrogen atoms omitted for clarity .............................................. 38  Figure 2.3. Molecular structure of two enantiomers of (R,R)-5 with ellipsoids at 50% probability level. Most hydrogen atoms omitted for clarity .............................................. 38  Figure 2.4. Identification of the peaks corresponding to the methylene and methyl groups in Zn-CH4H3 using 1H COSY NMR (C6D6, 25°C, 400 MHz)........................................... 40  Figure 2.5. Molecular structure of (±)-6 with ellipsoids at 50% probability level. Most hydrogen atoms omitted for clarity.................................................................................... 42  Figure 2.6.  Structural comparison of: a) (RR)-5, b) (±)-6 and c) L1ZnEt ....................... 44  Figure 2.7. 1H NMR spectrum of (±)-7 (C6D6, 298 K, 400 MHz)..................................... 45  Figure 2.8. 1H NMR spectrum of (R,R)-5 (C6D6, 25°C, 400 MHz) .................................. 46  Figure 2.9. 1H NMR (C6D6, 400 MHz) of (±)-5 after 10 min reaction at 298 K (top) and heating at 80 °C in C6D6 for 72 h (bottom)........................................................................ 48  Figure 2.10. Number assignment of selected atoms of 5-α and 5-β based on X-ray crystallography data ........................................................................................................... 49  Figure 2.11. 1D selective NOE experiments (C6D6, 298 K, 400 MHz) for a solution of (±)-5 containing minimal amount of 5-β (top) and a solution of (±)-5 containing 50% of 5-β (bottom), where the protons at C9 and C40 have been irradiated, respectively.......... 50 vii  Figure 2.12 1D selective NOE experiments (C6D6, 298 K, 400 MHz) for a solution of (±)-5 containing 50% of 5-β, where the cyclohexyl protons at C3 (top) and C33 (bottom) have been irradiated ........................................................................................................... 51  Figure 2.13. 1D selective NOE experiment (C6D6, 400 MHz) of a 50:50 mixture of 5-α and 5-β revealing through-space interaction between C3 and C9 protons for 5-α, and C33 and C40 protons for 5-β ..................................................................................................... 51  Figure 2.14. Molecular structure of (±)-8 with ellipsoids at 50% probability level.  Most hydrogen atoms omitted for clarity.................................................................................... 54  Figure 2.15. 1H NMR spectrum of (±)-8 (C6D6, 298 K, 300 MHz)................................... 55  Figure 2.16. Molecular structure of (±)-9•phenol with ellipsoids at 50% probability level. Most hydrogen atoms removed for clarity......................................................................... 57  Figure 2.17.  Addition of 2 equiv of pyridine to LZnEt (CD2Cl2, 298 K) at time = 0 min (bottom) and time = 60 min (top) ...................................................................................... 60 Figure 2.18. 1D NOE experiment of L1ZnEt (CD2Cl2, 298 K, 400 MHz). The irradiated methyl proton at 2.12 ppm shows through-space coupling with the methyl protons at 2.37 ppm .................................................................................................................................... 60  Figure 2.19 1H NMR spectrum of PLA (CDCl3, 298 K, 600 MHz) .................................. 65  Figure 2.20. Stereosequences of isotactic PLA induced by enantiomorphic site control .. 65  Figure 2.21.  Plot of [LA]/[TMB] for polymerization of 188 equiv rac-LA with (±)-9 in the presence of 1,3,5-trimethoxybenzene (TMB) internal standard.  The methine proton of LA was monitored via 1H-NMR spectroscopy (CD2Cl2, 298 K, 400 MHz). [(±)-9]o = 2.4 mM, [LA]o = 0.45 M) .................................................................................................. 67  Figure 2.22. Methine region of the 1H{1H} NMR spectrum of PLA (CDCl3, 298 K, 600 MHz) obtained from polymerization of rac-LA with (±)-9............................................... 68 viii LIST OF CHARTS    Chart 1.1. Molecular structure of lactic acid, lactide and poly(lactic acid) ......................... 2  Chart 1.2. Possible stereoisomers of lactide ........................................................................ 3  Chart 1.3. Microstructures of PLA ...................................................................................... 4  Chart 1.4. Historical and industrial catalysts for lactide ROP ............................................. 9  Chart 1.5. Metal-catalysts for ROP of lactide.................................................................... 13  Chart 1.6. Early zinc catalysts for LA ROP....................................................................... 17  Chart 1.7. Remarkable zinc catalysts................................................................................. 20  ix LIST OF SCHEMES    Scheme 1.1. Routes for industrial synthesis of PLA ........................................................... 6  Scheme 1.2 Coordination-insertion mechanism .................................................................. 8 Scheme 1.3. Schlenk equilibrium for Zn(II) species.......................................................... 11 Scheme 1.4. Major routes towards synthesis of heteroleptic monoorganozinc compounds ......................................................................................................................... 24  Scheme 2.1. Synthesis of Hillmyer’s zinc ethoxide complexes ........................................ 31 Scheme 2.2. Equilibrium between dimeric (predominant in the solid state) and monomeric species (predominant in solution)................................................................... 31  Scheme 2.3. Equilibrium observed when reacting trans-1,2-diaminocyclohexane with a salicylaldehyde (R = tBu, H).............................................................................................. 34  Scheme 2.4. Synthesis of proligands (±)-1, (±)-2 and (±)-3 ................................................ 35  Scheme 2.5. Resolution of (±)-trans- diaminocyclohexane .............................................. 36  Scheme 2.6. Synthesis of (±)-5........................................................................................... 37  Scheme 2.7. Synthesis of (±)-6........................................................................................... 41 Scheme 2.8 Synthesis of (±)-7............................................................................................ 45 Scheme 2.9. Synthesis of (±)-8 .......................................................................................... 53 Scheme 2.10. Synthesis of (±)-9 ........................................................................................ 56  Scheme 2.11. Proposed dissociation mechanism for the reaction of L1ZnEt and alcohols/ pyridine .............................................................................................................................. 62      x LIST OF ABBREVIATIONS    ASTM   American Society for Testing and Materials  BINAP   2,2’-bis(diphenylphosphino)-1,1’-binaphthyl  Bn    Benzyl, -CH2(C6H5)  t-Bu   Tert-butyl, -CMe3  br    Broad  Calcd   Calculated  COSY  Correlation spectroscopy  DACH  Trans-diaminocyclohexane  DMF   Dimethylformamide  d     Doublet  deg, (°)  Degree(s)  eq    Equation(s)  Et    Ethyl  g     Grams  GPC   Gel Permeation Chromatography  1H{1H}  Homonuclear decoupled proton  h     Hour(s)  i-Pr   iso-propyl, -CHMe2  J     Coupling constant in Hertz  LA    Lactide  Me    Methyl  xi Mn    Number average molecular weight  Mw    Weight average molecular weight  NMR   Nuclear magnetic resonance  NOE   Nuclear Overhauser Effect  ORTEP  Oak Ridge Thermal Ellipsoid Plot  PDI   Polydispersity index  Ph    Phenyl  PKa   Negative logarithm of the acid dissociation constant (Ka)  PLA   Poly(lactic) acid  Pm   Probability of forming a meso linkage  Pr    Probability of forming a racemic linkage  q     Quartet  ROP   Ring opening polymerization  SEC   Size-exclusion chromatography  THF   Tetrahydrofuran  xii ACKNOWLEDGEMENTS   I would like to thank my supervisor, Parisa Mehrkhodavandi, for her support during these two years in the group, and especially for convincing me to persevere on my research and studies when times were not easy. Also, among the past and present group members, I want to particularly thank Chris Wallis for giving me advices in chemistry when they were needed, and Amy Douglas for introducing me to the techniques of the laboratory while demonstrating the ligand synthesis to me. Also, thank you to Daniel for this good summer 2007 spent together in the lab.  Thanks to the UBC Chemistry department, in particular to Maria Ezhova from the NMR facility, for her help in the assignment of my compounds and the multiple NMR experiments she ran or helped me to run during my Master studies. Thanks to Brian Patrick as well, for solving the crystal structures that are contained in this work.  Although I have been far away from home, I have always benefited from the support of my family. Above all, thank you mum and dad for supporting me in my choices. Thanks to Silke, who was in my heart for a long time; she was definitely part of this experience. Finally, I am grateful to Qianyu who has been with me everyday of these last six months. You keep enlightening my days.        1 CHAPTER 1 – LITERATURE REVIEW  1. 1. Poly(lactic acid): a biodegradable polymer Chemistry of synthetic polymers has been extensively developed to the point where the commonly called “plastic” materials are omnipresent in today’s society. However, the last decades have seen a raise in concern towards the use of petrochemical-based polymers coinciding with the growing awareness for environment and more recently the significant increase in oil prices. Synthetic polymers have been dominating modern consumerism because of their extremely low cost of fabrication and the high control that can be exerted on their chemical and physical properties, but they present major drawbacks that can no longer wait to be ignored: the issue of disposal and degradability, and their fabrication from non- renewable resources (i.e. fossil fuels). In this respect, there is significant interest in developing biodegradable polymers that can be derived from biorenewable resources. “Biodegradable polymers” can be defined as “degradable plastics in which the degradation results from the action of naturally occurring microorganisms”,1 while biorenewable resources (also known as biomass) are generally recognized as “organic materials of recent biological origins”,2  i.e. sustainable natural resources. Let aside the naturally occurring polymers such as polysaccharides (starch, cellulose), polypeptides (gelatine) or polyhydroxyalkanoates, a category of biodegradable synthetic polymers that has received a great deal of attention is the polyesters made from lactones.3 More than any other synthetic polyester, poly(lactic acid) (PLA) is the polymer of choice that can be found in multiple practical and potential applications. Companies such as NatureWorks LLC® (a subsidiary from the multinational corporation Cargill, Inc.)4 have created a market for PLA for more than a decade now; the polymer finds major applications in the packaging 5 and biomedical applications such as drug delivery excipients and absorbable sutures.6 2 PLA is an aliphatic polyester derived from lactic acid. Natural production of lactic acid is observed in mammalian muscles during glycogenolysis and is involved in the Kreb’s cycle through pyruvic acid and acetyl-CoA.7 Although synthetic lactic acid is still produced from petrochemical feedstocks through the hydrolysis of lactonitrile, producing racemic lactic acid,8 the main production is obtained from biorenewable resources. Indeed, lactic acid is easily prepared in high yield by fermentation of molasses or potato starch or of dextrose from corn.9 As it can be seen from its molecular structure (Chart 1.1), lactic acid, formally named 2-hydroxycarboxylic acid, possesses a stereogenic center on the methylene carbon. Therefore two enantiomers of lactic acid can be found: D-(-)-lactic acid (also found under the name (R)- lactic acid) and L-(+)-lactic acid (or (S)-lactic acid); the latter is predominant in nature. By definition, the L isomer rotates the plane of polarized light clockwise while the D isomer rotates the plane of polarized light counterclockwise.  Chart 1.1. Molecular structure of lactic acid, lactide and poly(lactic acid).   Lactide (LA) is a six-membered-ring dimer with two lactic acid units; it is thus a molecule with two chiral centers and can be found in three stereoisomeric forms (Chart 1.2): D-lactide (also called (R,R)-lactide), L-lactide (or (S,S)-lactide) and the optically inactive meso-lactide (or (R,S)-lactide). PLA can be synthesized from lactic acid or lactide, which contain one and two stereogenic centers respectively, thus conferring to the polymer different microstructures. The polymer can have different physical properties depending on its content of D- and/or L-lactic acid units. The terminology referring to the different types of PLA is as 3 follows: poly(L-lactic acid) (PLLA) contains only L-lactic acid units; poly(D-lactic acid) (PDLA) is made of D-lactic acid units only; poly(D,L-lactic acid) (PDLLA or D,L-PLA) is a mixture of D- and L monomer units. Poly(L-lactic acid) and poly(D-lactic acid) are optically active whereas the racemic poly(D,L-lactic acid) shows no optical activity when it is pure (50% D-lactide).  Chart 1.2. Possible stereoisomers of lactide.    The tacticity of a polymer is the relative stereochemistry of adjacent chiral centers within a macromolecule.10  The regularity of the microstructure influences the degree to which it has rigid, crystalline long range order or flexible, amorphous long range disorder; therefore the tacticity influences the physical and mechanical properties of a polymer. A number of microstructures can be obtained from polymerization of lactide: isotactic, syndiotactic, atactic and heterotactic (Chart 1.3). Besides, microstructures containing stereoblocks of PLA or stereocomplexes of PLLA and PDLA can also be found. Isotactic PLA is formed from either 100% D- or 100% L-lactide, therefore the sequential stereogenic carbons have the same absolute configuration. On the other hand, pure syndiotactic PLA can be made from meso-lactide only; indeed, obtaining perfectly alternating configurations of the sequential stereocenters is possible only if the monomer itself contains both R- and S- chiral centers. Atactic PLA has a random distribution of configurations about the stereocenters, while heterotactic PLA has regions of stereo-homogeneity. Stereoblock PLA can be produced 4 in one way by sequential addition of D-LA followed by addition of L-LA upon complete conversion of D-LA; a second way is to polymerize rac-LA (also called D,L-LA, a racemic mixture of D- and L-lactide) with an enantiopure catalyst; the last way is when chain-transfer mechanism occurs during polymerization of rac-LA with a racemic mixture of a chiral catalyst.11  Chart 1.3. Microstructures of PLA.     PLA is elastic and flexible above its glass transition temperature (Tg = 50-57 °C for D,L-PLA) 12 and can go through melting/freezing cycles repeatedly without alteration of its physical properties. 13 Like other thermoplastics, it has low density (1.21-1.25 g/cm3), low processing costs and the ability to take on complex shapes relatively easily. Its melting point varies from 150 °C (PLLA) to 170-180 °C (PLLA), and up to 230 °C in the case of 5 stereocomplexes formed by blending PLLA and PDLA into melt spun fibers. 14  Fully amorphous PLA can be obtained by the inclusion of relatively high D-LA content (>20%) whereas highly crystalline material is obtained when the content in D-LA is low (<2%). When compared to other biodegradable polymers (poly(glycolic acid), poly(ε-caprolactone) and polyhydroxbutyrate), PLA seems to gather the best compromise in term of density (low), degradation behaviour, mechanical properties, and glass transition temperatures. 13 PLA is degraded in the environment through a two-step process. At first high molecular weight polyester chains are hydrolyzed to molecular oligomers; the degradation rate depends on temperature and moisture levels and can be accelerated by acids or bases. Then, under the action of microorganisms, these lower molecular weight chains are converted to carbon dioxide, water, and humus.15 Enzymatic degradation has been observed with lipases, such as proteinase K, which are efficient at breaking down low-molecular weight and some enzymes, such as the yeast Cyptococcus, which are able to degrade high-molecular weight PLA.16 Enzymes show a net preference for degradation of PLLA versus PDLA, with the highest degradation rates observed for amorphous PLA.17  1.2. Synthesis of PLA PLA can be obtained from two different routes: direct condensation of lactic acid and ring-opening polymerization (ROP) of the cyclic dimer lactide. The polymer was first synthesized by Carothers in 1929 as a low molecular weight product by heating lactic acid under vacuum.18 Indeed, direct condensation is an equilibrium reaction resulting in polymer of limited ultimate molecular weight, because of the difficulties in removing trace amounts of water in the late stages of polymerization. An azeotropic distillation process producing high molecular weigh PLA (Mw> 300,000) has however been reported,19 which makes use of an aprotic, high-boiling point solvent (Scheme 1.1) to overcome this problem. 6  Scheme 1.1. Routes for industrial synthesis of PLA.20    Ring-opening polymerization (ROP) of lactide has been recently the subject of extensive research.11, 21, 22, 23 It is a widely used methodology to prepare well-defined aliphatic polyesters of high molar mass and narrow molar mass distribution. Additionally, ROP allows diverse macromolecular architectures including graft, star-shaped, hyperbranched and dendrimers.24 From a thermodynamic viewpoint, lactide presents an interesting case of rare six-membered ring that exhibits a low negative enthalpy of activation (ΔH = -23 kJ/mol). Indeed, while three- and four membered rings are highly strained, the energy of ring strain diminishes to the point where ΔH becomes closer to zero (or even positive) for the large majority of six-membered rings, which makes them hard, if not impossible, to polymerize.25 X-ray crystallographic data of D-lactide (formally named (3R,6R)-3,6-dimethyl-l,4- dioxacyclohexane-2,5-dione)26 reveal a heterocycle in an irregular skew-boat conformation with the two ester moieties on opposite sides, nearly in the same plane. Assuming that the boat conformation is also favoured in solution, the relatively high absolute value of enthalpy 7 can be explained by the strain applied on the ring by the CH bonds oppositions and possibly by the bond angle distortions.  Metal-catalyzed ROP of lactide is by-far the most common method investigated in academic and industrial research, but other routes using nucleophilic catalysts are possible such as organocatalytic and enzyme-catalyzed polymerization.21 Among the metal-catalyzed mechanisms, the coordination-insertion mechanism is the promising polymerization route for LA that has received most attention compared to other routes like anionic27 and cationic28 polymerization. Due to their high reactivity, anionic and cationic initiations present some drawbacks that include susceptibility to impurity levels and side-reactions such as epimerization of the monomer stereogenic centers (i.e. racemisation) or transesterification.7b Other metal mediated polymerizations of lactide can occur through a coordination insertion mechanism (Scheme 1.2), which was first demonstrated by Kricheldorf and Teyssié about 20 years ago.29, 32 At first, a molecule of lactide is coordinated to the catalyst via dative bonding between the nucleophilic oxygen of the carbonyl and the Lewis acidic metal center. Then the monomer inserts into the metal-alkoxy bond via nucleophilic addition of the alkoxy, initiating, group on the carbonyl carbon. Finally, cleavage of the acyl bond results in ring- opening of the monomer unit.       Scheme 1.2 coordination-insertion mechanism 8    The goal of these catalysts is to achieve control over molecular weight and microstructures of the polymer. This is generally realized when the catalyst is able to promote living polymerization. Living polymerization is the term used when referring to a polymerization resulting in uniform molecular weight distribution, with a constant rate of monomer consumption throughout the polymerization. Chain transfer and chain termination should be absent from the polymerization, with, in many cases, a rate of chain initiation faster than the rate of chain propagation, i.e. constant number of kinetic-chain carriers (from the definition provided by the IUPAC Compendium of Chemical Terminology).30  A narrow molecular weigth distribution and a linear increase in polymer molecular weight with conversion are characteristic of a living polymerization. The molecular weigth distribution, also called polydispersity index (PDI), is defined as the weight average molecular weight divided by the number average molecular weight (Mw/Mn); Mw and Mn can be obtained by gel permeation chromatography (GPC). Some catalysts do not achieve a living polymerization because of chain transfer and chain termination. Transesterification, the major mechanism for chain transfer,31 can occur when one growing polymer chain binds to another leading to a random exchange of polymer ends (intermolecular transesterification), or when a polymer chain binds intramolecularly 9 leading to formation of cyclic polymer.32 Chain termination usually occurs when the catalyst reacts with impurities such as proton sources (e.g. water) introduced in the reaction mixture or from decomposition of the catalyst.  1.3. Metal catalysts for ROP of lactide Simple sodium, lithium and potassium alkoxides have shown some efficiency in lactide polymerization.33 However, these ionic species are often highly basic, resulting in epimerization of the stereogenic centers in the PLA backbone (i.e. low selectivity); metal alkoxides with a more covalent character have higher selectivity.  Chart 1.4. Historical and industrial catalysts for lactide ROP    Tin(II) bis(2-ethylhexanoate) (Chart 1.4), also found under the  tin octanoate, stannous octoate or Sn(Oct)2, is the industrial catalyst of choice for the ROP of lactide; commercially available, it is soluble in common organic solvents and melt monomers (for bulk polymerization) and is easy to handle. The highly active catalyst has shown reaction times from minutes to a few hours in bulk at 140-180 °C and yields high molecular weight polymers (up to 105 DA as sole catalyst or 106 DA in presence of an alcohol.34 Aluminum triisopropoxide (Chart 1.3) is another efficient catalyst for the ROP of LA. However, typical reaction times of several days in bulk at 125-180 °C and lower molecular weights make Al(OiPr)3 significantly less active than Sn(Oct)2.34 The aluminum alkoxide has been the subject of numerous mechanistic studies 35  due to a behaviour complicated by equilibria in solution involving species of varying nuclearity.  Aggregation phenomenon has 10 been held responsible for the induction period of few minutes systematically observed in ROP of lactide.  Lanthanides have also been investigated as metals in alkoxide catalysts. Among them, yttrium and lanthanum demonstrated great activity: Ln(OiPr)3 (Ln = Y, La) was able to polymerize 150 equivalents of lactide ([LA] = 0.2 M) in 5 minutes in dichloromethane at room temperature,36 with a rate much faster than the rate observed with Al(OiPr)3.32 Moreover, in situ UV spectroscopy did not reveal any induction period, in contrast to polymerization initiated by Al(OiPr)3. The mechanism has been reported to be more complicated than a simple coordination-insertion mechanism due to aggregation phenomenon.37 Calcium alkoxide Ca(OMe)2 achieves fast, complete conversion of lactide without induction time after 30 minutes at room temperature (with [LA]/[cat] = 1:100 and [LA] = 1 M); the PLA obtained had a very narrow molecular weight distribution (PDI = 1.03-1.07).38 It was found that formation of the catalyst in situ by alcoholysis of Ca[N(SiMe3)2]2(THF)2 in THF allows to minimize aggregation behaviour and avoid precipitation, thus leading to good control over the polymerization.  1.4. Single-site metal catalysts for lactide ROP Although first generation metal alkoxides have been successful initiators for ROP of lactide, complex aggregation and having more than one growing chain per metallic center complicate control over the polymerization. Therefore, single-site catalysts have received a great deal of attention from researchers trying to enhance catalytic activity and limit transesterification reactions. Single-site catalysts have the general formula LnMR where chain propagation occurs at the active metal center M; R is the initiating group and Ln are ancillary ligands. These non-labile ancillary ligands are not usually involved in the polymerization but 11 rather tune the properties of the metal center and its environment (such as introducing steric bulk) so as to avoid aggregation phenomena or undesired side-reactions. A number of single-site catalysts have been developed in the past 20 years, based on numerous combinations of metal and ligands. Among these metals, Mg, Zn, Al and Ca have shown particular interest due to their low toxicity, ready availability and their polymerization behaviour in term of activity and stereoselectivity. It is known, however, that the metal ions Mg2+, Zn2+ and Ca2+ are kinetically labile and may readily undergo ligand scrambling, such as in the case of Schlenk equilibrium (Scheme 1.3). Therefore, specific ligands have been designed to stabilize the catalytic species. Metals for polymerization catalysts have been mainly associated with O-donor ligands (only oxygen atom from the ancillary ligand donating to the metal), N-donor ligands (only nitrogen binding to the metal center), and N- and O- donor ligands (combination of nitrogen and oxygen donors on a polydentate ligand).  Scheme 1.3. Schlenk equilibrium for Zn(II) species    Aluminum complexes featuring methylenebisphenolate ligands (Chart 1.4) are examples of initiators containing a tetradentate O-donor ligand; they promote moderately active polymerization with very narrow molecular weight distribution.39 However, complex aggregation with O-donor ligands is often encountered due to the strong electron donating character of oxygen atoms.  Nitrogen atoms with their additional substituent are usually expected to bring more steric hindrance in N-donor ligands, and the nitrogen-metal bond is relatively inert towards monomer insertion when competing with highly active initiating groups such as an alkoxy moiety. The first historical catalytic systems featuring N-donor ligands are the 12 tetraphenylporphyrin aluminum alkoxides (Chart 1.4).40 The catalysts display control on the polymer molecular weight and narrow molecular weight distribution (PDI < 1.25). The drawback, however, is the low activity of the catalyst that requires heating at 100°C.  Interesting, tridentate N-donor ligand systems include the tin complex bearing a diamidoamino ligand (Chart 1.4); unlike its dimeric zinc counterpart, the intramolecularly stabilized stannylene exists as a monomer.41 It is also more active for bulk polymerization than the Zn complex; 95 % conversion of lactide ([LA]/[Sn] = 300) is obtained after 2 hours at 140 °C but at the expense of the molecular-weight distribution (PDI ~ 2). Trispyrazolyl complexes are catalytic systems of importance in the history of ROP of lactide (Chart 1.4). These tripodal monoanionic tridentate ligands provide large steric hindrance around the metal center to prevent aggregation. It was found that the calcium trispyrazolyl complexes are among the most active catalysts for ROP of LA discovered to date, far superior to their magnesium and zinc derivatives (possibly due to the polarity difference within the initiating M-X bond). Only 1 minute is necessary to polymerize 100 equivalent lactide at room temperature in THF although the polydispersity is broader (about 1.6) than for the other metals (1.1-1.25).42                   13 Chart 1.5. Metal-catalysts for ROP of lactide    Among the common bidentate N-donor ligands investigated for different catalytic systems in lactide ROP, β-diiminate ligands have taken a prominent place due to the exceptional stereochemical control and activity exhibited by some of their related metal complexes. Enantioselective polymerization, or stereochemical control, by single-site metal catalyst can occur through two different mechanisms. Chain-end control takes place when the stereogenic center of the last inserted monomer selectively incorporates the new enantiomer 14 of the monomer in the growing chain. On the other hand, when a chiral catalyst shows preference for only one enantiomer from a racemic mixture of monomer, the mechanism is called enantiomorphic site-control.45 In polymerization of rac-lactide, this translates into two possible tacticities: Provided the chiral center of the last inserted monomer is of R configuration and selects (R,R)-lactide, isotactic PLA is formed; if the (S,S) enantiomer is selected then the product of polymerization is heterotactic PLA. Although each mechanism independently favours stereoselective polymerization, combination of both of them (which is a frequent phenomenon) can be either destructive or constructive with respect to the overall stereoselectivity.43 Coates was first to use bulky β-diiminate ligands in ROP of lactide; the derived zinc complexes were found to stereoselectively polymerize rac-lactide via a chain-end control mechanism. 44  Especially, the complex [(BDI-1)ZnOiPr]2 (Chart 1.5) demonstrated great activity and stereoselectivity: heterotactic PLA was obtained after polymerizing rac-lactide for 30 minutes at 20°C in dichoromethane ([LA]/[Zn] = 200, [LA] = 0.4 M) with a 97% conversion rate and a very narrow polydispersity (PDI = 1.10). Chain-end mechanism allowed to reach a Pr of 0.90 (i.e. 90% probability of racemic linkage), which could be improved to 0.94 at 0 °C; no epimerization of monomer stereocenters was observed when L-lactide was polymerized to isotactic PLA. The magnesium complex [(BDI-1)MgOiPr]2 also showed high activity with 96% rac-lactide (500 equivalent, [LA] = 0.4 M) polymerized in less than 5 minutes at 20 °C. In fact, most studies of the relative reactivity of mononuclear complexes with identical supporting ligands but divergent metal ions have demonstrated that Mg2+ polymerizes lactide (LA) faster than Zn2+.42b-c, 45 Finally, a number of complexes featuring chelating ligands combining N-donors and O-donors have also been investigated. 21, 22, 23 Although relatively slow initiators, some chiral aluminum complexes bearing SALEN ligands have demonstrated excellent stereoselectivity 15 via enantiomorphic site control mechanism. The first to be reported for LA ROP was the complex R-(SalBinap)AlOMe (Chart 1.5), 46 followed few years later by the pure enantiomer of (SalBinap)AlOiPr.11 The latter polymerizes meso-lactide to syndiotactic PLA with a high probability of racemic linkage (Pr = 0.96), while the racemic version of the complex interestingly polymerizes meso and rac-lactide to heterotactic and isotactic stereoblock PLA, respectively.  In 2003, Hillmyer, Tolman et al. reported a very active zinc catalyst bearing a tridentate diaminophenolate ligand (Chart 1.5).47 The achiral zinc ethoxide complex L1ZnOEt (L1 = 2,4- di-tert-butyl-6-{[(2’-dimethylaminoethyl)methylamino]-methyl}phenolate) shows an activity 5 times faster than its β-diiminate counterpart,45 with a relatively narrow molecular weight distribution (PDI ~ 1.4). However, despite showing no epimerization when converting L- lactide to isotactic PLA, the catalyst presents no stereoselectivity and produces atactic polymer. This complex is of primary importance within the framework of this thesis for reasons that will be described later; therefore, from now on I will use the general notation L1ZnOEt, chosen by Hillmyer and Tolman, to refer to this complex throughout my thesis. When specifically referring to the dimeric form of this complex, the notation (L1ZnOEt)2 will be preferred. 16  1.5. Single-site zinc catalysts  Among the variety of metal-catalysts reviewed here, some of the zinc catalysts have demonstrated exceptional activity to polymerize lactide, sometimes with high degree of stereoselectivity.45, 62 These properties combined with low cost and biological tolerance have been major motivations for research to investigate more zinc catalysts in the recent years. Since 1959, several research groups48, 50-53 have been investigating a variety of zinc complexes for ROP of lactide, with the aim of optimizing the activity and stereoselectivity of the catalyst. Simple compounds such as zinc oxide 48, 49, 52 or zinc carbonates 50 were found to induce partial racemisation of L-lactide; it was also found that the stable, non hygroscopic salt zinc stearate follows the same trend.48 Diethyl zinc, despite showing a respectable activity, is not suited for bulk polymerization as it is a highly moisture-sensitive, flammable liquid.51, 52 Zinc powder, however, is considered as a good catalyst for polymerization of lactones to the point where it has found industrial applications;53 it is easy to handle as starting material, but its removal from the polymer product requires an ultrafiltration process,50, 54 , 55  and it promotes slow polymerization.56 Zinc lactate, a resorbable initiator, yields polymers with high conversion, activity, and molecular weight polymers; 57 Vert eventually proposed zinc lactate to be the actual initiator of lactide polymerization in presence of zinc metal. 57b A number of zinc salts, such as zinc L-mandelate, zinc glycolate,57a and salts of several amino acids were found to catalyze PLA in low molecular weight only (see Chart 1.5). Zinc halides, such as zinc chloride or zinc iodide, were also found to polymerize L-lactide, however they produce lower molecular weight poly(L-lactide) even at high temperatures.57a On the other hand, zinc bromide and zinc bis(2,2-dimethyl-3,5-heptanedionate) (Zn(DMH)2) were found to polymerize LA with high molecular weight without racemisation.58, 59  17  Chart 1.6. Early zinc catalysts for LA ROP   While stereocontrol induced by chiral metal catalysts in lactide polymerization had been demonstrated with some aluminum complexes before 1999,60 Coates’ group was first to report a stereoselective achiral zinc catalyst.61 The success of this catalyst (high activity and stereoselectivity) soon generated a number of studies performed on zinc complexes featuring β-diiminate ligands (Chart 1.7).45, 62 , 63 With the exception of entry 5, all the β-diiminate complexes reported in Table 1 (entries 1-4) showed efficient catalysis lactide ROP efficiently in chlorinated solvent at room temperature. These zinc complexes have been prepared either via an exchange reaction from the protonated ligands, or via substitution reaction from the lithium or potassium salt of the β-diiminate. Dimeric or monomeric structures were obtained, depending on the steric bulk of the β-diiminate ligand and the initiating group (Chart 1.5 for entry 1 and Chart 1.7 for entries 2-4). As the polymerizations were performed on catalysts featuring various co-ligands (amido, alkoxy, silyloxy, methyl lactacte, acetate and ethyl groups), it was found that polymerization initiated from silylalkoxy, acetate and ethyl groups are slower than with alkoxy ligands. Chain-end control, which is usually associated to 18 heterotactic PLA, was greatly influenced by modification of the aryl substituents, as indicated by the decreased heterotacticy observed at 20 °C with ethyl (79%), n-propyl (76%) compared to isopropyl groups (90%).45 On the other hand, modification of the aryl substituent from isopropyl (BDI-1) to ortho-methoxyphenyl (BDI-2) improved activity of the corresponding zinc complexes.64                     19  Table 1. Summary of comparative data for the polymerization of lactide initiated by various zinc catalysts.  20 Chart 1.7. Remarkable zinc catalysts.  These observations clearly illustrate the difficulties encountered by researchers who try to optimize both stereocontrol and activity of a catalyst. The other entries in Table 1 show the other most efficient single-site zinc catalysts that have been published to date. Jing’s β- diketone Schiff base zinc complex (entry 6) forms a macrocycle in solid state; it is capable of polymerizing 100 equivalent rac-lactide ([LA] = 0.61 M) in dichloromethane at 25°C in 30 minutes with a polydispersity of 1.09 and a remarkable Pr of 0.73.65 However, it remains unclear if all zinc centers participate in the ring-opening process. Among the zinc N- heterocyclic carbenes developed by Hillmyer, Tolman et al., the chlorinated dinuclear zinc 21 complex (entry 7) has demonstrated great activity; 130 equivalent rac-LA could be polymerized in 4 minutes at 25°C in dichloromethane, with a slight preference for racemic linkages between monomer units in the PLA chains (Pr = 0.6).66 Stereocontrol, in absence of chiral center in the N-heterocyclic ligand, is presumably induced by chain-end control mechanism. The same research group has developed two other highly active zinc catalysts based on diaminophenolate-based ligands. The first (entry 8) features two ethylenediamine arms and is dinuclear in solid state and solution.47 Polymerization of D,L-lactide ([LA]/[Zn] = 300) in dichoromethane at room temperature was obtained in 30 minutes with 90% conversion and a very narrow molecular weight distribution (PDI ~ 1.2). The characteristic features of a controlled polymerization were observed from kinetic studies and first-order dependency on monomer and initiator were in agreement with a bimetallic coordination- insertion mechanism.  The second zinc complex, (L1ZnOEt)2 (entry 9 in Table; Chart 1.4), contains only one ethylenediamine arm, therefore it is less prone to accommodate two zinc centers for one ligand unit; X-Ray crystallography and NMR techniques identified the complex as dimeric in solid state and monomeric in solution.47 Lactide was polymerized in presence of the catalyst with an exceptional rate, in a controlled fashion even at low catalyst yield (1500 equivalent lactide). Investigations from Lin’s group were led on several zinc complexes containing NNO- tridentate Schiff base ligands. The first study features a chiral zinc complex bearing a mono methylether Salen-type ligand, [(SalenMe)Zn(OBn)]2 (Chart 1.7), which was found to be moderately active towards lactide ROP (100 eq. L-lactide polymerized in 4 hours at 60 °C in toluene) despite producing a polymer with an extremely narrow molecular-weight distribution (PDI = 1.03); polymerization of 50 equivalent rac-lactide over a period of 24 hours in dichloromethane at 25 °C produced isotactic polymer with a Pr of 0.75.67 A Schiff base zinc complex studied in 2006 by the same group was found to exist as a dimer in solid state and an 22 equilibrium between dimer and monomer in solution (based on variable temperature NMR analyses).68 The protonated version of this achiral complex (Chart 1.7, entry 10, X = 4-H) is a highly active catalytic species (100 equivalent L-lactide converted to polymer in 30 minutes at 25°C in dicholoromethane) but shows weak stereoselectivity (Pr = 0.59 for 200 equivalent rac-lactide polymerized to heterotactic PLA at 25 °C in dicholoromethane). On the other hand, polymerization of the zinc complex bearing tert-butyl substituents (Chart 1.7, entry 10, X = 4,6-di-tBu) at -55 °C in the same solvent allowed to reach great stereocontrol (Pr = 0.91).  1.6. Organozinc compounds Organozinc compounds are essential reagents for the synthesis of some of the most active zinc catalysts for LA ROP discovered to date.47, 68, 69  Especially, formation of the organozinc complex L1ZnEt is an important step in the formation of the highly active zinc alkoxide L1ZnOEt reported by Hillmyer, Tolman et al.47 Since this thesis has a large emphasis on zinc chemistry related to zinc alkyl compounds, the following section gives an overview of different types of organozinc compounds that have been involved in the formation of catalysts for LA ROP. An insight on their crystallography data and, in some cases, their reactivity is provided. In the great majority of known organozinc compounds, zinc exists under the ionic form Zn2+ with an electron configuration of 3d104s0. Because zinc is relatively electropositive compared to carbon (Pauling electronegativity of zinc is 1.6 versus 2.5 for carbon), the zinc- carbon bonds are covalent with a rather polar character.70 Organozinc compounds of the type R2Zn are linear; the zinc-carbon bonds are formed by the overlap of two equivalent sp-hybridized molecular orbitals (combination of 4s and 4p atomic orbitals). When only one coordinate bond is formed, the complexes are trigonal planar with a sp2-hybridized zinc center. They possess one unoccupied hybrid orbital, which makes 23 the compounds coordinatively unsaturated. As a result, these rare three-coordinate zinc complexes exist when there is sufficient steric crowding around the zinc metal to prevent coordination to a fourth ligand. In many cases, the last hybridorbital is involved in the bonding; it is responsible for the sp3 hybridization at the zinc center and the preponderant occurrence of tetrahedral geometry among the organozinc compounds. Higher coordination (5 to 6) is achieved through the involvement of 4d orbitals (sp3d and sp3d2 hybridization).71  Diorganozinc compounds are among the first organometallic compounds to have been synthesized. Already in 1852, Edward Frankland prepared diethylzinc, the first organozinc compound known, by heating ethyl iodide with zinc metal in a Carius tube.72 Diethylzinc was found to be a volatile colourless liquid that inflames spontaneously upon exposure to air. As a liquid, X-ray crystallography could not be the appropriate method to determine the molecular structure of the compound. Other techniques such as Gas-phase Electron Diffraction or solution studies (molecular weight measurements, microwave titrations, dipole moment determinations and complexation reactions) proved to be more adapted to the nature of these species.71 C-Zn bond distance was found to be 1.950 (2) Ǻ for ZnEt2, which compares to 1.930(2) Ǻ  for ZnMe2 and 1.952(2) Ǻ for Zn(nPr2).73 Heteroleptic monoorganozinc compounds are of the type RZnY, where R is an organic group σ-bonded via a carbon atom to the zinc center and Y is an electronegative monoanionic heteroatom such as halogen, oxygen (alkoxide, aryloxide, enolate and carboxylate groups), nitrogen (primary and secondary amines), sulphur, phosphorus, arsenic and selenium.Error! Bookmark not defined.71 The acceptor character of the zinc and the donor character of the electronegative group are enhanced by their mutual bonding, resulting in many cases in aggregates with multi-centre bonding of the electronegative substituent with various centers. 24 Among the three major pathways for the synthesis of heteroleptic monoorganozinc compounds (Scheme 1.4), protonolysis of a dialkyl zinc (usually diethyl zinc) is the common route encountered in the synthesis of zinc catalysts for LA ROP via formation of an alkyl zinc complex. It involves the proton-induced cleavage of one of the Zn-C bonds by a heteroatom or compound bearing an acidic proton (e.g. inorganic acid, alcohols, primary and secondary amines). However, protonolysis of the first alkyl group greatly diminishes the reactivity of the remaining zinc-carbon bond, meaning that removal of the second alkyl group may require forcing the reaction conditions.74  Scheme 1. 4 Major routes towards synthesis of heteroleptic monoorganozinc compounds.     Heteroleptic monoorganozinc compounds tend to undergo Schlenk equilibria, especially in solution (Scheme 1.3). The equilibrium position depends on the nature of the ligands covalently bonded to zinc, the nature and polarity of the solvent, and the additional donor ligands. Therefore, structural information about material crystallized from such mixture is not always representative of the active species present in solution. A number of organozinc compounds with different structures have been used as intermediates for the synthesis of catalysts used in lactone polymerization. These specific heteroleptic monoorganozinc compounds can be classified as monoorganozinc cations, compounds with an oxygen-zinc σ bond or compounds with a nitrogen-zinc σ-bond. Structural information on representative examples among these complexes will be reported here to illustrate the above-mentioned subclasses of heteroleptic monoorganozinc compounds. 25 [MeZn(DAD)][B(C6F5)4] (DAD = [(MeC=NC6H3Pri2-2,6)2]) is an example of monoorganozinc cation found active as an initiator for ROP of ε-caprolactone (Figure 1.1).75 The catalyst is obtained from selective protonolysis of one of the methyl groups in Me2Zn with [DADH]+[B(C6F5)]-. X-ray crystallography reveals a solid state structure where the isolated [MeZn(DAD)]+ cations and [B(C6F5)4]- anions are packed in the crystal lattice. The two nitrogen atoms of the diazadiene ligand have equivalent bond lengths with zinc (2.035(2) and 2.045(2) Ǻ) and the methyl-zinc bond is slightly shorter (1.914(4) Ǻ) than in the starting material Me2Zn (1.930(2) Ǻ). The three bonds form a perfectly planar trigonal geometry (angle sum 359.6°) around the zinc center; it is the first structurally characterised example of a cationic zinc centre with such a low coordination number.  Figure  1.1. [MeZn(DAD)][B(C6F5)4]  A large number of monoorganozinc compounds contain an oxygen atom covalently bound to zinc, in addition to the C-Zn covalent bond. The monoorganozinc compound L1ZnEt reported by Hillmyer and Tolman,47 which has a specific importance within the framework of this thesis, is one of them. In L1ZnEt, L1 is a tridendate NNO ancillary ligand covalently binding to zinc through the zinc-oxygen bond (Figure 1.2). Because of the nitrogen atoms intramolecularly coordinated to zinc, the compound exists as a discrete monomeric molecule in solid state and adopts a distorted tetrahedral geometry. The Zn-O σ-bond was found to be 1.956(2) Ǻ long, while the Zn-C bond length is 1.997(4) Ǻ; the latter is logically different 26 from the bond length in the diorganozinc starting material Et2Zn (1.950(2) Ǻ) since the zinc center goes from sp to sp3 hybridization state.   Figure 1.2. Representation of the molecular structure of L1ZnEt, a discrete monoorganozinc compound bearing a Zn-O σ-bond,  with selected nonhydrogen atoms labeled and hydrogen atoms omitted for clarity (Hillmyer, Tolman et al. 2003)47    The dimeric organozinc complex [L3ZnEt]2 (Figure 1.3) reported by Lin et al.68 is a starting material in the formation of the corresponding alkoxide, initiator for LA ROP. X-ray single-crystal structure analysis revealed the dimeric behaviour of [L3ZnEt]2 with pentacoordination around each of the zinc centers. The latter are bridged through the oxygen atom of the Schiff base ligands; the geometry at each zinc center is distorted square pyramidal. The zinc-carbon covalent bond formed between the metal center and the ethyl group was found to measure 1.985(4) Ǻ, which falls in the range of the Zn-C bond distances observed in most tetrahedral monoorganozinc compounds. It is interesting to note that although Lin’s dimeric complex [L3ZnEt]2 presents close similarities with Hillmyer and Tolman’s L1ZnEt (phenolate moeity and N,N-dimethylated diamine), L1ZnEt is found as a mononuclear compound in solid state. It should be mentioned that the chloride groups are not responsible for the difference in aggregation behaviour. Indeed, Lin et al. reported the study of similar zinc complexes featuring ligands changing only by different functional groups (H, Br, tBu) in 27 ortho and para position on the phenolate moiety; the complexes were all found to be dimeric in the solid state. 68  Figure 1.3. Dimeric organozinc complex [L3ZnEt]2.  Monoorganozinc compounds with a N-Zn covalent bond have less tendency to form aggregates than their counterparts containing an O-Zn covalent bond. Among the β-diiminate catalysts investigated by Coates et al. in 2001, the group reported the synthesis of an organozinc compound featuring an ethyl group bound to the zinc center (Figure 1.4, (BDI- 1)ZnEt). Its characterization in the solid state showed that the bidentate monoanionic diketoiminate ligand is chelate-bound to zinc; the two Zn-N bonds have similar length (~1.96 Ǻ) and the Zn-C bond length (1.96 Ǻ) compares to distances measured in other related trigonal planar organozinc complexes. 76  The complex exhibits trigonal planar geometry around zinc, which translates by a flat six-membered ring formed by the metal and the ancillary ligand (Zn-N-C-C-C-N).   Figure 1.4. (BDI-1)ZnEt complex.  28 Another important N σ-bonded monoorganozinc compound is the tris(pyrazolyl)hydroborato-based system, which numerous derivatives, differing by the nature of their substituents, have been intensively studied by X-ray crystallography.42b, 77 Chisholm et al. reported the synthesis of a methyl zinc compound of this type (Figure 1.5, [TptBu]ZnMe), together with the activity of a tris(3-tert-butylpyrazolyl)hydroborato zinc alkoxide catalyst for LA ROP ([TptBu]ZnOSiMe3) 42b. The methyl zinc complex [TptBu]ZnMe is monomeric with a distorted tetrahedral coordination at zinc, and shows a C3 symmetric structure. The three nitrogen atoms bind to the metal center in an identical fashion (Zn-N distances ~2.09 Ǻ), and the Zn-C distance (1.965(5) Ǻ) compares with values previously reported by Parkin et al.78 Though, this complex was not tested towards alcoholysis, the tris(pyrazolyl)hydroborato- based ethyl zinc complex [TptBu]ZnEt werewere found to be kinetically inert to alcoholysis with tert-butanol and ethanol.87d  Figure 1.5. Tris(3-tert-butylpyrazolyl)hydroborato methyl zinc, [TptBu]ZnMe.  The examples mentioned above show that very few catalysts capable to polymerize lactide with high activity and selectivity have been discovered so far, most of which rely on chain-end control mechanism. We have attempted to develop a catalyst demonstrating high activity towards lactide ROP with stereocontrol induced via enantiomorphic site-control; the ligand framework was based on a chiral version of Hillmyer and Tolman’s catalyst L1ZnOEt,47 as the highly active catalyst showed no chain-end control in lactide ROP. 29 CHAPTER  2 - SYNTHESIS AND REACTIVITY OF SOME NEW ZINC COMPLEXES  Developing enantioselective catalysts is a major focus of research on lactide polymerization. Indeed, stereocontrol influences the tacticity and thus the physical properties of the polymer. As was discussed in the previous chapter, there have been a few active, stereoselective zinc catalysts reported in the past 10 years. Coates’s achiral β-diiminate zinc complexes are examples of such catalysts.44, 45 They generate highly heterotactic PLA via a chain-end control mechanism. However, chiral stereoselective zinc catalysts have attracted much less attention, with the exception of the catalysts reported by Chen et al.79 and Chisholm et al.63 The highly active achiral zinc catalyst developed by Hillmyer, Tolman et al. 47 inspired us to explore the efficacy of a chiral version of this complex, in term of enantioselective lactide polymerization via enantiomorphic-site control. The achiral catalyst produced strictly atactic PLA, indicating minimal chain-end control during the polymerization. Thus by developing a chiral version of the catalyst we hoped to take advantage of an enantiomorphic site control mechanism for lactide polymerization.  2.1. L1ZnOEt, a highly active zinc catalyst for ROP of lactide In 2003, Hillmyer et al. reported the synthesis of the zinc complex L1ZnOEt, with an achiral diaminophenolate ligand 2,4-di-tert-butyl-6-{[(2’-dimethylaminoethyl)methylamino]- methyl}phenolate. The complex was synthesized in a two-step process: reaction of the proligand L1H with diethyl zinc to form the ethyl zinc complex L1ZnEt, followed by an alcoholysis reaction with ethanol to yield the desired compound, L1ZnOEt (Scheme 2.1).   30 Scheme 2.1. Synthesis of Hillmyer’s zinc ethoxide complexes.47   Although the complex was dimeric in solid state (Scheme 2.2), crossover experiments with a related zinc complex ((L2ZnOEt)2, where R = Me), NMR pulse gradient spin-echo experiments (PGSE), and mass spectrometry identified a predominantly monomeric species in solution.  Scheme 2.2. Equilibrium between dimeric (predominant in the solid state) and monomeric species (predominant in solution).   While the zinc ethyl L1ZnEt was unsuccessful in polymerizing rac-lactide, polymerization with L1ZnOEt revealed the zinc ethoxide complex to be one of the most active 31 catalysts for lactide ROP at the time. High conversion was observed even with low catalysts loadings (up to a ratio [LA]/[Zn] = 1500), producing PLA with molecular weights up to 130 kg.mol-1 (Table 2.1).  Table 2.1. Results for the polymerization of rac-lactide by L1ZnOEt.  [LA]o/[L1ZnOEt] time (min) conversion (%) b Mn (kg/mol)c PDIc 650 5 96 67 1.42 1000 13 96 99 1.4 1500 18 93 130 1.34 a) Conditions: CH2Cl2,298 K, [LA]o = 1M, in a glovebox. b) Determined by 1H NMR spectroscopy. c) Determined by SEC using light scattering detection.  As lactide was polymerized with high catalyst loading ([LA]0/[L1ZnOEt]0  = 50) and quenched after 10 s (95% conversion), 1H NMR spectroscopy allowed identification of both ethoxy ester and hydroxy end groups on the polymer chain; this confirms that initiation of polymerization occurs by acyl cleavage, presumably through a coordination-insertion pathway. Kinetic experiments revealed an overall second-order rate law, first-order behaviour in both [LA] and [Zn], consistent with the coordination-insertion pathway. The microstructure of PLA obtained from rac-lactide was found to be atactic and polymerization of L-lactide yielded isotactic PLA without epimerization observed.  32 2.2. Compounds synthesis 2.2.1. Synthesis of the ligands In order to develop chiral analogues of Hillmyer’s system, L1ZnOEt, we synthesized a family of chiral tridentate ligands with trans-cyclohexyldiamine as a chiral backbone (Figure 2.1). (NNHOtBu)H (1), (NNMeOtBu)H (2) and (NNMeOH)H (3) have a combination of amine and phenol donors (NNO) connected by this chiral linker. The steric bulk of the ancillary ligand was tuned by controlling the substituents on the aryl ring as well as on the amine groups.  Figure 2.1. Chiral tridentate ancillary ligands (NNHOtBu)H (1), (NNMeOtBu)H (2) and (NNMeOH)H (3). The synthesis of the ligands in their racemic and enantiomerically pure versions was based on a preparation reported by Finney 80  that was improved and further developed in our laboratory81. Synthesis of the racemic ancillary ligand requires a seven-step synthesis due to the presence of the asymmetric diamine in the ligand backbone. The ligand synthesis is complicated by the fact that simple addition of salicylaldehyde to (±)-trans-1,2- diaminocyclohexane ((±)-DACH) results in formation of a mixture of the diamine, monoamine and di-imine (Scheme 2.2).84b, 84c   33 Scheme 2.3. Equilibrium observed when reacting trans-1,2-diaminocyclohexane with a salicylaldehyde (R = tBu, H.).    The synthesis of chiral unsymmetrical salens is a problem that has been known for a long time and a few strategies have been developed to circumvent the difficulties related to unwanted umsequilibriums. They have included chromatographic separation 82, the use of polymeric reagents82a, 83  and trapping of the monomeric species 84 a, 84b. HHowever,these methods target the synthesis of enantiomerically pure salen ligands; they are not suitable for the synthesis of our racemic ligands. To the best of our knowledge, the most consistent method to synthesize the racemic mono-Schiff base is to use a protecting group on (±)-DACH to prevent formation of the diimine before condensation with the salicylaldehyde80 (Scheme 2.4).         34 Scheme 2.4. Synthesis of pro-ligands 1a, 2a and 2b.    First, reaction of (±)-DACH with ethylacetimidate hydrochloride (Pinner salt) yields the corresponding imidazole ring. Hydrolysis of the heterocyclic diamine then allows protection of the amine group by monoacetylation. Subsequently, a dimethylated derivative is obtained by reductive amination with sodium cyanoborohydride and formaldehyde. Removal of the acetyl group is easily achieved by reflux under acidic conditions to yield the asymmetrically methylated DACH, which is converted to a Schiff base by condensation with 3, 5-di-tert-butoxysalicylaldehyde. Reduction of the imine with sodium borohydride produces the diaminophenol 1a. 2a is synthesized via a second reductive amination to methylate the 35 secondary amine in 10% overall yield. Compound 2b is synthesized in an analogous manner from 1b in 77 % overall yield.  Scheme 2.5. Resolution of (±)-trans- diaminocyclohexane    (±)-2a was synthesized from (±)-DACH as starting material, while its enantiopure analogue, (R,R)-2a, was prepared from (R,R)-DACH following a similar procedure. Starting material (R,R)-DACH was isolated by resolution of (±)-DACH using L-(+)-tartaric acid to form (R,R)-1,2-diammoniumcyclohexane mono-(+)-tartrate (Scheme 2.4) according to a known procedure.85 The diastereoisomeric salt is stable for several months, in contrast to the moderately air sensitive free amine.86 The salt was then directly reacted with the Pinner salt to produce the enantiopure imidazole compound; the procedure is eventually carried out further following an procedure identical to the one used to synthesize (±)-2a. All pro-ligands (1a, 2a, (R,R)-2a and (±)-2b) were recrystallized from acetonitrile and isolated as colorless crystals. Compounds (R,R)-2a, (±)-2a, 2b, 1b and its corresponding imine (NN=OH)H  have not been reported in the literature.  2.2.2. Zinc complexes (NNMeOtBu)ZnEt, 3a Reactions of (±)-2a and (R,R)-2a with one equivalent of diethyl zinc yield the ethyl zinc complexes (NNMeOtBu)ZnEt, (±)-3a and (R,R)-3a, respectively (298 K, 10 minutes). 36 Scheme 2.6. Synthesis of 3a.    The molecular structures of these complexes were determined by single crystal X-ray crystallography and showed in all cases mononuclear zinc complexes with a pseudo- tetrahedral geometry. The isolated structure for (±)-3a shows isomerism in the solid state; two mononuclear zinc complexes with (R,R) configuration at the cyclohexyl diamine are observed. Disorder observed in the structure shows the presence of the two other (S,S) stereoisomers. The two (R,R) enantiomers reflect the chirality at the zinc and central nitrogen atoms: (ZnR, NR) and (ZnS, NS). From the X-ray crystallography data obtained, the molecular structure of (R,R)-3a reveals only one compound in the crystal lattice (Figure 2.2). The differences between the angles C-Zn-N1 (126.22° in average) and N1-Zn-N2 (82.47° in average) for the two enantiomers of (±)-3a and the complex (R,R)-3a are illustrative of the distortion from an ideal tetrahedral typology (Table 2.1). The Zn-C bond lengths are very close between the 3 complexes (1.97-1.98 Ǻ); the distances are representative of Zn(II) alkyl bonds (1.85-2.01 Ǻ)68, 87 and slightly smaller than the Zn-C bond in Hillmyer and Tolman’s complex L1ZnEt (1.997 Ǻ).47 As demonstrated by Lin et al. in related complexes, modification of substituents of the ligand backbone can be responsible for such small difference in the Zn-C bond.68 37  Figure 2.2. Molecular structure of two enantiomers of (±)-3a with ellipsoids at 50% probability level.  Most hydrogen atoms omitted for clarity.   Figure 2.3. Molecular structure of two enantiomers of (R,R)-3a with ellipsoids at 50% probability level. Most hydrogen atoms omitted for clarity.   38 Table 2.2. Selected interatomic distances (Ǻ) and bond angles for zinc ethyl complexes.       (±)-3a (ZnS, NS) (±)-3a (ZnR, NR) (R,R)-3a L1ZnEt Bond lengths Zn-C1 1.978 (3) 1.983 (3) 1.971 (6) 1.997 (4) Zn-O1 2.118 (2) 2.111 (2) 1.927 (3) 1.956 (2) Zn-N1 2.190 (2) 2.194 (2) 2.152 (4) 2.147 (3) Zn-N2 1.934 (16) 1.937 (2) 2.185 (5) 2.128 (3)  Bond angles N1-Zn-N2 82.45(8) 81.76(8) 82.19 (15) 84.85 (13) N1-Zn-O1 97.71 (7) 95.75(7) 95.54 (14) 93.32 (12) N2-Zn-O1 105.56 (8) 108.36(8) 111.61 (16) 99.76 (12) C1-Zn-N1 130.04(11) 124.11 (8) 124.5 (2) 117.82 (16)  Although we observe only two isomers in the solid state structure of (±)-3a, it is important to note that a maximum of 2n = 4 isomers is possible, where n is the number of newly created chiral centers.88 Although (R,R)-3a contains a total of 4 chiral centers, only 2 of them, Zn and N, can have stereoisomers when the compound is formed.  X-ray crystallography data of the enantiopure (R,R)-3a revealed the presence of one compound only. NMR spectroscopy of the crystals of (±)-3a and (R,R)-3a in solution clearly shows the existence of another compound.  The presence of a secondary compound is neither observed in NMR spectroscopy of Hillmyer’s work,47 nor from data obtained from the same experiment repeated in our laboratory. 1H NMR data (C6D6, 298 K) reveal a characteristic set of peaks for the tridentate ligand and the coordinated ethyl group. While the ligand tert-butyl groups (1.84 and 1.45 ppm ) and the aromatic protons (7.59 and 6.85 ppm) show little difference in chemical shift compared to the free ancillary ligand, the methylene protons now show two 39 sharp, easily identifiable doublets (3.88 and 2.96 ppm) and the N-methyl peaks are also inequivalent and noticeably shifted (1.86, 1.85 and 1.56). The coordinated methyl group can be clearly identified as a triplet at 1.76 ppm. The methylene group Zn-CH2CH3 was distinguished from the ligand backbone using 1H NMR Correlation Spectroscopy (COSY, C6D6, 25 °C): on Figure 2.4, through-bond scalar coupling is observed between the methyl group (triplet) and the methylene group (multiplet).   Figure 2.4. Identification of the peaks corresponding to the methylene and methyl groups in Zn-CH2CH3 using 1H COSY NMR (C6D6, 25°C, 400 MHz)   Since these two complexes show identical 1H and 13C {1H} NMR spectra, we can rule out different isomers originating from the stereocenters at the cyclohexyl groups. Also, since 40 (±)-3a is obtained in nearly quantitative amount, we can affirm that (S,S)-3a is part of the racemic product mixture, although it is not visible by 1H or 13C {1H} NMR spectroscopy; this is in agreement with the X-ray data reported above.  (NNMeOH)ZnEt, 3b Reaction of (±)-2b, a less-sterically hindered version of (±)-2a, with diethyl zinc in pentane for 10 minutes at room temperature forms a white precipitate of racemic zinc complex (NNMeOH)ZnEt, (±)-3b. NMR spectroscopy (C6D6, 298 K) allows to distinguish a major and a minor compound in a 9 to 1 ratio. The set of peaks for the major compound unmistakeably features the methyl protons (three peaks at 1.83, 1.80 and 1.60 ppm) as well as the methylene protons (two doublets at 3.92 and 2.89 ppm). The aromatic region (4 protons) and the absence of tert-butyl groups in the 1H NMR spectrum are characteristic of the ancillary ligand 2b.  Scheme 2.7.Synthesis of 3b.   X-ray crystallography data for (±)-3b reveals two compounds of pseudo-tetrahedral geometry (Figure 2.5) with (R,R) configuration at the cyclohexyl carbons. Interestingly, the chirality at zinc and at the central amine of each stereoisomer is inverted relative to each other ((ZnR, NS) and (ZnS, NR)), in contrast to the chirality of the diastereoisomers in 3a (ZnR, NR) and (ZnS, NS)). As for (±)-3b, the compound is likely to exhibit 4 stereoisomers, from which 41 two have the (S,S) configuration at the cyclohexyl carbons. The bond angles and bond lengths of the two stereoisomers are almost identical between each other and in close range with the data found for the two stereoisomers of (±)-3a; the ethyl-zinc bond distances in (±)-3b are 1.95 and 1.98 Ǻ (Table 2.3).   Figure 2.5. Molecular structure of (±)-3b with ellipsoids at 50% probability level. Most hydrogen atoms omitted for clarity.   Table 2.3. Selected interatomic distances (Ǻ) and bond angles (°) for compound 3b   (±)-3b (ZnR, NS) (±)-3b (ZnS, NR) Bond lengths Zn-C1 1.980 (2) 1.950 (1) Zn-O1 1.950 (1) 1.930 (1) Zn-N1 2.121 (2) 2.127 (2) Zn-N2 2.145 (2) 2.161 (2)  Bond angles N1-Zn-N2 83.54 (6) 83.17 (6) N1-Zn-O1 96.29 (5) 96.06 (6) N2-Zn-O1 103.42 (6) 106.65 (6) C1-Zn-N1 119.16 (7) 124.11 (8)  42 In 3a and 3b, the phenol moiety of the tridentate ligand forms a claw-type shape with the cyclohexane ring, in such a way that the ethyl group is pointing away from the phenol ring (Figure 2.6). In L1ZnEt, the ethyl group and the phenol ring are closer to each other, resulting in less steric hindrance on the face opposed to the ethyl group; ; the claw-type shape formed by the diamine and the phenol ring seems to be “open”. This difference is more visible through the comparison of the angles N2-Zn-O1 and the C1-Zn-N1, which are significantly larger for 3a than for L1ZnEt; the difference is less pronounced for the less sterically hindered 3b. 43  Figure 2.6.  Structural comparison of: a) (R,R)-3a, b) (±)-3b and c) L1ZnEt.  (NNHOtBu)ZnEt, 3c Formation of (NNHOtBu)ZnEt, 3c, resulted from the reaction of (±)-1a with diethyl zinc in pentane (stirring at room temperature for 1 hour). 1H NMR spectroscopy (Figure 2.7) of the product shows the characteristic peaks corresponding to the ancillary ligand 1a with the 44 methylene doublets (3.97 and 3.15 ppm), the tert-butyl groups (1.89 and 1.47 ppm) and only two methyl groups belonging to the external dimethylated amine (1.88 and 1.52 ppm). In contrast to the 1H NMR spectra of 3a and 3b, no secondary set of peak with an identical pattern is observed, which suggests that a “β” compound is not present in 3c in solution.  Scheme 2.8. Synthesis of 3c.      Figure 2.7. 1H NMR spectrum of (±)-3c (C6D6, 298 K, 400 MHz).  45 Formation of a secondary isomer in solution for 3a As previously mentioned, 1H NMR spectra of (R,R)-3a and (±)-3a show a second set of peaks accounting for 10-15% of material with slightly different chemical shifts, in many points comparable to the main set of peak (Figure 2.8). Peaks corresponding to both aromatic protons (doublets at 7.59 and 6.96 ppm), as well as methylene (doublets at 3.48 and 2.90 ppm), methyl (singlets at 2.13, 2.08 and 1.67 ppm) and t-butyl (singlets at 1.90 and 1.47 ppm) protons can be distinguished by 1H NMR spectroscopy.   Figure 2.8. 1H NMR spectrum of (R,R)-3a (C6D6, 25°C, 400 MHz).   The major, 3a-α, and minor, 3a-β, products exist in a 85:15 ratio. Since 3a-β is observed for both (±)-3a and (R,R)-3a, it can be concluded that the minor compound observed in (±)-3a is not the enantiomer (S,S)-3a. Attempts to identify 3a-β were thus carried out with the racemic mixture (±)-3a. 46 In order to determine the identity of the secondary product, we studied the aggregation behaviour of our complexes in liquid state. Reaction of diethyl zinc with 2 equivalents of the proligand (±)-2a results in a product mixture containing a ratio close to (1:1) of (±)-2a and (±)-3a-α, with a minimal amount of (±)-3a-β. A homoleptic compound complex was not observed. Also, it has been shown that some zinc complex aggregates dissociate in the presence of coordinating solvents such as pyridine, lutidine or pyrrolidine89,90,91 and to some extent polar solvents such as DMF, chloroform or THF92.  However, the 1H NMR spectrum of (±)-3a in presence of two equivalents of pyridine for 14 hours did not show any significant difference in chemical shift for 3a-α compared to the spectrum of (±)-3a alone. Isolation of  3a-β was attempted through several means. Changing the reaction solvent to pentane, benzene, toluene or THF, as well as increasing the reaction time from 1 to 5 hours did not significantly modify the ratio of α to β. The reaction was also carried out at -90 °C for 10 minutes, and from -20 to +80 °C over 2.5 hours in d8-toluene during variable temperature studies without success to push the reaction towards further formation (at high temperature) or possible dissociation (at cold temperature) of 3a-β. On the other hand, heating (±)-3a to 80°C in d-benzene over a period of 72 hours resulted in a significant increase of 3a-β (50% of the product mixture, Figure 2.8). Similar results were later obtained by dissolving 3a in pyridine with stirring for 24 hours at room temperature. However, it can be noticed that the amount of 3a-β could not be increased to proportions greater than 50% of the product mixture. The presence of 3a-β was also observed in solution after analytically pure crystals of 3a had been dissolved in pentane. These observations lead to the hypothesis of a slow equilibrium existing in solution between the two compounds 3a-α and 3a-β.  47  Figure 2.9. 1H NMR (C6D6, 400 MHz) of (±)-3a after 10 min reaction at 25 °C (top) and heating at 80 °C in C6D6 for 72 h (bottom).  Though X-ray crystallography allowed identification of a second stereoisomer in solid state via the visualisation of its molecular structure, standard 1H and 13C{1H} NMR spectroscopy alone could not prove that the minor compound observed in solution was the same second compound observed in solid state. A series of NMR spectroscopy using different techniques (1D/2D NOE, short/long-range 2D COSY) was performed to identify 3a-β in solution. The data showed that all peaks that belong to 3a-α have an equivalent in the set of peaks that belongs to 3a-β. As it can be seen in Figure 2.8 for the aromatic and methylene protons, most of 3a-β peaks have chemical shifts slightly different from 3a-α peaks. In order to understand the reason for this difference in chemical shift, we decided to look more closely at the exact position of the cyclohexyl proton on carbon C3, relative to the protons of the N- methyl group C9 (Figure 2.10), using 1D selective NOE NMR spectroscopy. 48  Figure 2.10. Number assignment of selected atoms of 3a-α and 3a-β based on X-ray crystallography data.  The position of the two N-methyl groups at C9 and C10 were first identified in the NMR spectrum of a solution of 3a containing a minimal amount of 3a-β. By irradiating the proton at C9, a NOE effect was observed on the α methyl group C10 (Figure 2.11). In the same way, identification of the two β N-methyl groups at C39 and C40 was possible by irradiation of C40 in a solution of 3a containing 50% of 3a-β (prepared by reacting 3a for 48 hours at room temperature in pyridine); the NOE effect was visible at the methyl group C39 (singlet at 1.78, Figure 2.11). 49  Figure 2.11. 1D selective NOE experiments (C6D6, 298 K, 400 MHz) for a solution of 3a containing minimal amount of 3a-β (top) and a solution of 3a containing 50% of 3a-β (bottom), where the protons at C9 and C40 have been irradiated, respectively. Each identified peak has been numbered according to the carbon to which the absorbing 1H nucleus belongs.   Once the peaks corresponding to the methyl groups of the dimethylated amine for 3a- α and 3a-β had been identified, irradiation of cyclohexyl protons on C3 and C33 showed NOE effects with the opposite methyl groups on C9 and C40 respectively (Figure 2.11). Such through-space interaction is possible because the proton at C3 is in cis position relative to the methyl protons at C30 in 3a-α, while the cyclohexyl proton at C33 is in trans position relative to the methyl proton at C60 in 3a-β (Figure 2.11).  50  Figure 2.12 1D selective NOE experiments (C6D6, 298 K, 400 MHz) for a solution of 3a containing 50% of 3a-β, where the cyclohexyl protons at C3 (top) and C33 (bottom) have been irradiated.      51  Figure 2.13. 1D selective NOE experiment (C6D6, 400 MHz) of a 50:50 mixture of 3a-α and 3a-β revealing through-space interaction between C3 and C9 protons for 3a-α, and C33 and C40 protons for 3a-β.  In 3c, no secondary compound is observed in solution. Therefore, there may be a correlation between the presence of a methyl group on the central amine and the observation of a secondary compound in solution. Evidences obtained from NOE experiments also show that the methyl group has a different orientation relative to the cyclohexyl proton in C3 (3a-α) and C33 (3a-α) position, which is responsible for a difference in chemical shift between the two stereoisomers.  Although the secondary isomer is observed for all complexes bearing the (NNMeOR) ancillary ligands it does not affect reactivity.  52 2.3. Reaction of zinc ethyl complexes with alcohols 2.3.1. Formation of alkoxides via alcoholysis  Substitution of an alkyl ligand by an alkoxy group via alcoholysis has been used as a way to synthesize a number of zinc alkoxides, 62, 65, 65, 93, 94, 95, 96 including complexes very similar to ours.47, 67-69 Therefore, nucleophilic attack of alcohols on the zinc ethyl complexes 3a, 3b and 3c was expected to occur readily. However, reaction of methanol, ethanol and isopropanol with 3a and 3b over 24 hours at room temperature in toluene failed to give the corresponding zinc alkoxide products. Reactions carried out over one week between 3a and two equivalents of methanol, ethanol, and isopropanol in toluene at room temperature resulted in formation of free ancillary ligand, decomposition, and no reaction, respectively. Heating the reaction of 3a with isopropanol at 60 °C in d-benzene for 30 days resulted in formation of an unidentifiable mixture of products. Reaction of 3c with ethanol for 24 hours was equally unsuccessful. Though this non-reactivity was unexpected, such behaviour has been already reported in literature. 42b, 97, 98 The reaction of alcoholysis of (BDI)ZnEt was even once qualified as “unpredictable” by Coates, as he was observing different behaviour depending on the substituents on the complex’ ancillary ligand or the nature of the alkoxy group.97 One common alternative route to yield metal alkoxides is the synthesis of a zinc amide complex before reacting it with the desired alcohol.44, 45, 63, 99 We decided however first to investigate the reasons for the unreactivity of the complexes 3a, 3b and 3c towards alcoholysis.     53 Scheme 2.9. Synthesis of (±)-5a.   As we suspected that the pKa of the initiator species could have some influence on their reactivity, we decided to carry out the reaction of 3a with a more acidic species. Reaction of (±)-3a with HCl.Et2O for 1 hour at room temperature in pentane yielded a white precipitate of [NNMeOtbu]ZnCl , (±)-5a, in 95% yield (Scheme 2.9). The precipitate was filtered and washed before recrystallization from acetonitrile. X-ray crystallography analysis of colourless crystals revealed a pseudo-tetrahedral geometry around the zinc center, without dimerization or aggregation observed in the solid state (Figure 2.14). The angles O26-Zn-Cl (120.40(6) °) and N7-Zn-N8 (88.36(8) °) reflect the distorted tetrahedral geometry of the complex. The Zn-Cl bond distance is 2.2048(7) Ǻ, which is slightly smaller than the usual length encountered for Zn-Cl bonds in related coordination complexes (2.21-2.34 Ǻ).66,100,101 54  Figure 2.14. Molecular structure of (±)-5a with ellipsoids at 50% probability level.  Most hydrogen atoms omitted for clarity.    Table 2.4 Selected interatomic distances (Ǻ) and bond angles (°) for compound (±)-5a  Bond lengths   Bond angles Zn-N7 2.064(2)  O26-Zn-Cl 120.40(6) Zn-N8 2.067(2)  O26-Zn-N7 99.99(8) Zn-O26 1.8987(18)  Cl-Zn-N7   117.78(6) Zn-Cl 2.2048(7)  O26-Zn-N8 112.57(8)    Cl-Zn-N8 112.79(7)    N7-Zn-N8 88.36(8)  The 1H NMR spectrum of (±)-5a reveals the familiar pattern of our tridentate ligand, among which the two aromatic protons can be clearly distinguished at 7.61 and 6.77 ppm (singlets), the two methylene protons at 3.89 and 2.91 ppm (doublets), the two tert-butyl groups at 1.82 and 1.41 ppm (singlets) and the three peaks corresponding to the N-methyl groups at 2.01, 1.97 and 1.45 ppm (Figure 2.15). On the other hand, the absence of the characteristic triplet at 1.76 ppm for ZnCH2CH3 in 3a confirms the reaction of all the starting 55 material. Not surprisingly, a secondary compound is observed in minor amount in solution, following a pattern of peaks similar to the one that belongs to the major compound; based on the study of 3a-β previously mentioned, we can deduce that this minor species is likely to be a stereoisomer of  (±)-5a.   Figure 2.15. 1H NMR spectrum of (±)-5a (C6D6, 298 K, 300 MHz)  This experiment supports our hypothesis on the influence of pKa on the reactivity of protonated species towards 3a via protonolysis. Although the reactivity of alcohols towards 3a and its derived complexes was first unexpected, there have been a few examples of zinc alkyls exhibiting such behaviour reported in literature in recent years.42b, 64, 66 L1ZnEt, 3a, 3b and 3c do not significantly vary in term of electronic system so the electronic factor cannot be accounted for the major change in reactivity between L1ZnEt and our complexes. An explanation based on sterics may be that the cyclohexyl moeity combined with the bulk of the tert-butyl groups increases the steric hindrance around the zinc center; however, it was seen that the less bulky 3b (in which tert-butyl groups are absent) does not show different 56 reactivity compared to 3a when it reacted with alcohols in similar conditions. Comparison of the pKa of the different alcohols tested with 3a and HCl reveals a great difference in acidity; methanol, ethanol and isopropanol have pKa’s of 29.0, 29.8 and 30.3, respectively, while HCl has a pKa of 1.8 only (in DMSO102). In order to form the alkoxide complexes, a more acidic alcohol, such as phenol (pKa = 18.0), could therefore be a better candidate for the alcoholysis of 3a.  Reaction of one equivalent of phenol with (±)-3a in stoichiometric amounts at room temperature for 9 hours (Scheme 2.5) yielded (±)-4a in 96% yield. The enantiopure complex, (R,R)-4a, was generated with (R,R)-3a.  4a is one of the rare examples of well-defined polydentate zinc phenoxides that have been reported to date.103, 93   Scheme 2.10. Synthesis of (±)-4a.    1H NMR spectroscopy allows to recognize the distinctive three peaks for the N-methyl groups at 2.02, 1.94, and 1.43 ppm, while the phenoxy peaks are further downfield at 7.62, 7.39 and 6.85 ppm. The tert-butyl groups show two singlets at 1.85 and 1.42 ppm, and the methylene protons appear at 3.87 and 2.89 ppm (doublets). Once more, presence of a secondary compound is observed in solution. X-ray crystallography data confirm the molecular structure of (±)-4a and (R,R)- 4a, which exhibit a pseudo-tetrahedral geometry at the zinc center. As the molecular structure was obtained from reaction of (±)-3a with a slight 57 excess of phenol, a phenol adduct of (±)-4a, (±)-4a•phenol, was observed; the phenoxy ligand is in close contact with the second phenol proton (Figure 2.16).  The O(28)-H(36) and O(28)- O(36) bond lengths are 1.9 and 2.67 Å respectively.  The zinc-alkoxy distance, Zn-O28, is 1.920(2) Ǻ, which compares to other values reported in literature for zinc alkoxide ligands (1.85-1.98 Ǻ).104 The disparity between the angles around the zinc center (O26-Zn-O28: 119.04(10) °, N7-Zn-N8: 86.84(11) °) reflects the distortion in the tetrahedral geometry.   . Figure 2.16. Molecular structure of (±)-4a•phenol with ellipsoids at 50% probability level. Most hydrogen atoms removed for clarity           58 Table 2.5.Selected interatomic distances (Ǻ) and bond angles (°) for compound (±)-4a  Bond lengths   Bond angles Zn-N7 2.094(3)  O26-Zn-O28 119.04(10) Zn-N8 2.083(3)  O26-Zn-N7 98.56(11) Zn-O26 1.925(2)  O28-Zn-N7   120.60(11) Zn-O28 1.920(2)  O26-Zn-N8 120.35(11)    O28-Zn-N8 107.52(10)    N7-Zn-N8 86.84(11)  It was also found that 3a is totally inert towards water (under nitrogen atmosphere). This interesting property is easily explained by the relatively low acidity of water in organic solvent (pKa = 31.4 in DMSO).   The successful reaction of 3a with phenol validates our hypothesis: protonolysis of our complex occurs in presence of species that are sufficiently acidic. Since steric and electronic effects are not responsible for the discrepancy of reactivity between L1ZnOEt and 3a, further investigation was necessary to determinate the role played by the cyclohexyl moiety. Addition of the bulky, asymmetrical, cyclohexyl substituent on the ethylene diamine arm of Hillmyer’s complex not only created chirality in the ligand, but also introduced some rigidity in the ligand backbone. If the ethylene diamine arm in L1ZnOEt was labile enough to allow alcohol insertion on the metal center, the rigidity of 3a’s ligand backbone could greatly diminish the reactivity of 3a towards alcohol coordination. In order to study the degree of lability of the diamine arm in L1ZnOEt compared to 3a, we observed the behaviour of both complexes in presence of a relatively strong coordinating solvent, pyridine.  2.3.2. Mechanistic approach - Coordination to pyridine Reaction of L1ZnEt was carried out with 2 equivalents of pyridine overnight at room temperature. In order to remove excess free pyridine from the product mixture, the solution was dried in vacuo at room temperature for 24 hours; the isolated product was found to 59 contain only the zinc ethyl complex. From there, two conclusions could be drawn: either pyridine is not basic enough to coordinate to the zinc center, or the pyridine adduct of the complex is not stable under our experimental conditions. Parallel reaction with 3a instead of L1ZnEt led to identical results. We thus decided to run a second set of reactions with a different procedure in the removal of excess pyridine so as to discard the second hypothesis. Reaction of pyridine (2 equivalent) with L1ZnEt was this time run in situ for 60 minutes at room temperature, in CD2Cl2. In addition to the peaks belonging to the pyridine protons (doublet at 8.59 ppm, triplet at 7.68 and multiplet at 7.29 ppm), the 1H NMR spectrum of the reaction mixture clearly showed a broad signal at 2.24 ppm instead of the two singlets (2.38 and 2.09 ppm) that can be seen in the 1H NMR of L1ZnEt before pyridine addition (see Figure 2.17). In a 1D NOE NMR experiment run on L1ZnEt (CD2Cl2, 298 K), it was possible to see that a proton from the N-methyl group irradiated at 2.21 ppm resonates with another proton (peak at 2.40 ppm), from the methyl group of the same amine.  The peak for the non-irradiated methyl group is directed towards the bottom, possibly because of a proton exchange between the two N-methyl groups. These two singlets therefore correspond to the methyl protons of the terminal dimethylated amine (Figure 2.18). The peak for the central N-methyl does not change, which means that the coordination of the central amine is unaffected by the presence of pyridine, after 60 minutes reaction.  60   Figure 2.17.  Addition of 2 equiv of pyridine to LZnEt (CD2Cl2, 298 K) at time = 0 min (bottom) and time = 60 min (top).  Figure 2.18. 1D NOE experiment of L1ZnEt (CD2Cl2, 298 K, 400 MHz). The irradiated methyl proton at 2.12 ppm shows through-space coupling with the methyl protons at 2.37 ppm. 61 In situ reaction of 3a with 2 equivalents of pyridine at room temperature in CD2Cl2 was monitored by 1H NMR spectroscopy; the only difference observed after 60 minutes was a slight increase in amount of 3a-β. As the reaction time was increased to 14 hours, no more difference than an equilibration of the amount of 3a-α and 3a-β was observed. This experiment confirms that pyridine can displace the terminal amine in L1ZnEt, while the same in not possible with 3a; the unability of pyridine to coordinate to the zinc center of 3a can be explained by the rigidity conferred by the cyclohexyl moiety to the ligand backbone. We propose that the lability of the ethylene diamine arm allows coordination of alcohols on the zinc center in L1ZnEt through a dissociative mechanism (Scheme 2.11). We can envisage two mechanisms at work. First, species that readily dissociate (e.g. HCl), or are stable enough in their deprotonated form in organic solvent (e.g. phenol), can release protons in the reaction mixture. As the nucleophilic ethyl group binds to the available proton, the reaction is driven by the release of ethane gas from the product mixture (driving force). The deprotonated species can then take the available ligand position, previously occupied by the ethyl group, to coordinate to 3a. The second mechanism (Scheme 2.11) takes place when the dissociation of the dimethylated amine from the zinc center allows coordination of the protonated species to the zinc center.  Since the proton of the coordinated alcohol has become much more acidic, direct protonation of the ethyl group is possible. Eventually, ethane is released from the reaction mixture and the dimethylated amine is reattached to the metal center.      62 Scheme 2.11. Proposed dissociation mechanism for the reaction of L1ZnEt and alcohols/ pyridine. .  2.4. Polymerization  In the frameworks of this thesis, polymerization of lactide has been studied by NMR spectroscopy, one of the most useful techniques with IR and UV spectroscopy. Interpretation of spectra using different NMR techniques allowed to monitor the polymerization and determinate the tacticity of the polymer.  Monitoring the reaction is usually obtained by observing the formation of the product or the disappearance of the reagents. In this work, evaluation of the conversion rate was achieved by measuring the relative integration of the methine proton of lactide as it is converted to PLA, by 1H NMR spectroscopy. NMR spectra were obtained at regular time intervals over the course of the reaction. 1,3,5-trimethoxybenzene (TMB) was chosen as 63 internal standard so as to obtain quantitative data from the NMR spectra. Indeed, TMB is not expected to interfere in the polymerization process (initiator for the polymerization or chain transfer agent) or to compete for coordination sites on our catalysts, since it is a relatively weak ligand; its chemical shifts (3.77 and 6.09 ppm in CD2Cl2) also don’t overlap with the peaks of lactide or poly(lactic acid) in a 1H NMR spectrum. The degree of stereoselectivity of a catalyst is directly related to the microstructure of the polymer. The microstructure can be determined from 1H{1H}NMR spectroscopy data, using statistic tools such as the Bernoullian model.105 This mathematical model allows to calculate the probabilities related to the different stereosequences occurring in a polymer. Stereosequences are sequences of several monomer units of different configuration (triads have three monomer units, tetrads have four units, and so on for pentads and hexads) labelled according to the type of linkages in the monomer sequences. When two monomers have the same relative stereochemistry, the linkage is said meso; a racemic linkage indicates two consecutive monomers with an opposite stereochemistry. It is important to note that the Bernoullian model is intended for systems where the probabilities of a meso linkage, Pm, and the probability of a racemic linkage, Pr, are independent of the previous event. In other words, the chiral configuration of the next inserted monomer is independent of the chain end; this corresponds to polymerization catalyzed through enantiomorphic site control mechanism (as opposed to chain-end control mechanism). 1H NMR characterization of PLA offers a way to calculate Pm and Pr, thus to determine the degree of tacticity of a polymer. A 1H NMR spectrum of PLA reveals two multiplets, which correspond to the repetition of methine and methyl protons coupling with each other throughout the polymer chain. Since these two sets of peaks are the result of several overlapping quartets in the methine area (methine proton coupling with the adjacent methyl group) and doublets in the methyl area (methyl protons coupling with the adjacent 64 methine proton), it is convenient to be able to simplify the 1H NMR spectrum by suppressing the coupling between methine and methyl protons (Scheme 2.12). This is realized by homonuclear decoupled 1H NMR experiment (1H{1H}).   Figure 2.19 1H NMR spectrum of PLA (CDCl3, 298 K, 600 MHz).  1H{1H} NMR spectroscopy is a technique that refers to the constant irradiation of a region of the spectrum during the collection time; for PLA characterization, the region commonly chosen is the methyl area. Irradiation results in saturation of the excited state of the methyl protons, thus eliminating spin-spin coupling relaxation process with the methine protons. Since the coupling to the methyl protons is suppressed, the methine area is greatly simplified from a series of overlapping quartets to a series of distinct, yet partially overlapping, singlets. These singlets correspond to the different tetrad stereosequences: rrr, mmm, mrm, rmr, mmr, rmm, rrm, and mrr. In polymerization of rac-LA, only singlets for 65 mmm, mrm, rmr, mmr and rmm are observed; the mmr and rmm tetrads show two singlets but they are indistinguishable by 1H NMR spectroscopy (Figure 2.13).   Figure 2.20. Stereosequences of isotactic PLA induced by enantiomorphic site control   The relative integration of the tetrads allows to calculate Pm and Pr based on statistics from the Bernoullian model (Table 2.5), which determines the microstructure of the polymer, and, in some cases, the mechanism responsible for this specific microstructure.11 Pm = 1 indicates perfectly isotactic PLA, while Pr = 1 indicates perfectly syndiotactic PLA (from meso-LA) as well as heterotactic PLA (from rac-LA). A completely atactic polymer would have a value of 0.5 for Pm or Pr. Table 2.6. Probability of tetrad sequences in PLA according to the Bernoullian model.     Polymerization of LA catalyzed by 4a Although there was no success in forming the zinc ethoxide analogue of 3a, polymerization of lactide catalyzed by zinc phenoxide (±)-4a was attempted. Polymerization of 200 equivalents rac-lactide (0.31 M) with (±)-4a (1.61 mM) was carried out in 66 dichloromethane at room temperature for 40 hours. At the end of the reaction time, 1H NMR spectrum of the product mixture indicated 96% conversion (measured from integration of the methyl resonances of LA at the beginning and end of the reaction). Metal aryloxides catalysts for LA ROP with the aryloxide as initiator group are quite rare in literature.106  Table 2.7. Selected data for ring opening polymerization of rac-LA. Entry Cat LA/Cat [Cat] (mM) [LA] (M) time (h) Conversion a (%) P m b 1 (±)-4a 200 1.6 0.31 24 96 – 2 (±)-4a 200 1.6 0.31 40 100 54 3 (R,R) -4a 200 1.6 0.31 40 100 52 4 (±)-4a 188 2.4 0.45 25 96 – 5 L1ZnOEt d 1500 0.69 1.0 0.3 93 – All reactions carried out in CH2Cl2 at 298K. a Determined via integration of the methyl resonances of LA. b Determined from the methine region of the 1H{1H} NMR spectrum. d Reference 47.  The activity of (±)-4a was found to be considerably lower than the activity observed with L1ZnOEt (Table 2.7).47 In order to obtain more information about the activity of our catalyst, we monitored in situ by 1H NMR spectroscopy the polymerization of 188 equivalent rac-lactide catalyzed by (±)-4a (0.0024 M) over 25 hours at room temperature in CD2Cl2 (Figure 2.14). The plot of LA]/[TMB] vs. time for the polymerization of 188 equivalents lactide by (±)-4a shows a slow initiation period in the first 3 hours of reaction. This long induction period could be due to the formation of phenoxy-bridged polynuclear species 107  in equilibrium with mononuclear species. Then there is an increase in the polymerization activity before the polymerization slows down after 10 hours; the conversion rate is observed to increase by 1% only in the last 15 hours, reaching 96% after 25 hours reaction.  67 0 0.5 1 1.5 2 2.5 3 3.5 4 0 200 400 600 800 1000 1200 1400 1600 Time (mins) [L A ]/[ TM B]  Figure 2.21.  Plot of [LA]/[TMB] for polymerization of 188 equiv rac-LA with (±)-4a in the presence of 1,3,5-trimethoxybenzene (TMB) internal standard.  The methine proton of LA was monitored via 1H-NMR spectroscopy (CD2Cl2, 298 K, 400 MHz). [(±)-4a]o = 2.4 mM, [LA]o = 0.45 M.   Information on stereoselectivity was obtained by analysing the product of polymerization of rac-LA (200 eq.) with racemic (±)-4a and enantiopure (R,R)-4a catalysts (0.0016 M) after 40 hours (100% conversion achieved), by 1H{1H} NMR spectroscopy (CD3Cl, 298 K). The product mixture of the polymerization reaction was precipitated from methanol, then washed with cold methanol, before being thoroughly dried under vacuum and dissolved in CD3Cl for 1H{1H} analysis.   68  Figure 2.22. Methine region of the 1H{1H} NMR spectrum of PLA (CDCl3, 298 K, 600 MHz) obtained from polymerization of rac-LA with (±)-4a.   Through integration of the different stereosequences in the homonuclear decoupled 1H NMR spectrum (Figure 2.16), the probabilities of meso and racemic linkages, Pm and Pr respectively, were calculated. These calculations are based on the Bernoullian model mentioned earlier in the thesis. For ROP catalyzed by (±)-4a and (R,R)-4a, Pm was found to be 0.54 and 0.52, respectively. Though a distinct singlet for the methine region protons of PLA corresponding to the mmm tetrad can be observed (indicating a PLA sample predominantly isotactic PLA), the Pm value shows a net lack of preference for the catalyst towards the monomer inserted (D or L) in the polymer chain (assuming a enantiomorphic site control mechanism Polymerization of LA catalysed by 4a in both racemic and enantiomerically pure forms yielded atactic polymer in each case. Complex 4a is not an enantioselective catalyst for LA ROP.  69 CONCLUSION   The interesting reactivity of some zinc ethyl complexes (3a, 3b and 3c) bearing chiral tridentate ancillary ligands ((NNR’OR)H, with R = H, tBu and R’ = H, Me) were investigated. These chiral zinc complexes were designed on the model of an achiral zinc complex (L1ZnEt) previously reported in literature by Hillmyer and Tolman; L1ZnEt is the precatalyst for the synthesis of the achiral zinc ethoxide L1ZnOEt, which demonstrated great activity towards polymerization of lactide.  However, the complexes 3a, 3b and 3c demonstrated significant difference in reactivity towards protonolysis compared to their achiral zinc ethyl analogue. No alkoxide could be isolated from the reaction of 3a and 3b with methanol, ethanol and isopropanol. 3a is also not water sensitive. The difference in sterics and electronics between our complexes and L1ZnEt is not significant and cannot explain the major discrepancy in reactivity. On the other hand, it was found that 3a readily reacts with HCl to produce the corresponding zinc chloride (NNMeOtBu)ZnCl (5a); reaction with phenol also allowed isolation of the zinc phenoxide (NNMeOtBu)ZnOPh.(4a). These positive reactions suggest that the reactivity of the chiral zinc complexes is greatly influenced by the pKa of the alcohol they are reacted with. To explain this, we proposed that the ancillary ligand of the achiral complex L1ZnEt contains a labile diamine arm that can dissociate from the zinc center, thus allowing insertion of an alcohol to form the corresponding alkoxide. 3a’s ligand backbone, however, is more rigid and would not allow more basic alcohols to coordinate to zinc. Reaction of L1ZnEt and 3a with pyridine confirmed the lability of L1ZnEt’s ligand arm. Polymerization of rac-lactide catalyzed by 4a showed that the complex, in its racemic or enantiomerically pure version, has a slow activity and is not stereoselective. This work demonstrated that the ligand that was designed in our laboratory does not allow protonolysis of the corresponding zinc ethyl with more basic alcohols (such as ethanol), 70 likely because of the rigidity induced by the cyclohexane ring. It could be in part responsible for the slow polymerization of the derived zinc phenoxide catalyst. Future work includes redesigning a ligand, for which the chiral center should be positioned in such as place that the ligand does not lose the lability of the diamine arm.           71 Experimental section General considerations.  Unless otherwise indicated, all air- and/or water-sensitive reactions were carried out under dry nitrogen in a MBraun glovebox or using standard Schlenk techniques.  NMR spectra were recorded on Bruker Avance 300, 400, or 600 MHz spectrometers.  The chemical shifts in the 1H NMR spectra are given in ppm versus residual protons: δ 7.27 CDCl3, δ 7.16 C6D6, δ 5.34 CD2Cl2, δ 2.09 CD3C6D5 (methyl), δ 1.94 CD3CN. 13C{1H} NMR chemical shifts are given in ppm versus residual 13C atoms: δ 77.2 CDCl3, δ 128.39 C6D6. Mass spectrometry analyses were performed on Kratos MS-80 (low resolution), Bruker Biflex IV (high resolution). Elemental analysis data were obtained from a Carlo Erba Elemental Analyzer EA 1108. All measurements for X-Ray crystallographic data were made on a Bruker X8 APEX diffractometer with graphite monochromated Mo-Kα radiation.  Details of the X- ray structure determination of complexes is summarized in table 4. Materials. All solvents (pentane, toluene, methylene chloride, THF and diethyl ether) were degassed and dried using 3Å molecular sieves in an MBraun solvent purification system. Benzene and THF were further dried with Na/benzophenone, before being distilled and degassed prior to use. Deuterated solvents were either dried over calcium hydride (CD2Cl2, CD3Cl, CD3CN) or over Na/benzophenone (C6D6, toluene-d8). D,L-lactide was purchased from Alfa-Aesar and recrystallized consecutively from toluene then isopropanol.  Compounds (±)-1a and (R,R)-1a were prepared following literature procedures.85,108 All other compounds were obtained from Aldrich and used without further purification.  2,4-di-tert-butyl-6-(((2-(dimethylamino)cyclohexyl)(methyl)amino)methyl)phenol (±)- (NNMeOtBu)H, (±)-2a.  (±)-1a (19.9 g, 55.2 mmol) was dissolved in acetonitrile (500 mL). 37% w/w aqueous formaldehyde (22.6 g, 278 mmol) was added to the opaque orange mixture, 72 which was then stirred for 30 min at room temperature. Cooling the mixture to 0°C with an ice-bath allowed to slow down the exothermic reaction when sodium cyanotrihydroborate (7.03 g, 112 mmol) was added with constant stirring, in the fumehood. After 30 min stirring at room temperature, glacial acetic acid (10 mL) was added to the mixture dropwise, forming white foam at the surface of the mixture. After 5 h reaction at room temperature, methanol/CH2Cl2 (400 mL, 2% methanol by volume) was added and the mixture was extracted with 1 M NaOH (3 × 400 mL). The organic layer was concentrated to dryness and the resulting orange solid was recrystallized from acetonitrile to produce white crystals (14.9 g, 39.8 mmol, 72 % yield). 1H NMR (400 MHz, C6D6, 298 K): δ 11.24 (1H, s, OH), 7.56 (1H, d, J = 2.2 Hz , ArH), 7.06 (1H, d, J = 2.2 Hz, ArH), 3.81 (1H, d, J = 11.7 Hz, NCH2C), 3.10 (1H, br s, NCH2Ar), 2.24 (2H, m, CyH), 2.12 (6H, s, N(CH3)2) , 2.08 (3H, s, NCH3), 1.76 (9H, s, C(CH3)3)), 1.53 (4H, m, CyH), 1.42 (9H, m, C(CH3)3), 0.84 (4H, m, CyH). 13C{1H} NMR (101 MHz, C6D6, 298 K): δ 155.60 (ArC), 139.96 (ArC), 136.25 (ArC), 125.26 (ArC), 124.01 (ArC), 123.27 (ArC), 64.52 (CyC), 64.42 (CyC), 35.91 (N(CH3)2), 34.07 (NCH3), 32.51 (C(CH3)3), 30.58 (C(CH3)3), 26.25 (CyC), 26.08 (CyC), 24.22 (CyC), 22.41 (CyC). Anal. Calcd (found) for C24H42N2O: C 76.9 (77.22), H 11.3 (11.42), N 7.48 (7.77) %.  R,R-2,4-di-tert-butyl-6-(((2-(dimethylamino)cyclohexyl)(methyl)amino)methyl)phenol, (R,R)-(NNMeOtBu)H, (R,R)-2a. (R,R)-1a (6.37 g, 17.7 mmol) was dissolved in acetonitrile (150 mL). 37% w/w aqueous formaldehyde (7.29 g, 23.9 mmol) was added to the opaque orange mixture, which was then stirred for 30 min at room temperature. Cooling the mixture to 0°C with an ice-bath allowed to slow down the exothermic reaction when sodium cyanotrihydroborate (2.39 g, 38.0 mmol) was added with constant stirring, in the fumehood. After 30 min stirring at room temperature, glacial acetic acid (10 mL) was added to the mixture dropwise, forming white foam at the surface of the mixture. After 5 h reaction at 73 room temperature, methanol/CH2Cl2 (300 mL, 2% methanol by volume) was added and the mixture was extracted with 1 M NaOH (3 × 400 mL). The organic layer was concentrated to dryness and the resulting orange solid was recrystallized from acetonitrile to produce white crystals (5.99 g, 60.2 mmol, 90 % yield). 1H NMR (400 MHz, C6D6, 298K): δ 11.24 (1H, s, OH), 7.56 (1H, d, J = 2.2 Hz , ArH), 7.06 (1H, d, J = 2.2 Hz, ArH),  3.81 (1H, d, J = 11.7 Hz, NCH2Ar), 3.10 (1H, br s, NCH2Ar) 2.24 (2H, m, cyH) , 2.12 (6H, s, N(CH3)2) , 2.08 (3H, s, NCH3), 1.76 (9H, s, C(CH3)3), 1.53 (4H, m, cyH),  1.42 (9H, m, C(CH3)3), 0.84 (4H, m, CyH). 13C{1H} NMR (101 MHz, C6D6, 298 K): 155.62 (ArC), 139.99 (ArC), 136.28 (ArC), 125.22 (ArC), 124.04 (ArC), 123.29 (ArC), 69.51 (CyC), 67.40 (CyC), 35.82 (N(CH3)2), 34.89 (NCH3), 32.50 (C(CH3)3), 30.57 (C(CH3)3), 26.23 (CyC), 26.06 (CyC), 24.21 (CyC), 22.39 (CyC). Anal. Calcd (found) for C24H42N2O: C 76.95 (77.04), H 11.30 (11.35), N 7.48 (7.50) %.  (±)-(E)-2-((2-(dimethylamino)cyclohexylimino)methyl)phenol.  (±)-N,N-Dimethyl diaminocyclohexane (3.29 g, 23.1 mmol) was diluted in dry methanol (150 mL). Salicylaldehyde (2.37 g, 19.4 mmol) and molecular sieves were added to the mixture with stirring. The reaction was stopped after 5 hours stirring at room temperature on molecular sieves and under static nitrogen atmosphere. The orange product mixture was filtered, then the filtrate was dried under vacuum to afford a dark brown oil (5.27 g, 21.7 mmol, 94 % yield). 1H NMR (300 MHz, C6D6, 296 K): δ 14.02 (1H, m, OH), 7.90 (1H, s, N=CH), 7.13 (1H, 7.02, m, ArH), (7.04, 2H, m, ArH), 6.69 (1H, m, ArH), 2.76 (1H, dt, CyH), 2.32 (1H, dt, CyH), 2.06 (6H, s, N(CH3)2), 1.50 (5H, m, CyH), 0.94 (3H, m, CyH). 13C{1H} NMR (75 MHz, C6D6, 296 K): 163.72 (N=CH), 162.67 (ArC), 132.46 (ArC), 131.65 (ArC), 120.00 (ArC), 118.69 (ArC), 117.91 (ArC),  69.90 (CyC), 67.12 (CyC), 40.95 (N(CH3)2), 34.93 (CyC), 25.72 (CyC), 74 25.18 (CyC), 23.63 (CyC). HRMS (MALDI TOF), m/z: calcd for C15H23N2O (H+): 247.1810, found: 247.1808.  (±)-2-((2-(dimethylamino)cyclohexylamino)methyl)phenol.  (±)-(E)-2-((2- (dimethylamino)cyclohexylimino)methyl)phenol (3.75 g, 15.2 mmol) was dissolved in acetonitrile (150 mL) before ten equivalents of NaBH4 (5.79 g, 15.3 mmol) were added with stirring. After 30 min stirring at room temperature, a catalytic amount of acetic acid (3 mL) was slowly added.  The mixture was left reacting for 5 hours. The product mixture was then diluted in 2% methanol/DCM (150 mL) then washed three times with 1M NaOH (300 mL). The dark brown organic layer was water dried with Na2SO4, before all solvent was removed under vacuum by rotary evaporation and Schlenk line. A light brown solid was obtained in 76% yield (2.88 g, 11.5 mmol). 1H NMR (300 MHz, C6D6, 296 K): δ 7.20 (2H, m, ArH), 6.97 (1H, m, ArH), 6.82 (1H, m, ArH), 3.74 (2H, dd, J = 39.9, 13.9 Hz, N=CH), 2.49 (1H, br, NH), 2.09 (2H, m, CyH), 1.97 (6H, s, N(CH3)2), 1.50 (3H, m, CyH), 0.94 (5H, m, CyH). 13C{1H} NMR (75 MHz, C6D6, 296 K): 159.87 (ArC), 129.13 (ArC), 128.48 (ArC), 125.11 (ArC), 119.18 (ArC), 117.38 (ArC), 67.05 (CyC), 59.33 (CyC), 51.25 (NCH2C), 40.28 (N(CH3)2), 32.35 (CyC), 25.80 (CyC), 25.23 (CyC),  21.27 (CyC). HRMS (MALDI TOF), m/z: calcd for C15H25N2O (H+): 249.1962, found: 249.1967.  (±)-2-(((2-(dimethylamino)cyclohexyl)(methyl)amino)methyl)phenol, (±)-(NNMeOH)H, (±)-2b.  (±)-1b (2.62 g , 10.5 mmol) was dissolved in acetonitrile (200 mL). Aqueous formaldehyde (37 % by wt., 4.32 g, 53.2 mmol) was added to the reaction flask while stirring; the reaction was cooled in an ice-bath before sodium cyanotrihydroborate (1.55 g, 0.0247 mol) was weighed out in a closed vial and added to the mixture 15 min later. After 20 min stirring, a catalytic amount of acetic acid (2 mL) was slowly added. After 4 hours reaction, the product 75 was extracted with 2% methanol/CH2Cl2 (200 mL) and washed with 1M NaOH (3 x 300 mL). The organic phase was collected and dried from solvent by rotary evaporation and under vacuum to afford a beige solid. Recrystallization in acetonitrile allowed to isolate white crystal in 77% yield (2.13g, 8.12 mol). 1H NMR (400 MHz, C6D6, 298 K): δ 10.88 (1H, s, OH), 7.17 (2H, m, ArH), 6.99 (1H, m, ArH), 6.79 (1H, m, ArH), 3.73 (1H, d, J = 12.7 Hz, NCH2C), 2.86 (1H, d, J = 12.2 Hz, NCH2C), 2.28 (1H, dt,CyH), 2.15 (1H, dt,CyH), 2.09 (6H, s, N(CH3)2), 2.02 (3H, s, NCH3), 1.61 (4H, m, CyH), 0.82 (4H, m, CyH). 13C{1H} NMR (101 MHz, C6D6, 298 K): 159.14 (ArC), 130.42 (ArC), 129.25 (ArC), 124.69 (ArC), 118.29 (ArC), 117.56 (ArC), 64.74 (CyC), 64.04 (CyC), 52.67 (NCH2C), 38.73 (NCH3), 25.97 (CyC), 25.82 (CyC), 23.16 (CyC), 21.97 (CyC). HRMS (MALDI TOF), m:z: calcd for C16H27N2O (H+): 263.2119, found: 263.2123.  (±)-(NNMeOtBu)ZnEt, (±)-3a. Proligand (±)-2a (2.00 g, 5.34 mmol) was dissolved in benzene (10 mL) in a vial and cooled to -38 °C. Cold diethyl zinc (0.560 mL, 5.35 mmol) was added to the pro-ligand solution and the colorless mixture was left for 10 min at room temperature. A white solid was isolated by solvent evaporation under vacuum (2.47 g, 5.28 mmol, 99% yield) before recrystallization in TMS-ether led to formation of colorless crystals. 76   NMR assignment for a product mixture with high β content (50%): (±)-3a-α: 1H NMR (400 MHz, C6D6, 298 K): δ 7.59 (1H, d, J = 2.5 Hz, H15), 6.85 (1H, d, J = 2.4 Hz, H13), 3.88 (1H, d, J = 12.2 Hz, H11), 2.96 (1H, d, J = 12.5 Hz, H11), 2.51 (1H, td, J = 11.2, 3.8 Hz, H3), 1.86 (3H, br. s., H10), 1.85 (3H, br. s., H30), 1.84 (9H, s, H23-25), 1.76 (3H, t, J = 8.1 Hz, H29), 1.75 (1H, td, J = 10.96, 3.96 Hz, H4), 1.56 (3H, s, H9), 1.45 (9H, s, H19-21), 1.27-1.38 (3H, m, H1), 1.10-1.00 (2H, m, H6), 0.57 (2H, m, H2), 0.54 (2H, q, J = 8.22 Hz, H28), 0.35 (m, 2H, H5). 13C{1H}NMR (101 MHz, C6D6, 298 K): 165.91 (C17), 137.99 (C16), 134.26 (C14), 126.03 (C13), 124.64 (C15), 121.42 (C12), 63.96 (C4), 60.94 (C11), 56.82 (C3), 45.97 (C10), 38.95 (C9, C30), 36.35 (C22), 34.32, 34.52, 34.55, 32.81 (C19-21), 30.90 (C23-25), 25.29 (s, 1 C)  25.08 (s, 1 C), 24.91 (s, 1 C), 24.81 (s, 1 C), 22.89 (s, 1 C), 22.49 (s, 1 C), 22.29 (s, 1 C), 21.83 (s, 1 C), 14.58 (C29), -3.44 (C28).   (±)-3a-β: 1H NMR (400 MHz, C6D6, 298 K): 1H NMR (400 MHz): δ 7.59 (1H, d, J = 2.5 Hz, H45), 6.96 (1H, d, J = 2.4 Hz, H43), 3.48 (1H, d, J = 11.6 Hz, H41), 2.9 (1H, d, J = 11.9 Hz, 77 H41), 2.31 (1H,  td, J = 11.5, 3.8 Hz, H33), 2.13 (3H, s, H40), 2.08 (3H, s, H60), 1.91 (1H, td, J = 11.6, 3.65 Hz, C44), 1.90 (9H, s, C53-55), 1.67 (3H, s, H39), 1.62 (3H, d, J = 7.9 Hz, M09), 1.54 (3H, t, J = 7.92 Hz, H59), 1.47 (9H, s, C49-51), 1.4-1.6 (2H, m, H36), 0.71 (2H, m, , H35), 0.47 (2H, q, J = 7.92, H58), 041 (2H, m, H32). 13C{1H} NMR (101 MHz, C6D6, 298 K): δ 166.09 (C47), 138.29 (C46), 134.61 (C44), 126.40 (C43), 124.35 (C45), 123.36 (C42), 65.16 (C34), 64.32 (C33), 56.01 (C41), 44.77 (C40), 43.22 (C50), 39.14 (C39), 36.26 (C52), 34.55 (?), 34.32, 34.52, 32.75 (C49-51), 30.78 (C53-55), 25.29, 25.08, 24.91, 24.81, 22.49, 22.89, 22.29, 21.83, 14.38 (C59), -3.61 (C58). Anal. Calcd (found) for C26H46N2OZn: C 66.72 (66.91), H 9.91 (9.74), N 5.99 (5.92) %.  (R,R)-(NNMeOtBu)ZnEt, (R,R)-3a.  (R,R)-2a (1.60 g, 4.27 mmol) was dissolved in benzene (10 mL) in a vial and cooled to -38°C. Cold diethyl zinc (0.450 mL, 4.30 mmol) was added to this solution and the colorless mixture was left stirring for 10 min at room temperature. A white solid was isolated by solvent evaporation under vacuum (1.96 g, 4.18 mmol, 98% yield) before recrystallization in TMS-ether led to formation of colorless crystals. For a product mixture (R,R)-3a-α enriched (90%): 1H NMR (300 MHz, C6D6, 298 K): δ 7.62 (1H, d, J = 2.9 Hz, H15), 6.86 (1H, d, J = 2.9 Hz, H13), 3.90 (1H, d, J = 12.2 Hz, H11), 2.96 (1H, d, J = 12.2 Hz, H11), 2.52 (1H, td, J = 11.2, 3.8 Hz, H3), 1.87 (9H, s, H23-H25), 1.84 (3H, s, H10), 1.83 (3H, s, H30), 1.80 (3H, t, J = 8.1 Hz, H29), 1.68 (1H, td, J= 10.96, 3.96 Hz, H4), 1.57 (3H, s, H9), 1.47 (9H, s, H19-21), 1.21- 1.34 (3H, m, H1), 0.56 (2H, m, H2), 0.54 (2H, q, J = 8.22 Hz, H28), 0.34 (2H, m, H5). 13C{1H} NMR (75 MHz, C6D6, 298 K): 165.93 (C17), 138.09 (C16), 134.28 (C14), 126.38 (C13), 124.70 (C15), 121.41 (C12), 63.93 (C4), 60.95 (C11), 56.78 (C3), 45.94 (C10), 38.94 (C9, C30), 36.37 (C22), 34.47, 32.76 (C19-21), 30.89 (C23-25), 24.91, 24.80, 22.47, 21.80, 78 14.58 (C29), -3.61 (C28). Anal. Calcd (found) for C26H46N2OZn: C 66.72 (66.33), H 9.91 (9.94), N 5.99 (5.97) %.  (±)-(NNMeOH)ZnEt, (±)-3b. (±)-2b (1.18g, 4.40 mmol) was dissolved in pentane (5 mL) before the solution was cooled to -35 ºC. Diethyl zinc (0.541 mL, 5.17 mmol) was then added to the cold solution and stirred overnight to afford a white precipitate (1.49 g, 4.15 mmol, 94% yield). 1H NMR (400 MHz, C6H6, 298 K): δ 7.37 (1H, t, ArH), 7.22 (1H, d, ArH), 6.87 (1H, d, ArH), 6.68 (1H, t, ArH), 3.92 (1H, d, NCH2C), 2.89 (1H, d, NCH2C), 2.32 (1H, td, CyH), 1.84 (3H, s, N(CH3)2), 1.81 (3H, s, NCH3), 1.78 (3H, t, ZnCH2CH3), 1.65 (1H, td, CyH), 1.61 (3H, s, N(CH3)2), 1.17 (4H, m, CyH), 0.61 (2H, m, ZnCH2CH3), 0.58 (2H, m, CyH), 0.15 (2H, m, CyH). 13C{1H} NMR (101 MHz, C6D6, 298 K): 167.14 (ArH), 131.62 (ArH), 131.36 (ArH), 124.67 (ArH), 121.52 ArH), 115.16 (ArH), 64.20 (CyC), 59.57 (NCH2C), 57.01 (CyC), 45.73 (N(CH3)2), 39.13 ((N(CH3)2, NCH3), 38.49 (C(CH3)3), 24.69, 24.51, 22.46, 21.64, 14.62 (ZnCH2CH3), -3.21 (ZnCH2CH3). Anal. Calcd (found) for C18H30N2OZn: C 60.76 (60.52), H 8.50 (8.26), N 7.87 (8.12) %.  (±)-(NNHOtBu)ZnEt, (±)-3c. (±)-1a (0.30 g, 0.84 mmol) was dissolved in 3 mL toluene in a vial and cooled to -38°C. Cold diethyl zinc (0.090 mL, 0.86 mmol) was added to the ligand solution and the colourless mixture was left stirring with a magnetic stir bar for 70 min at room temperature. A white solid was isolated by solvent evaporation under vacuum (0.37 g, 0.81 mmol, 96% yield) before recrystallization in TMS-ether/toluene (colorless crystals).  1H NMR (C6H6, 400 MHz, 298 K): δ 7.62 (1H, d, J = 2.2 Hz, ArH), 6.86 (1H, d, J = 2.6 Hz, ArH), 3.97 (1H, dd, J = 12.1 Hz, NCH2C), 3.15 (1H, dd, J = 12.3 Hz, NCH2C), 2.17 (1H, m , CyH), 1.89 (9H, s, C(CH3)3), 1.88 (3H, s, N(CH3)2), 1.82 (3H, t, ZnCH2CH3), 1.73 (1H, td, CyH), 1.52 (3H, s, N(CH3)2), 1.47 (9H, s, C(CH3)3), 1.39 (2H, m, ZnCH2CH3), 0.18 (2H, m, 79 CyH). 13C{1H} NMR (101 MHz, C6D6, 298 K): 166.50 (ArC), 138.32 (ArC), 134.17 (ArC), 126.38 (ArC), 124.67 (ArC), 120.82 (ArC), 66.82 (CyC), 53.21, 51.01, 45.49 (N(CH3)2), 38.33 (N(CH3)2), 36.38 (C(CH3)3), 34.36, 32.76 (C(CH3)3), 30.80, 24.96, 21.17, 14.57 (ZnCH2CH3), -2.49 (ZnCH2CH3). Anal. Calcd (found) for C25H44N2OZn: C 66.14 (66.23), H 9.77 (9.47), N 6.17 (6.23) %.  (±)-(NNMeOtBu)ZnOPh, (±)-4a. (±)-3a (0.200 g, 0.486 mmol) was dissolved in 3 mL diethyl ether then left at -38°C for 10 min. Phenol (0.0432 g, 0.457 mmol) was added to the cold zinc solution and the reaction was left stirring for 20h. The white product mixture was vacuum filtered though a Fritz funnel and the white solid washed 3 times with pentane before being dried under vacuum (0.256 g, 0.405 mmol, 95% yield). Colorless crystals were obtained upon recrystallization from diethyl ether. 1H NMR (300 MHz, C6D6, 295 K): δ 7.62 (1H, d, J = 2.2 Hz, ArH), 7.39 (3H, br s, PhH), 6.85 (2H, m, PhH), 6.77 (1H, d, J = 2.2 Hz, ArH), 3.87 (1H, d, J = 12.4 Hz, NCH2C), 2.89 (1H, d, J = 12.4 Hz, NCH2C), 2.40 (1H, m, CyH), 2.02 (3H, s, N(CH3)2), 1.94 (3H, s, NCH3), 1.85 (9H, s, C(CH3)3), 1.43 (3H, s, N(CH3)2), 1.42 (9H, s, C(CH3)3), 0.49 (2H, m, CyH). 13C{1H} NMR (101 MHz, C6D6, 298 K): 165.26 (ArC), 138.91 (ArC), 136.24, 130.35 (ArC), 126.66 (ArC), 126.38 (ArC), 125.11 (ArC), 122.34 121.35 (ArC), 64.56, 60.92, 51.01, 57.29, 45.70 (N(CH3)2), 39.08 (N(CH3)2), 38.10, 36.34 (C(CH3)3), 34.48, 32.60 (C(CH3)3) 30.73 (C(CH3)3), 24.68, 24.62, 22.44, 22.25, 21.76. Anal. Calcd (found) for C30H46N2O2Zn: C 67.72 (67.53), H 8.71 (8.74), N 5.26 (5.17) %.  (R,R)-(NNMeOtBu)ZnOPh, (R,R)-4a. (R,R)-3a  (0.240 g, 0.513 mmol) was dissolved in 3 mL diethyl ether then left  at -38°C for 10 min. Phenol (0.0482 g, 0.512 mmol) was added to the cold zinc solution and the reaction was left stirring overnight. The white product mixture was vacuum filtered though a Fritz funnel and the white solid washed 3 times with pentane 80 before being dried under vacuum. (0.263 g, 0.493 mmol, 96 % yield). Colorless crystals could be obtained upon recrystallization from diethyl ether. 1H NMR (300 MHz C6D6, 295 K): δ 7.62 (1H, d, J = 2.2 Hz, ArH), 7.39 (3H, br s, PhH), 6.85 (2H, m, PhH), 6.77 (1H, d, J = 2.2 Hz, ArH), 3.87 (1H, d, J = 12.4 Hz, NCH2C), 2.89 (1H, d, J = 12.4 Hz, NCH2C), 2.40 (1H, m, CyH), 2.02 (3H, s, N(CH3)2), 1.94 (3H, s, NCH3), 1.85 (9H, s, C(CH3)3), 1.43 (3H, s, N(CH3)2), 1.42 (9H, s, C(CH3)3), 0.49 (2H, m, CyH). 13C{1H} NMR (101 MHz, C6D6, 298 K): 165.26 (ArC), 138.91 (ArC), 136.24, 130.35 (ArC), 126.66 (ArC), 126.38 (ArC), 125.11 (ArC), 122.34, 121.35 (ArC), 64.56, 60.92, 51.01, 57.29, 45.70 (N(CH3)2), 39.08 (N(CH3)2), 38.10, 36.34 (C(CH3)3), 34.48, 32.60 (C(CH3)3) 30.73 (C(CH3)3), 24.68, 24.62, 22.44, 22.25, 21.76. Anal. Calcd (found) for C30H46N2O2Zn: C 67.72 (67.45), H 8.71 (8.58), N 5.26 (5.13) %.  (±)-(NNMeOtBu)ZnCl, (±)-5a. (±)-3a (0.991 g, 2.12 mmol) was dissolved in approximately 5 mL pentane, then left at –38 °C for 10 min. 2M hydrochloric acid in ether (1.59 mL, 3.18 mmol) was slowly added to the cold zinc solution and the reaction was left stirring for an hour at room temperature. A white precipitate of zinc chloride immediately formed upon reaction, which was vacuum filtered on a Fritz funnel and washed with cold pentane. The white solid was then dried under high vacuum (0.952 g, 2.01 mmol, 95% yield). Colorless crystals were obtained upon recrystallization from acetonitrile. 1H NMR (300 MHz, C6D6, 295 K): δ 7.61 (1H, d, J = 2.93 Hz, ArH), 6.77 (1H, d, J = 2.93 Hz, ArH), 3.89 (1H, d, J = 12.4 Hz, NCH2C), 2.91 (1H, d, J = 12.4 Hz, NCH2C), 2.42 (1H, m, CyH), 2.02 (3H, s, N(CH3)2), 1.98 (3H, s, NCH3), 1.82 (9H, s, C(CH3)3), 1.45 (3H, s, N(CH3)2), 1.41 (9H, s, C(CH3)3), 0.88-1.25 (5H, m, CyH), 0.49 (3H, m, CyH), 0.20 (1H, m, CyH).  13C{1H} NMR (101 MHz, C6D6, 298 K): δ 165.43 (ArC), 139.15 (ArC), 136.39 (ArC), 126.90 (ArC), 125.37 (ArC), 121.52 (ArC), 64.72 (CyC), 61.16 (NCH2C), 57.61 (CyC), 46.76 (N(CH3)2), 39.37 (N(CH3)2), 39.32 (NCH3), 36.59 81 (C(CH3)3), 34.71, 32.83 (C(CH3)3), 31.02 (C(CH3)3), 24.94 (CyC), 24.83 (CyC), 22.76 (CyC), 21.98 (CyC) ppm. Anal. Calcd (found) for C24H41ClN2OZn: C 60.76 (60.77), H 8.71 (8.82), N 5.90 (5.95) %.  Lactide polymerization catalyzed by (±)-4a. A 0.0048 M solution of (±)-4a catalyst was made in deuteriated methylene chloride and stored in freezer. 65 mg D,L-lactide (0.45 mmol, 188 eq.) and 5 mg 1,3,5-trimethoxybenzene (0.030 mmol) were dissolved in 0.50 mL CD2Cl2 in a J-Young tube. 0.50 mL (0.0024 mmol, 1 eq.) of catalyst solution was added to the lactide solution and the NMR tube was immediately cooled in liquid nitrogen to prevent polymerization to occur before NMR data aquisition. The mixture was then warmed to room temperature just before starting acquisition of NMR data.  Polymerizations with (R,R)-4a were carried out in an analogous manner.  82 BIBLIOGRAPHY   [1] American Society for Testing Materials: Standard Specification for Compostable Plastics. . Annual Book of ASTM Standards; Philadelphia, PA, 1999; Part 08.03, ASTM D6400 – 04. [2] Brown, R. C. in: Biorenewable resources: Engineering New Products from Agriculture; Iowa State Press, Ames, 2003; p. 4. [3] Chandra, R.; Rutsgi, R. Prog. Polym. Sci., 1998, 23, 1273–1335. [4] Plastics, Additives and Compounding, 2005, 7 (4), 12. [5] Bastiol C. in: Wiley encyclopedia of packaging technology, 2nd Ed. John Wiley & Sons: New York, 1997; p. 77. [6] Griffith, G. L. Polym. Biomat. Acta Mater. 2000, 48, 263–277. [7] (a) Nelson, D. L.; Cox, M. M. in Lehninger Principles of Biochemistry, 4th Ed.; Worth Publishers: New York, 2000; p.523-525. (b) Garlotta, D.J. Polymers Environ. 2002, 9, 63. [8] Wasewar, K. L. Chem. Biochem. Eng. Q. 2005, 19 (2),159–172. [9] Vink E.T.H.1; Rabago K.R.; Glassner D.A.; Gruber P.R. Polymer Degradation and Stability 2003, 80, 403–419. [10] Chanda, M. Advanced Polymer Chemistry A Problem Solving Guide. Marcel Dekker Inc. NY, 2000. [11] Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc. 2002, 124, 1316. [12 ] Lu, L.; Mikos, A.G. Poly(lactic acid). in Polymer Data Handbook (ed. Mark, J.E.), Oxford University Press, New York, 1999, p.628. [13] Van de Velde, K.; Kiekens, P. Polymer Testing 2002, 21, 433-442. [14] Furuhashi, Y.; Kimura, Y.; Yoshie N; Yamane H. Polymer 2006, 47 (16), 5965-5972.  83  [15] Narayan, R. Degradation of Polymeric Materials. In: Science and Engineering of Composting: Design, Environmental, Microbiological and Utilization Aspects; Eds. Hoitink, H. A.; Keener, H. N. Renaissance Publications, Worthington , OH, 1993. [16] Masaki, K.; Kamini, N. R.; Ikeda, H.; Iefuji, H. Appl. Environ. Microbiol. 2005, 71, 7548-7550. [17] Reeve, M. S., McCarthy, S. P., Downey, M. J.,Gross, R. A., Macromolecules 1994, 27, 825-831. [18] Carothers, W. H., J. Am. Chem. Soc.1929, 51, 2548-2559. [19] (a) Enomoto, K.; Ajioka, M.; Yamaguchi, A. US patent 5 310 865; 1995. (b) Kashima, T.; Kameoka,  T.; Ajioka, M.; Yamaguchi A. US patent 5 428 126, 1995. (c) Ichikawa, F.; Kobayashi, M.; Ohta, M.; Yoshida, Y.; Obuchi, S.; Itoh,H. US patent 5 440 008; 1995. (d) Ohta M., Yoshida Y., Obuchi S. US patent 5 440 143; 1995. [20] Adapted from Lunt, J., Polym Degrad Stab  1998, 59, 145–152. [21] Dechy-Cabaret, O.; Martin-Vaca, B.; Bourissou, D. Chem. Rev. 2004, 104, 6147-6176. [22] O'Keefe, B. J.; Hillmyer, M. A.; Tolman, W. B. J. Chem. Soc., Dalton Trans. 2001, 2215-2224. [23] Wu J, Yu T-L; Chen C-T, Lin C-C. Coord Chem Rev. 2006, 250, 602–626. [24] Stridsberg, K. M.; Ryner, M.; Albertsson, A.-C. Adv. Polym. Sci. 2002,157, 41. [25] Duda, A.; Penczek, S. Macromolecules 1990, 23, 1636. [26] van Hummel, G. J.; Harkema, S.; Kohn, F. E.; Feijen, J. Acta. Crystallogr. Ser. B 1982, 38, 1679. [27] Kurcok, P.; Penczek, J.; Franek, J.; Kedlinski, Z. Macromolecules 1992, 25, 2285. [28] Kricheldorf, H. R.; Krieser, I. Makromol. Chem. 1986, 187, 1861. [29] Kricheldorf, H. R.; Berl, M.; Scharnagl, N. Macromolecules 1988, 21, 286-293.  84   [30] McNaught, A. D.; Wilkinson, A. Compendium of chemical terminoloy: IUPAC recommendations. Blackwell Science: Oxford, 1997. [31] Kricheldorf, H. R.; Kreiser-Saunders, I. Makromol. Chem. 1990, 191, 1057-1066. [32] Dubois, P.; Jacobs, C.; Jerome, R.; Teyssie, P. Macromolecules 1991, 24, 2266-2270. [33] Bero, M.; Dobrzynski, P.; Kasperczyk, J. J. Polym. Sci., Part A: Polym. Chem., 1999, 37, 4038. [34] Degée, P.; Dubois, P.; Jerôme, R.; Jacobsen, S.; Fritz, H.-G. Macromol. Symp. 1999, 144, 289. [35] Kowalski, A.; Duda, A.; Penczek, S. Macromolecules, 1998, 31, 2114. [36] Stevels, W. M.; Ankone, M. J. K.; Dijkstra, P. J.; Feijen, J. Macromolecules 1996, 29, 6132-6138. [37] a) Save, M.; Schappacher, M.; Soum, A. Macromol. Chem. Phys. 2002, 203, 889. b) Save, M.; Soum, A. Macromol. Chem. Phys. 2002, 203, 2591. [38] Zhong, Z.; Dijkstra, P. J.; Birg, C.; Westerhausen, M.; Feijen, J. Macromolecules 2001, 34 (12), 3863-3868. [39] Liu, Y.-C.; Ko, B.-T.; Lin, C.-C. Macromolecules 2001, 34, 6196. [40] Trofimoff, L.; Aida, T.; Inoue, S. Chem. Lett. 1987, 991. [41] (a) Faure, J.-L.; Gornitzka, H.; Reau, R.; Stalke, D.; Bertrand, G. Eur. J. Inorg. Chem. 1999, 2295. (b) Dumitrescu, A.; Martin-Vaca, B.; Gornitzka, H.; Cazaux, J.-B.; Bourissou, D.; Bertrand, G. Eur. J. Inorg. Chem. 2002, 1948. [42] (a) Chisholm, M. H.; Eilerts, N. W. Chem. Commun. 1996, 853. (b) Chisholm, M. H.; Eilerts, N. W.; Huffman, J. C.; Iyer, S. S.; Pacold, M.; Phomphrai, K. J. Am. Chem. Soc.  85  2000, 122, 11845. (c) Chisholm, M. H.; Gallucci, J.; Phomphrai, K. Chem. Commun. 2003, 48. [43] Chisholm, M. H.; Patmore, N. J.; Zhou, Z. Chem. Commun., 2005, 127. [44] Cheng, M.; Attygalle, A. B.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 1999, 121, 11583. [45] Chamberlain, M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc., 2001, 123, 3229. [46] Spassky, N.; Wisniewski, M.; Pluta, C.; Le Borgne, A. Macromol. Chem.Phys. 1996, 197, 2627-2637. [47] Williams, C. K.; Breyfogle, L. E.; Choi, S. K.; Nam, W.; Young, V. G.; Hillmyer, M. A.; Tolman, W. B. J. Am. Chem. Soc. 2003, 125, 11350. [48] Kricheldorf, H. R.; Boettcher, C. J. Macromol. Sci., Pure Appl. Chem. A 30, 441. [49] Kricheldorf, H. R.; Serra i Albet, A., Polym. Bull. Berlin, 1985, 14, 497. [50] Kleine, J.; Kleine, H. Makromol. Chem. 1959, 30, 23. [51] Bero, M.; Kasperczyk, J.; Jedlinski, Z. Makromol. Chem. 1990, 191,2287. [52] (a) Dittrich, W.; Schulz, R. C. Polymer 1983 24, 53. (b) Dittrich, W.; Schulz, R. C. Angew. Makromol. Chem. 1971, 15, 109. [53] Vert, M.; Schwach, G.; Engel, R.; Coudane, J. Control Release 1998, 53, 85–92. [54] Chabot, F.; Vert, M.; Chapelle, St.; Granger, P. Polymer 1983, 24, 53. [55] Schwach, G.; Coudane, J.; Engel, R.; Vert, M. Polym. Bull. Berlin,1994, 29, 617. [56] Vert, M.; Schwach, G.; Engel, R.; Coudane, J. Control Release 1998, 53, 85–92. [57] (a) Kricheldorf, H. R.; Damrau, D. O.Macromol. Chem. Phys.1997, 198,1753-1 766. (b) Schwach, G.; Coudane, J.; Engel, R.; Vert, M. Polymer International 1998, 46, 177-182. [58] Nijenhuis, A. J.; Grijpma, D.W. ; Pennings, A. J. Polymer 1991, 97.  86  [59] Nijenhuis, A. J.; Grijpma, D.W. ; Pennings, A. J. Macromolecules 1992, 25, 6419–6424. [60] Spassky, N.; Wisniewski, M.; Pluta, C.; Le Borgne, A. Macromol. Chem. Phys. 1996, 197, 2627. [61] Cheng, M.; Attygalle, A. B.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 1999, 121, 11583. [62] Chen, H.-Y.; Huang, B.-H.; Lin, C.C. Macromolecules 2005, 38 (13), 5400. [63] Chisholm, M. H.; Gallucci, J.; Phomphrai, K. Inorg. Chem., 2005, 44 (22), 8004 -8010. [64] Dowe, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.; Williams, D. J. Dalton Trans. 2004, 570. [65] Pang, X.; Chen, X.; Zhuang, X.; Jing, X. J. Polym. Sc. Part A: Polym. Chem. 2008, 46, (2), 643. [66] Jensen, T. R.; Schaller, C. P.; Hillmyer, M. A.; Tolman, W. B., J. Organomet. Chem. 2005, 690, (24-25), 5881-5891. [67] Wu, J. C. ; Huang, B. H. ; Hsueh, M. L.; Lai, S. L.;Lin, C. C. Polymer, 2005, 46, 9784- 9792. [68] Chen, H. Y.; Tang, H. Y.; Lin, C. C., Macromolecules 2006, 39 (16), 5583. [69] Williams, C. K.; Brooks, N. R.; Hillmyer, M. A.; Tolman, W. B. Chem. Commun. 2002, 2132. [70] Van Koeten, G. et al.Structural organozinc chemistry In: The Chemistry of Organozinc Compounds, Editors: Z. Rappoport and I. Marek, John Wiley & Sons, Chichester, 2006. [71]  Van der Kerk, G. J. M. Pure Appl. Chem., 1972, 30, (3-4), 389-408. [72] Frankland, E. Liebigs Ann. Chem., 1849, 71, 171. [73] Almenningen, A.; Helgaker, T. U.; Haaland, A.; Samdal, S. Acta Chem. Scand., Ser. A, 1982, 36, 159.  87  [74] Boersma, J. In Comprehensive Organometallic Chemistry Eds. G. Wilkinson, F. G. A. Stone and E. W. Abel, Vol. 2, Pergamon, Oxford, 1982, p. 823. [75] Hannant, M.. D.; Schormann, M.; Bochmann, M. J. Chem. Soc., Dalton Trans., 2002, 4071. [76] Allen, S. D.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W., J. Organomet. Chem. 2003, 683, (1), 137-148. [77] (a) Looney ,A.; Han, R.; Gorrell, I. B.; Cornebise, M.; Yoon, K.; Parkin, G.; Rheingold, A. L. Organometallics, 1995, 14, 274. (b). Alsfasser, R.; Powell, A. K.; Trofimenko, S.; Vahrenkamp, H. Chem. Ber., 1993, 126, 685. (c). Kisko, J. L.; Fillebeen, T.; Hascall, T.; Parkin, G. J. Organomet. Chem., 2000, 596, 22. (d). Dias, H. V. R.; Jin, W. Inorg. Chem., 2003, 42, 5034. [78] Gorrell, I. B.; Looney, A.; Parkin, G. J. Chem. Soc. Chem. Commun 1990, 220. [79] Chakraborty, D.; Chen, E. Y.-X. Organometallics 2003, 22, 769. [80] Mitchell, J. M.; Finney, N. S., Tetrahedron Lett, 2000, 41, 8431. [81] Improvements carried out by J. H.Y. Fang and A. F. Douglas. [82] (a) Fujii, Y.; Ebina, F.; Yanagisawa, M.; Matsuoka, H.; Kato, T.  J. Inorg. Organomet. Polym. 1994, 4, 273. (b) Konsler, R. G.; Karl, J.; Jacobsen, E. N. J. Am. Chem. Soc. 1998 120, 10780. (c) Annis, D. A.; Jacobsen, E. N. J. Am. Chem. Soc. 1999 121, 4147. [83] (a) Reger, T.S.; Janda, K.D. J. Am. Chem. Soc. 2000, 122, 6929. (b) Baxendale, L.R.; Lee and S., A.-L.; Ley, V. J. Chem. Soc., Perkin Trans. 1 2002, 1850. (c) Baleizao, C.; Gigante, B.; Sabater, M.J.; Garcia, H.; Corma, A.;Appl. Catal. A: Gen. 2002, 228, 279. (d) Canali, L.; Cowan, E.; Gibson, C.L.; Sherrington, D.C.; Deleuze, H. J. Chem. Soc., Chem. Commun. 1998, 63, 2561.  88  [84] (a) Campbell, E.J.; Nguyen, S.T. Tetrahedron Lett. 2001, 42, 1221. (b) McGarrigle, E. M.; Dalton, C. T.; Daly, A. M.; Gilheany, D. G. J. Mol. Catal. A:. Chem. 2005, 231, 205.(c) “Fang, Lagaditis, Douglas, unpublished results” [85] Jacobsen, EN; Zhang, W; Muci, AR ; Ecker, JR; Deng, L. J. Am. Chem.Soc.1991, 113, 7063-7064. [86] Gasbøl, F.; Steenbøl, P.; Sørensen, B. S. Acta Chem. Scand. 1972, 26, 3605. [87] (a) Corey, E. J.; Yuen, P.-W.; Hannon, F. J.; Wierda, D. A. J. Org. Chem. 1990, 55, 784. (b) Dowling, C.; Parkin, G. Polyhedron 1996, 15, 2463. (c) Hammes, B. S.; Carrano, C. J. Inorg. Chem. 1999, 38, 4593. (d) Looney, A.; Han, R.; Gorrell, I. B.; Cornebise, M.; Yoon, K.; Parkin, G.; Rheingold, A. L. Organometallics 1995, 14, 274. [88] The “maximum number of optical isomers for a compound is 2n, where n is the number of chiral atoms” (Fessenden, R. J.; Fessenden, J. S.  in Organic Chemistry, 5th Ed.; Cole Publishing Company, Pacific Grove, California, 1990, 157). [89] Abraham, R.J.; Fell, S.C.M.; Smith, K.M. Tetrahedron 1979, 35, 1759. [90] Leung, A. C. W.; MacLachlan, M. J. J. Mater. Chem., 2007, 17, 1923-1932. [91] Screen, T. E. O.; Lawton, K. B.; Wilson, G. S.; Dolney, N.; Ispasoiu, R.; Goodson, T. III, Martin, S. J.; Bradley, D. D. C.; Anderson, H. L. J. Mater. Chem. 2001, 11, 312–320. [92] Gürol, I.; Durmu, M.; Ahsen, V.; Nyokong, T. Dalton Trans. 2007, 3782–3791. [93] Herrmann, J.-St.; Luinstra, G. A.; Roesky, P.W. J. Organomet. Chem. 2004, 689 (16), 2720-2725. [94] Allen, S. D.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J. Organomet. Chem. 2003, 683, 137-148. [95] Bochmann, M. et al. J. Organomet. Chem.  2008, 693, 1494–1501.  89  [96] (a) Chisholm M. H. et al. Dalton Trans., 2003, 406 – 412. (b) Carpentier, J. F. et al Inorg. Chem. 2007, 46, 328-340. [97] Allen, S. D.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J. Organomet. Chem. 2003, 683, 137-148. [98] Cole, S. C.; Coles, M. P.; Hitchcock, P. B. Organometallics 2004, 23, 5159-5168. [99] Chisholm, M. H.; Gallucci, J. C.; Zhen, H. H.; Huffman, J. C. Inorg. Chem. 2001, 40 (19), 5051. [100] Jeong, J. H. et al. Polyhedron 2008, 27, 319–324. [101] Williams, C. K.; Breyfogle, L. E..; Young, V. G.; Hillmyer, M. A.; Tolman, W. B. Dalton Trans., 2006, 928–936. [102] Bordwell, F. G.  Acc. Chem. Res. 1988, 21, 456-463.  [103] (a) Moers, O.; Blaschette, A.; Jones, P. G. Z. Anorg. Allg. Chem. 2001, 627, 1081. (b) Trösch, A.; Vahrenkamp, H. Inorg. Chem. 2001, 40, 2305-2311. (c) Garner, D.K.; Allred, R.A.; Tubbs, K.J.; Arif, A.M.; Berreau, L.M. Inorg. Chem. 2002, 41, 3533. (d) Hlavinka, M. L.; McNevin, M. J.; Shoemaker, R. K.; Hagadorn, J. R. Inorg. Chem., 2006, 45 (4), 1815 -1822. [104] (a) Abufarag, A.; Vahrenkamp, H. Inorg. Chem. 1995, 34, 2207. (b) Trösch, A.; Vahrenkamp, H. Inorg. Chem., 2001, 40, 2305. (c) Darensbourg, D. J. et al. J. Am. Chem. Soc. 1999, 121, 107-116 (d) Darensbourg, D. J., Wildeson, J. R., Yarbrough, J. C., Reibenspies, J. H.  J. Am. Chem. Soc. 2000, 122, 12487-12496. [105] Seymour, S.B.; Carraher, C.E. in: Polymer Chemistry—An Introduction, 2nd Edition, Marcel Dekker, New York, 1988.  90  [106] (a) Giesbrecht, G. R.; Whitener G. D.;Arnold J.J. Chem. Soc., Dalton Trans., 2001, 923–927. (b) Aubrecht, K. B.; Chang, K.; Hillmyer, M. A.; Tolman, W. B., J. Polym. Sc. Part A: Polym. Chem. 2001, 39, (2), 284-293. [107] (a) M. M. Olmstead, M. M.;P. P. Power, P. P.; S. C. Shoner, S. C. J. Am. Chem. Soc. 1991, 113, 3379 – 3385. (b) Dinger, M. B.; Scott, M. J. Inorg. Chem. 2000, 39, 1238- 1254 (c) Erxleben, A.  Inorg. Chem. 2001, 40, 208 –213. (c) Huang, B. H.; Lin, C. N.;Hsueh, M. L.; Athar T.; Lin, C. C. Polymer, 2006, 47, 6622. [108] Douglas, A. F.; Patrick, B. O.; Mehrkhodavandi, P. Angew. Chem. Int. Edit. 2008, 47, 2290-2293.                 91  APPENDIX 1   X-Ray Crystallography Data Table A.1. Crystal and refinement data for complexes (±)-3a, (R,R)-3a, (±)-3b, (±)-4a•phenol, and (±)-5a. (±)-3a (R,R)-3a (±)-3b (±)-4a•phenol (±)-5a empirical formula C26H46N2OZn C26H46N2OZn C18H30N2OZn C36H52N2O3Zn C24H41N2OZnCl fw 468.02 468.02 355.81 626.17 474.41 T (K) 178 178 178 178 178 a (Å) 12.4011(6) 11.494(3) 9.1548(7) 16.281(2) 8.7634(8) b (Å) 13.7314(8) 14.886(3) 12.2840(9) 15.2933(2) 30.501(3) c (Å) 17.1106(7) 16.076(3) 15.7667(12) 27.503(4) 9.3308(8) a (deg) 67.654(1) 90.0 87.630(4) 90.0 90.0 b (deg) 88.060(2) 90.0 89.522(4) 90.0 90.839 g (deg) 78.655(1) 90.0 87.107(4) 90.0 90.0 volume (Å3) 2639.5(2) 2639.5(2) 1769.3(2) 6848.2(2) 2493.8(4) Z 4 4 4 8 4 crystal system triclinic orthorhombic triclinic orthorhombic monoclinic space group P -1  P 212121 P -1 P bca P 21/n dcalc (g/cm3) 1.178 1.130 1.336 1.215 1.264 -1) 9.49 9.10 13.92 7.53 11.09 2qmax (deg) 53.6 50.0 56.1 49.9 55.9 absorption correction (Tmin, Tmax) 0.693, 0.789 0.498, 0.947 0.627, 0.846 0.562, 0.942 0.710, 0.885 total no. of reflections 25992 12202 52234 31010 25614 no. of indep reflections (Rint) 11212 (0.034) 4769 (0.064) 10744 (0.035) 5979 (0.082) 5969 (0.041)residuals (refined on F , all data): R1 a; wR2 b 0.074; 0.118 0.099; 0.108 0.050; 0.079 0.102; 0.107 0.065; 0.097 GOF 1.03 0.99 0.97 1.02 1.12 no. observations [I > 2s(I)] 7707 3269 8092 3744 4669 residuals (refined on F): R1; wR2 0.044; 0.106 0.056; 0.094 0.032; 0.076 0.050; 0.092 0.046; 0.092  a R1 = Σ ||Fo| - |Fc|| / Σ |Fo|; b wR2 = (Σw(Fo2 - Fc2)2/Σ w(Fo2)2)1/2.   

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 14 1
China 4 12
City Views Downloads
Seattle 6 0
Newport Beach 4 0
Beijing 3 0
Unknown 2 0
Shanghai 1 0
Redmond 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items