UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Indium complexes and their role in the ring-opening polymerization of lactide Douglas, Amy Frances 2008

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

Item Metadata

Download

Media
ubc_2008_spring_douglas_amy.pdf [ 1.93MB ]
Metadata
JSON: 1.0059291.json
JSON-LD: 1.0059291+ld.json
RDF/XML (Pretty): 1.0059291.xml
RDF/JSON: 1.0059291+rdf.json
Turtle: 1.0059291+rdf-turtle.txt
N-Triples: 1.0059291+rdf-ntriples.txt
Original Record: 1.0059291 +original-record.json
Full Text
1.0059291.txt
Citation
1.0059291.ris

Full Text

  INDIUM COMPLEXES AND THEIR ROLE IN THE RING-OPENING  POLYMERIZATION OF LACTIDE  by AMY FRANCES DOUGLAS B.Sc., The University of Victoria, 2005    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)     April 2008  ? Amy Frances Douglas, 2008 ii  ABSTRACT     The synthesis and characterization of a  series of chiral  indium complexes bearing a tridentate NNO ligand are reported. The ligand 2-[[[(dimethylamino)cyclohexyl]amino]methyl]-4,6-bis(tert-butyl) phenol (H2NNO) was synthesized via a previously published procedure and bound  to  indium  by  both  a  protonolysis  and  salt  metathesis  route.  A  dimethylated  indium complex  (NNO)InMe2  (1)  was  isolated  by  reaction  of  InMe3  with  H2NNO.  A  one-pot  salt metathesis  route  was  used  to  produce  a  unique  mixed-bridge  dinuclear  indium  complex [(NNO)InCl]2(?-OEt)(?-Cl)  (3) from a mixture of indium trichloride, potassium ethoxide and the  monopotassiated  salt  of  the  ligand,  KH(NNO).  Direct  reaction  of  KH(NNO)  and  indium trichloride  resulted  in  the  formation  of  (NNO)InCl2  (4)  which  was  carried  forward  to  3  by reaction with sodium ethoxide.   The complex 3 is active for the ROP of ?-butyrolactone ?-caprolactone and lactide and is the first reported indium-based catalyst for lactide or ?-butyrolactone ROP. Kinetic studies of 3 for ROP of LA revealed that catalyst was well-behaved, and that the rate was first order with regard  to  lactide  and  catalyst.  The  enthalpy  and  entropy  of  activation  for  the  ROP  were experimentally  determined.  Polymer  produced  by  ROP  by  3  has  narrow  molecular  weight distribution and a good correlation is seen between the observed moleular weight and monomer loading. A mechanism was proposed for 3 acting as a catalyst for the ROP of lactide; however further experiments are required to confirm this mechanism. Polymer samples isolated from the ROP of rac-lactide by rac-3 show isotactic enrichment. It is postulated that the chiral catalyst 3 is exerting stereocontrol via an enantiomorphic site control mechanism.  iii  TABLE OF CONTENTS  Abstract  ......................................................................................................................... ii Table of Contents ..........................................................................................................iii List of Tables ................................................................................................................ iv List of Figures ............................................................................................................... vi List of Charts  ............................................................................................................... vii List of Schemes ...........................................................................................................viii List of Abbreviations  .................................................................................................... ix Acknowledgments ........................................................................................................ xii Chapter 1. General Introduction 1.1 Introduction to Poly(Lactic Acid) ............................................................................. 1 1.2 Synthesis of PLA Industrially.................................................................................... 6 1.3 Ring Opening Polymerization of LA......................................................................... 8 1.4 Stereoselectivity in Polymerization.......................................................................... 21 1.5 Steroselectivity in ROP LA..................................................................................... 24 1.6 General Chemistry of Indium and Group 13 ........................................................... 32  Chapter 2. The Synthesis of Indium Complexes 2.1 Introduction ............................................................................................................ 37 2.2 Results  ................................................................................................................... 39 2.3 Conclusion ............................................................................................................. 56 2.4 Experimental .......................................................................................................... 57  Chapter 3. Kinetic Studies and Characterization of Polymers 3.1 Introduction ............................................................................................................ 62 3.2 Results  ................................................................................................................... 68 3.3 Conlusions and Future Work .................................................................................. 82 3.4 Experimental .......................................................................................................... 84  Bibliography................................................................................................................ 87 Appendix  ..................................................................................................................... 96  iv  LIST OF TABLES   Table 1.1. Probability of tetrad sequences in PLA based on Bernoullian statistics  ........ 29 Table 2.1. Selected interatomic distances (  ) for compound 1 ...................................... 41 Table 2.2. Selected bond angles (?) for compound 1...................................................... 41 Table 2.3. Selected interatomic distances (?) for compound 2 ...................................... 43 Table 2.4. Selected bond angles (?) for compound 2  ..................................................... 43 Table 2.5. pKa values for simple organic compounds .................................................... 46 Table 2.6. Selected interatomic distances (?) for compound 3 ...................................... 48 Table 2.7. Selected bond angles (?) for compound 3  ..................................................... 48 Table 2.8. Selected interatomic distances (?) for compound 4?Py ................................ 50 Table 2.9. Selected bond angles (?) for compound 4?Py ................................................ 50 Table 2.10. Selected interatomic distances (?) for compound 5 .................................... 52 Table 2.11. Selected bond angles (?) for compound 5  ................................................... 52 Table 2.12. Selected interatomic distances (?) for compound 6..................................... 54 Table 2.13. Selected bond angles (?) for compound 6  ................................................... 54 Table 3.1. Equations for calibrating temperatures on NMR spectrometer  ..................... 65 Table 3.2. GPC data for the polymerization of ?-caprolactone and ?-butyrolactone by 3 .............................................................................................................................. 70  Table 3.3. Experimental results for polymerization of LA using 3 at various  temperatures  ................................................................................................................ 73  Table 3.4. Pm and Pr values for polymer samples from the polymerizations of rac-LA at 0 ?C and 25 ?C using 3 as a catalyst  .................................................................... 79  v  LIST OF FIGURES   Figure 1.1 Schematic representations of two possible forms of propagation errors: (a) enantiomorphic site control error, (b) chain end control error................................... 22 Figure 1.2. Schematic representation of propagation mechanisms (a) Bernoullian (b) first-order Markov  .................................................................................................. 24 Figure 1.3. 1H NMR characteristics of PLA .................................................................. 26 Figure 1.4. Indium catalysts for the polymerization of ?-caprolactone  .......................... 35 Figure 2.1. ORTEP view of (NNO)InMe2 (1). Most hydrogen atoms are omitted for clarity and thermal ellipsoids are drawn at 50% probability.  ................................... 41 Figure 2.2. ORTEP view of (NNO)InMe2?InCl(Me)2 (2).  Most  hydrogen  atoms and solvent molecules (pentane) are omitted for clarity and thermal ellipsoids are shown at 50% probability. ............................................................................................ 43 Figure 2.3. ORTEP view of [(NNO)(Cl)In]2(?-OEt)(?-Cl) (3).  Hydrogen  atoms omitted for clarity and thermal ellipsoids are shown at 35% probability  ....................... 48 Figure  2.4.  ORTEP  view  of  (NNO)(Cl2)In?Py  (4?Py).  Hydrogen  atoms  are omitted for clarity and thermal ellipsoids are shown at 50% probability  ....................... 50 Figure  2.5.  ORTEP  view  of  [(NNO)(Cl)In(?-OH)]2  (5).  Most  hydrogen  atoms omitted for clarity, and thermal ellipsoids shown at 50% probability ............................ 52 Figure 2.6. ORTEP view of  [(NNO)(Cl)In]2(?-OH)(?-Cl) (6). Hydrogen atoms are omitted for clarity and thermal ellipsoids are shown at 50% probability. ................. 53 Figure 3.1. Internal standard (1,3,5-trimethoxybenzene) for the polymerization of LA and its 1H NMR characteristics (CD2Cl2, 298K, 400MHz) ...................................... 64 Figure 3.2. Plot of the observed molecular weight (Mn = ?) and molecular weight distributions  (PDI  =  ?)  of  PLA  as  a  function  of  added  monomer  (calculated values for the molecular weights are shown using the line) ........................................... 70 Figure 3.3. Plot of Ln([LA]/[TMB]) versus time for two sequential additions of LA  (?  =  1st  addition  of  100  equivalents;  ?  =  2nd  addition  of  100  equivalents) (CD2Cl2, 298K, 400MHz) ............................................................................................. 72  Figure  3.4.  Dependence  of  the  observed  polymerization  rate  upon  the concentration of 3  ........................................................................................................ 73  Figure 3.5. Eyring plot for the polymerization of LA using 3  ....................................... 74  vi  LIST OF FIGURES CONTINUED     Figure 3.6. 1H NMR spectrum of the methine region of oligomeric PLA (CD2Cl2, 298 K,  400MHz) ......................................................................................................... 75  Figure 3.7. 1H{1H} and 13C{1H} NMR of PLA (CDCl3, 298 K, 600 MHz) from (a) polymerization of rac-LA using rac-3 (b) polymerization of rac-LA using RR,RR-3, and (c) polymerization of L-LA using rac-3.  ............................................................ 78  Figure 3.8. The observed rate of polymerization of LA as a function of the enantiomeric composition of the catalyst ...................................................................... 80    vii  LIST OF CHARTS   Chart 1.1. Molecular structure of poly(lactic acid) .......................................................... 1 Chart 1.2. Isomers of lactide generated from DL-lactic acid ........................................... 3 Chart 1.3: Organocatalytic reagents for the polymerization of lactide ........................... 14 Chart  1.4:  Important  and  common  ligand  motifs  in  catalysts  for  the polymerization of lactide  ............................................................................................. 19 Chart 1.5. Possible microstructures of PLA .................................................................. 25 Chart 1.6: Stereosequences of PLA: (a) syndiotactic PLA (b) heterotactic PLA (c) isotactic PLA (d) isotactic PLA with stereoerror (enantiomorphic site control) (e) isotactic PLA with stereoerror (chain end control) ........................................................ 28 Chart  1.7:  Representative  catalysts  for  (a)  enantiomorphic  site  control  and  (b) chain end control in the stereoselective polymerization of lactide.................................. 31     viii  LIST OF SCHEMES Scheme  1.1:  Transesterification  mechanism  for  the  polymerization  of  lactide (both intramolecular and intermolecular)......................................................................... 5 Scheme 1.2: The isomers of lactic acid and its direct condensation to form PLA ............ 7  Scheme 1.3: Metal-based industrial preparation of PLA ................................................. 8  Scheme 1.4: Proposed mechanisms for the anionic ring-opening of lactide  .................. 10 Scheme 1.5: Proposed mechanism for the cationic ring-opening of lactide ................... 12 Scheme  1.6.  Mechanism  for  polymerization  of  LA  by  nucleophilic  activation  of  the monomer....................................................................................................................... 15 Scheme 1.7: Metal-catalyzed coordination-insertion mechanism for the ROP of LA  .... 17 Scheme 2.1. Proposed routes to the synthetic target....................................................... 38 Scheme  2.2.  Synthesis  of  2-[[[(dimethylamino)cyclohexyl]amino]methyl]-4,6-bis(tert-butyl)phenol ..................................................................................................... 39 Scheme 2.3. Protonolysis routes to indium complexes .................................................. 44 Scheme 2.4. Synthesis of compounds 3 and 4................................................................ 47 Scheme 2.5. Synthesis of [(NNO)InCl(?-OH)]2 ............................................................ 51 Scheme 2.6. Proposed route to the formation of [(NNO)(Cl)In]2(?-OH)(?-Cl) (6) ........ 54 Scheme 3.1. Dissociation of 3 to generate the active catalyst 7?LA .............................. 68 Scheme 3.2. Monomer scope for the polymerization of lactones by 3. .......................... 69 Scheme 3.3. Coordination-insertion mechanism for ROP of LA by 7?LA  .................... 76 Scheme 3.4 Possible indium catalysts for the polymerization of LA ............................. 83   ix  LIST OF ABBREVIATIONS  Anal.      Analysis br      broad BINAP      2,2?-bis(diphenylphosphino)-1,1?-binaphthyl Bn      benzyl, -CH2(C6H5) i-Bu      iso-butyl, -CH2CH(CH3)2  n-Bu      n-butyl, -CH2CH2CH2CH3  t-Bu      tert-butyl, -CMe3  BuLi      n-butyllithium Calc?d      calculated COSY      Correlated Spectroscopy d      doublet  deg, (?)      degree(s) DEPT      Distortionless Enhancement by Polarization Transfer DMAP      4-Dimethylaminopyridine ESI      Electrospray Ionization equiv      equivalent(s) eq      equation(s) fac      facial EI      Electron Impact Et      ethyl   g      grams GPC      Gel Permeation Chromatography h      hour(s) h      Planck?s constant 1H{1H}      homonuclear decoupled proton  HETCOR    Heteronuclear Correlated Spectroscopy x  H2NNO      2-[[[(dimethylamino)cyclohexyl]amino]methyl]-4,6-bis(tert-      butyl)phenol HRMS      High Resolution Mass Spectrometry ITO      Indium tin oxide J      coupling constant in Hertz kb      Boltzmann?s constant kobs      observed rate constant for polymerization reactions K      equilibrium constant LA    Lactide LED    Light emitting diode LRMS    Low Resolution Mass Spectrometry   m      multiplet(s) MALDI-TOF    Matrix Assisted Laser Desorption-Time of Flight Mn      number average molecular weight Mw      weight average molecular weight Me      methyl MeLi      methyllithium Mes      mesityl, 2,4,6-trimethylphenyl mer      meridional  MWD      molecular weight distribution MS      mass spectroscopy NaOtBu      sodium-t-butoxide NOESY      Nuclear Overhause Enhancement Spectroscopy NHC      N-Heterocyclic Carbene NMR      nuclear magnetic resonance NNO-      monodeprotonated H2NNO NNO2-      doubly-deprotonated H2NNO obs      observed ORTEP      Oak Ridge Thermal Ellipsoid Plot xi  OTf      triflate,  CF3SO3- 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 ppm      parts per million PPY      4-pyrrolidinopyridine Pr       Probability of forming a racemic linkage i-Pr      iso-propyl, -CHMe2  n-Pr      n-propyl, -CH2CH2CH3 py      pyridine q      quartet R      Universal gas constant rac      racemic ROP      ring-opening polymerization s      singlet t      triplet TBD      Triazabicyclodecene THF      tetrahydrofuran TMB      1,3,5-trimethoxybenzene TMS      trimethylsilyl ?      chemical shift downfield from tetramethylsilane in ppm xii  ACKNOWLEDGMENTS     I would like to take this opportunity to thank a number of people who made this work possible. Thank you to my supervisor Parisa Mehrkhodavandi for her support and enthusiasm for this work. Thank you to the past and present members of the Mehrkhodavandi Group. I want to especially  thank  both  Insun  Yu  and  Jennifer  Fang.  Insun  for  growing  crystals  of  the (NNO)InMe2 complex and Jennifer for her tireless work in setting-up the lab. Thank you to the Gates group, especially Bronwyn Gillon and Kevin Noonan, for the use of their GPC instrument and  the time they spent helping me figure out how to use it.    Thank  you  to  the  all  the  UBC  Chemistry  Department  staff,  especially  Brian Snapkauskas and the mech shop guys who always found time to fix the numerous problems we had in setting up a new lab. A particular thank you should go to Brian Patrick for solving the crystal structures that are contained in this work. Especially for the extra time he took to solve the data for the structure of the catalyst. I would like to thank the NMR staff Nick Burlinson, Maria  Ezhova  and  Zorana  Danilovic  for  their  help  with  both  the  kinetic  studies  and  the homonuclear decoupled NMR.   Thank you to my family (my parents John and Frieda, and sisters Carole and Lauren) for your faith in my abilities, and the endless support. Last, but not least, I?d like to thank Tim for taking a chance and asking ?Do you want to grab a cup of coffee?? I couldn?t have done this without you (or the coffee).  1  CHAPTER 1. GENERAL INTRODUCTION  1.1 Introduction to Poly(Lactic acid)   In  recent  years  biodegradable  polymers  and  polymers  from  biorenewable  resources have  received  a  great  deal  of  attention  as  alternatives  to  traditional  petrochemical  based polymers. In particular poly(lactic acid) (PLA) has received a large amount of industrial interest (Chart 1.1). Cargill-Dow LLC and other companies have developed the infrastructure for both the manufacture and processing of PLA.1 This has caused a significant drop in the cost of the plastic and has led to an increased use of PLA in a number of applications such as the textile industry,  the  packaging  industry  for  both  food  and  horticultural  applications,  and  in  the pharmaceutical and biotechnology industries.2  Chart 1.1. Molecular structure of poly(lactic acid)     PLA?s many applications stem from its biodegradability and bioassimilability. Current industrial uses for PLA include its use in  ?green? disposable packages in the food and beverage industry. Under suitable conditions  such as composting the  plastic  containers  will  degrade to carbon  dioxide  and  water,  preventing  the  plastic  from  entering  the  traditional  waste  streams. PLA has been used in the medical industry for many years; one of its first applications was in 2  sutures for surgical procedures (a composite of PLA and poly(glycolic acid) is used). The sutures need  not  be  removed  and  are  gradually  degraded  by  the  body?s  natural  pathways.  Similarly, screws made of PLA are used for the fixation of bones or joints. Because the screws degrade over time, secondary surgeries are  not  required  in order to remove  the  unnecessary  hardware after the desired healing process has been achieved.    Current research, both industrial and academic, has focused on using PLA in controlled drug delivery by biodegradable devices. The drug is encapsulated in a polymer matrix that is implanted subcutaneously; the active drug is slowly and continually released over time as the polymer is degraded. Alternatively, nano or microparticles of drug-encapsulating polymers have also been explored for use in drugs administered either orally or nasally. In this case the polymer formulation might allow for more specific targeting of the drug to a particular active site.    PLA  microstructure  has  a  strong  influence  on  the  polymer  properties.  There  are  a number  of  possible  microstructures  for  PLA  due  to  the  different  ways  of  incorporating  the isomers of lactide  (LA) (Chart 1.2) into the polymer chain.  Poly(L-lactic acid) (PLLA), and poly(D-lactic acid) (PDLA) are both crystalline polymers with melting points of 180 ?C, while atactic PLA produced from the polymerization of DL-lactide is an amorphous polymer with a melting point of 130 ?C.  Stereocomplexes of PLA, a physical mixture of PLLA and PDLA, can have even higher melting points at 210 ?C. The processing methods for these polymers can also influence their properties.  3  Chart 1.2. Isomers of lactide generated from DL-lactic acid.      Degradation  of  PLA  occurs  through  two  mechanisms:  hydrolysis  of  the  polyester chain,  and  enzymatic  degradation.  Acidic  or  basic  conditions  can  be  used  to  catalyze  the hydrolysis; elevating both the temperature and humidity increases the hydrolysis rate. The rate of hydrolysis  is  dependent  on  a  number  of  factors,  including  the  molecular  weight  and  the crystallinity of the polymer. Once the polymer has been hydrolyzed into small oligomeric chains (< 40 000 g mol-1) bacteria and other microorganisms can further degrade the polymer to carbon dioxide  and  water  via  the  Kreb?s  cycle.    Although  PLA  degrades  relatively  easily  when compared to traditional plastics, further research into the infrastructure required for large-scale degradation of the plastic, as well as for monomer recovery/recycling is required. To that end, research has begun in order to find enzymes suitable for degrading PLA.    Lipases, such as proteinase K, are active in the biodegradation of PLA. Proteinase K is very  efficient  at  breaking  down  PLA;  however,  it  is  not  active  for  the  degradation  of  high molecular  weight  PLA.  A  two-step  degradation  pathway  involving  both  hydrolysis  and enzymatic  digestion  of  the  polymer  is  required  for  high  molecular  weight  PLA.1  Recently enzymes  from  a  yeast  Cryptococcus  have  been  isolated  and  used  to  degrade  high  molecular weight PLA, as well as other biodegradable polymers such as poly(caprolactone).3 It has been shown that enzymes have a strong preference for the degradation of PLLA versus PDLA. The 4  highest rates were observed for the degradation of the amorphous regions in samples of PLA that contained  atactic  PLA.4  Further  research  is  required  into  the  degradability  of  other microstructures of PLA.   Achieving  control  over  polymer  formulations  and  microstructures  is  often accomplished by using a living catalyst. As defined by the IUPAC Compendium of Chemical Terminology, a living polymerization is ?A chain polymerization from which chain transfer and chain termination are absent. In many cases, the rate of chain initiation is fast compared with the rate  of  chain  propagation,  so  that  the  number  of  kinetic-chain  carriers  is  essentially  constant throughout the polymerization.?5 If a polymerization is truly living, and follows these criteria, the molecular weight distribution of the polymer chains is uniform, and the rate of consumption of monomer is constant throughout the polymerization. It also follows that the molecular weight of the polymer will increase linearly with conversion until one hundred percent conversion is achieved.    If  no  termination  processes  are  present  in  the  system,  it  is  possible  to  restart  the polymerization after complete conversion via addition of more monomer.    There are several measureable characteristics for a living polymerization. The narrow molecular weight distribution (MWD) can be determined by the measurement of the molecular weights of the  polymer chains. MWD or polydispersity index (PDI) is defined  as the  weight average molecular weight divided by the number average molecular weight (Mw/Mn). Both the Mw and the Mn can be determined by gel permeation chromatography or GPC. GPC can also be used in conversion experiments to determine the molecular weights of polymer samples. The rate of consumption of monomer is usually demonstrated through kinetic experiments.   Transesterification is the major chain transfer mechanism in the polymerization of LA (Scheme 1.1).6 For intermolecular transesterification two growing polymer chains come together 5  and a random exchange  of polymer ends occurs.  This leads to a  scrambling  in the molecular weights  observed  for  the  polymers,  and  a  broadening  of  the  PDI.  Alternatively  the transesterification can occur through an intramolecular process causing the formation of cyclic polymers.7  In  the  case  of  both  inter-  and  intra-  transesterification,  lactate  subunits  from  the cleavage of a lactide unit can be  identified  by MALDI-TOF (matrix assisted  laser  desorption ionization-time of flight) mass spectrometry.8 Chain termination  processes for polymerization usually result from decomposition of the catalyst, or by introduction of a proton source, such as water, into the reaction mixture.  Scheme  1.1:  Transesterification  mechanism  for  the  polymerization  of  lactide  (both intramolecular and intermolecular).      If  the  chain  transfer  and  termination  mechanisms  can  be  prevented,  the  system  is living. A living system is desirable because it allows for precise control of polymer molecular weight  via  control  of  catalyst  loadings  and  reaction  times.  Polymer  compositions  can  be controlled through copolymerizations of different monomers; block copolymers can be formed 6  via control of the monomer feedstock. Due to this precise control, developing living systems is one  of  the  major  goals  in  polymer  research.  One  area  of  success  in  this  field  is  in  the development of  new  organometallic  complexes as  catalysts for  the  polymerization of  olefins. This has led to an increased knowledge base in the area of catalyst and ligand design, as well as a huge increase in research into developing  new catalyst  systems. In the body  of this work  the development of a catalyst for the living polymerization of lactide will be discussed.  1.2 Synthesis of PLA Industrially   PLA can be synthesized directly from the condensation of lactic acid. Lactic acid is produced  commercially  by  fermentation  of  food-crops  or  from  petrochemical  feedstocks. Biological based lactic acid is exclusively the L-isomer (Scheme 1.2) as it is fermented from the naturally occurring sugars. In contrast petrochemical based lactic acid is a racemic mixture of D- and L-lactic acid, and is prepared by the oxidation and functionalization of ethylene.2 In recent years the fermentation process has become more cost effective and it is now the favored method for producing lactic acid. Since lactic acid can be isolated by the fermentation of crops such as corn or other vegetation, it is considered a biorenewable feedstock.  7   Scheme 1.2: The isomers of lactic acid and its direct condensation to form PLA     One industrial route to PLA has been developed by  Mitsui Toatsu Chemicals. They have optimized, and patented an azeotropic dehydrative process that uses a condensation route to produce PLA.1 This process uses high temperatures and high boiling solvents in order to remove water  and  drive  the  polymerization  through  to  completion  producing  high  molecular  weight polymers.   The second industrial route to PLA, used by Cargill-Dow LLC involves metal-based ring-opening  polymerization  of  lactide  (LA),  and  is  outlined  in  Scheme  1.3.1  Lactic  acid undergoes a condensation to form low  molecular  weight  prepolymer. The prepolymer is  then depolymerized  using  a  tin  based  metal  catalyst  to  produce  lactide.  After  purification  by distillation, the lactide undergoes ring-opening polymerization (ROP) catalyzed by a different tin catalyst to form high molecular weight PLA. There are several advantages to this polymerization method. No solvents are used as all processes are carried out in the melt (m.p. of lactide is 116-119 ?C). This makes the Cargill-Dow process a green alternative to the solvent-based process used  by  Mitsui  Toatsu  Chemicals.  The  metal-based  catalysis  allows  for  good  control  over 8  polymer properties such as molecular weight. Since Cargill-Dow uses biologically synthesized lactic acid, the polymer produced by this technique is almost exclusively poly(L-lactic acid).  Scheme 1.3: Metal-based industrial preparation of PLA.   1.3 Ring-opening polymerization of LA   The industrial routes to synthesize PLA are not the only possible routes to formation of PLA;  there  are  many  routes  to  ROP  lactide  including  enzymatic,  anionic,  cationic, organocatalytic, and metal catalyzed polymerizations.9 The ring-opening process is favorable in all of these cases due to relief of a larger than normal ring-strain for lactide when compared to other  six-membered  rings.  In  this  case  the  high  degree  of  ring-strain  is  due  to  the  unusual geometric conformation imposed on the ring by the presence of the two ester groups in the ring.10 The ester groups are nearly planar forcing the ring into an irregular skew-boat conformation with the methine protons at the axial and the methyl substituents at the equatorial positions of the structure.  The ring-opening of the strained lactide drives the polymerization; standard enthalpy 9  of polymerization is ?23 kJ mol-1  and standard entropy of polymerization is ?40 J K-1  mol-1 for the formation of poly(L-lactic acid). The negative entropic value indicates a critical temperature, calculated to be 557  K (a value of 913K  was obtained for polymerizations  carried out in  the bulk), at which the polymerization will become non-spontaneous. 11   The  enzymatic  route  for  the  ring-opening  polymerization  of  lactide  uses  lipases  to catalyze the polymerization.12-14 Using an enzyme to catalyze the polymerization circumvents the need for purification of the polymer in order to remove metallic catalysts, which is necessary if the  polymer  is  to  be  used  for  medical  or  food  packaging  applications.  Most  lipases  that  are effective  in  catalyzing  the  ROP  of  LA  produce  high  molecular  weight  polymers  with  small values  for  the  polydispersity  index  (1.1-1.2).  Lipase  PS  produces  polymer  with  the  highest conversion rate and molecular weights. Lipase catalyzed polymerizations are carried out in the melt, avoiding the use of organic solvents; however, the polymerizations require lengthy reaction times (7-14 days) at elevated temperatures (80?130 ?C).13 More recently, the use of this route along with multifunctional alcohols has produced branched polymers of lactide. Branched PLA differs  from  linear  PLA  in  its  thermal,  physical  and  mechanical  properties.  It  has  also  been shown  that  branching  increases  the  rate  at  which  PLA  is  degraded  by  both  enzymatic degradation and base hydrolysis. 12   Anionic polymerization  of  lactide was  one  of the earlier methods developed for  the controlled polymerization of lactide. Starting with the pioneering work of Kricheldorf in the late 1980?s and 1990?s it was discovered that strong bases such as alkyl lithium reagents and highly basic  alkoxides  were  active  for  the  polymerization  of  lactide  at  room  temperature.6,15  Two mechanisms were proposed (Scheme 1.4). The first involves deprotonation of the monomer. The methine proton is removed from the lactide, forming an enolate type structure, which can then 10  ring  open.  The  remaining  anionic  chain  acts  as  a  propagating  species  and  continues  the polymerization. The second mechanism involves nucleophilic attack by the base on lactide and subsequent ring opening of the monomer. In this case initiator fragments would be incorporated as chain ends in the growing chain. This is not observed for initiators such as n-butyl lithium or potassium tert-butoxide. The easily identifiable tert-butyl ester end-group is not observable by 1H NMR spectroscopy and in the case of the butyl reagent the evolution of butane is observed. This evidence supports the deprotonation route for the polymerization of lactide by anionic initiators.  Scheme 1.4: Proposed mechanisms for the anionic ring-opening of lactide     Competitive processes occur during anionic polymerization. The polymerizations are low  yielding,  and  the  correlation  between  observed  molecular  weight  versus  monomer  to initiator ratio is non-linear. As the polymerization proceeds, the polydispersity of the polymer samples  increases.  This  is  an  indication  that  transesterification  is  occurring  during polymerization. For anionic polymerizations, this process is more prevalent at longer reaction times and elevated temperatures.   11    Racemization  of  the  lactide  monomer  is  a  phenomenon  observed  during  anionic polymerization of lactide.6 It has been shown that for both strong and relatively weak bases can cause  racemization  of  the  monomer.  As  shown  by  the  deprotonation  mechanism  for  anionic polymerization  (Scheme  1.4),  an  equilibrium  exists  between  the  protonated  and  deprotonated monomer, which allows for scrambling of the stereogenic centre. In the polymerization of L-lactide, racemization of the monomer occurs throughout the polymerization and it increases with longer  reaction  times  as  observed  by  the  optical  rotation  of  polarized  light  by  the  polymer samples  isolated  at  different  times.  This  indicates  that  the  chain  ends  also  play  a  role  in  the racemization of the monomer. An increase in reaction temperature also leads to an increase in the scrambling of the stereocentres.  Lower temperatures and initiator loadings can minimize the amount of racemization, although these limitations consequently increase the reaction times.    Cationic polymerization of lactide has also been explored. Kricheldorf laid down the groundwork in the field with his investigation of methanesulfonic acid and methyl triflate as catalysts  for  the  ring-opening  polymerization  of  lactide.16  He  found  that  the  cationic  catalyst could readily polymerize lactide, with no epimerization of monomer stereocentres. He also found that  the  counter  anion  for  the  polymerization  played  an  important  role  in  whether  the  cation would  polymerize  lactide.  Simple  Lewis  acids  such  as  BF3?OEt2,  or  zinc  fluoride  were  not successful  as  catalysts,  and  neither  were  a  variety  of  organic  acids  (trifluoroacetic  acid, methanesulfonic  acid,  and  fluorosulfonic  acid).    A  number  of  alkyl  sulfonates  were  tried  as catalysts of these methyl triflate was the only active catalyst.17 The reactions required several days  at  room  temperature  in  order  to  reach  completion,  and  if  the  temperature  was  elevated above 50 ?C racemization of the monomer became an issue. Another critical problem with the cationic  approach  was  lack  in  correlation  between  polymer  molecular  weight  and  the  initial catalyst  loading;  the  relative  viscosities  of  the  polymer  samples  showed  no  significant 12  difference  for  samples  made  from  different  catalyst  loadings.  These  results  indicate  a  severe limitation in the use of cationic catalysts for the polymerization of lactide.     The proposed mechanism for the cationic polymerization of LA follows an activated monomer route Scheme 1.5. In this case the acid activates the carbonyl of the monomer making it susceptible to nucleophilic attack. The ring is opened via cleavage of the acyl bond. This ring-opened product can then act as the nucleophile for propagation of the polymerization.16  Scheme 1.5: Proposed mechanism for the cationic ring-opening of lactide     More  recently  the  scope  of  cationic  polymerization  has  been  expanded  to  lactone monomers  other  than  lactide.    Both  ?-caprolactone  and  ?-valerolactone  were  polymerized  by combining HCl and an alcohol (butanol and more complex multifunctional alcohols).18,  19 The polymer samples displayed narrow molecular weight distributions and high yields of polymer for short  reaction  times; furthermore,  adjustment  of the  monomer  to  initiator  ratio  controlled  the molecular  weight  of  the  polymers.  The  catalyst  system  was  also  used  to  produce  block copolymers of the two monomers. Considering the simplicity of the reagents for catalysis, the cationic  route  to  synthesizing  polyesters  seems  highly  desirable;  however,  several  limitations still  exist.  Lactide  is  less  readily  polymerized  than  either  caprolactone  or  valerolactone.  To address this, Bourissou and co-workers re-examined the earlier work carried out by Kricheldorf, 13  and optimized his system for the polymerization of lactide.   Bourissou?s group uses a combination of a protic reagent such as simple alcohol and trifluoromethanesulfonic acid in order to polymerize lactide.20 This differs from the work carried out by Kricheldorf. In Kricheldorf?s case he did not add a secondary protic agent as a co-catalyst. The system gives polymer samples with high conversions, low molecular weight distributions (PDI < 1.5), short reaction times (< 4h for 125 equivalents) and a linear relationship is observed between molecular weight and catalyst loadings.  By 1H NMR analysis of the polymers it can be seen that the end groups for the polymer are consistent with acyl-bond cleavage; namely, both ester and hydroxyl chain ends were observed.    Another  rapidly  expanding  area  in  the  polymerization  of  lactide  is  in  the  use  of organocatalysts. Where an organocatalyst is defined as an organic compound that in the absence of a metal species acts as a catalyst for a given reaction. Both pyridine-based nucleophiles, and N-heterocyclic carbenes (NHCs) have demonstrated activity in the ring-opening polymerization of lactide (Chart 1.2).8, 21-24 All of the organocatalytic catalysts have the characteristics of a living system;  they  show  narrow  MWD,  linear  conversion  of  monomer  to  polymer,  and  a  good correlation between the molecular weight of the polymers and the monomer to initiator ratios.  14   Chart 1.3: Organocatalytic reagents for the polymerization of lactide    The polymerization mechanism for the organocatalysis follows an activated monomer pathway  (Scheme  1.6).24  The  first  step  in  the  mechanism  is  the  formation  of  a  base-lactide complex  by  attack  of  the  carbonyl  of  the  monomer  by  the  strong  organic  nucleophile.  The nucleophilic attack causes the lactide ring to open via an acyl bond cleavage. The lactide-base complex  then  undergoes  nucleophilic  attack  by  either  the  original  alcohol  initiator  or  by  the terminal  ?-hydroxy  group  on  the  end  of  a  growing  polymer  chain,  releasing  the  original organocatalyst.  The  chain  is  propagated  by  further  nucleophilic  attack  of  the  chain  end  ?-hydroxy group on activated lactide monomers. 15   Scheme 1.6. Mechanism for polymerization of LA by nucleophilic activation of the monomer     If  an  initiator  and  a  carbene  are  used  in  the  polymerization  of  LA,  linear  PLA  is produced.22,  23 Either an alcohol can be added to the reaction mixture or a masked carbene (in Chart  1.2  the  alcohol  adduct  of  the  saturated  carbene)  can  be  used. In  this  case,  the  masked carbene acts as the source for both the initiator and the base; dissociation of the adduct results in the  formation  of  ROH  and  the  unsaturated  carbene.22  When  an  initiator  is  absent  from  the reaction mixture and the NHCs are used cyclic oligomers or polymers are formed exclusively.8 In the case of the NHC polymerization in the absence of an initiator a zwitterion is formed upon ring opening, with the positively charged carbene bound to one end of the chain. The chain can propagate  by  the  mechanism  outlined  above.  However,  termination  of  chain  growth  occurs through a slightly different mechanism. Coulombic forces between the chain ends cause them to come together, releasing the NHC and a cyclic polymer. The NHC can then initiate the growth of a new polymer chain.   The  reaction  times  for  polymerization  of  lactide  using  carbenes  are  rapid.  In  a 16  representative reaction conversion of 200 equivalents of LA to PLA is achieved in two minutes, making carbenes the most active known catalysts for the polymerization of lactide. Interestingly, the  very  sterically  bulky  carbenes  also  show  some  degree  of  stereoselectivity  during polymerization.23    The  final  mechanism  to  be  discussed  is  the  metal  mediated  ring-opening polymerization of lactide. This is perhaps the most widely explored area in the catalysis of LA polymerization. It has been found that a number of metals are capable of polymerizing lactide including magnesium25-33, calcium,28, 34-37 yttrium,38-49 the lanthanides,50-56 tin,57-60 iron,61-64 zinc,25, 26,  28,  29,  31,  55,  65  aluminum,7,  25,  40,  55,  66-74  and  germanium.75,  76  Simple  metal  alkoxides  of  iron64, tin,77yttrium,78,  79  the  lanthanides,78,  79  calcium,80  and  aluminum7  are  effective  catalysts. Unfortunately  the  polymers  produced  from  these  catalysts  have  large  molecular  weight distributions, and analysis by MALDI-TOF MS shows that transesterification is occurring during polymerization. Due to these disadvantages, non-labile ancillary ligands were used in order to introduce steric bulk at the metal centre in the hopes of preventing the deleterious side reactions. A general formula for these catalysts is LnMR where Ln are ancillary ligands that do not play an active role in the polymerization, but do strongly influence the metal centre and its behaviour in the polymerization. M is a Lewis acidic metal centre, and R is the initiating group. In general the most effective initiator group is an alkoxide, but alkyls, amides, and halides have been found to initiate the polymerization. However, it was not conclusively shown that protic impurities in the monomer such as lactic acid, alcohols or water were not acting as the initiator in these cases.    These  metal  mediated  polymerizations  are  thought  to  occur  through  a  coordination insertion  mechanism  (Scheme  1.7).  The  first  step  in  the  mechanism  is  coordination  of  one molecule  of  lactide  to  the  active  metal  centre  via  a  dative  bond  between  the  oxygen  of  the 17  carbonyl and the metal. This is followed by insertion of the monomer into the metal alkoxy bond via a nucleophilic attack at the carbon of the carbonyl by the initiator group (the alkoxy group). The ring is then opened via cleavage of the acyl bond. In the case of an alkoxy initiator this leads to an ester on the polymer chain end, and  an alcohol end group if hydrolysis  of  the catalyst occurs.  Experimental  evidence  for  this  mechanism  was  first  shown  in  the  late  1980?s  by  the groups of Kricheldorf and Teyssi?. 7, 81   Scheme 1.7: Metal-catalyzed coordination-insertion mechanism for the ROP of LA.     Some of the first ancillary ligand systems developed for the polymerization of lactide were  the  metalloporphyrin  systems  synthesized  by  Inoue  and  co-workers  in  the  late  1980?s (Chart 1.4).71,  82 These systems gave polymers of narrow molecular weight distribution, in high yield.  Polymerization  of  ?-valerolactone  showed  that  the  reaction  showed  second-order dependence on the catalyst concentration. A mechanism by which two aluminum centres play a role in the polymerization of the monomer was proposed. In this mechanism the initiator group of one catalyst acts as nucleophile in an attack on the carbonyl of the monomer, which is being 18  activated by the other equivalent of the catalyst. In this case the aluminum centre acts as a Lewis acid to activate the carbonyl of the monomer. The reaction itself is not very rapid (the conversion of  100  equivalents  of  lactide  went  to  completion  after  96  h)  and  the  reaction  requires  high temperatures (100 ?C) to proceed.   It was postulated that a more flexible ligand might allow for a unimolecular pathway for chain growth, because of the less constrained geometry at the metal centre. The incoming monomer and the initiator have to have a cis-geometry in order for the coordination-insertion mechanism  to  be  viable.  In  the  case  of  the  porphyrinato  ligands  the  metal  is  completely encompassed  by  a  very  rigid  equatorial  ligand,  leaving  only  two  trans  axial  sites  for  the coordination of the initiator and the monomer.   To  this  end,  the  Bertrand  group  developed  aluminum,  zinc,  samarium  and  tin diamidoamino complexes for the polymerization of lactide.68,  83-85 The complexes of aluminum have trigonal monopyramidal geometry. A similar geometry was seen for the tin analogue while the zinc and samarium complexes were dimeric. The tin system was found to be the most active when bulk copolymerizations with glycolide were attempted. However, the PDI for the polymer samples were large (3 for polymerizations at 180 ?C, and 2 for polymerizations at 140 ?C).   19  Chart 1.4:  Important and common ligand motifs in catalysts for the polymerization of lactide     A  bulkier  tridentate  ligand  was  used  to  prevent  transesterification  reactions.  The sterically bulky trispyrazolyl-hydroborate ligand was used on magnesium, calcium and zinc to form  complexes  for  the  polymerizations  of  lactide.  26,  28,  30,  34  It  was  found  that  the  calcium analogue was the most active, with the others falling into the series Ca > Mg > Zn. This was attributed to the polarity of the metal-initiator bond. The calcium analogue could polymerize 100 equivalents  of  lactide  in  one  minute  at  room  temperature.  The  trisindazolyl-hydroborate analogues of zinc and magnesium were also synthesized. The change in the ligand caused an increase in the activity  of the catalysts. The  systems showed a first  order  dependence for  the polymerization upon the concentration of the catalyst, but the PDI values were quite high for the polymer samples (1.6-1.7).    Salen based catalysts have been widely studied for the polymerization of LA. Inspired by the porphyrinato systems of Inoue, the range of salen catalysts includes aluminum, 67, 69, 73, 74, 86 20  yttrium,40 tin,87 and zinc analogues.88 Changes in the ligand backbone can play a major role in the activity of the catalyst. It was found that when the backbone is flexible the rate of polymerization is higher than when a rigid backbone is used. 69 Steric bulk at the R1 position of the ligand (Chart 1.4)  slows  the  rate  of  polymerization  but  enhances  the  stereoselectivity.  The  polymerizations rates are first-order in catalyst concentration, and the polymer samples contain ester end groups indicating  a  coordination  insertion  mechanism  with  acyl  bond  cleavage  (Scheme  1.7).  The polymers  have  narrow  molecular  weight  distributions  and  a  linear  relationship  is  observed between the conversion and molecular weight, indicating the salen catalysts are living.    The  salan  analogues  of  these  complexes  have  also  been  synthesized  (the  imine  is reduced to a secondary amine or reductively aminated to give the tertiary amine) and are active for polymerization of lactide.66  A half-salan based ligand developed by Hillmyer and Tolman was a landmark catalyst for the polymerization of LA; it was the most active catalyst at the time of the report.65 The high activity of the tridentate system was attributed to facile coordination of LA to the less coordinatively saturated metal centre.   ?-Diiminate complexes have also received a large amount of attention as catalysts for polymerization. Analogues of magnesium,25, 89, 90 tin,91 iron63 and zinc25, 89, 90, 92 catalysts have been reported. In general, these catalysts display the characteristics of living systems. The magnesium ?-diiminate catalysts are among the most active catalyst currently known for the metal-mediated ring-opening polymerization of LA; the  zinc  systems are  also  highly active and exhibit  some stereoselectivity in the polymerization. 65 21  1.4 Stereoselectivity in polymerization   As  mentioned  earlier  in  the  discussion  of  the  biodegradability  of  PLA,  the microstructure of a polymer can have a very significant impact on its macroscopic properties. There  are  a  number  of  different  microstructures  that  are  possible  for  polymers.  The stereoselectivity in olefin polymerizations has been extensively studied since Natta discovered that metal-based catalyst could control the stereoregularity of olefin polymerization.93 To date a large number of systems are available for the controlled polymerizations of olefins. For olefin polymerization one variance in microstructure is based on regioisomerism, which is based on the number of head-to-head, head-to-tail and tail-to-tail linkages in the polymer backbone. Different microstructures  can  also  be  based  on  the  stereochemical  configuration  of  the  polymer. If  the given monomer for a polymerization is optically active (i.e. it is chiral) or pro-chiral the resulting polymer can be optically active. In this case the relative configuration of chiral centres in the polymer backbone determines the microstructure. A chain that contains sequential stereocentres of the same relative stereochemistry is isotactic while a chain that contains sequential alternating stereocentres is syndiotactic. If a chain exhibits no regularity in the relative stereocentres then the polymer is atactic.    NMR  spectroscopy  of  polymers  is  a  very  powerful  technique  in  the  determination  of polymer microstructure.94 Short polymer sequences such as triads, tetrads, pentads, or hexads (a triad is a sequence of three monomer units, a tetrad is a sequence of four monomer units, etc) can be  assigned  to  their  various  stereosequences  based  on  their  distinct  chemical  shifts.  The stereosequences  are  labeled  according  to  the  type  of  linkages  in  the  monomer  sequences.  A linkage between monomers can be either meso or racemic: a meso linkage indicates that the two monomers  have  the  same  relative  stereochemistry;  a  racemic  linkage  indicates  that  the  two 22  adjacent  monomers  have  opposite  stereochemistry  (Figure  1.1).  Both  1H  NMR  and  13C  {1H} NMR  have  been  used  extensively  to  analyze  polymers,  and  with  the  advances  in  two-dimensional techniques COSY, HETCOR, and NOESY have become increasingly used in the analysis of polymers. Nuclei such as 13C, 31P, and 15N nuclei are particularly useful for assigning stereosequences;  their  wide  chemical  shift  ranges  allow  for  better  resolution  of  peaks,  and assignment of more complex microstructures.    Figure  1.1  Schematic  representations  of  two  possible  forms  of  propagation  errors:  (a) enantiomorphic site control error, (b) chain end control error.    Two mechanisms are possible for stereoselective polymerizations by single-site metal-based  catalysis:  enantiomorphic  site  control  and  chain-end  control.  In  enantiomorphic  site control a chiral catalyst will selectively polymerize one enantiomer of a monomer preferentially over the opposite enantiomer. In chain-end control the stereoselectivity of the polymerization is controlled  by  the  chirality  of  the  last  incorporated  monomer  in  the  growing  chain.  The mechanism by which stereocontrol is achieved is hard to determine; the polymers resulting from the two mechanisms can have very similar microstructures. As well, the two mechanisms can act in tandem to provide the observed stereocontrol. This interaction between chain end control and site control can either be constructive (increasing the stereoselectivity) or destructive (decreasing the stereoselectivity).       23    Stereoerrors occur during polymerization for systems that are not ideal. Occasionally the  ?wrong?  enantiomer  is  incorporated.  The  way  these  stereoerrors  are  propagated  in  the growing polymer chain can provide invaluable evidence for determining which mechanism is responsible for stereoselectivity.  In the case of enantiomorphic site control once an error occurs it  is  corrected  by  the  selectivity  of  the  catalyst,  and  the  ?right?  enantiomer  continues  to  be polymerized, while in chain end control, the enantiomer being polymerized is switched once an error occurs (Figure 1). Errors introduced during chain-end control can lead to stereoblocks the length of which are determined by the degree of selectivity exhibited by the catalyst; the more selective the catalyst the larger the length of the isotactic blocks.     Mathematical  models  have  been  used  in  order  to  understand  the  way  in  which polymers propagate. From these models probabilities  for the different stereosequences  can  be calculated. The two most common models are the Bernoulli trial and first-order Markov steps. By definition a Bernoullian system involves two possible events each with a given probability of occurring (Figure 1.2).  In this case, the two events are either a meso or racemic linkage being formed with  the probability for each of these  events being Pm and  Pr.  These probabilities  are independent of the previous event; the chirality of the next incorporated monomer is independent of  the  chain  end.  This  corresponds  to  a  system  in  which  enantiomorphic  site  control  is  the method by which stereoselectivity is achieved. In contrast, for the Markov model the probability of the next event occurring is dependent on the current state of the system; therefore for a two-event  system  (the  formation  of  either  a  meso  or  racemic  linkage)  there  are  four  different probabilities Pm/m, Pm/r, Pr/r. and Pr/m. The  Markov  model is  used to describe probabilities in a chain-end  controlled  polymerization.  Higher  order  Markov  models  take  into  account  the stereochemistry of additional chain units. 24   Figure 1.2. Schematic representation of propagation mechanisms (a) Bernoullian (b) first-order Markov    1.5 Stereoselectivity in the polymerization of LA Lactide contains two stereocentres; therefore a large number of microstructures are possible in the ROP of lactide. In addition to isotactic, syndiotactic, and atactic polymers, microstructures containing  heterotactic  PLA,  stereoblocks  of  PLA,  as  well  as  stereocomplexes  of  PLLA  and PDLA can be formed (Chart 1.4).70 It is possible to form isotactic, heterotactic, stereoblock, and stereocomplexes of PLA from the polymerization of rac-lactide. In the case of heterotactic PLA the monomer being polymerized must alternate after each ring-opening event.  Syndiotactic PLA 25  can  only  be  synthesized  from  meso-LA;  the  only  way  to  obtain  perfectly  alternating stereocentres in the backbone is if the monomer itself contains both an R, and an S-centre.  The catalyst  then  has  to  preferentially  ring-open  one  stereocentre  of  the  meso-lactide  in  order  to obtain the syndiotactic polymer.95 If the catalyst alternates which acyl bond is cleaved for meso-LA  then  a  heterotactic  microstructure  is  obtained.92  Isotactic  PLA  can  be  obtained  by polymerizing  enantiopure  monomer  or  by  the  chiral  resolution  of  rac-LA  using  an enantioselective catalyst. It has been shown that rac-Salen catalysts can be used to polymerize rac-LA to produce stereocomplexes of PLLA and PDLA. In this case the R-catalyst polymerizes D-LA preferentially to form PDLA, and the S-catalyst polymerizes L-LA to form PLLA.71-73, 82   Chart 1.5. Possible microstructures of PLA      Three  methods  are  possible  for  the  formation  of  stereoblock  PLA.  Control  of  the feedstock by sequential addition of first D-LA followed by L-LA upon complete conversion of the  D-LA  can  produce  stereoblock  isotactic  PLA.  The  polymerization  of  rac-lactide  with 26  enantiopure  catalyst  can  also  produce  isotactic  stereoblock  PLA,  although  in  this  case  the microstructure  is  most  likely  a  tapered  stereoblock  with  the  percent  composition  of  the  two monomers changing over the length of the chain. Alternatively, a chain transfer mechanism has been proposed to explain the experimental observation of the formation of stereoblock PLA from the polymerization of rac-LA with rac-catalyst.  Coates proposes that rapid chain transfer occurs between two enantiomers of the catalyst. He uses the relative ratio of different stereosequences in the polymer backbone to justify the proposed mechanism.40   The 1H NMR spectrum of a sample of PLA contains two sets of peaks corresponding to the  methine  and  methyl  protons.  The  methine  protons  only  couple  to  the  adjacent  methyl protons,  and  are  observed  as  a  number  of  overlapping  quartets,  while  the  methyl  region  is composed of overlapping doublets. Homonuclear  decoupling experiments  are  commonly used for simplifying the 1H NMR spectra of PLA samples. In these experiments the methyl region is constantly irradiated in pulses during the collection time. This constant irradiation saturates the excited state for the methyl protons, eliminating spin-spin coupling relaxation processes with the methine  protons.  Without  the  coupling  to  the  methyl  protons  the  overlapping  quartets  are simplified to a series of  singlets in the methine region. These singlets correspond to different tetrad sequences: rrr, mmm, mrm, rmr, mmr, rmm, rrm, and mrr.   Figure 1.3. 1H NMR characteristics of PLA.  27    Polymer samples that are obtained from the polymerization of rac-LA can contain five of these singlets: mmm, mrm, rmr, mmr, and rmm. It should be noted that two peaks are observed for the mmr, and the rmm tetrads but the two stereosequences are indistinguishable by 1H NMR. The  rrr,  rrm,  and  mrr  are  only  observed  if  meso-LA  is  polymerized  or  if  scrambling  of  the stereocentres  occurs  during  polymerization  either  by  racemization  of  the  monomer  or  by scrambling  of  the  chains  via  transesterifcation.  The  homonuclear  decoupled  1H  NMR  of heterotactic PLA contains two peaks of the same relative intensity corresponding to the rmr and mrm  stereosequences,  while  syndiotactic  and  isotactic  PLA  contain  only  one  resonance corresponding  to  stereosequences  of  rrr  and  mmm  respectively.  The  rmr,  mmr,  and  rmm sequences are present in atactic PLA, or in isotactically enriched PLA that contains stereoerrors (Chart  1.5).    Stereoerrors  in  syndiotactic  or  heterotactic  PLA  will  give  the  rrm,  mrr stereosequences. By examining the relative integration of the tetrads it is possible to determine both  the  overall  microstructure  of  the  polymer  sample  and  the  mechanism  by  which stereoselectivity is occurring.  28  Chart 1.6: Stereosequences of PLA: (a) syndiotactic PLA (b) heterotactic PLA (c) isotactic PLA (d)  isotactic  PLA  with  stereoerror  (enantiomorphic  site  control)  (e)  isotactic  PLA  with stereoerror (chain end control)     As  was  discussed  previously,  the  way  in  which  stereoerrors  are  propagated  during polymerization can elucidate the mechanism behind the stereoselectivity of a given catalyst. In the case of PLA if the stereocontrol is due to enantiomorphic site control the ratio for the relative integration of the tetrads will be 1:2:1:1 for mmr, mrm, rmr, and rmm (Chart 5). When chain end control is the mechanism by which stereocontrol is achieved isotactic stereoblocks are formed and the rmr tetrad is absent while the remaining sequences are found in a ratio of 1:1:1 (mmr, mrm, rmm). 40, 94 29    The relative integration of the different stereosequences in the polymer can also be used to  determine  the  probability  of  a  meso  (Pm)  or  racemic  (Pr)  linkage  forming  during  the polymerization (Table 1.1).25, 94 These values can then be correlated to the overall microstructure: Pm values close to one indicate entirely isotactic PLA, and Pr values close to one indicate entirely syndiotactic PLA. Heterotactic PLA  also has a  Pr value  of  one  since  linkages  formed during polymerization of rac-lactide, and not the inherent linkages in the monomer contribute to the value of Pr. The assignment of r and m linkages in the polymer backbone is shown in Chart 5, and is based on connections between lactic acid subunits in the polymer chain (not lactide units).  Table 1.1. Probability of tetrad sequences in PLA based on Bernoullian Statistics.   Probability Tetrad  rac-LA  meso-LA [mmm]  Pm2 + PrPm/2  0 [mmr]  PrPm/2  0 [rmm]  PrPm/2  0 [rmr]  Pr2/2  (Pm2 + PrPm)/2 [rrr]  0  Pr2 + PrPm/2 [rrm]  0  PrPm/2 [mrr]  0  PrPm/2 [mrm]  (Pr2 + PrPm)/2  Pm2/2     Analysis of the methine and carbonyl regions of the 13C NMR spectrum has also been used to determine the microstructure of PLA samples; the carbon NMR is used to confirm the assignment from the homonuclear decoupled 1H spectrum. In the 13C NMR spectrum the hexad stereosequences are resolved in the carbonyl region, and the most commonly used assignments were  made  by  Kasperczyk.15  The  resolution  in  the  methine  region  is  slightly  lower  although assignment at the tetrad level can be achieved. The Pr and Pm values can also be calculated from 30  the  intensity  of  the  13C  NMR  peaks  analogous  to  the  methods  used  for  the  homonuclear decoupled spectra.   A number of catalysts have been shown to be stereoselective in the polymerization of lactide. Several chiral catalysts have been shown to exhibit enantiomorphic site control in the polymerization of rac-LA to produce isotactic PLA. These include the chiral aluminum salen complexes  used  by  Coates  et  al.  In  this  case  a  binapthyl  backbone  was  used  to  create  a symmetric salen ligand, Feijen et al later used a diaminocyclohexane moiety in the salen ligand to produce another stereoselective catalyst.67, 73 In contrast achiral salen,69, 74 and salan complexes have been shown to  be stereoselective  in the polymerization of lactide via chain end control. Bulky  ?-diiminate  zinc  complexes25,  29  and  the  sterically  bulky  calcium  complexes  of trispyrazolyl  hydroborate34  are  also  stereoselective  via  a  chain-end  control  mechanism.  Other complexes  with  bulky  ligands  such  as  the  trisphenolate  titanium96  or  the  amine  trisphenolate germanium complexes75 have also demonstrated good stereoselctivity via chain end control. An interesting  thiol  containing  salen  type  complex  of  scandium  synthesized  by  Okuda  and  co-workers,  shows  chain-end  stereocontrol  of  the  polymerization  through  fluxionality  in  the ancillary ligand.56 The catalysts that exhibit chain end control usually produce heterotactic PLA from the polymerization of rac-LA. 25, 29, 34, 42, 56, 92, 96 31   Chart 1.7: Representative catalysts for (a) enantiomorphic site control and (b) chain end control in the stereoselective polymerization of lactide.     The previously discussed NHC organocatalysts also exhibit moderate stereoselectivity in  the  polymerization  of  LA.23  The  Pm  values  for  the  systems  range  from  0.55  to  0.59  for reactions carried out at 25 ?C depending on the steric bulk of the substituents on the nitrogens of the carbene, with the highest values being obtained for mesityl substitution. When the reaction is cooled down to -70 ?C this stereoselectivity increases up to 0.9 for the value of Pm indicating that almost entirely isotactic segments of PLA have been formed. A chiral carbene was used in order to  see  if  enantiomorphic  site  control  could  play  a  role  in  the  stereoselectivity  of  the polymerization  but  no  greater  control  was  observed;  this  supports  a  chain-end  control mechanism for the stereoselectivity of all the catalysts including the chiral carbene.    32  1.6 General Chemistry of Indium and Group 13 Metals   The success of aluminum catalysts for the stereoselective and controlled ROP of LA has  led  us  to  investigate  the  activity  of  indium  complexes  in  this  role.  However,  group  13 elements are varied in their reactivity and properties.97  Boron is perhaps the most obvious outlier when it comes to Group 13. It is a covalently bonded non-metallic insulator, while aluminum, gallium, indium and thallium are soft metals with low melting points, and low resistivity. The valence electronic configuration for the group is ns2np1. For Ga and In the valence orbitals also contain a filled d-orbital, and Tl possesses a filled f-orbital. Boron is also amphoteric in nature, while aluminum and gallium are only minimally amphoteric, and indium is even less so. The group displays an unusual trend in ionization energies (B > Tl > Ga > Al > In). For Groups 1 and 2 an increase in ionization energy is observed with an increase in atomic number; however, for Group 13 this trend is interrupted at both gallium and thallium due to d-block contraction and f-block contraction respectively.  The electronegativiy of the atoms increases from Al to Tl.   For  compounds  of  Group  13  elements  the  most  common  oxidation  state  is  the  +3 oxidation state. Thallium is the exception with its most common oxidation state being +1.  This is attributed to a decrease in stability of the +3 oxidation state with increased atomic size; the energy required to involve the ns2 electrons in bonding is greater than the energy released in the formation of the bonds.  The relative stability of the  +1 oxidation state for  Group 13 metals increases down the group: Al < Ga < In < Tl. Tl tends to behave more like the alkali metals in its reactivity. It is highly basic and upon exposure to air it forms TlOH and hydrogen gas.   The Lewis acidity of the Group 13 metals follows the general trend: AlX3 > GaX3 > InX3 (X = halogen) with AlCl3 being the most Lewis acidic. Depending on the nature of the atom that is acting as an electron donor (N, O, or S) the acceptor ability of the metal changes. In terms 33  of polarizability B and Al are considered hard acceptors, and Ga and In are considered soft. For donation from nitrogen or oxygen containing compounds the above trend is observed, but when sulfur-containing  donors  are  used  the  large  more  polarizable  gallium  becomes  the  strongest acceptor  (GaX3  >  AlX3  >  BX3  (X  =  Cl,  Br)).  For  neutral  adducts  of  the  trihalides  the stoichiometry for monodentate ligands is generally 1:1 for Al and Ga (MX3L). However for In the most common stoichiometry is 1:3 (MX3L3).    Elemental indium is naturally occurring, but was not isolated in large amounts until the 20th century. It is the least abundant of the Group 13 elements; it is found in the earth?s crust with an abundance of 0.05 ppm.98 China and Canada are two of the top exporters of indium, which is isolated as a by-product in the smelt processing of zinc. Demand for indium has increased in recent years due to its extensive use in the materials industry. Indium tin oxide (ITO) films are used  in  light  emitting  diodes  (LEDs)  and  accounted  for  84%  of  the  global  consumption  of indium. Low molecular weight indium complexes such as trimethylindium have been used in the semiconductor  industry  as  precursors  in  metal  vapor  deposition  of  thin  films.    The  price  for elemental indium has risen from $170 US/kg in 2003 up to $918 US/kg in 2006.   The coordination chemistry of indium is  not as well known as that of aluminum or even  gallium.  The  natural  abundance  (8.3%  by  weight  of  the  Earth?s  crust)  and  low  cost  of aluminum have played a role in the amount of research in aluminum chemistry.97 This along with its notable success in many catalytic processes has led to extensive research into many of the aspects of aluminum chemistry. In contrast, the relatively expensive and less available indium has  not  received  such  interest;  although,  it  has  garnered  notice  recently  as  a  robust  and interesting Lewis acid for organic catalytic transformations.    From the late 1990?s to early 2000?s an exponential growth was seen in the number of 34  organic transformations carried out using InX3 (X = Cl, Br, I, or OTf) as a Lewis acid catalyst. 99-102 Indium has been used in many different types of reactions including additions, cyclizations, aromatic electrophilic substitutions, rearrangements, and coupling reactions. The transformations can usually be carried out under mild conditions with high yields and in some cases very high selectivity. One of the most exciting aspects of this chemistry is the notable water tolerance of the indium salts.100 This allows in some cases for the reaction to be carried out in water or for the catalyst to be recovered.     Very few discrete indium complexes have been used to carry out the aforementioned transformations. A chiral (S)-Binol indium catalyst has been used for the asymmetric allylation of carbonyl compounds. The chiral catalyst gives good enantioselectivity, and is tolerant of small amounts of water in the system (7.4 equivalents).100 This promising start to indium complexes playing a role in enantioselective catalysis prompted us to propose a chiral indium complex for the polymerization of lactones as the goal for this work.    A  recently  published  example  by  Huang  and  co-workers  demonstrated  that  indium compounds could be active in the polymerization of caprolactone.103 Ancillary ligands based on a substituted pyrrole ((2-dimethylaminomethyl)pyrrolate) were used to encapsulate the metal. The chosen  initiator  groups  were  methyl,  chloride,  and  pyrrole  ligands.  Crystal  structures  were obtained for two of the complexes (the methyl and pyrrole derivative). These showed that both complexes had distorted trigonal bipyramidal geometry.  35   Figure 1.4. Indium catalysts for the polymerization of ?-caprolactone.    All  three  compounds  successfully  catalyzed  the  ring-opening  polymerization  of  ?-caprolactone.  To  our  knowledge  this  is  the  only  example  of  an  indium  catalyst  ring-opening polymerizing lactones, prior to the results presented in this thesis. ?-Caprolactone is perhaps the easiest of the lactones to polymerize; there is no steric bulk on the ring to slow coordination, and the ROP is more thermodynamically favorable.  No mention is made of attempts to polymerize other  monomers,  and  complexes  with  better  initiator  groups  such  as  alkoxides  were  not synthesized.    The  polymer  samples  were  analyzed  by  GPC  in  order  to  determine  their  molecular weight and molecular weight distributions. The lowest molecular weight distribution was seen for the chloride-initiated polymerization at room temperature (1.45) while the largest distribution was observed for the pyrrole-initiated polymerization (1.99).  These values could be improved upon lowering the temperature at which the polymerization occurred (0 ?C), and worsened by elevating the reaction temperature (60 ?C). The yield of polymer was high, but the molecular weights do not match expected values for metal mediated single site chain growth; the values are much lower than expected. 36  Goals of this Work    The  goals  of  this  work  were  to  expand  upon  the  known  chemistry  of  indium  by developing new chiral indium complexes, and to expand the scope of Group 13 complexes that act as catalysts for the polymerization of lactide. It was hoped that incorporation of indium in the catalyst would result in a highly active, functional group tolerant, robust catalyst. To this end a series  of  chiral  tridentate  indium  complexes  were  synthesized,  and  their  activity  towards  the polymerization of lactide was explored.104 37  CHAPTER 2. THE SYNTHESIS OF INDIUM COMPLEXES. 2.1 Introduction   As discussed in Chapter 1, a metal catalyst for the ROP of lactide the compound must have the general formula of LnMR where Ln represents a non-labile ancillary ligand set, and R represents an initiator group. The ancillary ligand that we chose for our indium catalysts was 2-[[[(dimethylamino)cyclohexyl]amino]methyl]-4,6-bis(tert-butyl)phenol  (H2NNO),  previously reported by Finney, et al.105   The tridentate ligand is a chiral version of the ligand used by Hillmyer and Tolman in their highly active achiral zinc catalyst.65 It has been demonstrated that chiral aluminum catalysts have exhibited a marked stereoselectivity for the ROP of rac-lactide. However, these aluminum-based systems demonstrated low activity compared to the zinc systems.67 Therefore we hoped to combine these characteristics and develop a highly active and stereoselective Group 13 catalyst by using a chiral tridentate pro-ligand.   Two different synthetic approaches for the formation of indium complexes are known. The first method is via protonolysis of either a trialkyl106-109 or triamido-indium110-112 compound via elimination of an alkane or an amine, respectively. The second route is via a salt metathesis reaction of the sodium, lithium, or potassium salts of the ligand and an indium trihalide.103, 113-115 In this work both routes were attempted with varying degrees of success (Scheme 2.1). In order to generate a  catalyst, we anticipated that  an  alkoxide  would have  to  be  introduced at a  later stage via reaction with either an alcohol (protonolysis route) or a metal alkoxide (salt metathesis route). 38   Scheme 2.1. Proposed routes to the synthetic target.  39  2.2 Results   The  synthesis  of  H2NNO  was  modified  from  the  reported  literature  synthesis.105 Purification steps were added at the imine and the amine stage in order to obtain a pure product, and the use of sodium cyanoborohydride was eliminated in the final step of the ligand synthesis. Instead the more cost effective reagent sodium borohydride was used for the reduction of the imine to a secondary amine (Scheme 2.2).   Scheme  2.2.  Synthesis  of  2-[[[(dimethylamino)cyclohexyl]amino]methyl]-4,6-bis(tert-butyl) phenol     Ligand synthesis involves the formation of an asymmetric diamine from (?)-1,2-trans-diaminocycloxhexane  ((?)-DACH),  via  a  protection/deprotection  route.  An  imidazole  ring  is formed from the reaction of the (?)-DACH with the Pinner salt ethylacetimidate hydrochloride. 40  This  product  is  hydrolyzed  to  give  an  asymmetric  acylated  product.  The  primary  amine  is methylated  in  a  reductive  amination  using  sodium  cyanoborohydride  and  formaldehyde.  The acyl protecting group is removed under acidic conditions to reform a primary amine, which is functionalized via a condensation reaction to form an imine. The final step in the synthesis is to reduce the imine to a secondary amine via reduction with sodium borohydride. The enantiopure analogue of the ligand RR-H2NNO was also prepared. In this case RR-DACH was isolated from (?)-DACH by formation of the diastereomeric salt with L-tartaric acid, using the methodology developed by Jacobsen and co-workers.116   The  first  attempts  to  form  this  complex  from  the  prepared  pro-ligand,  H2NNO, followed  a  protonolysis  route,  adapted  from  literature.114  Starting  from  indium  trichloride, trimethyl  indium  was  formed  in  situ  from  reaction  with  three  equivalents  of  methyl  lithium (Scheme 2.3). Addition of H2NNO to this mixture formed a dimethylated species, (NNO)InMe2 (1) as opposed to the expected monomethylated indium complex (for simplicity NNO- is defined as  the  monodeprotonated  ligand  NNHO-  where  the  secondary  amine  remains  protonated). Complex 1 was isolated in a 40% yield as a white solid. The 1H NMR and 13C NMR spectra of 1 showed the expected signals for both the bound HNNO as well as the two methyl groups bound to indium, with characteristic upfield signals near 0 ppm. Crystals of 1 suitable for single crystal X-ray  diffraction  were  grown  from  diethyl  ether  (-35  ?C);  the  ORTEP  diagram  is  shown  in Figure 2.1. The molecule shows a distorted square-based pyramidal geometry, with the tridentate ligand bound in a meridional fashion.  41   Figure 2.1. ORTEP view of (NNO)InMe2 (1). Most hydrogen atoms are omitted for clarity and thermal ellipsoids are drawn at 50% probability. Table 2.1. Selected interatomic distances (?) for compound 1. Bond  Length (?) N1-In1   2.503(2) N2-In1   2.3549(18) C1-In1                      2) C2-In1                     2) O1-In1                     15) C4-N1                       3) C5-N1                      1(3)  Table 2.2. Selected bond angles (?) for compound 1.     Angle (?)     Angle (?) C2-In1-O1              41(8)  C1-In1-N2                9) C2-In1-C1              15(10)  C2-In1-N1                8) O1-In1-C1              15(9)  O1-In1-N1              60(6) C2-In1-N2              96(7)  C1-In1-N1              51(9) O1-In1-N2              66(6)  N2-In1-N1                6)  42     Interestingly,  the  secondary  amine  does  not  deprotonate  to  form  a  monomethyl complex. This lack of reactivity could be due to the low acidity of the secondary amine, or steric constraints which do not allow the proton of the secondary amine and the ?CH3 to become close enough to eliminate methane. When complex 1 is heated in either coordinating (THF) or non-coordinating (toluene) solvents under static vacuum no further reaction is observed.  Preliminary attempts  were  made  to  form  an  indium  alkoxide  complex  from  1.  When  the  compound  was reacted with either one or two equivalents of a secondary alcohol isopropanol no reaction was observed.  Even  when  using  a  large  excess  of  ethanol  at  room  temperature  for  two  hours  no indium-alkoxide product was observed, and only starting material was recovered. These results confirm the lower polarity and greater covalent nature of the indium-carbon bond and show that the indium-bound methyl fragment cannot act as a strong enough base in the reaction to form the alkoxide.    In our attempts to synthesize 1 an interesting by-product was isolated. Early attempts to recrystallize 1 from pentane resulted in the formation of the compound (NNO)InMe2?InClMe2 (2). From X-ray analysis of a single crystal of the compound it was determined that a mixed species had been formed from the combination of 1 and partially reacted Me2InCl. In the crystal structure it can be seen that dinuclear species had been formed from the combination of 1 and partially reacted Me2InCl. The structure shows the ligand phenolic oxygen acting as a bridging ligand between the two indium atoms (Figure 2). The second metal centre (In2) has a distorted tetrahedral geometry while the metal coordinated to the NNO ligand (In1) has a distorted square-based pyramidal geometry, as was seen in the crystal structure of 1. The distance between In1 and Cl1 (3.774 ?) falls just outside the sum of the van der Waal radii for indium and chlorine 43  (3.75  ?)  indicating  that  there  may  be  some  interaction  or  attractive  forces  between  the  two atoms.   Figure  2.2.  ORTEP  view  of  (NNO)InMe2?InCl(Me)2  (2).  Most  hydrogen  atoms  and  solvent molecules (pentane) are omitted for clarity and thermal ellipsoids are shown at 50% probability. Table 2.3. Selected interatomic distances (?) for compound 2. Bond  Length (?)  Bond  Length (?) N1-In1                     4)  C2-In1                     5) N2-In1                     3)  C3-In2                     5) O1-In2                      3)  C4-In2                      5) O1-In1                      3)  C5-N1                       7) In2-Cl1                     18)  C6-N1                       7) C1-In1                     4)        Table 2.4. Selected bond angles (?) for compound 2.    Angle (?)     Angle (?) In2-O1-In1  120.49(11)  O1-In1-N1  142.27(13) C1-In1-C2  125.2(2)  N2-In1-N1  72.25(13) C1-In1-O1  91.46(15)  C4-In2-C3  126.3(2) C2-In1-O1  104.62(16)  C4-In2-O1  103.28(18) C1-In1-N2  142.22(17)  C3-In2-O1  108.60(19) C2-In1-N2  92.38(17)  C4-In2-Cl1  109.17(19) O1-In1-N2  81.67(11)  C3-In2-Cl1  108.55(16) C1-In1-N1  92.82(18)  O1-In2-Cl1  96.88(8) C2-In1-N1  103.28(18)       44     Alternative protonolysis routes were also attempted (Scheme 2.3). The trialkyl starting material tris(isobutyl) indium (III) was also synthesized from isobutylmagnesium bromide and indium trichloride. The bulky starting material was then reacted with H2NNO. It was hoped that the  steric  crowding  at  the  metal  centre  would  provide  the  driving  force  for  the  second deprotonation of the ligand causing elimination of isobutane. Unfortunately, the reaction never produced a clean product, and resulted in an intractable mixture of products. This was perhaps due to impure starting material; Schlenk equilibria are known to occur during metal-Grignard reactions resulting in a mixture of mono-, di- and tri- alkylated salts of the metal.  117  Scheme 2.3. Protonolysis routes to indium complexes     Since both the alkyl starting materials did not follow the desired synthetic route indium starting  material  with  potentially  more  basic  substituents  capable  of  deprotonating  both  the phenolic  and  secondary  amine  protons  was  synthesized.  The  starting  material  tris-45  (bistrimethylsilylamido)  indium (III)  was  synthesized  from  the  sodium  salt  of  the  amide  and indium  trichloride  using  a  published  procedure.118  The  reaction  of  the  indium  triamide  with H2NNO resulted in a mixture of products under a variety of conditions, and a single product was never isolated from the mixture.   At this point, the alternative salt metathesis route for the synthesis of the target complex was attempted. The first step in this route, as shown in Scheme 1, is the deprotonation of the ligand to form a salt of the pro-ligand.  The most common route used to deprotonate compounds for  inorganic  synthesis  is  through  the  use  of  n-butyllithium  or  hydrides  of  either  sodium  or potassium. A number of reagents were used as bases to deprotonate the pro-ligand including n-butyllithium,  potassium  hydride,  and  benzylpotassium.  Despite  a  number  of  attempts  to deprotonate  the  ligand  only  the  monodeprotonated  pro-ligand,  HNNO,  was  isolated.  This strongly suggests that the central nitrogen of the pro-ligand is simply not acidic enough to be deprotonated.   The related pKa values for a number of organic compounds are shown in Table 2.5.119 It can be seen that a secondary amines (i-Pr2NH and TMS2NH) have a pKa of approximately 30-40, while toluene (the conjugate acid of benzylpotassium) has a pKa of 41 or 43 depending upon which  solvent  is  used  to  measure  the  value.  From  this  data,  one  might  assume  that  benzyl potassium could in fact deprotonate the secondary amine of the H2NNO pro-ligand. Possibly the formation of the charged NNO2- compound is disfavored because of the destabilizing effect of having  two  anions  close  together  on  the  molecule.  Alternatively,  the  weak  acidity  of  the secondary amine could be attributed to a highly stable six membered ring being formed upon hydrogen bonding of the proton of the secondary amine to the phenolic oxygen. This ?proton sponge  effect?  might  increase  the  stability  of  the  monodeprotonated  state  and  cause  a  lower 46  acidity for the amine.  Table 2.5. pKa values for simple organic compounds     pKa * Entry  Substrate  THF  H2O  DMSO 1  MeOH     15.5  27.9 2  t-BuOH    17.0  29.4 3  NH3    38  41 4  i-Pr2NH  36     5  TMS2NH  26    30 6  Ph2NH      25 7  PhCH3    41  43 8  CH4    48  56 9  (Me)2CH2     51                 *Values >14 for water and >35 for DMSO were extrapolated by various methods     In  one  attempt  to  deprotonate  H2NNO  in  diethyl  ether  using  two  equivalents  of benzylpotassium.  A  mixture  of  the  monopotassiated  salt,  KH(NNO),  and  potassium  ethoxide was isolated. The potassium ethoxide  was produced in  situ from the deprotonation of  diethyl ether.  This  pathway  to  the  formation  of  alkoxides  is  known  and  is  thought  to  occur  via deprotonation  of  the  ?-carbon  of  the  diethylether  with  elimination  of  ethylene.120  When  the mixture  of  potassium  ethoxide  and  KH(NNO)  was  carried  forward  by  reaction  with  indium trichloride  a  bridged  dinuclear  indium  complex  [(NNO)InCl]2(?-OEt)(?-Cl)  (3)  was  formed (Scheme 2.4). The bridged complex shows the incorporation of one equivalent of ethoxide per two indium atoms (Scheme 4). This reproducible reaction was carried out in up to a 2 g scale and gave overall yields of 60-70% (based on the amount of InCl3 used). 47  Scheme 2.4. Synthesis of compounds 3 and 4.    The solid state structure of 3 was obtained from single crystal X-ray diffraction (Figure 2.3). The structure shows chloride and ethoxide ligands bridging the two indium atoms. To our knowledge, this is the first example of this  type of a  mixed-bridge system for  indium.  Each indium  centre  has  pseudo-octahedral  geometry  and  a  terminal  chloride  ligand.  The  ligand  is acting  as  a  monoanionic  tridentate  ligand  and  is  bound  in  a  facial  fashion.  Although  the compound was formed from a racemic mixture of the pro-ligand, only homochiral complexes are formed. Although both homochiral enantiomers are observed in the lattice of the crystal only the structure of RR,RR-3 is shown. The 1H and 13C NMR spectra show only one set of peaks for the ligand  backbone,  and  ethoxide  group,  and  provides  no  evidence  for  a  mixed  RR,SS-3  being formed. When enantiopure H2NNO is used to synthesize 3 identical NMR chemical shifts are observed. 48   Figure  2.3.  ORTEP  view  of  [(NNO)(Cl)In]2(?-OEt)(?-Cl)  (3).  Hydrogen  atoms  omitted  for clarity and thermal ellipsoids are shown at 35% probability.  Table 2.6. Selected interatomic distances (?) for compound 3. Bond  Length (?)  Bond  Length (?) N1-In1  2.334(10)  Cl1-In1                    19(3) N2-In1  2.257(8)  Cl2-In2                    28(3) N3-In2  2.354(10)  Cl3-In2                    36(4) N4-In2                     69(9)  Cl3-In1                    67(3) O1-In1                     84(7)  C14-N1                    39(16) O2-In2                     50(8)  C15-N1                     93(16) O3-In1                     12(8)  C37-N3                     12(18) O3-In2                     29(8)  C38-N3                     01(17)  Table 2.7. Selected bond angles (?) for compound 3.    Angle (?)     Angle (?)     Angle (?) In1-O3-In2             8.1(3)  N1-In1-Cl1              .0(3)  N4-In2-N3               9(4) In2-Cl3-In1             62(9)  O1-In1-Cl3             6.8(2)  O2-In2-Cl2             7(2) O1-In1-O3               0(3)  O3-In1-Cl3              .1(2)  O3-In2-Cl2  94.8(2) O1-In1-N2               2(3)  N2-In1-Cl3             3.5(3)  N4-In2-Cl2             .4(3) O3-In1-N2               8(3)  N1-In1-Cl3              .2(3)  N3-In2-Cl2              7(3) O1-In1-N1              2.1(3)  Cl1-In1-Cl3             .89(10)  O2-In2-Cl3             .1(2) O3-In1-N1              1.4(3)  O2-In2-O3               .0(3)  O3-In2-Cl3              6(2) N2-In1-N1               0(4)  O2-In2-N4               .0(3)  N4-In2-Cl3              5(3) O1-In1-Cl1              7(2)  O3-In2-N4               .8(4)  N3-In2-Cl3              5(3) O3-In1-Cl1              9(2)  O2-In2-N3              2.8(4)  Cl2-In2-Cl3             30(12) N2-In1-Cl1             1.4(3)  O3-In2-N3              1.8(4)       49     A rational synthetic route to 3 was explored (Scheme 2.4). In this route the KHNNO was formed via direct reaction of the H2NNO with one equivalent of benzylpotassium in toluene and reacted with indium trichloride to form (NNO)InCl2 (4). When the reaction is carried out in a weakly or non-coordinating solvent (i.e. diethyl ether or toluene) multiple isomers of the product are seen by 1H NMR spectroscopy.  However, one isomer of the product is formed if THF or pyridine is used as the solvent for the reaction. In the spectrum of the pyridine adduct of 4 (4?Py) the relative integration of the aromatic protons  of the pyridine molecule  to the diastereotopic methylene protons of the backbone (-NH-CH2-Ar) indicate a ratio of two pyridine molecules to one molecule of 4. Upon crystallization of 4?Py from a toluene/dichlormethane solution, the ratio is reduced to 1:1 and a single-crystal X-ray structure was obtained (Figure 2.4). The indium atom is in a pseudo-octahedral geometry, with the tridentate ligand bound in a facial fashion. The two chloride ligands are in a cis-geometry. Reaction of 4 with one equivalent of sodium ethoxide results in the formation of 3 in a 67% overall yield based on H2NNO. 50    Figure 2.4. ORTEP view of (NNO)(Cl2)In?Py (4?Py). Hydrogen atoms are omitted for clarity and thermal ellipsoids are shown at 50% probability. Table 2.8. Selected interatomic distances (?) for compound 4?Py. Bond  Length (?)  Bond  Length (?) N1-In1                    3415(12)  Cl1-In1                     4) N2-In1                      12)  Cl2-In1                    4) N3-In1                      13)  C1-N1                       19) O1-In1                      10)  C2-N1                       18)  Table 2.9. Selected bond angles (?) for compound 4?Py.    Angle (?)     Angle (?) O1-In1-N2                84.80(4)  N3-In1-Cl2               3) O1-In1-N3                84.12(4)  N1-In1-Cl2  94.50(3) N2-In1-N3               94.63(4)  O1-In1-Cl1             66(3) O1-In1-N1                98.75(4)  N2-In1-Cl1               3) N2-In1-N1               77.60(4)  N3-In1-Cl1               3) N3-In1-N1               171.37(4)  N1-In1-Cl1               3) O1-In1-Cl2              91.13(3)  Cl2-In1-Cl1              14) N2-In1-Cl2              170.41(3)        51     A  solution  of  the  dinuclear  complex  3  is  stable  at  room  temperature  under  inert atmosphere  in  a  variety  of  solvents.  The  compound  appears  to  dissociate  into  a  number  of isomers upon reaction with coordinating compounds such as pyridine or methyl-(S)-lactate (an acyclic analogue of lactide).  However, upon exposure to water the complex rapidly transforms to a bis-hydroxy complex [(NNO)InCl(?-OH)]2 (Scheme 2.5). A single crystal suitable for X-ray diffraction was obtained by crystallization of 5 from acetonitrile (Figure 2.5).  Scheme 2.5. Synthesis of [(NNO)InCl(?-OH)]2  52   Figure 2.5. ORTEP view of [(NNO)(Cl)In(?-OH)]2 (5). Most hydrogen atoms omitted for clarity, and thermal ellipsoids shown at 50% probability. Table 2.10. Selected interatomic distances (?) for compound 5. Bond  Length (?)  Bond  Length (?) N1-In1                      5)  Cl1-In1                     15) N2-In1                      4)  In1-O2*                   4) O1-In1                      3)  C7-N2                       7) O2-In1                      4)  C8-N2                       7) O2-In1*                    4)        Table 2.11. Selected bond angles (?) for compound 5.    Angle (?)     Angle (?) In1-O2-In1*           67(19)  O2-In1-N2              77(15) O1-In1-O2               96(15)  N1-In1-N2                15) O1-In1-N1               15(16)  O2*-In1-N2              15) O2-In1-N1              20(15)  O1-In1-Cl1               11) O1-In1-O2*             92(15)  O2-In1-Cl              82(12) O2-In1-O2*             33(19)  N1-In1-Cl1             74(12) N1-In1-O2*            67(17)  O2*-In1-Cl1             11) O1-In1-N2               83(15)  N2-In1-Cl1               12)    The solid state structure  of  5 shows  a  dinuclear indium complex doubly bridged  by hydroxy ligands. The two ligands are bound on opposing faces of the molecule, allowing for a 53  centre of inversion. The ligands are facially bound to the indium atoms in a pseudo-octahedral geometry. Both this bridging motif and the reaction of dichloride indium complexes with water to form bridged hydroxy complexes have been reported previously in literature.115  Complex 5 can be synthesized reproducibly in a quantitative yield.   A similar reaction of 4 with adventitious water, resulted in the isolation of crystals of [(NNO)(Cl)In]2(?-OH)(?-Cl) (6) (Scheme 2.6). When a sealed NMR tube containing a sample of 4  in  deuterated  benzene  was  left  on  the  bench-top  over  a  month  a  slow  air  leak  led  to  the crystallization  of  6  (Figure  6).    In  the  complex  the  two  metal  centres  are  bound  by  facially coordinated ligands in a pseudo-octahedral geometry. The structure is very similar to that seen for 3.  It is possible that this type of structure could be an intermediate in the conversion of 3 to 5. However, no attempts were made to develop a rational synthesis for 6.  Figure 2.6. ORTEP view of  [(NNO)(Cl)In]2(?-OH)(?-Cl) (6). Hydrogen atoms are omitted for clarity and thermal ellipsoids are shown at 50% probability.  54  Table 2.12. Selected interatomic distances (?) for compound 6. Bond  Length (?)  Bond  Length (?) N1-In1                      3)  Cl1-In1                     9) N2-In1                      3)  Cl2-In2                     9) N3-In2                      2)  Cl3-In2                     8) N4-In2                      3)  Cl3-In1                     8) O1-In1                      2)  C3-N1                       4) O2-In2                      2)  C4-N1                       4) O3-In2                      2)  C26-N3                      4) O3-In1                      2)  C27-N3                      4)  Table 2.13. Selected bond angles ( ? ) for compound 6.    Angle (?)     Angle (?)     Angle (?) In2-O3-In1             84(9)  N1-In1-Cl1               7)  N4-In2-N3                9) In2-Cl3-In1              2)  O1-In1-Cl3             90(6)  O2-In2-Cl2               6) O1-In1-O3                8)  O3-In1-Cl3               6)  O3-In2-Cl2               6) O1-In1-N2                9)  N2-In1-Cl3               7)  N4-In2-Cl2             51(7) O3-In1-N2                9)  N1-In1-Cl3               7)  N3-In2-Cl2               7) O1-In1-N1              41(9)  Cl1-In1-Cl3              3)  O2-In2-Cl3             92(6) O3-In1-N1              84(9)  O2-In2-O3                8)  O3-In2-Cl3               6) N2-In1-N1                9)  O2-In2-N4                9)  N4-In2-Cl3               7) O1-In1-Cl1               6)  O3-In2-N4                9)  N3-In2-Cl3               7) O3-In1-Cl1               6)  O2-In2-N3              96(9)  Cl2-In2-Cl3              3) N2-In1-Cl1             42(7)  O3-In2-N3              25(9)        Scheme 2.6. Proposed route to the formation of [(NNO)(Cl)In]2(?-OH)(?-Cl) (6)  55     Attempts  were  made  to  synthesize  a  mononuclear  indium  alkoxide  complex  from complexes 4 and 3 (Scheme 2.7). To this end the dichloride indium complex 4 was reacted with both one and two equivalents of sodium ethoxide. As stated previously (Scheme 2.4) the reaction with  one  equivalent  of  sodium  ethoxide  gives  the  dinuclear  complex  3  in  high  yields. Interestingly, the reaction of 4 with two equivalents of sodium ethoxide yields 3 as the major product.  Other  attempts  with  more  basic  alkoxides  (potassium  tert-butoxide  or  sodium  tert-butoxide) led to multiple products. Attempts  directly install  a second alkoxide group on  to 3 using  potassium  tert-butoxide  as  a  base  led  to  isolation  of  the  starting  material  as  the  major component, which suggests that 3 is more thermodynamically favorable and stable than either the  mononuclear  (NNO)InCl(OEt)  or  the  dinuclear  [(NNO)InCl(?-OEt)]2  complex  which contain one ethoxide per indium atom. 56  2.3 Conclusion   A number of new indium complexes were synthesized and isolated. The complexes all contained a tridentate N,N,O-based ligand. The ligand is chiral containing (?)1,2-trans-DACH moiety  in  the  backbone.  It  is  the  first  time  to  our  knowledge  that  this  compound,  2-[[[(dimethylamino)cyclohexyl]amino]methyl]-4,6-bis(tert-butyl)phenol,  has  been  used  as  a ligand.  Crystal structures were obtained for four isolated products as well as two impurities. The majority of these structures show octahedral geometry at the metal centre (the exception being the  dimethylated  indium  complexes  2  and  3).  The  complexes  also  tend  to  form  dinuclear structures with either oxygen based bridging  ligand  (hydroxy-,  or alkoxy  ligands)  or chloride bridging ligand. This is a common motif in other indium chemistry. Complex 3 was shown to convert cleanly to complex 5 in the presence of water.    The  ligand  did  not  bind  as  expected  to  the  metal  centres.  It  was  hoped  that  the tridentate ligand would be dianionic in nature, with both the phenolic proton and the proton of the secondary amine being acidic enough to be removed either prior to attachment to the metal (salt metathesis route) or upon attachment to the metal (protonolysis).  In all the cases observed here the ligand was monoanionic. The phenolic proton was deprotonated via both routes, but the proton of the secondary amine could not be removed via either route.   A dinuclear indium ethoxide complex 3, [(NNO)(Cl)In]2(?-OEt)(?-Cl), was obtained via the direct reaction of a mixture of salts, KH(NNO) and KOEt, with indium trichloride. The synthetic methodology was further explored and  it was found that the complex could also be obtained  by  direct  reaction  of  (NNO)InCl2  and  NaOEt.  In  fact  this  complex  seems  to  be unusually thermodynamically stable and is the major product in numerous attempts to synthesize a mononuclear indium complex.  57  2.4 Experimental General Considerations.   Unless otherwise indicated, all air- and/or water-sensitive reactions were carried out under dry nitrogen using either an MBraun glovebox or standard Schlenk line techniques.  NMR spectra were recorded on a Bruker Avance 400MHz spectrometer.  1H NMR chemical shifts are given in ppm  versus  residual  protons  in  deuterated  solvents  as  follows:  ?  7.27  CDCl3,  ?  7.16  C6D6.  13C{1H} NMR chemical shifts are given in ppm versus residual 13C in solvents as follows: ? 77.2 CDCl3, ? 128.4 C6D6 Materials.   Solvents (pentane, THF, toluene, dichloromethane, and diethyl ether) were degassed and dried using 3? molecular sieves in an mBraun Solvent Purification System.  The  THF was  further dried over sodium and distilled under N2. CD2Cl2 and CDCl3 were dried over CaH2, and degassed through a series of freeze-pump-thaw cycles. C6D6 was dried over sodium and degassed using a series of freeze-pump-thaw cycles. InCl3 was purchased from Strem Chemicals and used without further  purification.  For  the  enantiopure  catalyst  the  (?)-trans-1,2-diaminocyclohexane  was resolved using Jacobsen?s method,116 and then carried forward through the same procedures as used  for  the  rac-catalyst.  The  benzyl  potassium  was  synthesized  using  a  modified  literature procedure using n-butyl lithium (Aldrich), potassium tert-butoxide (Alfa Aesar) and toluene.121 All other compounds were obtained from Aldrich and used without further purification. The  literature  preparation  of  H2NNO  was  modified  as  follows:  two  purification  steps  were added:  the  imine  was  recrystallized  from  warm  acetonitrile,  and  the  ligand  itself  was recrystallized from acetonitrile when it was formed.  58  Synthesis of (NNO)InMe2 (1) A 125 mL round bottom flask was charged with InCl3 (0.100 g; 0.45 mmol) in 15 mL of Et2O. 1.6M MeLi in Et2O (0.85 mL, 0.45 mmol) was added dropwise to the stirring mixture, which was stirred at room temperature for a further 20 minutes. H2NNO (0.163 g; 0.45 mmol) was then added to the mixture as a solution in Et2O (15 mL). The reaction was heated at 30 ?C for 24 h, cooled to room temperature and filtered through celite. Concentration of the filtrate to ca. 15 mL resulted in the formation of a white precipitate that was isolated by filtration and washed with pentane (3 x 2 mL) and dried under high vacuum to yield 1 ( 92 mg; 0.18 mmol; 41% yield). 1H NMR (400 MHz, CDCl3): ? 7.22 (1H, d, J=3 Hz, ArH), 6.82 (1H, d, J=3 Hz , ArH), 4.31 (1H, t, J=10 Hz, R2N-CH-CH2), 3.59 (1H, d, J=7 Hz, NH-CH2-Ar) 2.61(1H, d, J=12 Hz, NH-CH2-Ar), 2.47(1H, m, -CH2- of DACH), 2.21, (6H, br s, N-(CH3)2), 1.90 (3H, m, -CH2- of DACH), 1.61 (1H, t, J=10 Hz, -CH2- of DACH), 1.43 (9H, s, t-Bu), 1.28 (9H, s, t-Bu), 1.24 (1H, m, -CH2- of DACH), -0.15 (3H, s, In-CH3), -0.35 (3H, s, In-CH3). 13C{1H} NMR (75 MHz, CDCl3): ?163.9, 138.3, 133.0, 124.3, 123.7, 122.8, 66.8, 58.1, 53.8, 35.4, 33.9, 32.6, 32.0, 29.7, 25.2, 24.8, 21.5, -7.0, -7.2. EI-LRMS (m/z) calc?d for C25H45InN2O  [M+] 504.26; found: 504. Synthesis of [(NNO)InCl]2(?-OEt)(?-Cl) (3).  A 500 mL round bottom flask was charged with H2NNO (1.000 g; 2.775 mmol) in 100 mL of Et2O and cooled to -34 ?C.  Benzyl potassium (0.722 g; 5.55 mmol) in 25 mL Et2O was cooled to -34  ?C  and  added  drop-wise  to  the  stirring  solution.    The  reaction  was  warmed  to  room temperature and stirred for a further 48 h. An off-white solid was isolated by filtration using a cintered frit, and dried under high vacuum for approximately 2 h. A 500 mL round bottom flask was charged with InCl3 (0.608 g; 2.75 mmol) in 125 mL Et2O to this stirring mixture was added the solid from the previous reaction as a slurry in 25 mL of Et2O. The mixture was stirred at 59  room temperature for 48 h, and filtered through celite. The filtrate was concentrated to dryness, and the resulting  residue was taken  up  in pentane  resulting in a white  precipitate which was isolated by vacuum filtration, and dried under high vacuum for 2 h to yield 3 (0.99g; 0.90mmol; 65%yield  based  on  InCl3  as  the  limiting  reagent).  1H  NMR  (400MHz,  C6D6):  ?  7.61  (1H,  d, J=2.3Hz, ArH), 6.8 (1H, d, J=2.2Hz, ArH), 5.26 (1H, d, J=13.6Hz, NH-CH2-Ar), 4.69 (1h, m, -OCH2CH3), 3.59 (1H, d, J=13.6Hz, NH-CH2-Ar), 2.84 (1H, d, J=11.0Hz, R2N-CH-CH2-), 2.66 (1H, m, R2N-CH-CH2), 2.52 (3H, s, -N-CH3), 2.42 (1H, br m, -CH2- of DACH), 2.20 (1H, d, J=4.3Hz, -CH2- of DACH), 1.95(2H, br m, -CH2- of DACH), 1.83 (9H, s, -tBu), 1.78 (3H, s, -N-CH3), 1.56 (1H, t, J=6.7Hz, -O-CH2-CH3) 1.40 (9H, s, -tBu) 1.20 (2H, br m, -CH2- of DACH), 1.06 (1H, d, J=11.5Hz, -CH2- of DACH) 0.61 (2H, br m, -CH2- of DACH) 0.16 (1H, br m, -CH2- of DACH). 13C{1H} NMR (75 MHz, C6D6): ? 163.9,  140.3,  137.4, 127.1,  125.4,  120.0, 65.6, 64.1, 53.3, 51.9, 51.2, 44.8, 39.0, 36.7, 34.9, 32.9, 31.5, 31.4, 25.4, 25.3, 22.2, 20.6. Anal. Calc'd (found) for C48H83Cl3In2N4O3: C 52.40 (51.96), H 7.60 (7.53), N 5.09 (5.03) %. X-ray quality crystals were grown from cold toluene over 1 week. Synthesis of (NNO)InCl2 (4). A 250 mL round bottom flask was charged with a magnetic stir bar and a solution of H2NNO (1.000 g, 2.78 mmol) in 20 mL cold toluene (-34 ?C). KCH2Ph (0.361 g, 2.78 mmol) in 20 mL cold toluene was added dropwise as a slurry to the stirred solution. The reaction was warmed to room temperature and stirred for 16 h. The mixture was concentrated to dryness and the residue was taken up in 10 mL of pentane. A white solid precipitated from pentane and was isolated by filtration  then  dried  under  high  vacuum  for  2h  to  yield  KHNNO  (0.908  g,  2.28  mmol,  82% yield). 250 mL round bottom flask was charged with a slurry of InCl3 in 30 mL cold THF (-34 ?C). KHNNO was added dropwise as a solution in 20 mL cold THF (-34 ?C) to the stirring InCl3 60  mixture. The reaction was warmed to room temperature and stirred for 16 h then filtered through celite. The filtrate was concentrated to dryness taken up in 10 mL pentane. A white solid that precipitated out of solution was isolated by filtration and dried under high vacuum for 2 h to yield 4  (1.24g, 2.27mmol, 99% yield, 81% yield over  two steps).  1H  NMR  (400MHz, C6D6): ? 7.58 (1H, d, J=2.1Hz, ArH), 6.75 (1H, d, J=1.9Hz, ArH), 5.26 (1H, d, J=13.0Hz, NH-CH2-Ar), 3.48 (1H, d, J=12.7Hz, NH-CH2-Ar), 2.58 (1H, t, J=12Hz, -R2N-CH-CH2- of DACH), 2.40 (3H, s, -N-CH3), 1.71 (9H, s, -tBu), 1.67 (3H, s, -NCH3), 1.35 (9H, s, -tBu), 1.21 (1H, d, J=10.6Hz,  -CH2-  of  DACH),  1.12 (1H,  d,  J=11.5Hz,  -CH2- of  DACH)  1.00  (1H,  d,  J=12.2Hz,  -CH2-  of DACH),  0.61  (3H,  m,  -CH2-  of  DACH),  0.11  (1H,  q,  J=12.1Hz,  -CH2-  of  DACH).  13C{1H} NMR (75 MHz, C6D6): ? 163.92, 140.05, 137.85, 126.83, 125.43, 120.35, 65.71, 54.11, 51.72, 44.81,  38.67,  36.68,  32.89,  31.33,  31.05,  25.23,  24.94,  22.13.  Anal.  Calc?d  (found)  for C23H39Cl2InN2O: C 50.66 (50.43), H 7.21 (7.17), N 5.14 (5.04) %. EI-LRMS (m/z) calc?d for C25H39Cl2InN2O [M+] 544.15; found: 544. The pyridine adduct was formed by dissolving 4 in pyridine, with subsequent removal of the volatile solvent. X-ray quality crystals of 4?Py were isolated by recrystallization from a toluene/dichloromethane mixture at room temperature.  Alternative synthesis of 3 from 4. A 125 mL round bottom flask was charged with NaOEt (37.5mg; 0.551mmol) in 20 mL toluene. A  solution  of  4  (0.300g,  0.551mmol)  in  15mL  toluene  was  added  dropwise  to  this  stirred mixture.  The  mixture  was  stirred  at  room  temperature  for  16  h,  filtered  through  celite,  and concentrated to dryness. The residue was taken up in 10 mL pentane from which a white solid precipitated. The solid was isolated by filtration to yield 1 (0.2551g, 0.23mmol, 83% yield, 67% over  three  steps).  1H  NMR  (400MHz,  C6D6):  ?  7.61  (1H,  d,  J=2.3Hz,  ArH),  6.8  (1H,  d, 61  J=2.2Hz,  ArH),  5.25  (1H,  d,  J=13.8Hz,  NH-CH2-Ar),  4.70  (1h,  m,  -OCH2CH3),  3.59  (1H,  d, J=13.6Hz, NH-CH2-Ar), 2.83 (1H, d, 11.0Hz, R2N-CH-CH2-), 2.66 (1H, m, R2N-CH-CH2), 2.51 (3H,  s,  -N-CH3),  2.42  (1H,  br  m,  -CH2-  of  DACH),  2.20  (1H,  d,  4.3Hz,  -CH2-  of  DACH), 1.95(2H, br m, -CH2- of DACH), 1.83 (9H, s, -tBu), 1.78 (3H, s, -N-CH3), 1.60 (1H, t, 6.6Hz, -O-CH2-CH3) 1.39 (9H, s, -tBu) 1.22 (2H, br m, -CH2- of DACH), 1.06 (1H, d, J=11.5Hz, -CH2- of DACH) 0.61 (2H, br m, -CH2- of DACH) 0.16 (1H, br m, -CH2- of DACH). 13C{1H} NMR (100MHz, C6D6): ? 163.9, 140.3, 137.4, 127.1, 125.4, 120.0, 65.6, 64.1, 53.3, 51.9, 44.8, 39.0, 36.7, 35.2, 34.9, 32.9, 31.5, 31.4, 25.4, 25.3, 22.2, 20.6. Synthesis of [(NNO)InCl(?-OH)]2 (5)  Compound 1 was removed from an inert atmosphere and exposed to moist air for 48 h to yield the air and moisture stable compound 2 in quantitative yield. 1H NMR (400mHz, C6D6): ? 7.71 (1H, d, J= 2.2Hz, ArH), 6.85 (1H, d, J=2.2Hz, ArH), 4.83 (1H, d, J=13.3H, -N-CH2-Ar), 3.52 (1H, d, J=12.7Hz, -N-CH2-Ar) 3.04 (1H, d, J=10.6Hz, -N-CH-CH2-), 2.82 (1H, t, -N-CH-CH2-), 2.59 (3H, s, -N-CH3), 2.43 (1H, m, -CH2- of DACH), 2.02 (2H, br m, -CH2- of DACH) 1.97 (9H, s, -tBu), 1.81 ( 3H, s, -N-CH3), 1.52 (9H, s, -tBu), 1.28 (2H, br m, -CH2- of DACH), 1.13 ( 1H, d, J=12.7Hz, -CH2- of DACH), 0.72 (3H, br m, -CH2- of DACH) 0.23 (1H, m, -CH2- of DACH). 13C{1H} (C6D6,100MHz): ? 163.8, 139.8, 137.5, 127.3, 125.3, 121.2, 65.7, 53.2, 51.9, 51.1, 44.9, 38.7,  36.7,  34.9,  32.9,  31.5,  31.4,  31.3,  25.3,  25.3,  22.1.  ESI-LRMS  (m/z)  calc?d  for C46H80Cl2In2N4O4  [MK+]  1091.33;  found:  1091.7.  X-ray  quality  crystals  were  grown  from acetonitrile at room temperature over the period of 1 week. 62  CHAPTER 3. KINETIC INVESTIGATIONS/POLYMER CHARACTERIZATION 3.1 Introduction   Monitoring reactions as they occur  can provide useful data about how  and why  the reaction proceeds. In most cases, monitoring a reaction involves observing the formation of the product or the disappearance of the reagents. A simple reaction of a reagent, A, being converted to product, B, is shown below.    In  this  case  the  rate  of  the  reaction  could  be  observed  in  two  ways:  one  could  observe  the increase  in  concentration  of  B  or  the  decrease  in  the  concentration  of  A.  Assuming  that  the reaction is first order with respect to A, it follows:  If we combine the two equations:    where [A]o is the concentration of A at time zero. If the assumption holds that the rate is first 63  order with respect to the concentration of A, then a plot of the natural logarithm of [A] versus time should be linear with the following equation.  According to the Eyring equation the relationship between the natural logarithm of the observed rate (k) and the inverse of the temperature at which the reaction is carried out should be linear. The enthalpy (?H?) and the entropy of (?S?) activation are calculated from the slope of the line and the y-intercept respectively.122  Combining these values and observations can lead to useful insights about the mechanism of the reaction. In order to obtain data, one must  choose the  appropriate  spectroscopic technique to monitor the reaction.   The  time-scale  for  the  reaction  often  determines  which  technique  will  be  used  to monitor the process.  If the process is on a femtosecond time scale then you cannot observe it using a technique that allows for collection of data every millisecond. Fortunately for polymer chemists the observable reaction time for most polymerizations is on the order of minutes, hours, or  days.    This  means  that  a  number  of  techniques  are  useful  including  UV,  IR,  and  NMR spectroscopy. Observation of the 1H nuclei by NMR spectroscopy remains one of the most useful techniques.   For the polymerization of lactide kinetic experiments are usually carried out using 1H 64  NMR spectroscopy. The relative integration of the methine proton of the monomer is monitored over the course of the reaction. In CD2Cl2 at 298 K the methine proton of LA is a quartet located at 5.07 ppm, while the methine protons of the polymer fall in a range from 5.12 - 5.22 ppm. This chemical shift difference is large enough to resolve the monomer peaks from the peaks for PLA.  As  the  polymerization  proceeds  a  series  of  1H  NMR  spectra  are  obtained  at  regular  time intervals. By measuring the integration of the LA methine proton as it is converted to PLA the rate for the overall reaction can be determined.   In order to obtain quantitative data from the NMR spectra an internal standard must be added to the reaction mixture. Several characteristics are important for an internal standard. It must not interfere with either the method of data collection, or the chemical processes that are occurring. For the polymerization of lactide 1,3,5-trimethoxybenzene (TMB) was chosen as the internal standard. It is reasonable to assume that TMB will not interfere with the polymerization, as it is unlikely to act as an initiator for the polymerization or as a chain transfer agent during polymerization. TMB is a relatively weak ligand and should not compete for coordination sites on the metal when in the presence of stronger ligands (ie. the ancillary ligand, or LA).    Figure 3.1. Internal standard (1,3,5-trimethoxybenzene) for the polymerization of LA and its 1H NMR characteristics (CD2Cl2, 298K, 400MHz)   65    The chemical shifts for the internal standard must not overlap or interfere with any of the peaks that are being observed. TMB has only two peaks in the 1H NMR spectrum (Figure 1). The first peak is a singlet at 3.77 ppm for the methyl protons. The second peak is also a singlet at 6.09 ppm corresponding to the aromatic protons. The two signals neatly bracket the region of interest for PLA but do not interfere with any of the peaks for the polymer. The peak for the methyl  protons  was  used  for  calibrating  the  integration.  Since  the  methyl  peaks  in  TMB constitute 9 equivalent protons, a very small amount of the internal standard can be used relative to LA (molar ratio of 1:16 for TMB to LA).    In  addition  to  the  need  for  an  internal  standard,  the  temperature  at  which  the polymerizations are performed must be known exactly in order to obtain high quality data for the construction of an Eyring plot. To this end, the temperature settings for the NMR probe must be calibrated  at  each  temperature.  The  temperatures  are  calibrated  using  the  values  for  the difference  between  the  chemical  shifts  (??)  of  the  two  peaks  in  a  given  standard  sample (provided by Bruker) and the equations given in Table 3.1.123 Two standards are used for the desired temperature range; methanol is used for the lower temperature calibrations (230 ? 300 K) and ethylene glycol is used for the higher temperature calibration (300-400 K).      Table 3.1. Equations for calibrating temperatures on NMR spectrometer     Analysis of the polymer samples is carried out using gel permeation chromatography (GPC).  GPC  is  a  form  of  size  exclusion  chromatography  (SEC)  a  separation  technique  that separates compounds based on their size or molecular weight. The chromatography column is 66  packed with small particles of silica or polymers. These particles contain pores into which solute and  solvent  molecules  can  diffuse.  Molecules  that  are  larger  than  the  average  pore  size  are excluded and pass through the column quickly, while molecules that are smaller than the pore size are taken up in the pores and retained for longer periods on the column. Molecules of an intermediate  size  enter  some  but  not  all  of  the  pores  depending  on  their  size.  In  this  way molecules of different molecular weight are separated as they pass through the column. Multiple pieces of data can be determined by GPC analysis of polymers related to the samples molecular weight.  These include the number average molecular weight (Mn), the weight average molecular weight (Mw), and the polydispersity index (PDI) or molecular weight distribution (MWD). PDI is defined as Mw/Mn.124   A  calibration  curve  is  normally  used  for  determining  the  molecular  weight  of  PLA samples. A series of polystyrene samples of known molecular weights are run on the instrument to construct a calibration curve. The samples of PLA are then analyzed and assigned a molecular weight based on where their retention time falls on the curve. This does not provide an absolute value  for  molecular  weight  (detection  by  laser  light  scattering  provides  absolute  molecular weights), instead it is an approximation of the molecular weight.  Laser light scattering detection for analysis of PLA is not generally used because of PLA?s poor ability to scatter light (PLA has a  low  differential  refractive  index  or  dn/dc)  and  because  the  methodology  has  not  been investigated as fully for polyesters as it has for other large molecules such as proteins.    All of the methods and techniques mentioned above were used, along with 1H, 1H{1H}, and 13C{1H} NMR characterization  of PLA  samples,  in order  to  describe the  ROP of LA by complex  3.  The  rate  of  polymerization  was  monitored  by  1H  NMR  spectroscopy  and  the molecular  weight  and  polydispersity  of  the  polymers  were  determined  by  GPC.  The  NMR 67  spectra  of  the  polymers  were  used  to  assign  the  microstructure  of  the  PLA  samples  and  the mechanism  by  which  the  catalyst  was  exerting  stereocontrol  over  the  polymerization.10468  3.2 Results   The  dinuclear  complex  3  is  an  active  catalyst  for  the  ring-opening  polymerization  of lactide.  In  polymerizations  monitored  to  90%  conversion  by  1H  NMR  spectroscopy  (25  ?C, CD2Cl2), 200 equivalents of rac-LA are converted to PLA in 30 min. This result is comparable to some of the most active metal-based catalysts reported to date and is significantly faster than aluminum  salen  complexes,  which  usually  require  several  hours  at  elevated  temperatures  to convert LA under otherwise similar conditions.    It  was  previously  shown  that  the  dinuclear  complex  3  dissociates  in  the  presence  of coordinating compounds.  Therefore, we propose that at high concentrations of LA the dinuclear complex  will  split  into  two  mononuclear  fragments  (NNO)InCl2?LA  (4?LA)  and (NNO)InCl(OEt)?LA (7?LA). A control experiment shows that the mononuclear complex 4 is not a catalyst for the polymerization of lactide. Therefore, the mononuclear fragment 7?LA is proposed  as  the  active  catalyst  when  the  pre-catalyst  3  is  used  for  the  ring-opening polymerization of lactide (Scheme 3.1).   Scheme 3.1. Dissociation of 3 to generate the active catalyst 7?LA.     69    Complex 3 is also active for other lactone monomers including ?-caprolactone, and ?-butyrolactone  (Scheme  3.2).  The  polymerizations  were  performed  in  THF  and  complete conversion  was  achieved  in  less  than  24  hours  ([M]/[3]  =  500).    Preliminary  GPC  data  on samples of the polymers show that the polymers have a narrow molecular weight distribution, and a close correlation is seen between the monomer loading and the molecular weight (Table 3.2). After attempts to establish the substrate scope, the work was directed towards elucidating the reactivity of 3 as a catalyst for LA polymerization. Kinetic studies were performed in order to determine the reaction order of the polymerization with respect to both the monomer and the catalyst and to determine the activation parameters (?H? and ?S?).  Scheme 3.2. Monomer scope for the polymerization of lactones by 3.   70  Table 3.2. GPC data for the polymerization of ?-caprolactone and ?-butyrolactone by 3.     GPC analysis of isolated polymer samples shows a linear relationship between added monomer ([LA]/[3]) and the observed molecular weight (Mn) (Figure 3.2). Mn closely follows the calculated molecular weight (illustrated by the straight line). The values for Mn(calc?d) were calculated  assuming  a  single  chain  growing  per  molecule  of  3  with  complete  conversion  of monomer  to  polymer.  This  is  in  agreement  with  7?LA  being  the  active  catalyst  for polymerization.   Figure 3.2. Plot of the observed molecular weight (Mn = ?) and molecular weight distributions (PDI = ?) of PLA as a function of added monomer (calculated values for the molecular weights are shown using the line).  71    The  PDI  values  obtained  for  the  polymer  samples  are  quite  low  (from  1.09-1.14) indicating  that  the  system  is  well  behaved.  The  small  PDI  values  make  it  unlikely  that transesterification processes, which result in larger values for PDI, are playing a significant role in  the  polymerization.  The  data  for  the  polymer  molecular  weights  and  molecular  weight distributions are indicative of a living system. However, two deviations in the data should be noted for the low monomer loading (100 equivalents): a relatively high value for the PDI (1.23) and  a  deviation  between  the  calculated  and  observed  Mn  values  (Mn(obs)  =  28  000  g  mol-1; Mn(calc?d) = 14 000 g mol-1).   A  high  Mn  value  suggests  that  fewer  active  sites  are  catalyzing  the  polymerization, while  the  high  PDI  value  indicates  that  the  number  of  active  sites  is  not  constant  during polymerization.  This  apparent  change  in  catalyst  concentration  could  be  attributed  to  an equilibrium between an active mononuclear catalyst and a dormant dinuclear species, similar to the  one  represented  in  Scheme  3.1.  This  would  cause  a  decrease  in  the  number  of  growing catalyst chains and a subsequent increase in the molecular weight and PDI.    Examination  of  the  reaction  in  situ  provides  further  evidence  for  the  proposed mechanism. It was seen for the sequential addition of two aliquots of LA (100 equivalents each), that when the natural logarithm of the relative integration of the methine proton of LA is plotted versus time an initial curved region is observed indicating an induction period (Figure 3.3). This induction period is attributed to the equilibrium between 3 and the mononuclear species 4?LA and  7?LA.  After  the  induction  period,  the  plot  becomes  linear  and  first-order  behavior  is observed  until  approximately  90%  conversion  is  achieved.  Above  90%  conversion,  the  slope decays  exponentially  until  the  reaction  is  complete.  Upon  addition  of  the  second  aliquot  of monomer, no induction period is observed. 72   Figure 3.3. Plot of Ln([LA]/[TMB])  versus time for two  sequential additions of LA (?  =  1st addition of 100 equivalents; ? = 2nd addition of 100 equivalents) (CD2Cl2, 298K, 400MHz)     The rate of polymerization was obtained from the slope of the line for the linear region of the graph. The rates of polymerization before (k = 0.56(0.18) s-1 M-1) and after the second addition of monomer (k = 0.59(0.18) s-1M-1) were the same within error.  The fact that the rate remains constant is a strong indication that the number of active catalyst sites did not decrease upon complete conversion to polymer. This information, along with the low PDI values and the linear  relationship  between  molecular  weight  and  catalyst  loading,  indicates  that  the  catalyst system is living under the reaction conditions.    The rate of polymerization of LA is first order with respect to [3] (Figure 3.4). The rate also  displayed  first-order  behavior  towards  LA one  can  therefore  propose  an  overall  second-order rate law.     73  This  simplified  rate  law  is  an  adequate  description  of  the  system  for  the  bulk  of  the polymerization;  however,  it  does  not  take  into  account  the  observed  induction  period for  the catalyst or the decay in rate observed above 90% conversion.    Figure 3.4. Dependence of the observed polymerization rate upon the concentration of 3.  Table 3.3. Experimental results for polymerization of LA using 3 at various temperatures     74   Figure 3.5. Eyring plot for the polymerization of LA using 3    To  elucidate  the  mechanism  of  polymerization  an  Eyring  plot  was  constructed  by carrying  out  a  series  of  polymerizations  from  -5  to  32  ?C  in  CD2Cl2  (Figure  3.5).  When  the natural  logarithm  of  the  observed  rate  was  plotted  versus  the  inverse  of  the  temperature,  the graph was linear. The enthalpy of activation (?H? = 49(2) kJmol-1) is small while the entropy of activation is large and negative (?S? = -140(12) JK-1mol-1). These data indicate that the transition state  is  highly  ordered  and  support  an  associative  coordination  insertion  mechanism  for  the polymerization as was proposed previously in literature (Scheme 1.7).   The coordination-insertion mechanism proposes  an acyl  bond  cleavage  for the  ring-opening polymerization of LA. If the catalysis follows this mechanism then the two ends of the chain should be an alcohol and an ester (upon hydrolysis of the metal complex). If the ethoxide in 7?LA is acting as the initiator for polymerization an ethyl ester should be observed in the 1H NMR spectrum of the polymer. As seen in Figure 3.6, an ethyl ester is observed at 4.25 ppm in the 1H NMR of an oligomeric sample of PLA ([LA]/[3] = 40).  75    Figure 3.6. 1H NMR spectrum of the methine region of oligomeric PLA (CD2Cl2, 298 K,  400MHz)    We propose that 7?LA initiates the ROP of LA by a coordination-insertion mechanism (Scheme 3.3). After 7?LA has been formed it can enter into the reaction cycle. The first step is nucleophilic attack by the alkoxide on one of the carbonyl carbons; the data presented so far supports an associative transition state. This is followed by ring-opening of the LA via formation of a new metal-oxygen bond. The new alkoxide (or growing polymer chain) can then continue the cycle upon coordination of another monomer.  76   Scheme 3.3. Coordination-insertion mechanism for ROP of LA by 7?LA.      A modification to the general mechanism is required to account for the decay in the observed rate that was noted above 90% conversion for the in situ monitoring of ln([LA]/[TMB]) versus time by 1H NMR (Figure 3.3). Since 4?LA is present in solution it can be postulated that a dinuclear species might reform during the polymerization; 4 would then be competing with LA for coordination sites at the metal.  The dinuclear species and the mononuclear species 4?LA are in an equilibrium that favors the dinuclear species at low [LA] (ie. above 90% conversion).  The equilibrium  would  account  for  a  lower  number  of  active  sites  being  present  at  higher conversions, leading to a decrease in the rate of polymerization. The dinuclear complex could be considered a resting state for the catalyst; the addition of more lactide causes the equilibrium to shift  to  favor  the  lactide  adduct  which  re-enters  the  catalytic  cycle  and  recommences polymerization. This was demonstrated by the sequential addition of monomer aliquots (Figure 3.3).   A major goal of this research was to develop a stereoselective catalyst for the ROP of LA. In order to determine the stereoselectivity of the catalyst the 1H{1H} and 13C{1H} NMR 77  spectra were obtained for a series  of  polymer samples. The temperature, solvents  and  optical purity of both monomer and catalyst were all varied for this series of polymerizations in the hope that  the  mechanism  for  stereocontrol  in  the  polymerization  might  be  elucidated.  The  1H{1H} NMR of the polymer samples reveal a series of peaks corresponding to the mrm, rmr, mmr/rmm, and mmm tetrads (Figure 3.7). The presence of all four peaks in large amounts indicates that the catalyst  is  not  achieving  significant  stereoselectivity;  however,  the  polymer  does  show  some isotactic enrichment. The relative integration of the mmm tetrad is much larger than it would be for atactic PLA. The ratio of the stereoerror sequences mmr, mrm, rmr, and rmm was found to be roughly 1:2:1:1. As was explained in Section1.5, these values are indicative of enantiomorphic site control mechanism for the stereoselectivity of the catalyst. The methine region of 13C{1H} NMR spectra of the polymer samples also shows the presence of stereoerror sequences and is consistent  with  those  observed  in  the  1H{1H}  NMR.  The  13C{1H}  NMR  confirms  a microstructure  of  isotactically  enriched  PLA. The  peak  at  69.4  ppm  would  have  an  intensity almost equal to the one observed at 69.2 ppm if the sample was completely atactic.    A polymerization of L-LA was performed using rac-3, and the 1H{1H} and 13C{1H} NMR  spectra  were  obtained  (Figure  3.7(c)).  The  homonuclear  decoupled  spectrum  shows  a distinct  singlet  for  the  methine  region  protons  of  PLA  corresponding  to  the  mmm  tetrad. Similarly the methine region of the 13C{1H} NMR spectrum shows one distinct singlet. This is the expected result, as no stereoerror sequences should be formed during the polymerization of enantiopure LA unless racemization of the monomer is occurring.     78     Figure  3.7.  1H{1H}  and  13C{1H}  NMR  of  PLA  (CDCl3,  298  K,  600  MHz)  from  (a) polymerization  of  rac-LA  using  rac-3  (b)  polymerization  of  rac-LA  using  RR,RR-3,  and  (c) polymerization of L-LA using rac-3.         The relative integration of the stereosequences were also used to calculate the values of Pm and Pr for the catalyst (Table 1.1).  The polymerizations were performed in different solvents at room temperature and at 0 ?C (Table 3.4). For a completely random atactic polymer the value 79  for Pm should be 0.5. All of the polymer samples catalyzed by rac-3 show values of Pm between 0.53 and 0.62 indicating that the polymers are isotactically enriched. The lowest values for Pm corresponds to polymerization carried out in THF, while the highest values are seen for the low temperature experiments.    Table 3.4. Pm and Pr values for polymer samples from the polymerizations of rac-LA at 0 ?C and 25 ?C using 3 as a catalyst.  Entry  [LA]/[3]  t (h)  Solvent  Temp   (?C)  Pm  Pr  Pr + Pm 1  205  24  CH2Cl2  25  0.59  0.38  0.97 2  200  18  CH2Cl2  0  0.62  0.36  0.98 3  200  48  Toluene  25  0.60  0.36  0.96 4  200  48  Toluene  0  0.62  0.34  0.96 5  200  24  THF  25  0.53  0.44  0.97 6[a]  182  18  CH2Cl2  25  0.43  0.46  0.89          [a] This sample was prepared with RR,RR-3       Results obtained using RR,RR-3 as the catalyst show very different results. The Pm value (0.43) is significantly lower than that observed for the analogous reaction with rac-3 (0.59), and the Pr value also increased (from 0.38 to 0.46).  According to the Bernoullian model the value of Pr + Pm must equal one. If one looks at the Pr + Pm values for entries 1-5 it is clear that the values are all close to one. The value for the sixth entry shows a significant deviation from 1 at 0.89. This indicates that the Bernoullian model has failed for this set of polymerization conditions. Further investigations into the role of the enantiopurity of the catalyst in the polymerization were required to explain this result.   A kinetic study was performed using varying ratios of RR,RR and SS,SS-3. A physical mixture  of  rac-  and  RR,RR-3  stock  solutions  was  used  to  control  the  enantiopurity  of  the 80  catalyst.  The  rates  were  observed  using  1H  NMR  spectroscopy,  and  the  results  are  shown  in Figure 3.8. As seen in the figure, the rates show a distinctly non-linear behavior with the highest rate being observed for the polymerization of rac-LA by rac-3.        Figure  3.8.  The  observed  rate  of  polymerization  of  LA  as  a  function  of  the  enantiomeric composition of the catalyst.     Qualitatively, the variance in rate with enantiopurity of the catalyst is consistent with a site control mechanism. If both enantiomers of a stereoselective catalyst are present in solution then  each  enantiomer  will  rapidly  polymerize  its  preferred  enantiomer  of  the  monomer. However, as the monomer composition changes from a perfect 50/50 mixture of enantiomers the rate  is  expected  to  decrease.  A  concentration  gradient  will  develop  in  the  mixture;  the enantiomer of the catalyst that is in excess (RR,RR-3) will deplete its monomer of choice more quickly than the other enantiomer of the catalyst can consumes its enantiomer. As the RR,RR-catalyst runs out of its monomer, the occurrence of stereoerrors in the polymer chain become more frequent and the polymerization slows down. 81    If the mismatch in chirality of the monomer and catalyst is the cause for the observed decrease in rates then the relative rate for the polymerization of enantiopure LA by RR,RR-3 should either be much slower than it is for rac-LA, or much faster depending if the enantiopure LA  is  the  ?right?  enantiomer.  Therefore,  the  polymerization  of  L-LA  by  RR,RR-3  was attempted. The observed rate for polymerization of L-LA by RR,RR-3  was roughly one order of magnitude slower (kobs = 4.4 x 10-5 s-1) than the corresponding rate for polymerization of rac-LA by RR,RR-3 (kobs = 2.4 x 10-4 s-1) when the same catalyst and monomer concentration is used. This clearly indicates that there is a mismatch between the chirality of the catalyst and that of the monomer. Therefore, it is possible that  the  RR,RR-3 enantiomer  of the catalyst preferentially polymerizes  D-LA,  and  that  the  SS,SS-3  enantiomer  of  the  catalyst  prefers  L-LA.    This assumption  could  be  tested  either  by  synthesizing  SS,SS-3  and  testing  its  rate  in  the polymerization  of  L-LA,  or  by  purchasing  D-LA  and  performing  the  polymerization  with RR,RR-3.   The above data suggests that the mechanism for stereocontrol exhibited by 3 over the polymerization of LA has a significant component of enantiomorphic site control. However, it is by no means conclusive that chain-end control is not playing a role in the stereoselectivity of the catalyst,  as  the  stereoselectivity  is  not  high  enough  to  obtain  strong  data  (ie  Pm  values approaching  one  for  isotactic  PLA).  More  selective  catalysts  must  be  developed  in  order  to obtain higher stereoselectivity, and control over the polymerization.   82  3.3 Conclusions and Future Work    Complex rac-3 has been shown to be a highly active catalyst for the polymerization of LA.  The  catalyst  is  more  active  than  previously  discovered  aluminum-based  compounds, although direct comparisons are difficult as there have been no examples of similar tridentate aluminum  catalysts  for  LA  ROP  reported.  Complex  3  is  also  the  first  known  indium  based catalyst for the polymerization of lactide and ?-butyrolactone, and the second indium catalyst active for the polymerization for the polymerization of ?-caprolactone.   The system was characterized using a number of techniques including analysis of the isolated polymer samples by GPC and NMR, and by monitoring the reaction kinetics using 1H NMR.  The  system  was  living;  a  linear  relationship  between  molecular  weight  and  monomer loading  was  observed.  The  polymer  samples  obtained  from  the  polymerization  had  narrow molecular weight distributions. Analysis of the polymer samples by 1H{1H} and 13 C{1H} NMR spectroscopy showed a degree of isotactic enrichment in the polymer chain. This is attributed to a site-control mechanism due to the chiral nature of the tridentate ligand.     Further investigation into the behaviour of the dinuclear catalyst (3) is required. First, a set  of  kinetic  experiments  should  be  performed  with  varying  [4].  If  the  dinuclear  species  is hindering the  catlalytic  process either at the early stages during initiation  or  at low [LA],  an overall decrease in the rate of polymerization should be observed with higher [4].  Alternatively a  similar  mononuclear  catalyst  (of  the  type  LnInR)  should  be  synthesized;  the  study  of  its behaviour in the polymerizations of LA would help to confirm some of the asumptions made about  the  dinculear  system;  namely,  that  the  initiation  period  is  due  to  dissociation  of  the dinuclear complex.     In order to improve catalyst selectivity a bulkier  tridentate chiral  catalyst should  be 83  synthesized. Because of the design of the ligand it is relatively facile to change substituents at both the amines as well as the phenol. From the crystal structures of complexes 1 - 6 it can be seen  that  the  indium  atom  is  not  very  sterically  crowded,  especially  where  the  dimethylated amine is bound to indium; therefore, we propose to introduce more steric bulk at that location in order to enhance the stereoselectivity of the catalyst. To this end an undergraduate student in our group has synthesized a terminally arylated pro-ligand of H2NNO (Scheme 3.4). Attempts are currently being made to synthesize an inium catalyst using this pro-ligand.   Scheme 3.4 Possible indium catalysts for the polymerization of LA  84   3.4 Experimental Materials Solvents (THF, toluene, dichloromethane) were degassed and dried using 3? molecular sieves in an mBraun Solvent Purification System.  The THF was further dried over sodium and distilled under N2. CD2Cl2 and CDCl3 were dried over CaH2, and degassed through a series of freeze-pump-thaw  cycles.  DL-lactide  and  L-lactide  was  purchased  from  Alfa  Aesar  and  purified  by recrystallization  from  iso-propanol,  and  toluene.  ?-butyrolactone  and  ?-caprolactone  were purchased from Aldrich and Alfa Aesar respectively. They were dried over CaH2, distilled under N2,  and  degassed  through  a  series  of  freeze-pump  cycles.  1,3,5-Trimethoxybenzene  was purchased from Aldrich and was used without further purification. General Methods Unless otherwise indicated, all air- and/or water-sensitive reactions were carried out under dry nitrogen using either an mBraun glovebox or standard Schlenk line techniques.  NMR spectra were  recorded  on  either  a  Bruker  Avance  400  MHz  or  600  MHz  spectrometer.    1H  NMR chemical shifts are given in ppm versus residual protons in deuterated solvents as follows: ? 7.27 CDCl3, ? 5.32 CD2Cl2.  13C{1H} NMR chemical shifts are given in ppm versus residual 13C in solvents  as  follows:  ?  77.2  CDCl3.  Molecular  weights  were  estimated  by  triple  detection  gel permeation chromatography (GPC - LLS) using a Waters liquid chromatograph equipped with a Waters 515 HPLC pump, Waters 717 plus  autosampler, Waters Styragel columns (4.6 ? 300 mm) HR5E, HR4 and HR2, Waters 2410 differential refractometer, Wyatt tristar miniDAWN (laser light scattering detector) and a Wyatt ViscoStar viscometer. A flow rate of 0.5 mL min?1 85  was  used  and  samples  were  dissolved  in  THF  (ca.  1  mg  mL?1).  Narrow  molecular  weight polystyrene standards were used for calibration purposes. NMR scale polymerization of rac-lactide with 3 In a teflon sealed NMR tube, 0.50 mL of 3 in CD2Cl2 (0.0048 M, 0.0024 mmol) was added to a solution of rac-lactide (66 mg; 0.47 mmol) and 1,3,5-trimethoxybenzene (5 mg; 0.03 mmol) in 0.48 mL of CD2Cl2. This mixture was immediately cooled in liquid nitrogen for transport to the instrument.  The  NMR  tube  was  warmed  to  room  temperature  before  it  was  inserted  into  the instrument (400 MHz Avance Bruker Spectrometer).  The polymerization was monitored to ca. 95% conversion.  Large-scale polymerization of rac-lactide using 3 Rac-lactide (129.3 mg, 0.92mmol) was dissolved in 6 mL of CH2Cl2 in a vial and stirred using a magnetic stir bar. To this solution 1mL of a stock solution of 3 in CH2Cl2 was added (0.0046 M; 0.0046 mmol). The reaction was allowed to proceed for 16 h and then concentrated to dryness. The resulting polymer was dissolved in a minimum amount of CH2Cl2 and added to cold wet methanol  (0  ?C,  7  mL).  The  polymer  precipitated  and  was  isolated  by  centrifugation.  The supernatant was decanted off and the polymer was dried under high vacuum for 2 hours. Large-scale polymerizations of ?-butyrolactone and ?-caprolactone using 3. Butyrolactone (391 mg, 4.5 mmol) was dissolved in 6 mL of THF  in a vial and stirred using a magnetic stir bar. To this solution 1mL of 3 in THF was added (10 mg; 0.01 mmol). The reaction was allowed to proceed  at room temperature  for 16  h and then concentrated to  dryness. The resulting  polymer  was  dissolved  in  a  minimum  amount  of  CH2Cl2  and  added  to  cold  wet methanol  (0  ?C,  15  mL).  The  polymer  precipitated  and  was  isolated  by  centrifugation.  The 86  supernatant was decanted off and the polymer was dried under high vacuum for 2 hours. Similar conditions were used for the polymerization of ?-caprolactone by 3.  87  BIBLIOGRAPHY  1.  Drumright, R. E., Gruber, P. R.,Henton, D. E., Adv. Mater. 2000, 12, 1841-1846. 2.  Gupta, B., Revagade, N.,Hilborn, J., Prog. Polym. Sci. 2007, 32, 455. 3.  Masaki, K., Kamini, N. R., Ikeda, H.,Iefuji, H., Appl. Environ. Microbiol. 2005, 71,   7548-7550. 4.  Reeve, M. S., McCarthy, S. P., Downey, M. J.,Gross, R. A., Macromolecules 1994, 27,   825-831. 5.  McNaught, A. D.,Wilkinson, A., Compendium of chemical terminoloy: IUPAC   recommendations. Blackwell Science: Oxford, 1997. 6.  Kricheldorf, H. R.,Kreiser-Saunders, I., Makromol. Chem. 1990, 191, 1057-66. 7.  Dubois, P., Jacobs, C., Jerome, R.,Teyssie, P., Macromolecules 1991, 24, 2266-70. 8.  Culkin, D. A., Jeong, W., Csihony, S., Gomez, E. D., Balsara, N. P., Hedrick, J.   L.,Waymouth, R. M., Angew. Chem., Int. Ed. 2007, 46, 2627-2630. 9.  Dechy-Cabaret, O., Martin-Vaca, B.,Bourissou, D., Chem. Rev. 2004, 104, 6147-6176. 10.  Van Hummel, G. J., Harkema, S., Kohn, F. E.,Feijen, J., Acta Crystallogr., Sect. B   1982, B38, 1679-81. 11.  Duda, A.,Penczek, S., Macromolecules 1990, 23, 1636-1639. 12.  Numata, K., Srivastava, R. K., Finne-Wistrand, A., Albertsson, A. C., Doi, Y.,Abe, H.,   Biomacromolecules 2007, 8, 3115-3125. 13.  Shuichi Matsumura, K. M. K. T., Macromol. Rapid Commun. 1997, 18, 477-482. 14.  Dong, H., Wang, H.-D., Cao, S.-G.,Shen, J.-C., Biotechnol. Lett. 1998, 20, 905-908. 15.  Kasperczyk, J. E., Macromolecules 1995, 28, 3937-3939. 16.  Kricheldorf, H. R.,Dunsing, R., Makromol. Chem. 1986, 187, 1611-25. 88  17.  Kricheldorf, H. R.,Kreiser, I., Makromol. Chem. 1987, 188, 1861-73. 18.  Lou, X., Detrembleur, C.,Jerome, R., Macromolecules 2002, 35, 1190-1195. 19.  Shibasaki, Y., Sanada, H., Yokoi, M., Sanda, F.,Endo, T., Macromolecules 2000, 33,   4316-4320. 20.  Bourissou, D., Martin-Vaca, B., Dumitrescu, A., Graullier, M.,Lacombe, F.,   Macromolecules 2005, 38, 9993-9998. 21.  Pratt, R. C., Lohmeijer, B. G. G., Long, D. A., Waymouth, R. M.,Hedrick, J. L., J. Am.   Chem. Soc. 2006, 128, 4556-4557. 22.  Csihony, S., Culkin, D. A., Sentman, A. C., Dove, A. P., Waymouth, R. M.,Hedrick, J.   L., J. Am. Chem. Soc. 2005, 127, 9079-9084. 23.  Dove, A. P., Li, H., Pratt, R. C., Lohmeijer, B. G. G., Culkin, D. A., Waymouth, R.   M.,Hedrick, J. L., Chem. Commun. 2006, 2881-2883. 24.  Fredrik Nederberg, E. F. C. M. M. T. G. J. L. H., Angew. Chem. Int. Ed. 2001, 40,   2712-2715. 25.  Chamberlain, B. M., Cheng, M., Moore, D. R., Ovitt, T. M., Lobkovsky, E. B.,Coates,   G. W., J. Am. Chem. Soc. 2001, 123, 3229-3238. 26.  Chisholm, M. H., Eilerts, N. W., Huffman, J. C., Iyer, S. S., Pacold, M.,Phomphrai, K.,   J. Am. Chem. Soc. 2000, 122, 11845-11854. 27.  Tang, H.-Y., Chen, H.-Y., Huang, J.-H.,Lin, C.-C., Macromolecules 2007, 40, 8855-  8860. 28.  Chisholm, M. H., Gallucci, J.,Phomphrai, K., Chem. Commun. 2003, 48-49. 29.  Chisholm, M. H., Gallucci, J. C.,Phomphrai, K., Inorg. Chem. 2005, 44, 8004-8010. 30.  Lian, B., Thomas, C. M., Casagrande, O. L., Jr., Roisnel, T.,Carpentier, J.-F.,   Polyhedron 2007, 26, 3817-3824. 89  31.  Wu, J.-C., Huang, B.-H., Hsueh, M.-L., Lai, S.-L.,Lin, C.-C., Polymer 2005, 46, 9784-  9792. 32.  Ejfler, J., Kobylka, M., Jerzykiewicz, L. B.,Sobota, P., Dalton Trans. 2005, 2047-2050. 33.  Chivers, T., Fedorchuk, C.,Parvez, M., Organometallics 2005, 24, 580-586. 34.  Chisholm, M. H., Gallucci, J. C.,Phomphrai, K., Inorg. Chem. 2004, 43, 6717-25. 35.  Chen, H.-Y., Tang, H.-Y.,Lin, C.-C., Polymer 2007, 48, 2257-2262. 36.  Chisholm, M. H., Gallucci, J. C.,Phomphrai, K., Inorg. Chem. 2004, 43, 6717-6725. 37.  Zhong, Z., Schneiderbauer, S., Dijkstra, P. J., Westerhausen, M.,Feijen, J., Polym. Bull.   2003, 51, 175-182. 38.  Abderramane Amgoune, C. M. T. S. I. T. R. J.-F. C., Angew. Chem. Int. Ed. 2006, 45,   2782-2784. 39.  Abderramane Amgoune, C. M. T. J.-F. C., Macromol. Rapid Commun. 2007, 28, 693-  697. 40.  Ovitt, T. M.,Coates, G. W., J. Am. Chem. Soc. 2002, 124, 1316-1326. 41.  Liu, X., Shang, X., Tang, T., Hu, N., Pei, F., Cui, D., Chen, X.,Jing, X.,   Organometallics 2007, 26, 2747-2757. 42.  Cai, C.-X., Amgoune, A., Lehmann, C. W.,Carpentier, J.-F., Chem. Commun. 2004,   330-331. 43.  Shang, X., Liu, X.,Cui, D., J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 5662-5672. 44.  Miao, W., Li, S., Zhang, H., Cui, D., Wang, Y.,Huang, B., J. Organomet. Chem. 2007,   692, 4828-4834. 45.  Platel, R. H., Hodgson, L. M., White, A. J. P.,Williams, C. K., Organometallics 2007,   26, 4955-4963. 46.  Yang, Y., Li, S., Cui, D., Chen, X.,Jing, X., Organometallics 2007, 26, 671-678. 90  47.  Hodgson, L. M., White, A. J. P.,Williams, C. K., J. Polym. Sci., Part A: Polym. Chem.   2006, 44, 6646-6651. 48.  Alaaeddine, A., Amgoune, A., Thomas, C. M., Dagorne, S., Bellemin-Laponnaz,   S.,Carpentier, J.-F., Eur. J. Inorg. Chem. 2006, 3652-3658. 49.  Westmoreland, I.,Arnold, J., Dalton Trans. 2006, 4155-4163. 50.  Skvortsov, G. G., Yakovenko, M. V., Castro, P. M., Fukin, G. K., Cherkasov, A. V.,   Carpentier, J.-F.,Trifonov, A. A., Eur. J. Inorg. Chem. 2007, 3260-3267. 51.  Miao, W., Li, S., Cui, D.,Huang, B., J. Organomet. Chem. 2007, 692, 3823-3834. 52.  Heck, R., Schulz, E., Collin, J.,Carpentier, J.-F., J. Mol. Catal. A: Chem. 2007, 268,   163-168. 53.  Amgoune, A., Thomas, C. M., Roisnel, T.,Carpentier, J.-F., Chem. Eur. J. 2005, 12,   169-179. 54.  Ma, H., Spaniol, T. P.,Okuda, J., Dalton Trans. 2003, 4770-4780. 55.  O'Keefe, B. J., Hillmyer, M. A.,Tolman, W. B., J. Chem. Soc., Dalton Trans. 2001,   2215-2224. 56.  Haiyan Ma, T. P. S. J. O., Angew. Chem. Int. Ed. 2006, 45, 7818-7821. 57.  Dove, A. P., Gibson, V. C., Marshall, E. L., Rzepa, H. S., White, A. J. P.,Williams, D.   J., J. Am. Chem. Soc. 2006, 128, 9834-9843. 58.  Kowalski, A., Libiszowski, J., Biela, T., Cypryk, M., Duda, A.,Penczek, S.,   Macromolecules 2005, 38, 8170-8176. 59.  Nimitsiriwat, N., Marshall, E. L., Gibson, V. C., Elsegood, M. R. J.,Dale, S. H., J. Am.   Chem. Soc. 2004, 126, 13598-13599. 60.  Lahcini, M., Castro, P. M., Kalmi, M., Leskelae, M.,Repo, T., Organometallics 2004,   23, 4547-4549. 91  61.  Wang, X., Liao, K., Quan, D.,Wu, Q., Macromolecules 2005, 38, 4611-4617. 62.  McGuinness, D. S., Marshall, E. L., Gibson, V. C.,Steed, J. W., J. Polym. Sci., Part A:   Polym. Chem. 2003, 41, 3798-3803. 63.  Gibson, V. C., Marshall, E. L., Navarro-Llobet, D., White, A. J. P.,Williams, D. J., J.   Chem. Soc., Dalton Trans. 2002, 4321-4322. 64.  O'Keefe, B. J., Monnier, S. M., Hillmyer, M. A.,Tolman, W. B., J. Am. Chem. Soc.   2001, 123, 339-340. 65.  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, 11350-11359. 66.  Hormnirun, P., Marshall, E. L., Gibson, V. C., White, A. J. P.,Williams, D. J., J. Am.   Chem. Soc. 2004, 126, 2688-2689. 67.  Zhong, Z., Dijkstra, P. J.,Feijen, J., J. Am. Chem. Soc. 2003, 125, 11291-11298. 68.  Emig, N., Nguyen, H., Krautscheid, H., Reau, R., Cazaux, J. B.,Bertrand, G.,   Organometallics 1998, 17, 3599-3608. 69.  Hormnirun, P., Marshall, E. L., Gibson, V. C., Pugh, R. I.,White, A. J. P., Proc. Natl.   Acad. Sci. U. S. A. 2006, 103, 15343-15348. 70.  Ovitt, T. M.,Coates, G. W., J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 4686-4692. 71.  Trofimoff, L., Aida, T.,Inoue, S., Chem. Lett. 1987, 991-4. 72.  Radano, C. P., Baker, G. L.,Smith, M. R., III, J. Am. Chem. Soc. 2000, 122, 1552-  1553. 73.  Zhong, Z., Dijkstra, P. J.,Feijen, J., Angew. Chem., Int. Ed. 2002, 41, 4510-4513. 74.  Nomura, N., Ishii, R., Yamamoto, Y.,Kondo, T., Chem. Eur. J. 2007, 13, 4433-4451. 75.  Chmura, A. J., Chuck, C. J., Davidson, M. G., Jones, M. D., Lunn, M. D., Bull, S.   D.,Mahon, M. F., Angew. Chem., Int. Ed. 2007, 46, 2280-2283. 92  76.  Finne, A., Reema,Albertsson, A.-C., J. Polym. Sci., Part A: Polym. Chem. 2003, 41,   3074-3082. 77.  Kowalski, A., Duda, A.,Penczek, S., Macromolecules 2000, 33, 7359-7370. 78.  Stevels, W. M., Ankone, M. J. K., Dijkstra, P. J.,Feijen, J., Macromolecules 1996, 29,   6132-6138. 79.  Stevels, W. M., Ankone, M. J. K., Dijkstra, P. J.,Feijen, J., Macromolecules 1996, 29,   3332-3. 80.  Zhong, Z., Dijkstra, P. J., Birg, C., Westerhausen, M.,Feijen, J., Macromolecules 2001,   34, 3863-3868. 81.  Hans R. Kricheldorf, M. B., Nico Scharnagl, Macromolecules 1988, 21, 286-293. 82.  Aida, T.,Inoue, S., Acc. Chem. Res. 1996, 29, 39-48. 83.  Emig, N., Reau, R., Krautscheid, H., Fenske, D.,Bertrand, G., J. Am. Chem. Soc. 1996,   118, 5822-5823. 84.  Dumitrescu, A., Martin-Vaca, B., Gornitzka, H., Cazaux, J.-B., Bourissou,   D.,Bertrand, G., Eur. J. Inorg. Chem. 2002, 1948-1951. 85.  Faure, J.-L., Gornitzka, H., Reau, R., Stalke, D.,Bertrand, G., Eur. J. Inorg. Chem.   1999, 2295-2299. 86.  Chisholm, M. H., Patmore, N. J.,Zhou, Z., Chem. Commun. 2005, 127-129. 87.  Agustin, D., Rima, G., Gornitzka, H.,Barrau, J., J. Organomet. Chem. 1999, 592, 1-10. 88.  Thakur, S. S., Chen, S.-W., Li, W., Shin, C.-K., Kim, S.-J., Koo, Y.-M.,Kim, G.-J., J.   Organomet. Chem. 2006, 691, 1862. 89.  Chisholm Malcolm, H., Gallucci, J.,Phomphrai, K., Inorg. Chem. 2002, 41, 2785-94. 90.  Chisholm, M. H., Huffman, J. C.,Phomphrai, K., J. Chem. Soc., Dalton Trans. 2001,   222-224. 93  91.  Dove, A. P., Gibson, V. C., Marshall, E. L., White, A. J. P.,Williams, D. J., Chem.   Commun. 2001, 283-284. 92.  Cheng, M., Attygalle, A. B., Lobkovsky, E. B.,Coates, G. W., J. Am. Chem. Soc. 1999,   121, 11583-11584. 93.  Natta, G., Pino, P., Corradini, P., Danusso, F., Mantica, E., Mazzanti, G.,Moraglio, G.,   J. Am. Chem. Soc. 1955, 77, 1708-1710. 94.  Bovey, F. A.,Mirau, P. A., NMR of Polymers. Academic Press: San Diego, 1996. 95.  Ovitt, T. M.,Coates, G. W., J. Am. Chem. Soc. 1999, 121, 4072-4073. 96.  Russell, S. K., Gamble, C. L., Gibbins, K. J., Juhl, K. C. S., Mitchell, W. S., III,   Tumas, A. J.,Hofmeister, G. E., Macromolecules 2005, 38, 10336-10340. 97.  Greenwood, N. N.,Earnshaw, A., Chemistry of the Elements. 2nd Ed. Butterworth-  Heinemann: Boston, 1997. 98.  Carlin, J. F. J., Mineral Commodities Studies: Indium. In U.S. Geological Survey:   2006; pp 82-83. 99.  Fringuelli, F., Piermatti, O., Pizzo, F.,Vaccaro, L., Curr. Org. Chem. 2003, 7, 1661. 100.  Loh, T.-P.,Chua, G.-L., Chem. Commun. 2006, 2739-2749. 101.  Teo, Y. C.,Loh, T. P., Org. Lett. 2005, 7, 2539-2541. 102.  Kamlesh K. Chauhan, C. G. F., J. Chem. Soc., Perkin Trans. 1 2000, 3015-3019. 103.  Hsieh, I. P., Huang, C.-H., Lee, H. M., Kuo, P.-C., Huang, J.-H., Lee, H.-I., Cheng, J.-  T.,Lee, G.-H., Inorg. Chim. Acta 2006, 359, 497. 104.  Douglas, A., F., Patrick, B., O.,Mehrkhodavandi, P., Angew. Chem. Int. Ed. Engl.   2008, 47, 2290-3. 105.  Mitchell, J. M.,Finney, N. S., Tetrahedron Lett. 2000, 41, 8431. 106.  Xu, C., Baum, T. H., Guzei, I.,Rheingold, A. L., Inorg. Chem. 2000, 39, 2008-2010. 94  107.  Chou, T.-Y., Chi, Y., Huang, S.-F., Liu, C.-S., Carty, A. J., Scoles, L.,Udachin, K. A.,   Inorg. Chem. 2003, 42, 6041-6049. 108.  Lewinski, J., Zachara, J.,Starowieyski, K. B., J. Chem. Soc., Dalton Trans. 1997, 4217-  4222. 109.  Schumann, H., Kaufmann, J.,Dechert, S., Z. Anorg. Allg. Chem. 2004, 630, 1999-2005. 110.  Suh, S.,Hoffman, D. M., J. Am. Chem. Soc. 2000, 122, 9396-9404. 111.  Kim, J., Bott, S. G.,Hoffman, D. M., Inorg. Chem. 1998, 37, 3835-3841. 112.  Sussek, H., Stark, O., Devi, A., Pritzkow, H.,Fischer, R. A., J. Organomet. Chem.   2000, 602, 29. 113.  Lin, Z., Jiang, F., Chen, L., Yuan, D.,Hong, M., Inorg. Chem. 2005, 44, 73-6. 114.  Zhou, Y.,Richeson, D. S., Organometallics 1995, 14, 3558-61. 115.  Stender, M., Eichler, B. E., Hardman, N. J., Power, P. P., Prust, J., Noltemeyer,   M.,Roesky, H. W., Inorg. Chem. 2001, 40, 2794-9. 116.  Eric N. Jacobsen, W. Z., Alexander R. Muci, James R. Ecker, Li Deng, J. Am. Chem.   Soc. 1991, 113, 7063-7064. 117.  Holm, T., Tetrahedron Lett. 1966, 3329-36. 118.  Burger, H., Cichon, J., Goetze, U., Wannagat, U.,Wismar, H. J., J. Organomet. Chem.   1971, 33, 1. 119.  Evans pKa Table. http://www2.lsdiv.harvard.edu/labs/evans/pdf/evans_pKa_table.pdf 120.  Schlosser, M., Organometallics in Synthesis: A Manual, 2nd ed. Wiley: Chichester,   2002; p 292. 121.  Schlosser, M.,Hartmann, J., Angew. Chem. 1973, 85, 544-5. 122.  Atkins, P.,de Paula, J., Physical Chemistry, 7th ed. W.H. Freeman and Company: New   York, 2002. 95   123.  Berger, S.,Siegmar, B., 200 and More NMR Experiments: A Practical Course.   WILEY-VCH: Weinham, 2004; p 141-148. 124.  Skoog, D. A.,Holler, J. F., Principles of Instrumental Analysis, 5th ed. Nelson   Thomson Learning: Toronto, 1998.   96  APPENDIX 1    X-Ray Crystallography Data  Table A.1. Selected crystal structure data for 1, 2, and 3     1  2  3 Formula  C25H45N2OIn  C29.5H57N2OIn2Cl  C62H97N4O3In2Cl3 mol wt  504.45  720.86  1282.43 Crystal System  orthorhombic  monoclinic  monoclinic Space Group  P bca  C 2/c    P 21/c   a/?  19.6863(12)   20.904(3)  16.905(2) b/?  13.1562(8)  13.314(2)   20.967(2) c/?  20.1573(10)   25.413(5)  18.428(1) ?/?  90  90  90 ?/?  90   111.765(9)  94.020(4)  ?/?  90  90  90 V/?3  5220.68(5)  6569(2)  6515(1) ?calc/g cm-3  1.284  1.458  1.307 Z  8  8  4 ? (Mo-K?) / mm-1  9.22  15.08  8.74 80087/6271  38177/8249  42181/8470  Unique (Rint)  (0.059)  (0.05)  (0.062) Residuals (refined on F2, all data): R1; wR2 0.049; 0.062  0.073; 0.106  0.122; 0.181 Residuals (refined on F): R1; wR2 0.027; 0.054  0.045; 0.095  0.081; 0.168         97    Table A.2. Selected crystal data for 4?Py, 5, and 6    4?Py  5  6 Formula  C35H52N3OInCl2  C59H86N6O4In2Cl2  C61H93N4O3In2Cl3 mol wt  716.52  1135.79  1266.38 Crystal System  monoclinic  primitive  triclinic Space Group  P 21/c   P -1    P -1  a/?  11.2696(11)  8.5090(8)  11.5642(5)  b/?  20.749(2)  9.6905(9)  16.8754(8) c/?  15.4251(14)  16.916(2)  18.1223(8) a/?  90  87.023(6)  92.263(2) b/?  100.875(5)  88.076(6)  106.473(2) g/?  90  89.371(6)  109.942(2) V/?3  3542.0(6)  1392.1(2)  3151.5(2) rcalc/g cm-3  1.344  1.355  1.335 Z  4  1  2 m (Mo-Ka) / mm-1  8.48  9.69  9.03 38088/8462  14217/4962   48876/11134 Collected/ Unique (Rint)  (0.021)  (0.044)  (0.048) Residuals (refined on F2, all data): R1; wR2  0.026; 0.050  0.079; 0.164  0.065; 0.063 Residuals (refined on F): R1; wR2  0.020; 0.047  0.070; 0.159  0.034; 0.055  

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 11 0
Canada 8 0
China 7 4
Germany 5 10
India 5 0
France 3 0
Indonesia 1 0
Poland 1 0
Taiwan 1 0
City Views Downloads
Unknown 14 11
Washington 6 0
Shenzhen 4 4
Beijing 3 0
Vancouver 2 0
Olpe 2 1
Ashburn 2 0
Essen 1 0
Mountain View 1 0
Richibucto 1 0
Bonn 1 0
Chicago 1 0
Bandung 1 0

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

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0059291/manifest

Comment

Related Items