Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of group 4 and 5 complexes with N,O chelating supporting ligands as catalysts for the alpha-alkylation… Lauzon, Jean Michel 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_fall_lauzon_jean michel.pdf [ 9.55MB ]
Metadata
JSON: 24-1.0073860.json
JSON-LD: 24-1.0073860-ld.json
RDF/XML (Pretty): 24-1.0073860-rdf.xml
RDF/JSON: 24-1.0073860-rdf.json
Turtle: 24-1.0073860-turtle.txt
N-Triples: 24-1.0073860-rdf-ntriples.txt
Original Record: 24-1.0073860-source.json
Full Text
24-1.0073860-fulltext.txt
Citation
24-1.0073860.ris

Full Text

DEVELOPMENT OF GROUP 4 AND 5 COMPLEXES WITH N,O CHELATING SUPPORTING LIGANDS AS CATALYSTS FOR THE α-ALKYLATION OF AMINES  by Jean Michel Lauzon  B.Sc., Mount Allison University, 2007  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  May 2013  © Jean Michel Lauzon, 2013  Abstract  The use of stoichiometric, catalytic and theoretical methods in the development of an early transition metal catalyst for the α-alkylation of amines is described herein.  The  investigation is primarily focused on a series of mono(amidate) complexes of tantalum with varying steric and electronic properties. The amidate binding mode and catalytic activity of these complexes is significantly influenced by sterics. Corresponding bis(amidate) complexes are less active as catalysts for the α-alkylation of amines but offers a platform to study the hemilability of amidate ligands as well as tantalaziridine formation in these systems. A model 5membered metallacycle is synthesized and characterized. Isotopic labeling studies with the most active mono(amidate) precatalyst reveal off-cycle reactions and suggest that tantalaziridine formation is rapid and reversible. Preliminary kinetic investigations implicate alkene insertion as the turnover limiting step, consistent with stoichiometric investigations. In addition, the use of radical probes in ligand backbones and an alkene substrate contradicts a one electron mechanism. Quantum chemical calculations are used to develop a theoretical model of the proposed catalytic cycle. The hemi-lability of amidate ligands is highlighted with the optimization of both κ1(O) and κ2(N,O) minima and transition states. Here, protonolysis is calculated to be the turnover limiting step with small changes in geometry having a significant effect on the potential energy surface. The unlikelihood of a radical mechanism is supported by the computations of triplet species. A survey of established steric parameters has been completed for asymmetric amidate ligands to be used as a predictive tool for catalyst design. The calculated values can be related to the catalytic activity of mono(amidate) and axially chiral tantalum precatalysts.  ii  Diamide and diurea proligands featuring a neutral chalcogen atom tether are installed on zirconium and tantalum. The zirconium species form well-defined κ4(N,N,O,O) complexes with fluxional behaviour observed for the tantalum species in solution. No evidence of bonding is observed between the chalcogen donor and any metal centre. Fundamental differences in the redox potentials for ligands and complexes are investigated using cyclic voltammetry. The tantalum complexes are found to catalyze the α-alkylation of amines with the zirconium species being competent precatalysts for hydroamination.  iii  Preface  Parts of the research conducted for this thesis were carried out collaboratively with other members of the Schafer and Fryzuk research groups. I, in consultation with my supervisors, Dr. Laurel Schafer and Dr. Michael Fryzuk, designed all of the experiments described herein. I have performed all of these experiments save the following specific instances. Byproduct 13 was isolated and characterized by Dr. Patrick Eisenberger (Chapter 2). The primary and secondary aminoalkene substrates used for the hydroamination screening presented in Chapter 4 were synthesized by Eugene Chong.  Electrochemical analysis (Chapter 4) was performed in  collaboration with Nathan R. Halcovitch, including set-up of the instrumentation. Initial studies for isotopically labeled amidate ligands (Chapter 5) were performed by Alexandru Vlasceanu, an undergraduate researcher, under my supervision. However, the experimental design was solely mine. The data for the solid-state molecular structures presented herein was collected by Neal Yonson, Jacky Yim or Scott Ryken while I performed the final refinements. A portion of Section 2.1.2 and 2.1.3 has been reported in Dalton Transactions published by The Royal Society of Chemistry as: Lauzon, J. M. P.; Schafer, L. L. “Tantallaaziridines: from synthesis to catalytic applications” Dalton Trans. 2012, 41, 11539–11550. Figure 2.10 has also appeared in Angewandte Chemie International Edition published by Wiley-VCH: Eisenberger, P.; Ayinla, R. O.; Lauzon, J. M. P.; Schafer, L. L. “Tantalum–Amidate Complexes for the Hydroaminoalkylation of Secondary Amines: Enhanced Substrate Scope and Enantioselective Chiral Amine Synthesis” Angew. Chem., Int. Ed. 2009, 48, 8361–8365.  iv  Table of Contents  Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents ...........................................................................................................................v List of Tables ..................................................................................................................................x List of Figures .............................................................................................................................. xii List of Schemes ........................................................................................................................... xxi List of Abbreviations and Acronyms ................................................................................... xxviii Acknowledgements ................................................................................................................ xxxiii Dedication ............................................................................................................................... xxxiv Chapter 1: The Direct α-Alkylation of Amines ..........................................................................1 1.1  Historic Overview ............................................................................................................. 1  1.2  Stoichiometric Zirconium Reagents ................................................................................. 5  1.3  A Zirconium Precatalyst for the Intramolecular α-Alkylation of Amines...................... 13  1.4  Titanium Precatalysts for the α-Alkylation of Amines ................................................... 16  1.5  Group 5 Metal Catalysts ................................................................................................. 20  1.6  Late Transition Metal Catalysts ...................................................................................... 23  1.7  Scope of Thesis ............................................................................................................... 26  Chapter 2: Tantalum Amidate Complexes as Precatalysts for the α-Alkylation of Amines ........................................................................................................................................................29 2.1  Introduction..................................................................................................................... 29  2.1.1  Amidates as Ligands for Early Transition Metal Complexes ............................... 29  v  2.1.2  Synthesis, Structure, and Characterization of Tantalaziridines............................. 32  2.1.3  Stoichiometric Reactivity of Tantalaziridines ....................................................... 41  2.1.4  Scope of Chapter ................................................................................................... 47  2.2  Results and Discussion ................................................................................................... 48  2.2.1  Synthesis and Characterization of Mono(amidate) Tantalum Complexes ............ 48  2.2.2  Catalytic Reactivity of Mono(amidate) Tantalum Complexes ............................. 55  2.2.3  Synthesis and Characterization of Bis(amidate) Tantalum Complexes ................ 59  2.2.3.1  Characterization of a κ1(O), κ2(N,O) Bis(amidate) Tantalum Complex ........ 60  2.2.3.2  Synthesis and Characterization of a Bis(amidate) Tantalaziridine ................ 66  2.2.3.3  Reactivity of Tantalaziridine 9 with Unsaturated Molecules ......................... 70  2.2.3.4  Attempted Synthesis of an Electronically Modified Bis(amidate)  Tantalaziridine ................................................................................................................ 72 2.2.4  Probing the Mechanism of Mono(amidate) Tantalum Catalyzed α-Alkylation of  Amines ............................................................................................................................... 75 2.2.4.1  Byproduct Formation ..................................................................................... 76  2.2.4.2  Isotopic Labeling Studies ............................................................................... 77  2.2.4.3  Reaction Kinetics ........................................................................................... 82  2.2.5  Probing a One Electron Mechanism for the α-Alkylation of Amines ................... 89  2.2.5.1  Cyclopropyl-based Mono(amidate) Precatalysts ........................................... 90  2.2.5.2  Substrate Containing an Ultrafast Radical Probe ........................................... 92  2.3  Conclusions..................................................................................................................... 95  2.4  Experimental ................................................................................................................... 99  2.4.1  Materials and Methods .......................................................................................... 99  vi  2.4.2  Synthesis and Characterization of Tantalum Amidate Complexes ..................... 100  2.4.3  General Procedure for Screening of Mono(amidate) Precatalysts ...................... 111  2.4.4  Isotopic Labeling Studies .................................................................................... 112  2.4.5  Kinetic Investigations .......................................................................................... 112  Chapter 3: Computational Modeling of the Catalytic Cycle for the α-Alkylation of Amines ......................................................................................................................................................115 3.1  Introduction................................................................................................................... 115  3.1.1  Computational Investigations Involving Amidate Species ................................. 115  3.1.2  Modeling of the Intramolecular α-Alkylation of Amines ................................... 117  3.1.3  Scope of Chapter ................................................................................................. 119  3.2  Results and Discussion ................................................................................................. 122  3.2.1  Tantalaziridine Formation ................................................................................... 122  3.2.2  Dimethylamine as a Model Substrate.................................................................. 125  3.2.2.1  Insertion of 1-Octene into Tantalaziridine ................................................... 125  3.2.2.2  Protonolysis of 5-Membered Metallacycle by N-Methylaniline .................. 128  3.2.2.3  β-Hydrogen Abstraction to Release Product and Reform Tantalaziridine... 130  3.2.2.4  Potential Energy Surface .............................................................................. 133  3.2.3  Potential Energy Surface using N-Methylaniline as a Substrate ......................... 136  3.2.3.1  Namidate cis to the Naziridine .............................................................................. 136  3.2.3.2  Oamidate cis to the Naziridine .............................................................................. 140  3.2.3.3  Alternative Pathways using N-methylaniline as a Substrate ........................ 147  3.2.4  Triplet Species as Possible Transition States and Intermediates ......................... 152  3.2.5  Development of a Steric Parameter for Tantalum Precatalysts ........................... 157  vii  3.3  Conclusions................................................................................................................... 168  3.4  Experimental ................................................................................................................. 171  3.4.1  Computational Methods ...................................................................................... 171  Chapter 4: Tethered Amidate and Ureate Ligands with a Neutral Donor Bridge .............173 4.1  Introduction................................................................................................................... 173  4.1.1  Tethered Diamido Ligands with Neutral Donors ................................................ 173  4.1.2  Scope of Chapter ................................................................................................. 179  4.2  Results and Discussion ................................................................................................. 180  4.2.1  Tethered Bis(amidate) Complexes of Zr and Ta ................................................. 180  4.2.2  Tethered Bis(ureate) Complexes of Zr and Ta .................................................... 187  4.2.3  Electrochemical Analysis of Tethered Bis(amidate) and Bis(ureate) Complexes .... ............................................................................................................................. 198  4.2.4  Catalytic Activity of Tethered Bis(amidate) and Bis(ureate) Complexes ........... 204  4.3  Conclusions................................................................................................................... 208  4.4  Experimental ................................................................................................................. 210  4.4.1  Materials and Methods ........................................................................................ 210  4.4.2  Synthesis and Characterization of Proligands and Complexes ........................... 210  4.4.3  General Procedures for Catalytic Screening ....................................................... 221  4.4.4  General Procedure for Electrochemical Analysis ............................................... 222  Chapter 5: Summary and Future Directions .........................................................................224 5.1  Summary ....................................................................................................................... 224  5.2  Future Directions .......................................................................................................... 228  5.2.1  In Situ Catalyst Monitoring ................................................................................. 228  viii  5.2.2 5.3  Alternate Mechanistic Pathways ......................................................................... 232  Concluding Remarks .................................................................................................... 235  References ...................................................................................................................................237 Appendices ..................................................................................................................................244 Appendix A X-Ray Crystallographic Data ............................................................................ 244 Appendix B Selected NMR Spectra ....................................................................................... 251 Appendix C Supplemental Computational Data .................................................................... 272 C.1  Alternate Mechanistic Pathways for the α-Alkylation of Amines ...................... 272  C.2  Supplemental Data for Steric Parameter Calculations ........................................ 276  Appendix D Supplemental Electrochemical Data .................................................................. 278  ix  List of Tables  Table 2.1 Crystallographic data and relevant metrical parameters for 1 – 5. Simplified illustration showing idealized geometry given for reference. ....................................................... 53 Table 2.2 Screening of mono(amidate) tantalum precatalysts for the α-alkylation of Nmethylaniline with 1-octene.  Data for Ta(NMe2)539 and [Cl3Ta(NMePh)2]240 is shown for  comparative purposes.................................................................................................................... 56 Table 2.3 Selected bond lengths (Å) and angles (°) for complex 9. ............................................ 68 Table 2.4 Selected bond lengths (Å) and angles (°) for complex 10. .......................................... 71 Table 2.5 Selected bond lengths (Å) and angles (°) for complex 11. .......................................... 74 Table 2.6 Comparison of deuterium incorporation or retention in 6-d for group 5 catalysts. ..... 80 Table 2.7 Comparison of the initial rates of consumption of N-methylaniline to the absolute initial concentration of species present in the reaction mixture. Initial concentrations of Nmethylaniline are given as representative examples. .................................................................... 87 Table 3.1 Reactivity data for a series of axially chiral tantalum precatalysts. ........................... 164 Table 4.1 Selected bond lengths (Å) and angles (°) for complexes 28 and 29. ......................... 184 Table 4.2 Selected bond lengths (Å) and angles (°) for complexes 30 and 31. ......................... 185 Table 4.3 Selected bond lengths (Å) and angles (°) for complexes 32 and 33. ......................... 192 Table 4.4 Selected bond lengths (Å) and angles (°) for complexes 30 and 31. ......................... 195 Table 4.5 Catalytic screening of complexes 28 – 35 for the intermolecular α-alkylation of amines (α-alk, top) and intramolecular hydroamination of primary (1° HA, middle) and secondary amines (2° HA, bottom).  Conversions determined by 1H NMR spectroscopy using 1,3,5-  trimethoxybenzene as an internal standard. ................................................................................ 205  x  Table A.1 Crystallographic parameters for mono(amidate) complexes (Chapter 2). ................ 244 Table A.2 Crystallographic parameters for bis and tris(amidate) complexes (Chapter 2). ........ 246 Table A.3 Crystallographic parameters for diamide and diurea proligands (Chapter 4). .......... 247 Table A.4 Crystallographic parameters for tethered bis(amidate) complexes (Chapter 4). ....... 249 Table A.5 Crystallographic parameters for tethered bis(ureate) complexes (Chapter 4)........... 250  xi  List of Figures  Figure 1.1 Titanium precatalysts for the direct α-alkylation of amines. ...................................... 17 Figure 1.2 2-Aminopyridinate titanium catalysts for the α-alkylation of amines. ....................... 20 Figure 1.3 Enantioselective tantalum α-alkylation of amines catalysts supported by biphenyl and binaphthyl bis(amidate) ligands. ................................................................................................... 22 Figure 2.1 Possible binding modes for amidate ligands. ............................................................. 30 Figure 2.2 General forms of three related compounds: tantalaziridine (left), η2-imine complex (middle) and iminoacyl complex (right). ...................................................................................... 33 Figure 2.3 Metallacycle intermediates for the insertion of benzaldehyde into the tantalaziridine.87 ............................................................................................................................ 46 Figure 2.4 Amide proligands used for synthesizing mono(amidate) tantalum complexes. ......... 49 Figure 2.5 ORTEP representations of the solid-state molecular structures for 1, 2, 3, 4 and 5 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. ............... 52 Figure 2.6 Section of the 1H NMR spectra (400 MHz, 25 °C) for the thermolysis of 3. The bottom spectrum was collected before heating; the top spectrum was collected after heating to 130 °C for 36 hours. ...................................................................................................................... 59 Figure 2.7 ORTEP representation of the solid-state molecular structure of 7 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. ...................................... 61 Figure 2.8 Variable temperature (-80.9 – 101.3 °C) 400 MHz 1H NMR study of 7. .................. 62 Figure 2.9 ORTEP representation of the solid-state molecular structure of 8 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. ...................................... 65  xii  Figure 2.10 ORTEP representation of the solid-state molecular structure of 9 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms and most of the carbon atoms of the ligands omitted for clarity. ........................................................................................................................ 68 Figure 2.11 ORTEP representation of the solid-state molecular structure of 10 drawn at 50% probability for thermal ellipsoids. Most hydrogen atoms omitted for clarity. .............................. 71 Figure 2.12 ORTEP representation of the solid-state molecular structure of 11 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. ...................................... 74 Figure 2.13 Consumption of amine as a function of time for the α-alkylation of N-methylaniline and is -d3 and -d4 deuterated analogues with 1-octene catalyzed by 3. Solid lines depict the leastsquares fits (R2 = 0.971, 0.982, 0.980) to the data points. ............................................................ 81 Figure 2.14 Consumption of amine as a function of time for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Solid line depicts the least-squares fit (R2 = 0.994) to the data points. ............................................................................................................................................ 83 Figure 2.15 Consumption of amine as a function of time for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Solid line depicts the least-squares fit (R2 = 0.991) to the data points. ............................................................................................................................................ 84 Figure 2.16 Consumption of amine as a function of time for the α-alkylation of N-methylaniline with 1-octene catalyzed by varying concentrations of 3. Solid lines depict the least-squares fit to the data points. .............................................................................................................................. 85 Figure 2.17 Observed initial rates of consumption of N-methylaniline as a function of catalyst concentration of 3 (ccatalyst). Solid line depicts the least-squares fit (R2 = 0.947) to the data points. ....................................................................................................................................................... 85  xiii  Figure 2.18 Observed initial rates of consumption of N-methylaniline and initial rates of formation of 6 as a function of initial alkene concentration. Solid and dashed lines depict the least-squares fits (R2 = 0.985 and 0.993 respectively) to the data points. .................................... 88 Figure 2.19 ORTEP representations of the solid-state molecular structures of 15 (left) and 16 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. .... 91 Figure 2.20 1H NMR spectra (300 MHz, 25 °C) of the α-alkylation of N-methylaniline with cyclopropyl-based alkene 17 catalyzed by precatalyst 3. Bottom spectrum corresponds to time zero; top spectrum corresponds to heating at 130 °C for 24 hours. .............................................. 95 Figure 2.21 Typical kinetics experiment monitoring the conversion of N-methylaniline (δH = 6.37) and 1-octene (δH = 5.00, 5.78) to product 6 (δH = 6.46) catalyzed by 3 with integrations relative to 1,3,5-trimethoxybenzene (δH = 6.14). Time zero spectrum is shown at the bottom with each subsequent spectrum taken at 1 hour intervals up to 10 hours. .................................. 113 Figure 3.1 Possible geometric isomers of a bis(amidate) bis(amido) titanium complex. The highlighted isomer is crystallographically observed and is calculated to have the lowest ground state energy. Relative free energies (ΔG) are reported in kcal/mol. .......................................... 116 Figure 3.2 Axially chiral bis(amidate) bis(amido) zirconium precatalyst for the enantioselective hydroamination of aminoalkenes. ............................................................................................... 117 Figure 3.3 Optimized geometry for C–H activation transition state TS(3/I). Most hydrogen atoms omitted for clarity. Colouration: C = gray, H = white, N = blue, O = red, Ta = purple .. 123 Figure 3.4 Optimized geometries for intermediates I (left) and II (right). Most hydrogen atoms omitted for clarity. ...................................................................................................................... 124 Figure 3.5 Optimized geometry for transition state TS(III/IV). Hydrogen atoms omitted for clarity. ......................................................................................................................................... 127  xiv  Figure 3.6 Optimized geometry for 5-membered metallacycle intermediate IV. Hydrogen atoms omitted for clarity. ...................................................................................................................... 128 Figure 3.7 Optimized geometry for transition state TS(V/VI). Most hydrogen atoms omitted for clarity. ......................................................................................................................................... 129 Figure 3.8 Optimized geometries for bis(amido) intermediates VI (left) and VII (right). Hydrogen atoms and select carbon atoms form the alkyl chain omitted for clarity. Simplified representations (above) are giving for reference. ....................................................................... 132 Figure 3.9 Optimized geometry for transition state TS(VII/VIII). Most hydrogen atoms omitted for clarity. .................................................................................................................................... 133 Figure 3.10 Potential energy surface (PES) for the α-alkylation of dimethylamine with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol. ....................................... 135 Figure 3.11 Optimized geometries for transition states TS(XI/XII) (left), TS(XIII/XIV) (middle), and TS(XV/XVI) (right) viewed along the axial plane. Most hydrogen atoms and select carbon atoms form the alkyl chain omitted for clarity...................................................... 137 Figure 3.12 Potential energy surface (PES) for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol. ....................................... 139 Figure 3.13 Optimized geometries for tantalaziridines X (left) and XVII (right). Hydrogen atoms omitted for clarity. ............................................................................................................ 140 Figure 3.14 Optimized geometries for intermediate XVIII (left) and transition state TS(XVIII/XIX) (right). Hydrogen atoms omitted for clarity. ................................................... 141 Figure 3.15 Optimized geometries for intermediate XX (left) and transition state TS(XX/XXI) (right). Most hydrogen atoms omitted for clarity. ...................................................................... 142  xv  Figure 3.16 Potential energy surface for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol. ....................................... 143 Figure 3.17 Comparison of the two pathways found for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol. ........................ 145 Figure 3.18 Optimized geometry for a 5-membered metallacycle intermediate with a κ1(O) amidate ligand orthogonal to the metallacycle. Hydrogen atoms omitted for clarity. ................ 148 Figure 3.19 Optimized geometry for a protonolysis transition state with a κ1(O) amidate ligand orthogonal to the N,O amidate plane. Most hydrogen atoms omitted for clarity. ...................... 149 Figure 3.20 Optimized geometry for an alkene insertion transition state with a κ1(O) amidate ligand orthogonal to the N,O amidate plane. Hydrogen atoms omitted for clarity. ................... 150 Figure 3.21 Potential energy surfaces showing the effects of axial amido substitution. Relative free energies (ΔG) are reported in kcal/mol. Colouration: none = black dash, bottom = red, top = blue, both = green. ...................................................................................................................... 151 Figure 3.22 Optimized geometry of tantalaziridine X in the triplet state. Select carbon atoms from the amidate backbone and all hydrogen atoms have been removed for clarity. ................ 153 Figure 3.23 Molecular orbital surfaces for the SOMOs (HOMO, top left; HOMO-1, top right) calculated for the triplet version of X. A view without molecular orbitals is given for reference (bottom)....................................................................................................................................... 154 Figure 3.24 Molecular orbital surfaces for the SOMOs (HOMO, top left; HOMO-1, top right) calculated for the triplet version of TS(XI/XII). A view without molecular orbitals is given for reference (bottom). ...................................................................................................................... 155 Figure 3.25 Sphere with radius of arbitrary size drawn from the centre of the tantalum atom in complex 3. The buried volume is the amount of sphere’s volume occupied by the ligand. ...... 158  xvi  Figure 3.26 Solid-G output for complex 3. Shadow for amidate ligand shown in blue. Shadows for dimethylamido ligands shown in green, yellow, purple, and red (not visible). .................... 159 Figure 3.27 Buried volumes (%Vbur) calculated for mono(amidate) tantalum precatalysts. ...... 161 Figure 3.28 Solid-G parameters GL and Gspace calculated for mono(amidate) tantalum precatalysts. ................................................................................................................................. 162 Figure 3.29 Axially chiral tantalum precatalysts for the enantioselective α-alkylation of amines. ..................................................................................................................................................... 163 Figure 3.30 Buried volumes (%Vbur) calculated for axially chiral tantalum precatalysts. ......... 165 Figure 3.31 Solid-G parameters GL and Gspace calculated for axially chiral tantalum precatalysts. ..................................................................................................................................................... 167 Figure 4.1 Diamido ligands with an aryl tether featuring a neutral oxygen donor. ................... 173 Figure 4.2 Diamido proligands with an aryl tether featuring a neutral sulfur donor. ................ 175 Figure 4.3 Diamine proligands featuring a neutral phosphorus donor. ..................................... 176 Figure 4.4 Synthesis of a side-on bound dinitrogen unit bridging a zirconium dimer supported by a diamido ligand with a neutral phosphorus donor.171 ................................................................ 177 Figure 4.5 Protonation of a side-on bridged dinitrogen moiety in a zirconium dimer supported by a diamido ligand with a neutral phosphorus donor.171 ................................................................ 178 Figure 4.6 ORTEP representations of the solid-state molecular structures for oxy- and thiobridged diamide proligands drawn at 50% probability for thermal ellipsoids. Only one part of a disordered tBu group for the oxy-bridged species is shown for clarity. Hydrogen atoms omitted for clarity. .................................................................................................................................... 182 Figure 4.7 ORTEP representation of the solid state molecular structure of 28 (left) and 29 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. ............. 183  xvii  Figure 4.8 ORTEP representation of the solid-state molecular structure of 30 (left) and 31 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. ............. 185 Figure 4.9 Variable temperature (25 – 90 °C) 400 MHz 1H NMR study of 30. ........................ 187 Figure 4.10 ORTEP representations of the solid-state molecular structures for oxy- and thiobridged diurea proligands drawn at 50% probability for thermal ellipsoids. Only one part of a disordered N(iPr)2 group for the thio-bridged species is shown for clarity. Hydrogen atoms omitted for clarity. ...................................................................................................................... 189 Figure 4.11 ORTEP representation of the solid-state molecular structures for 32•HNMe2 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. ........................ 190 Figure 4.12 ORTEP representation of the solid-state molecular structure of 32 (left) and 33 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. .. 191 Figure 4.13 ORTEP representation of the solid-state molecular structure of 34 (left) and 35 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity. .. 194 Figure 4.14 Variable temperature 400 MHz 1H NMR study of 34. Select spectra taken at -70, 25 and 80 °C shown.179 .................................................................................................................... 197 Figure 4.15 Cyclic voltammogram for 28 at a scan rate of 100 mV/s (0.005 mol L-1 in THF, 0.1 mol L-1 TBAPF6). Electrodes as follows: working, Pt button; counter, glassy carbon; reference, Ag wire........................................................................................................................................ 201 Figure 4.16 Cyclic voltammograms for 30 and 31 at a scan rate of 50 mV/s (0.005 mol L-1 in THF, 0.1 mol L-1 TBAPF6). Electrodes as follows: working, Pt button; counter, glassy carbon; reference, Ag wire. ...................................................................................................................... 202 Figure 4.17 Cyclic voltammograms for 34 (0.005 mol L-1 in THF) using 0.1 mol L-1 TBAPF6 (left, scan rate = 100 mV/s) and 0.1 mol L-1 TBABPh4 (right, scan rate = 500 mV/s) as the  xviii  electrolyte. Electrodes as follows: working, Pt button; counter, glassy carbon; reference, Ag wire. ............................................................................................................................................ 203 Figure 4.18 Variable temperature 400 MHz 1H NMR study of 34. ........................................... 220 Figure 5.1 15N spectrum (40 MHz, 25 °C) of 15N-7................................................................... 231 Figure C.1 Optimized geometry for an alternate transition state for the formation of a tantalaziridine. Most hydrogen atoms omitted for clarity. .......................................................... 272 Figure C.2 Optimized geometries for the intermediates depicted in Scheme C.1. Most hydrogen atoms omitted for clarity. ............................................................................................................ 274 Figure C.3 Optimized geometry for a tantalaziridine with the azametallacyclopropane moiety perpendicular to the plane of the amidate ligand. Hydrogen atoms omitted for clarity. ............ 275 Figure C.4 Optimized geometry for a transition state with the insertion of 1-octene into the azametallacyclopropane moiety perpendicular to the plane of the amidate ligand. Hydrogen atoms omitted for clarity. ............................................................................................................ 276 Figure C.5 Buried volumes (%Vbur) calculated for axially chiral tantalum precatalysts. .......... 276 Figure C.6 Solid-G parameters GL and Gspace calculated for axially chiral tantalum precatalysts. ..................................................................................................................................................... 277 Figure D.7 Cyclic voltammograms for N,N’-(2,2’-oxybis(2,1-phenylene))bis(tert-butylamide). ..................................................................................................................................................... 278 Figure D.8 Cyclic voltammograms for 28. ................................................................................ 279 Figure D.9 Cyclic voltammograms for 30. ................................................................................ 279 Figure D.10 Cyclic voltammograms for 31. .............................................................................. 280 Figure D.11 Molecular orbital surfaces for the LUMOs calculated for 30 (left) and 31 (right).280  xix  Figure D.12 Cyclic voltammograms for 1,1’-(oxybis(2,1-phenylene))bis(3,3-diisopropylurea). Scans start from -2.8 V towards positive potentials. .................................................................. 281 Figure D.13 Cyclic voltammogram for 34. ................................................................................ 282 Figure D.14 Cyclic voltammogram for 34 using TBABPh4 as the electrolyte. ......................... 282  xx  List of Schemes  Scheme 1.1 The intra- and intermolecular direct α-alkylation of amines. ..................................... 2 Scheme 1.2 Deuterium scrambling of N-deuterated dimethylamine via a 3-membered metallacycle. ................................................................................................................................... 3 Scheme 1.3 Proposed mechanism for the α-alkylation of amines catalyzed by early transition metals. ............................................................................................................................................. 4 Scheme 1.4 Insertion of a carbon-element bond into a zirconaziridine Zr–C bond. ...................... 5 Scheme 1.5 Insertion of terminal alkenes into a zirconaziridine Zr–C bond to yield chiral amines.11.......................................................................................................................................... 6 Scheme 1.6 Insertion of terminal alkynes into a zirconaziridine Zr–C bond.11 ............................. 7 Scheme 1.7 α-Alkylation of tetrahydroquinoline via insertion of alkynes into a zirconaziridine Zr–C bond.13 ................................................................................................................................... 7 Scheme 1.8 One-pot enantioselective synthesis of allylic amines via insertion of alkynes into an in situ generated zirconaziridine Zr–C bond.17 ............................................................................... 8 Scheme 1.9 Synthesis of N-heterocycles from amine substrates derived from insertion of terminal alkenes with pendant electrophiles into a zirconaziridine Zr–C bond.19 .......................... 9 Scheme 1.10 Synthesis of azetidines from amine substrates with an alkyl iodide functionality derived from insertion of terminal alkenes into a zirconaziridine Zr–C bond.18 ............................ 9 Scheme 1.11 Cyclization of unsaturated hydrazones via insertion into a zirconaziridine Zr–C bond.22 ........................................................................................................................................... 10 Scheme 1.12 Synthesis of substituted pyrroles via CO insertion into the Zr–C bond of a 5membered metallacycle.23 ............................................................................................................. 11  xxi  Scheme 1.13 Organic products synthesized via insertion of carbon-heteroatom double bonds into zirconaziridines. ............................................................................................................................ 12 Scheme 1.14 Intramolecular α-alkylation of amines using bis(pyridonate) zirconium precatalyst. ....................................................................................................................................................... 14 Scheme 1.15 Proposed mechanism for the intramolecular α-alkylation of primary aminoalkenes using a bis(pyridonate) zirconium precatalyst. ............................................................................. 15 Scheme 1.16 Formation of a bimetallic Ti species with a bridging metallaziridine. ................... 16 Scheme 1.17 Screening of titanium catalysts for intramolecular α-alkylation of amines. ........... 17 Scheme 1.18 The α-alkylation of N-aryl amines by terminal olefins, catalyzed by a titanium catalyst. ......................................................................................................................................... 18 Scheme 1.19 α-Alkylation of N-methylaniline with styrene catalyzed by a dinuclear sulfamide titanium complexes. Products formed in approximately a 5:4:1 ratio. ........................................ 19 Scheme 1.20 Tantalum α-alkylation of amine catalysts supported by amido and chloride ligands. ....................................................................................................................................................... 21 Scheme 1.21 Enantioselective niobium and tantalum α-alkylation of amines catalysts supported by silyl-substituted bis(naphtholate) ligands. ............................................................................... 23 Scheme 1.22 Direct α-alkylation of amines with late transition metal catalysts. ......................... 24 Scheme 1.23 Generalized scheme for direct α-alkylation of amines using late metal catalysts. . 25 Scheme 2.1 Generalized synthesis of amide proligands. ............................................................. 29 Scheme 2.2 Synthesis of bis(amidate) imido complexes of Ti29 and Zr.57................................... 31 Scheme 2.3 Insertion of isocyanate into the Zr–C bonds of a bis(amidate) dibenzyl zirconium complex.62 ..................................................................................................................................... 32  xxii  Scheme 2.4 Common methods for synthesizing tantalaziridines. Insertion into two Ta–R bonds (top); reaction of imine with Ta(III) species (middle); β-hydrogen abstraction from amido ligand (bottom)......................................................................................................................................... 34 Scheme 2.5 Reaction of an in situ generated Ta(III) species with pyridine. The heterocycle experiences a loss of aromaticity when the tantalaziridine is installed.71 ..................................... 35 Scheme 2.6 Reduction of a tantalum diiodo tris(amido) species to result in β-hydride elimination from the neopentyl ligand to give a tantalum hydride as well as a metallaziridine (top). Equilibrium between 3- and 5-membered metallacycles (bottom).74 ........................................... 36 Scheme 2.7 Spontaneous elimination of diethylamine to give a tris(amido)tantalaziridine complex.75-76.................................................................................................................................. 37 Scheme 2.8 Decomposition of a mixed amido-, alkyl-tantalum half sandwich complex to give a tantalaziridine and eliminate methane.78 ....................................................................................... 38 Scheme 2.9 Two possible pathways for tantalaziridine formation from N-donor arm of a tantalum N-heterocyclic carbene complex. Calculations and labeling experiments show alkylidene formation followed by C–H addition is the preferred pathway.80 ............................... 39 Scheme 2.10 Proton transfer to a carbene ligand to generate a tantalaziridine in a hydrotris(3,5dimethylpyrazoyl)borate complex.81............................................................................................. 40 Scheme 2.11 Spontaneous elimination of dicyclohexylamine to give a tris(amido)tantalaziridine complex.77 ..................................................................................................................................... 41 Scheme 2.12 Insertion of 2,6-dimethylphenylisonitrile into Ta–C bond of tantalaziridine to form an azacyclotantalabutanes (top).82 Decomposition of a related chloro species to give a tantalum imido complex and an unsaturated organic molecule (bottom).83 ................................................ 42  xxiii  Scheme 2.13 Insertion of nitriles into the tantalaziridine Ta–C bond to give a 5-membered metallacycle with two amido contacts.88 ...................................................................................... 43 Scheme 2.14 Insertion of ethylene (top) or alkynes (bottom) into a tantalaziridine to form a 5membered metallacycle.84 ............................................................................................................. 44 Scheme 2.15 Reaction of a tantalum hydride with CO2 to give a bimetallic methylene diolate complex. The tantalaziridine moiety does not participate in the reaction.74................................. 45 Scheme 2.16 Organic products isolated after aqueous workup of the insertion of cyclohexanone, benzaldehyde and cyclohexylisocyanate into a tantalaziridine chloride complex.87 .................... 46 Scheme 2.17 Protonolysis synthesis of mono(amidate) tantalum complexes 1 – 4. .................... 49 Scheme 2.18 Protonolysis synthesis of mono(amidate) tantalum complex 5. ............................. 50 Scheme 2.19 Thermolysis of precatalyst 3. .................................................................................. 58 Scheme 2.20 Protonolysis synthesis of bis(amidate) complex 7. ................................................. 60 Scheme 2.21 Proposed equilibrium between coordination modes of 7 illustrating the dynamic exchange of amidate binding modes observed on the NMR time scale. Labeling of amidate N and O for illustrative purposes only. ............................................................................................. 63 Scheme 2.22 Protonolysis synthesis of bis(amidate) product 7 and tris(amidate) product 8. ...... 64 Scheme 2.23 Protonolysis synthesis of bis(amidate) complex 9. ................................................. 67 Scheme 2.24 Insertion of acetonitrile into Ta–C bond of 9 to generate 10. ................................. 71 Scheme 2.25 Protonolysis synthesis of bis(amidate) complex 11. ............................................... 73 Scheme 2.26 Byproducts 12 and 13 formed from the α-alkylation of 1-octene and dimethylamine liberated from the catalyst. ............................................................................................................ 76 Scheme 2.27 Reaction of 3 with 1-ocetene to form both branched (12) and linear (14) monoalkylated α-alkylation of amines byproducts. ............................................................................... 77  xxiv  Scheme 2.28 Proposed mechanism for tantalum catalyzed α-alkylation of amines showing reactions off the catalytic pathway that have been previously studied using deuterium labeling. ... ....................................................................................................................................................... 78 Scheme 2.29 Incorporation of H into the labeled methyl group of N-methylaniline-d3. ............. 79 Scheme 2.30 Radically induced ring opening of cyclopropyl ring. ............................................. 90 Scheme 2.31 Synthesis of cyclopropyl-based amide proligands.................................................. 90 Scheme 2.32 Formation of 8-membered metallacycles from the stepwise radical insertion of a cyclopropyl-based alkene into a tantalaziridine. ........................................................................... 93 Scheme 2.33 Synthesis of 17 as a substrate for the α-alkylation of amines. ................................ 93 Scheme 2.34 α-Alkylation of N-methylaniline with cyclopropyl-based alkene substrate 17. ..... 94 Scheme 3.1 Calculated intermediates for the catalytic cycle of the intramolecular α-alkylation of aminoalkenes using Ti(NMe2)4 as the precatalyst.103 ................................................................. 118 Scheme 3.2 Proposed catalytic cycle for the α-alkylation of N-methylaniline with 1-octene catalyzed by complex 3. Generalized transition states as well as intermediates are depicted for the first turnover. ......................................................................................................................... 120 Scheme 3.3 Intermediates and transition state for the C–H activation of a dimethylamido ligand of complex 3 to form tantalaziridine I. Relative free energies (ΔG) are reported in kcal/mol. .. 123 Scheme 3.4 Intermediates and transition state for the insertion of 1-octene into the Ta–C bond of tantalaziridine II. Relative free energies (ΔG) are reported in kcal/mol. ................................... 126 Scheme 3.5 Intermediates and transition state for the protonolysis of the Ta–C bond of metallacycle IV by N-methylaniline. Relative free energies (ΔG) are reported in kcal/mol...... 129  xxv  Scheme 3.6 Intermediates and transition state for the β-hydrogen abstraction of a methyl C–H bond of bis(amido) VII to release amine 12. Relative free energies (ΔG) are reported in kcal/mol. ..................................................................................................................................................... 131 Scheme 4.1 Two possible isomers for group 4 metal complexes supported by diamido ligands with an aryl tether featuring a neutral oxygen donor. ................................................................. 174 Scheme 4.2 Synthesis of a tantalaziridine supported by a diamido ligand with a neutral sulfur donor.166 ...................................................................................................................................... 176 Scheme 4.3 Synthesis of Zr and Ta complexes supported by a tethered bis(amidate) or bis(ureate) ligand with a neutral chalcogen donor. ..................................................................... 179 Scheme 4.4 Synthesis of diamide proligands. ............................................................................ 181 Scheme 4.5 Protonolysis synthesis of bis(amidate) zirconium complexes 28 and 29. .............. 183 Scheme 4.6 Protonolysis synthesis of bis(amidate) zirconium complexes 30 and 31. .............. 184 Scheme 4.7 Synthesis of diurea proligands. ............................................................................... 188 Scheme 4.8 Protonolysis synthesis of bis(ureate) zirconium complex 32•HNMe2. .................. 190 Scheme 4.9 Protonolysis synthesis of bis(ureate) zirconium complexes 32 and 33. ................. 191 Scheme 4.10 Protonolysis synthesis of bis(ureate) tantalum complexes 34 and 35. ................. 194 Scheme 4.11 Reduction of tantalum amidate complexes to yield η2-iminoacyl ligated tantalum oxo species. ................................................................................................................................. 199 Scheme 5.1 Proposed synthetic route for 15N-enriched amide proligands. ................................ 229 Scheme 5.2 Synthesis of isotopically enriched 15N-phenylpivalamide. ..................................... 230 Scheme 5.3 Proposed synthesis of 2,6-disubstituted  15  N-enriched amide proligands.  Mesitylbromide used as an example starting material. ............................................................... 232  xxvi  Scheme 5.4 Proposed alternative protonolysis transition state for the α-alkylation of amines involving the participation of an additional equivalent of amine substrate. ............................... 233 Scheme 5.5 Proposed alternative protonolysis pathway for the α-alkylation of amines involving a κ1(O) amidate ligand as a proton shuttle. .................................................................................... 235 Scheme C.1 Intermediates and transitions state for the C–H activation of an alternative dimethylamido ligand. Free energies (ΔG) are reported in kcal/mol and are calculated relative to the κ1(O) species depicted (top numbers) and to the corresponding κ2(N,O) species (bottom numbers). .................................................................................................................................... 273  xxvii  List of Abbreviations and Acronyms Å  angstrom (10-10 m)  A  ampere  anal.  analysis  Ar  aryl  aq  aqueous  B3LYP  Becke 3-parameter Lee-Yang-Parr functional  BINAP  2,2’-bis(diphenylphosphino)-1,1’-binaphthyl  Si-binaphth  (R)-3,3'-bis[methyl(diphenyl)silyl]-2,2’-dihydroxy-1,1’-binaphthylate  Bn  benzyl  br.  broad  Bu  butyl  c  concentration  C  Celsius  cal  calorie  calcd.  calculated  cat.  catalyst  CMD  concerted metalation-deprotonation  conv.  conversion  Cp  cyclopentadienyl  Cp*  pentamethylcyclopentadienyl  CV  cyclic voltammetry  Cy  cyclohexyl  xxviii  d  doublet  d  deuterium  δ  chemical shift  °  degrees  DCM  dichloromethane  DFT  density functional theory  DIB  (diacetoxy)iodobenzene  DMEDA  N,N’-dimethylethane-1,2-diamine  ΔG  Gibbs free energy  ΔG‡  Gibbs free energy of activation  Δν  maximum peak separation (NMR spectroscopy)  DME  1,2-dimethoxyethane  DMSO  dimethylsulfoxide  η  hapticity  E  chalcogen  EA  elemental analysis  ee  enantiomeric excess  EI  electron impact  ESI  electrospray ionization  Et  ethyl  fac  facial  Fc  ferrocene  Fw  formula weight  xxix  G  solid angle  GC  gas chromatography  gem  geminal  h  hours  HOMO  highest occupied molecular orbitals  Hz  Hertz  IEFPCM  integral equation formalization polarizable continuum model  Ind  indenyl  IR  infra red  IRC  intrinsic reaction coordinate  i  Pr  isopropyl  J  coupling constant  κ  denticity  keq  equilibrium constant  KIE  kinetic isotope effect  L  supporting ligand  LUMO  lowest unoccupied molecular orbital  m  multiplet  M  metal  M+  molecular ion  μ  bridging ligand  μ  absorption coefficient (X-ray crystallography)  mer  meridional  xxx  mol  mole  mol%  mole percent  Me  methyl  Mes  mesityl  MS  mass spectrometry  MP2  second order Møller-Plesset perturbation theory  m/z  mass-to-charge ratio  NBO  natural bond orbital  NMR  nuclear magnetic resonance  no.  number  Np  neopentyl  ORTEP  Oak Ridge thermal ellipsoid plot  PES  potential energy surface  Ph  phenyl  ppm  parts per million  p-Tol  para-tolyl  Pr  propyl  psi  pounds per square inch  py  pyridine  q  quartet  R  organic substituent  R2  coefficient of determination (statistics)  ρ  density  xxxi  reflns  reflections  s  singlet  SOMO  singly occupied molecular orbital  Σ  summation  t  triplet  t  Bu  tert-butyl  TBABPh4  tetrabutylammonium tetraphenylborate  TBAPF6  tetrabutylammonium hexafluorophosphate  Tc  coalescence temperature  TEMPO  (2,2,6,6-tetramethylpiperidin-1-yl)oxyl  THF  tetrahydrofuran  TMS  tetramethylsilane  TOF  turnover frequency  %Vbur  percent buried volume  V  Volt  VT  variable temperature  xxxii  Acknowledgements  I would like to sincerely thank all of the people and organizations that have contributed to this body of work. Firstly, I am indebted to Dr. Michael Wolf and Dr. Pierre Kennepohl for their critique of this thesis prior to submission. Your comments and suggestions greatly improved the quality of this work. I would also like to thank the following agencies for funding over my postgraduate tenure: the University of British Columbia, the Natural Sciences and Engineering Research Council of Canada, the Killam Trusts and Boehringer Ingelheim Ltd. It is with the greatest pleasure that I thank all of the labmates I have had over the years in both the Schafer and Fryzuk research groups. You bring the workplace to life and it has been an honor working with you all. Special mention goes to Pippa Payne, Andreas Wagner, Jacky Yim and Scott Ryken, the most talented, and thorough, editing team I have ever worked with. This document would not be in existence today without the help of these terrific coworkers and friends. Also I would like to thank Dr. Patrick Eisenberger for being an excellent mentor and role model both academically and personally. I have also had the pleasure of sharing my entire academic career with Nathan Halcovitch, a selfless collaborator and timeless friend. Thank you Ken Love, for without you there would be no glove boxes or DIY lessons. Finally, and most importantly, I need to thank my supervisors Dr. Laurel Schafer and Dr. Michael Fryzuk for allowing me the opportunity to learn with them.  Your mentorship is  invaluable and I have made great strides not only as a chemist but as a person under your guidance. Thank you both for your tireless efforts.  xxxiii  Dedication  This thesis is dedicated to my brother, Ryan Lauzon, for without him I would never push myself to, and beyond, my limits.  This thesis is dedicated to my mother, Ellenjane Mooney-Lauzon, for without her I would have never known unconditional love and unwavering support.  This thesis is dedicated to my father, Jean Paul Lauzon, for without him I would have not learned the merits and satisfaction of independent thought.  xxxiv  Chapter 1: The Direct α-Alkylation of Amines  1.1  Historic Overview The development of atom-economic methods to form carbon-element bonds is a  fundamental challenge faced by synthetic chemists.1  Due to increasing focus on the  environmental impact of industrial processes, both chemical producers and government regulators are interested in maximizing resource efficiency.2 This is especially true for the pharmaceutical and agrochemical industries where most of the innovation has been focused on decreasing the amount of waste released to the environment through remediation rather than reducing the amount of total waste produced. With the refining of technologies used to minimize emitted waste, the cost to eliminate the remaining pollutants is becomingly increasingly expensive.2  As a result, reducing inefficiencies in the production process by using atom-  economic strategies is becoming a viable alternative to established chemical practices. To this end, hydrofunctionalization reactions are particularly attractive.3  Formal  additions of an element-hydrogen bond across an unsaturated carbon-carbon bond are a general set of reactions that can be used to increase chemical complexity using hydrocarbon feedstocks without generating wasteful byproducts. One such reaction, the direct α-alkylation of amines or hydroaminoalkylation (Scheme 1.1), exploits such transformations to catalytically functionalize amines using alkene substrates.  1  Scheme 1.1 The intra- and intermolecular direct α-alkylation of amines.  The utility of small, nitrogen-containing molecules in a variety of both academic and industrial applications has led to intense research into the efficient synthesis of amines.4 The αalkylation of amines is attractive as it avoids the use of protecting groups and stoichiometric coreagents such as bases or oxidants. Coupled with the use of transition metal catalysts and unactivated hydrocarbon feedstocks, this synthetic tool is approaching the ideal of a so-called “green” process.5 However, in order to be considered a competitive option for amine synthesis, the direct α-alkylation of amines must be able to match current methods in terms of selectivity and substrate scope.6 An initial report from Maspero and Clerici in 1980 disclosed the catalytic alkylation of secondary dialkyl amines with terminal alkenes using homoleptic dimethylamido precatalysts.7 The exclusive formation of the branched product (up to 38%) was achieved over a 20 hour period at 160 °C. Zr, Nb, and Ta based precatalysts were found to be the most effective while Ti, V, Mo, and Sn complexes afforded only traces of product. Three years later, Nugent and coworkers demonstrated that reversible metalation of dimethylamido ligands occurs for homoleptic Zr, Nb, Ta, and W species at elevated temperatures (140 – 180 °C).8 This was accomplished by monitoring the disappearance of the N–D infrared (IR) stretch in N2  deuterodimethylamine and subsequent incorporation of deuterium into one of the methyl groups (Scheme 1.2).  Scheme 1.2 Deuterium scrambling of N-deuterated dimethylamine via a 3-membered metallacycle.  This monodeuteration is thought to occur via deuteration of the M–C bond of a threemembered metallacycle intermediate. It was found that the azametallacyclopropane moiety could also react with terminal alkenes, resulting in the α-alkylation observed by Maspero. Based on these findings, a proposed mechanism was put forth (Scheme 1.3).8  3  Scheme 1.3 Proposed mechanism for the α-alkylation of amines catalyzed by early transition metals.  Initially, a reversible C–H activation of an amido ligand generates an equivalent of amine and forms a three-membered metallacycle (A). An equivalent of alkene substrate can then insert into the reactive metal-carbon bond generating a more stable five-membered metallacycle (B). Both Maspero and Nugent observed the branched product exclusively, suggesting that the insertion regioselectivity arises from minimization of alkene steric bulk at the metal centre. Then the M–C bond is protonated by an equivalent of incoming amine. The resultant bis(amido) species then undergoes a β-hydrogen abstraction to release the product and regenerate the active metallacycle (A).  4  Since the initial reports of the α-alkylation of amines, a number of stoichiometric and catalytic systems based on transition metals have been developed. For instance, group 4 and 5 complexes containing a three-membered metallacycle, a metallaziridine, have been shown to react stoichiometrically with a variety of unsaturated molecules, effectively functionalizing amines α to nitrogen.9-10 The most successful systems are based on the insertion of electrophiles into the Zr–C bond of zirconaziridines which, upon aqueous work-up, yield amines of increased chemical complexity.  1.2  Stoichiometric Zirconium Reagents The application of zirconaziridines in organic synthesis is largely based on the insertion  of C–E unsaturations (E = C, N, O) into the reactive zirconium-carbon bond to generate expanded 5-membered metallacycles (Scheme 1.4).10 These organometallic compounds can be quenched with protic solvent to release the organic fragment from the Zr centre. Using this methodology, a variety of products can be synthesized including chiral amines,11 allylic amines,11-17 heterocycles,11, 18-19 diamines,11 amino alcohols,11 amino amides,20 amino amidines,21 and amino acid esters.20  Scheme 1.4 Insertion of a carbon-element bond into a zirconaziridine Zr–C bond.  5  The carbon-carbon unsaturation found in alkenes and alkynes is the most common functional group inserted into the reactive Zr–C bond and can occur in an inter- or intramolecular fashion. In line with the observations of Maspero and Nugent, Buchwald and coworkers found the insertion of terminal alkenes into zirconaziridines to be highly regioselective yielding only metallacycles substituted at position 4 (Scheme 1.5).11 Gratifyingly, these reactions were also found to be diastereoselective, with only one diastereomer observed in the chiral amine obtained upon aqueous workup. When norbornene is used, the only internal alkene substrate found in the literature for this reaction, both diastereomers are produced in a 2:1 ratio.18  Scheme 1.5 Insertion of terminal alkenes into a zirconaziridine Zr–C bond to yield chiral amines.11  Alkynes are also suitable substrates for intermolecular insertion into zirconaziridines to form 5-membered metallacycles.  Though no exclusive regioselectivity is observed upon  insertion of terminal alkynes, the 3,5-isomer, and subsequently the linear product, is favoured in all cases (Scheme 1.6).11 This is in direct contrast to the exclusive 3,4-regioisomer observed with terminal alkene insertion.  6  Scheme 1.6 Insertion of terminal alkynes into a zirconaziridine Zr–C bond.11  For asymmetric internal alkynes, the less bulky of the two groups is observed in the 5 position, presumably due to the steric demands of the metal centre. A one-pot procedure for the α-alkylation of amines with alkynes to generate allylic amines was developed based on this insertion chemistry.11 Interestingly, Whitby and coworkers used this technology to functionalize tetrahydroquinoline with a variety of alkenes, alkynes and allenes (Scheme 1.7, 2,5-dimethylhex3-yne is shown as an example substrate).12-13  Scheme 1.7 α-Alkylation of tetrahydroquinoline via insertion of alkynes into a zirconaziridine Zr–C bond.13  7  By  substituting  the  cyclopentadienyl  ligands  for  the  related  tethered  ethylenebis(tetrahydroindenyl) ligand (Scheme 1.8), enantioselective insertions into the zirconium-carbon bond are possible.17 Insertion of a C–C triple bond into the in situ generated zirconaziridine yields an enantioenriched product upon quenching of the reaction with methanol. Though previously discussed regioselectivity issues are still present, the allylic amine products are formed in good yields with excellent enantioselectivities (90 – 99% ee).  Scheme 1.8 One-pot enantioselective synthesis of allylic amines via insertion of alkynes into an in situ generated zirconaziridine Zr–C bond.17  The insertion of C–C unsaturations into zirconaziridines not only offers the ability to functionalize existing heterocycles but to generate amines that undergo cyclization to give Ncontaining heterocycles of varying ring sizes. This is achieved by inserting terminal alkenes with pendant electrophiles, such as halides or oxiranes, which do not react with the zirconaziridine.19  The resultant α-functionalized amines cyclize after work-up, generating  diastereomerically pure pyrrolidines, piperidines, and azepanes (Scheme 1.9).  8  Scheme 1.9 Synthesis of N-heterocycles from amine substrates derived from insertion of terminal alkenes with pendant electrophiles into a zirconaziridine Zr–C bond.19  Interestingly, if the electrophile is installed by cleaving the Zr–C and Zr–N bonds with iodine instead of protons, intramolecular cyclization of the amine results in the formation of 4membered azetidines (Scheme 1.10).18 As with the aforementioned 5-, 6-, and 7-membered heterocycle precursors, the regio- and diastereoselective insertion of the terminal alkene into the zirconaziridine results in diastereomerically pure azetidine products.  Scheme 1.10 Synthesis of azetidines from amine substrates with an alkyl iodide functionality derived from insertion of terminal alkenes into a zirconaziridine Zr–C bond.18  9  Using  hydrazone  substrates,  Livinghouse  and  coworkers  demonstrated  that  intramolecular insertion of alkenes and alkynes into zirconaziridines leads to the formation of disubstituted hydrocarbon rings (Scheme 1.11).22 This technique is compatible with a range of alkyl-, aryl-, and silyl-substituted C–C unsaturations and the short tether length eliminates the regioselectivity issue previously observed for alkyne insertion.11  Scheme 1.11 Cyclization of unsaturated hydrazones via insertion into a zirconaziridine Zr–C bond.22  The reactivity of the 5-membered metallacycle generated after alkyne insertion into a zirconaziridine has been exploited by Buchwald and coworkers to synthesize substituted pyrroles.23 Inserting carbon monoxide into the reactive Zr–C bond yields a 6-membered ring that rearranges to form a bicyclic zirconoxirane, presumably due to the oxophilic nature of zirconium (Scheme 1.12). Aromatization of the pyrrole ring occurs upon protonolysis of the zirconium species, leaving a trisubstituted 5-membered heterocycle. Though tolerant of a wide range of functional groups, high CO pressures (up to 1500 psi (pounds per square inch)) are required.  10  Scheme 1.12 Synthesis of substituted pyrroles via CO insertion into the Zr–C bond of a 5membered metallacycle.23  The concept of using the Zr–C bond of the 5-membered metallacycle for further reactivity was extended by Mori and coworkers. By exchanging the Zr–C bond for a Cu–C bond via transmetallation, allylated allylic amines24-25 as well as substituted pyrroles26 could be formed in moderate to good yields. In addition to C–C unsaturations, carbon-heteroatom double bonds have been shown to insert into zirconaziridines (Scheme 1.13).10 In all cases, the heteroatom is bound to zirconium in the resultant 5-membered metallacycle.  11  Scheme 1.13 Organic products synthesized via insertion of carbon-heteroatom double bonds into zirconaziridines.  The insertion of imines (C=N) and aldehydes (C=O) ultimately results in diamine or amino alcohol formation upon quenching with protic solvents while treatment with ethyl chloroformate results in imidazolidinone or oxazolidinone products.23  Though exclusive  insertion at the Zr–C bond is commonly observed, urea products isolated after isocyanate insertion into zirconaziridines suggest a competing pathway with the Zr–N bond (Scheme 1.13).20 However, using the bulky ethylenebis(tetrahydroindenyl) ligand results in the exclusive 12  formation of the amino amide product. Other heterocumulenes such as carbodiimides are also competent substrates for insertion into zirconaziridines, yielding α-amino amidines after aqueous workup.21 Unfortunately, the insertion of CO2 into the reactive organometallic bonds resulted in either no reaction or a stable zirconium complex resistant to protic media.20 In contrast, the CO2 synthon ethylene carbonate inserts into the Zr–C bond to form a spirocyclic zirconium complex which, upon treatment with methanol in tetrahydrofuran (THF), generates an amino acid ester. Interestingly, quenching in benzene results in an amino acid methyl ester via transesterification (Scheme 1.13). A variety of functionalized amine products can be generated by inserting carbon-element unsaturations into zirconaziridines and then quenching the resultant 5-membered metallacycles with protic solvents. Despite the wealth of reactivity presented, none of the aforementioned systems have displayed catalytic activity. Though these transformations are synthetically useful, the use of zirconium in stoichiometric quantities has practical limitations. To this end, a number of group 4 catalysts for the α-alkylation of amines have been reported for both intra- and intermolecular reactivity.  1.3  A Zirconium Precatalyst for the Intramolecular α-Alkylation of Amines In line with early observations from Maspero and Nugent, Zr-based metal complexes can  acts as catalysts for the direct α-alkylation of amines. Recently, our group reported the use of a bis(pyridonate) zirconium complex in the intramolecular α-alkylation of aminoalkenes (Scheme 1.14).27  13  Scheme 1.14 Intramolecular α-alkylation of amines using bis(pyridonate) zirconium precatalyst.  The reported bis(pyridonate) system affords exclusively exo cyclized products in good yields and does not rely on gem-disubstituent effects for catalysis to proceed. However, this system is not compatible with secondary amines and only modest diastereoselectivities are achieved.  In addition, chemoselectivity is a challenge as hydroamination is a competitive  pathway at lower catalyst loadings. The mechanistic proposal for this reaction involves the formation of bridging imido dimers (Scheme 1.15), supported by the observations that secondary amines cannot be used as substrates and an increase in catalyst loading significantly increases the chemoselectivity for αalkylation over hydroamination.  Initially the dimethylamido ligands are protonated by an  equivalent of amine substrate to generate a zirconium-imido complex. This imido complex is in equilibrium with a zirconium dimer species containing a bridging imido unit, believed to be the active catalyst. In contrast, group 4 dimeric complexes are thought to be off-cycle, inactive species in hydroamination catalysis.27 C–H activation at the methylene group α to the nitrogen yields a metallaziridine bearing a reactive zirconium-carbon bond, analogous to Nugent’s proposed mechanism.  Insertion of the alkene into the Zr–C bond gives a five-membered  metallacycle which is protonated by an incoming amine substrate to release the cyclized product and regenerate the bridging imido Zr dimer.  14  Scheme 1.15 Proposed mechanism for the intramolecular α-alkylation of primary aminoalkenes using a bis(pyridonate) zirconium precatalyst.  The formation of the bimetallic bridging metallaziridine intermediate is supported by the isolation of a related Ti system. Reacting a similar bis(amidate) bis(dimethylamido) titanium species with 1.5 equivalents of benzylamine yields a dinuclear complex bridged by one amidate ligand, one imido ligand and a metallaziridine (Scheme 1.16). A bond length of 1.368(4) Å suggests significant double bond character in the N–C fragment of the metallaziridine, consistent with the assignment of an η2-imine moiety.10 Interestingly, the starting amidate complex is a  15  broadly applicable hydroamination catalyst but does not catalyze the direct α-alkylation of amines.28-32  Scheme 1.16 Formation of a bimetallic Ti species with a bridging metallaziridine.  Moving from the smaller Ti to Zr allows for a larger coordination sphere and therefore a more readily accessible metal centre. Additionally, the bulk imposed by pyridonate ligands is further removed from the metal centre compared to amidates which facilitates the formation of the bridging bimetallic species required for the catalytic α-alkylation of amines.  Indeed,  bis(amidate) zirconium complexes have shown exclusive formation of the hydroamination product.  1.4  Titanium Precatalysts for the α-Alkylation of Amines Notably, a catalyst screen of other group 4 dimethylamido complexes showed the  homoleptic Ti(NMe2)4 favours α-alkylation of amines over the hydroamination of primary aminoalkenes (18:1).27  Initially observed as a byproduct during catalyst screening for  hydroamination,33 Doye and coworkers screened a series of Ti catalysts (Figure 1.1) with selected substrates that would innately favour α-alkylation over hydroamination. The primary  16  aminoalkene hept-6-en-1-amine was selected due to the more challenging 7-membered ring product formed via hydroamination.  Figure 1.1 Titanium precatalysts for the direct α-alkylation of amines.  Disappointingly, even over extended reaction times (72 hours) at elevated temperatures (160 °C), Ti(NMe2)4 gives the intramolecular α-alkylation product in less than 50% yield (Scheme 1.17).34 When tetrabenzyltitanium is used as the precatalyst, a marked improvement in yield is observed.35 This trend is consistent for other primary aminoalkene substrates as well. Unfortunately, the bis(indenyl)titanium alkyl complex showed no reactivity with this class of intramolecular substrates.  Scheme 1.17 Screening of titanium catalysts for intramolecular α-alkylation of amines.  17  As titanium systems usually do not catalyze the hydroamination of secondary aminoalkene substrates, N-arylated aminoalkenes were also selected for screening.  These  substrates show less reactivity for the α-alkylation of amines than the primary aminoalkene analogues, with a maximum yield of 26% being achieved by the tetrakis(dimethylamido) ligated species.35  Interestingly, double bond migration, attributed to a competing C–H activation  pathway, is observed with the bis(indenyl) species.36 Inspired by the work of Hartwig and Herzon (vide infra), N-arylated amines were found to react with terminal alkenes in the presence of titanium complexes to give the corresponding αalkylation product in moderate to good yields (Scheme 1.18). This represents the first report of group 4 systems catalyzing the intermolecular α-alkylation of amines.34  Scheme 1.18 The α-alkylation of N-aryl amines by terminal olefins, catalyzed by a titanium catalyst.  As with the intramolecular transformation, Ti(NMe2)4 is found to be the least active catalyst, with the tetrabenzyl and bis(indenyl) species displaying higher reactivity.35-36 This increase in activity leads to a reduction of the reaction temperature, to as low as 80 °C, paving the way for compatibility with substrates that polymerize at higher temperatures, such as styrene. Indeed, at 105 °C, the bis(indenyl) titanium precatalyst catalyzes the α-alkylation of a range of substituted N-methylanilines with substituted styrenes in good to excellent yields.  18  However, these intermolecular Ti systems are plagued by regioselectivity issues. While the intramolecular systems afforded exclusively the exo product, the intermolecular version, especially with styrene substrates, can have branched to linear product ratios as low as 3:1.36 This selectivity issue is emphasized with a dinuclear sulfamidate titanium precatalyst, with the observation of an additional bis-alkylated product (Scheme 1.19).37  Scheme 1.19 α-Alkylation of N-methylaniline with styrene catalyzed by a dinuclear sulfamide titanium complexes. Products formed in approximately a 5:4:1 ratio.  In addition to regioselectivity issues, these Ti systems are restricted to N-arylated amine substrates. With regard to alkene substrates, only terminal alkenes are tolerated, except for the highly activated norbornene; even this strained alkene offers much lower yields than terminal alkyl alkenes or aryl-substituted styrenes. Recently, a report was published describing new Ti complexes supported by 2aminopyridinate ligands that can catalyze the intermolecular α-alkylation of amines.38 The complexes are generated in situ using either 1 or 2 molar equivalents of ligand though it should be noted that with 1 equivalent of ligand a mixture of titanium species are observed, including Ti(NMe2)4 (Figure 1.2). 19  Figure 1.2 2-Aminopyridinate titanium catalysts for the α-alkylation of amines.  These complexes, akin to the aforementioned Ti catalysts, are compatible with arylsubstituted styrenes and N-methylanilines.  Interestingly, the α-alkylation of select dialkyl  amines is also reported, albeit in low yields with mild regioselectivities. It is worth noting that these reactions are carried out at high temperatures (140 °C) using n-hexane as a solvent (boiling point 69 °C) for long reaction times (96 hours). Unfortunately, these group 4 systems are still plagued by selectivity shortcomings as well as limited substrate scope.  1.5  Group 5 Metal Catalysts As predicted by the initial reports, group 5 metal complexes show the most promise as  broadly applicable catalysts for the α-alkylation of amines.7-8  Using the same homoleptic  pentakis(dimethylamido)tantalum complex as Maspero and Nugent, Hartwig and Herzon showed in 2007 that the α-alkylation of N-aryl amines with terminal alkenes could be achieved in yields up to 96% (Scheme 1.20).39 Although these reactions were performed at high temperatures (160 – 165 °C) over long reaction times (27 – 67 hours), this report revived the catalytic α-alkylation of amines as a viable pathway for amine synthesis. Interestingly, the branched product was exclusively observed in all cases except for substrates with a silyl group adjacent to the double bond. In 2008, the same researchers published a report highlighting a more electrophilic catalyst  20  bearing chloride ligands (Scheme 1.20).40 This new system allowed for reactivity with dialkyl amines and terminal alkenes in 24 –36 hours at slightly reduced temperatures (150 °C). More impressively, the α-alkylation of N-methylaniline by 1-octene could be achieved at temperatures as low as 90 °C. During this investigation, a wide variety of chelating amine and alcohol ligands were reacted with Ta(NMe2)5 to form in situ precatalysts.41 However, none of the resulting precatalysts were more effective than either the parent homoleptic tantalum complex or chloride dimer.  Scheme 1.20 Tantalum α-alkylation of amine catalysts supported by amido and chloride ligands.  Since these initial reports, a number of group 5 catalysts for the α-alkylation of amines have been reported. In 2010 and 2011, Zi and coworkers described synthesizing a series of bis(amidate) ligands with axially chiral biaryl backbones and installed them on vanadium, niobium and tantalum.42-43 These complexes were found to be proficient at catalyzing the intramolecular α-alkylation of aminoalkenes. Interestingly, the vanadium systems were the most reactive, with excellent yields and low to good enantioselectivities (11 – 76% ee) after 24 hours at 140 °C. However, neither the vanadium nor niobium systems were found to catalyze the  21  intermolecular α-alkylation of N-methylaniline with terminal alkenes. In contrast, when the tantalum congeners (Figure 1.3) are used the reaction proceeds with yields greater than 80% and up to 93% ee. Unfortunately, temperatures of 130 – 160 °C are required as well as multi-day reaction times (48 – 120 hours).  Figure 1.3 Enantioselective tantalum α-alkylation of amines catalysts supported by biphenyl and binaphthyl bis(amidate) ligands.  In 2011, silylated bis(naphtholate) complexes of niobium and tantalum reported by Hultzsch and coworkers were shown to catalyze the α-alkylation of amines (Scheme 1.21).44-45 In contrast to the finding of Zi and coworkers, both Nb- and Ta-based systems are competent catalysts for intra- and intermolecular reactivity. The tantalum systems offer better reactivity for intramolecular substrates, achieving full conversion to product in 11 – 36 hours at 100 – 140 °C with 0 – 81% ee. The opposite is true for intermolecular reactivity, with the Nb systems completing the reaction in almost half the time required for their Ta congeners. Though reaction temperatures are still high at 140 °C, the niobium catalysts can achieve maximum conversion with N-aryl amines and terminal alkenes in less than 15 hours with 59 – 98% ee. While decreasing the temperature increases the enantioselectivities, it is accompanied by a drastic lengthening of reaction times.  22  Scheme 1.21 Enantioselective niobium and tantalum α-alkylation of amines catalysts supported by silyl-substituted bis(naphtholate) ligands.  1.6  Late Transition Metal Catalysts Though the initial reports by Maspero and Nugent did not involve any late transition  metals, ruthenium and iridium catalysts have been shown to activate C–H bonds α to nitrogen in amine substrates and add across alkene unsaturations (Scheme 1.22), more akin to the proposed mechanism for early transition metals.  23  Scheme 1.22 Direct α-alkylation of amines with late transition metal catalysts.  In 1998, Jun and coworkers reported a ruthenium carbonyl system, Ru3(CO)12, that adds benzyl sp3 C–H bonds α to a nitrogen atom across carbon-carbon double bonds (Scheme 1.22).46 A variety of alkenes can be used for this transformation including styrene and internal alkenes. However, this process is limited to amines bearing a pyridyl directing group in the 2 position, as coordination of the substrate to the ruthenium centre is necessary for catalytic turnover and requires temperatures of 130 °C. A generalized proposed mechanism for this transformation is outlined in Scheme 1.23. Pre-coordination of the pyridine nitrogen to the metal centre facilitates the oxidative addition of the C–H bond to give a metal hydride. Insertion of the alkene into the metal-hydride bond followed by reductive elimination yields the product and regenerates the catalysts. Based on this proposed mechanism, the C–C bond will be formed exclusively at the terminal sp2 carbon. Interestingly, when 2- or 3-hexene is employed, an isomerization to the linear 1-hexyl-metal 24  complex is observed to avoid steric congestion, ultimately generating the same product observed when using 1-hexene as a substrate.46-47  Scheme 1.23 Generalized scheme for direct α-alkylation of amines using late metal catalysts.  This precatalyst was subsequently shown by Murai and coworkers to catalyze the alkylation of N-2-pyridinyl alkylamines (Scheme 1.22).48 Like Jun, they found that the pyridine donor is required for the reaction to proceed and are therefore restricted to specific amine substrates. Interestingly, double alkylations are possible when the amine substrates bear two methylene groups α to the nitrogen. This leads to a variable mixture of mono- and bis-alkylated products depending on the substrate. Though the alkene substrate scope is comparable to the Jun system, the use of ethylene as a feedstock is a noteworthy addition.  25  In addition to the ruthenium systems, Shibata and coworkers communicated the use of a cationic iridium catalyst bearing a chiral bis-phosphine ligand for the stereoselective α-alkylation of amines.49 Again requiring the pendant pyridine group for amine substrate activation, reaction with terminal alkenes generates the products in good yields with up to 90% ee. In addition to introducing enantioselectivity, they were able to significantly reduce the reaction temperature to 85 °C albeit at the expense of longer reaction times. The aforementioned ruthenium and iridium complexes are competent catalysts for the direct α-alkylation of amines but are plagued by the requisite pendant pyridyl group on the amine substrate. This, along with high cost and toxicity issues, presents a major challenge in the development of a more general late transition metal catalyst for this process.  1.7  Scope of Thesis As a reaction, the α-alkylation of amines is still in its infancy. To date, only a handful of  complexes have been discovered that can catalyze this C–C bond forming reaction (vide supra). Though recent progress has been made to improve substrate scope and reducing harsh reaction conditions, there remains much room for improvement. To this end, a systematic study has been undertaken to focus on the development of a catalyst for the α-alkylation of amines based on an early transition metal centre supported by N,O-chelating ligands using both experimental and theoretical methods. In Chapter 2, a series of mono(amidate) tantalum precatalysts with varying degrees of steric bulk and electronic properties is presented. Trends in catalytic reactivity are related to metric parameters found in the solid state as well as solution phase spectroscopic data. Based on catalytic performance and complex stability, a lead compound is identified for further study. In  26  addition to the mono(amidate) precatalysts, a number of bis(amidate) complexes are also discussed.  Though not as catalytically active as their mono(amidate) counterparts, these  complexes offer the opportunity to study the fundamental nature of these amidate complexes. For instance, the hemi-lability of amidate ligands is probed using variable temperature NMR studies of a κ1(O), κ2(N,O) ligated tantalum species. Model catalytic intermediates including a tantalaziridine (A, Scheme 1.3) and the resultant insertion product (B) have also been synthesized.  These fundamental stoichiometric studies aid in the development of future  generations of tantalum amidate catalysts. In addition, isotopic labeling experiments probe the equilibrium governing tantalaziridine formation from the precatalyst and off-cycle pathways. Preliminary kinetic studies to determine the order of the catalyst and the substrates in the rate law are also performed. The possibility of one electron processes in the catalytic cycle are explored using ultrafast radical probes. To deepen the understanding of the catalytic cycle, the first theoretical model for group 5 catalyzed α-alkylation of amines was developed (Chapter 3). A full set of intermediates and transition states were found using quantum chemical calculations, furthering the understanding of specific steps in the catalytic cycle. Alternate reaction pathways based on catalyst geometry and the possibility of a one electron mechanism were also explored.  Efforts towards the  development of a steric parameter that can aid in predictive catalyst design are also reported. In order to build the library of α-alkylation of amines catalysts available for study, a series of diamide proligands containing a neutral donor atom tether was synthesized and installed on zirconium and tantalum. Chapter 4 focuses on the characterization and reactivity of these complexes for both the α-alkylation of amines and intramolecular hydroamination. Inspired by other work in the Schafer group, corresponding bis(ureate) complexes were also synthesized,  27  characterized and tested as catalysts. Additionally, the fundamental differences between the amidate and ureate complexes are explored using electrochemical techniques. Finally, overall conclusions drawn from and a summary of the research presented herein will be presented in Chapter 5.  This chapter also contains proposed avenues of research  stemming from the knowledge derived herein.  Included are ideas for the spectroscopic  monitoring of reactions in situ using isotopically labeled amidate ligands as well as alternative mechanistic pathways for the α-alkylation of amines catalyzed by mono(amidate) tantalum catalysts.  28  Chapter 2: Tantalum Amidate Complexes as Precatalysts for the αAlkylation of Amines  2.1 2.1.1  Introduction Amidates as Ligands for Early Transition Metal Complexes Amidate ligands are a set of N,O-chelating ligands that have been shown to support a  variety of transition metal complexes.50 The amide proligands are easily synthesized in high yields from inexpensive, commercially available primary amines and acid chlorides (Scheme 2.1). This modular synthesis allows for the systematic investigation of both electronic and steric effects. Amidate ligands can adopt a variety of binding modes (Figure 2.1),51 however, the κ1(O) and κ2(N,O) arrangements are preferred by oxophilic early transition metals. Previous work has shown that κ2(N,O) bound amidates are not susceptible to unwanted reactivity at the ligand.51 As a result of these previous studies, amidate ligands have been proven to support discrete early transition metal complexes that can be used as robust catalysts and can stabilize reactive species when used as auxiliary ligands.  Scheme 2.1 Generalized synthesis of amide proligands.  29  Figure 2.1 Possible binding modes for amidate ligands.  For example, tris(amidate) yttrium complexes are highly active initiators for the ringopening polymerization of ε-caprolactone while related bis(amidate) complexes have been shown to catalyze the hydroamination of aminoalkenes as well as the amidation of aldehydes.5254  Interestingly, bis(amidate) yttrium complexes can catalyze the hydroamination of both  primary and secondary aminoalkenes.53 This is in contrast to the bis(amidate) complexes of group 4 precatalysts that are found to be incompatible with secondary aminoalkenes due to an alternative [2+2] cycloaddition mechanism.29-32,  55-58  A key feature of the latter proposed  mechanism is an imido intermediate as the active precatalyst.59  Interestingly, bis(amidate)  species containing the reactive imido moiety can be synthesized by reacting bis(amidate) bis(amido) complexes with primary amines in the presence of a neutral donor (Scheme 2.2).29, 57 These compounds are stable at ambient temperatures and have been characterized in both solution phase and the solid state. Gratifyingly, they show comparable hydroamination reactivity when compared to the parent bis(amidate) bis(amido) complex.  30  Scheme 2.2 Synthesis of bis(amidate) imido complexes of Ti29 and Zr.57  In addition to stabilizing reactive imido species, amidate ligands have also been shown to support titanium, zirconium and hafnium complexes containing chloride and benzyl ligands.60-63 The fundamental reactivity of the metal-carbon bonds found in the bis(amidate) benzyl complexes was investigated by reacting these species with alkyl and aryl isocyanates.62 In all cases, the N–C isocyanate unsaturation was inserted into the M–C bond as expected. However, the reaction with 2,6-dimethylphenyl isocyanate produced two η2-iminoacyl moieties (Scheme 2.3), a functional group closely associated with zirconaziridines.  31  Scheme 2.3 Insertion of isocyanate into the Zr–C bonds of a bis(amidate) dibenzyl zirconium complex.62  The ability of amidate ligands to stabilize reactive organometallic species containing strained metallacycles and promote the isolation of catalytic intermediates makes the amidate ligand set an attractive starting point for the development of a group 5 catalyst for the intermolecular α-alkylation of amines. Based on the success of tantalum-based catalysts in the literature,7-8, 39-40 the chemistry of tantalaziridines, a key reactive intermediate, was identified as a reactivity profile that required further investigation.  2.1.2  Synthesis, Structure, and Characterization of Tantalaziridines Tantalaziridines are an interesting class of organometallic compounds that feature a  strained three-membered ring composed of one tantalum, one nitrogen and one carbon atom. They have two limiting resonance structures: a dianionic, 4 electron donor ligand as an azametallacyclopropane and a neutral 2 electron donor η2-imine ligand with the tantalum atom interacting with the π system of the C=N bond (Figure 2.2). The related, monoanionic iminoacyl moiety is an alternative accessible metallacycle that can be a useful reagent for a variety of  32  transformations,64 including their application as precursors for the synthesis of tantalaziridines (Scheme 2.4).  Figure 2.2 General forms of three related compounds: tantalaziridine (left), η2-imine complex (middle) and iminoacyl complex (right).  In the tantalaziridine moiety, the carbon atom has two substituents bound to it and the nitrogen atom has one. The synthesis of such tantalaziridines occurs through 3 major routes (Scheme 2.4): insertion of isonitriles into two existing tantalum-ligand bonds, reduction of C=N π bonds with a tantalum(III) species and β-hydrogen abstraction from a Ta–NCH2R group to eliminate a neutral compound and form a Ta–C bond.  33  Scheme 2.4 Common methods for synthesizing tantalaziridines. Insertion into two Ta–R bonds (top); reaction of imine with Ta(III) species (middle); β-hydrogen abstraction from amido ligand (bottom).  As with tantalum iminoacyl complexes, the most common method of synthesis of tantalaziridines is the insertion of isonitriles into an existing tantalum-element bond. Insertion of 1 equivalent of isonitrile generates an iminoacyl intermediate, which then reacts with another ligand bound to the tantalum to produce the tantalaziridine (Scheme 2.4, top).65-69 In certain cases, largely depending on the steric bulk of the isonitrile, it is possible to isolate the iminoacyl intermediates as gentle heating is required to give the desired tantalaziridine complex.70 More typically, the iminoacyl intermediates are never observed, but are proposed as intermediates en route to the tantalaziridine products.66-67 Another method for generating such fused bicyclic tantalaziridines involves reacting aromatic heterocycles, such as pyridine and its derivatives, with Ta(III) species to generate the η2-N,C binding motif (Scheme 2.4, middle).  Wolczanski and coworkers have shown that  siloxide supported dichloride tantalum(V) species bearing pyrrolide or piperidyl ligands can be reduced to give tantalum(III) species, and then undergo reaction with pyridine to generate a 34  tantalaziridine complex (Scheme 2.5).71 Similar tantalum siloxide compounds without the amido donor have been rigorously characterized in the solution phase and in the solid state.72  Scheme 2.5 Reaction of an in situ generated Ta(III) species with pyridine. The heterocycle experiences a loss of aromaticity when the tantalaziridine is installed.71  Based on frontier molecular orbital arguments, tantalum backdonation into the π* C=N orbital of the heterocycle substrate occurs to generate the new Ta–C and Ta–N bonds.73 Notably, as evidenced from the solid-state structure, the aromaticity of the pyridine ring is broken and it is bound to the tantalum as a tantalaziridine rather than an η2-imine. The C–N bond distance of 1.43 Å is suggestive of a single bond, supported by the 13C NMR shift of δc = 82. The planarity of the pyridine ring is also slightly distorted as the nitrogen atom becomes more pyramidal. Further mirroring these reactivity trends, related fused tantalaziridine complexes can be accessed with prolonged reaction times with 2,6-lutidine substrates to give highly substituted tantalaziridine products.73 Such compounds demonstrate the preference for tantalum to be in its highest oxidation state with a tantalaziridine moiety rather than the formal reduced Ta(III) species with an η2-imine interaction. Interestingly,  Cummins  and  coworkers  have  shown  that  when  a  bulky  tris(amido)diiodotantalum complex is treated with a reducing solution of magnesium anthracide 35  in THF, an intramolecular oxidative addition of an -sp3 hybridized C–H bond results giving a tantalaziridine and hydride bound tantalum (V) product (Scheme 2.6).74 Most importantly, this species is in equilibrium with a less strained 5-membered metallacycle, resulting from the reversible C–H oxidative addition and subsequent activation of another sp3 hybridized bond to give a less strained 5-membered metallacycle.  Scheme 2.6 Reduction of a tantalum diiodo tris(amido) species to result in β-hydride elimination from the neopentyl ligand to give a tantalum hydride as well as a metallaziridine (top). Equilibrium between 3- and 5-membered metallacycles (bottom).74  Another well-established method of tantalaziridine formation involves β–H abstraction of an sp3 hybridized C-H bond to eliminate an equivalent of alkane or amine from the tantalum coordination sphere. This phenomenon was first disclosed when TaCl5 was reacted with lithium diethylamide, presumably to generate a pentakis(diethylamido) product (Scheme 2.7).75-76 However, spectroscopic evidence, including the diagnostic signal at δC = 64.0 in the spectrum was suggestive of an sp3 hybridized carbon directly bound to tantalum.  13  C NMR  Indeed, the  integrations from the 1H NMR spectrum as well as the mass spectrum of the compound showed 36  there to be one less diethylamido ligand than expected. The A3B coupling pattern in the 1H NMR spectrum (δA = 1.91, δB = 2.28 ppm, JAB = 5.0 Hz) confirmed the presence of a –CHCH3 moiety while a strong, distinctive band in the IR spectrum at 1235 cm-1 suggested a C–N single bond. These factors led to the assignment of the first reported tantalum complex featuring a 3membered azametallapropane ring. While these early reports suggested the formation of such metallacycles, it took 18 years before a tantalaziridine generated via β–H abstraction could be crystallographically characterized.77  Scheme 2.7 Spontaneous elimination of diethylamine to give a tris(amido)tantalaziridine complex.75-76  Following up on this synthetic strategy, it was established that alkane elimination is a high-yielding and reliable synthetic approach for the synthesis of early transition metallaziridines generally.71,  78-79  For example, Bercaw and coworkers established that Cp*Ta(NMe2)Me3 is  stable and isolable at temperatures below 0 °C, but at room temperature undergoes β–H abstraction to generate methane and a tantalaziridine (Scheme 2.8).78 The process was slow enough to perform a kinetic analysis on the system, finding that the decomposition obeys firstorder kinetics.  Deuterium labeling of the amidomethyl groups resulted in the exclusive  production of CH3D and a large primary kinetic isotope effect was observed (kH/kD ≈ 9). Monitoring of the decomposition of the perdeutero and perprotio species in the same reaction  37  produced CH4 and CD4 with no crossover products observed, suggestive of a unimolecular process.  Scheme 2.8 Decomposition of a mixed amido-, alkyl-tantalum half sandwich complex to give a tantalaziridine and eliminate methane.78  A similar decomposition process, the C–H activation of the donor arm of a Nheterocyclic carbene diamido ligand, has been studied by computational methods.80 In this case the calculations suggest that the lower energy pathway involves a tantalum alkylidene intermediate, generated by expulsion of methane from the model system, which mediates C–H activation from the ligand methylene group rather than direct β–H abstraction (Scheme 2.9). This result was further supported by deuterium labeling studies as well as the observed reactivity of the ligand with a tantalum complex with a pre-installed alkylidene moiety.  38  Scheme 2.9 Two possible pathways for tantalaziridine formation from N-donor arm of a tantalum N-heterocyclic carbene complex. Calculations and labeling experiments show alkylidene formation followed by C–H addition is the preferred pathway.80  The use of early metal alkylidenes as reactive species suitable for inducing β–H abstraction has precedence, as the Boncella group used a tris(pyrazolyl)borate supported tantalum alkylidene as a suitable reagent for transmetallation with potassium methylanilide (Scheme 2.10). Here the tantalaziridine product was the only product observed, resulting from an intramolecular proton transfer from the newly installed N-CH3 group to the alkylidene carbon.81  39  Scheme 2.10 Proton transfer to a carbene ligand to generate a tantalaziridine in a hydrotris(3,5dimethylpyrazoyl)borate complex.81  Analogous to alkane elimination, tantalaziridines can be synthesized via the spontaneous elimination of amines.  Indeed, the first reported tantalaziridines were synthesized by this  method (vide supra).75-76 Gambarotta and coworkers have also reported the β–H abstraction of an amido ligand to eliminate dicyclohexylamine and form a tantalaziridine (Scheme 2.11).77 However, this result is observed with only 4 equivalents of lithium amide salt added to TaCl 5 in contrast to the 5 equivalents of Sugiyama and Bradley. The presence of a chloride should alleviate some of the steric strain in the tantalum coordination sphere but the added bulk of the cyclohexyl substituted amido ligands is great enough to force the system to eliminate amine and generate a tantalaziridine.  40  Scheme 2.11 Spontaneous elimination of dicyclohexylamine to give a tris(amido)tantalaziridine complex.77  While tantalaziridines themselves are an interesting class of compound, exploring the reactivity of these species to form novel organic and organometallic compounds is of interest in the development of synthetic approaches for the preparation of amines and amine derivatives.  2.1.3  Stoichiometric Reactivity of Tantalaziridines The tantalaziridine moiety features a reactive metal-carbon bond that offers the  opportunity for further functionalization. Insertion of unsaturated molecules into the tantalumcarbon bond extends the metallacycle from a 3- to a 4- or 5-membered ring, analogous to the stoichiometric reactivity observed for zirconaziridines in Section 1.2. Intramolecular reactions with alkyl ligands bound to tantalum have also been documented. However, not all of these insertion products are stable, with some species rearranging to form imido ligands or expelling small organic molecules. The tantalaziridine Ta–C bond has been shown to react with isonitriles to generate azacyclotantalabutanes82 that, when supported by ancillary chloro ligands, can rearrange to give tantalum-imido species (Scheme 2.12).83 Similar imido-containing products are obtained when dichloro tantalaziridine pentamethylcyclopentadienyl complexes are mono-alkylated with Grignard reagents.84 Compounds containing both iminoacyl and tantalaziridine ligands have  41  been shown to undergo similar reactions to form species containing both imido and amido, or iminoacyl, ligands.85-86  Scheme 2.12 Insertion of 2,6-dimethylphenylisonitrile into Ta–C bond of tantalaziridine to form an azacyclotantalabutanes (top).82 Decomposition of a related chloro species to give a tantalum imido complex and an unsaturated organic molecule (bottom).83  Carbon-heteroatom multiple bonds, such as the C=O double bonds of cyclohexanone and benzaldehyde, can be inserted into the Ta–C bond to give a 5-membered metallacycle containing both Ta–N and Ta–O contacts.87  Alternatively nitriles can insert to generate 5-membered  metallacycles with two Ta-N contacts (Scheme 2.13) where the intermediate imine structure undergoes a 1,3-hydride shift to generate the observed alkene product.88  42  Scheme 2.13 Insertion of nitriles into the tantalaziridine Ta–C bond to give a 5-membered metallacycle with two amido contacts.88  Alternatively, carbon-carbon multiple bonds can also be inserted into the tantalaziridine fragment, creating 5-membered azametallacycles (Scheme 2.14). Both alkenes, such as ethylene, and alkynes, such as acetylene, phenylacetylene and trimethylsilylacetylene, have been shown to be competent substrates for this type of insertion.84  In a Cp supported Ta complex, the  azatantalacyclopentane ring generated by the insertion of ethylene is not stable in solution, and in the presence of light decomposes to a tantalum imido species in a matter of hours, releasing alkene products.  However, the stability of the azatantalacyclopentene (Scheme 2.14) has  improved stability and has been isolated and rigorously characterized.84  43  Scheme 2.14 Insertion of ethylene (top) or alkynes (bottom) into a tantalaziridine to form a 5membered metallacycle.84  In cases where tantalaziridine complexes are formed via –H elimination, resulting in mixed tantalaziridine hydrides, the Cummins group has shown that the C–O multiple bonds of CO2 react preferentially with the tantalum hydride moiety rather than the Ta–C bond of the tantalaziridine to generate a bridged methylene diolate complex (Scheme 2.15).74  44  Scheme 2.15 Reaction of a tantalum hydride with CO2 to give a bimetallic methylene diolate complex. The tantalaziridine moiety does not participate in the reaction.74  Analogous to the zirconaziridine systems discussed in Section 1.2, stoichiometric quantities of tantalaziridines can be reacted with ketone or aldehyde substrates to generate βamino alcohols upon quenching with protic solvents (Scheme 2.16).87 The aldehyde reaction proceed in high yields (> 85%) with a 3:1 preference for the syn diastereomer due to minimized steric interaction between the phenyl groups in the trans metallacyclic intermediate (Figure 2.3). Furthermore, isocyanate substrates have been shown to generate α-amino amides and ureas in a 2:1 ratio.87 While isocyanate insertion into the Ta–C bond of the tantalaziridine results in the preparation of α-amino amides, the significant amount of urea product suggests that insertion into the Ta-N bond of the tantalaziridine is a competitive reaction profile with this substrate.  45  Scheme 2.16 Organic products isolated after aqueous workup of the insertion of cyclohexanone, benzaldehyde and cyclohexylisocyanate into a tantalaziridine chloride complex.87  Figure 2.3 Metallacycle intermediates for the insertion of benzaldehyde into the tantalaziridine.87  While such organic products can be useful small molecule building blocks, the stoichiometric generation of reactive tantalum containing intermediates and extensive by-  46  products formed upon quenching of the reaction present challenges for the application of this approach in large scale synthesis. Despite the stoichiometric work discussed herein, no catalytic systems based on tantalaziridines were known prior to the work contained in this thesis.  2.1.4  Scope of Chapter This chapter focuses on the ability of tantalum complexes supported by amidate ligands  to catalyze the α-alkylation of amines. Solution phase and solid-state data will be presented for mono(amidate) tantalum precatalysts in Section 2.2.1, highlighting the effect of steric bulk on these systems. Section 2.2.2 will correlate trends in catalytic activity with the bulk at the 2 and 6 position of the N-aryl substituent of the amidate ligand.  Changes in reactivity based on  electronic perturbations of the ligand as well as amidate binding mode (κ1(O) or κ2(N,O)) will also be discussed. A series of bis(amidate) complexes have also been synthesized and will be explored in Section 2.2.3. Though less efficient at catalyzing the α-alkylation of amines, these species offer the opportunity to study the fundamental chemistry of tantalum amidate systems. For instance, the hemi-labile nature of amidate ligands has been probed using a variable temperature NMR spectroscopy study of a bis(amidate) complex containing both κ1(O) and κ2(N,O) bound amidate ligands. Additionally, the room temperature formation of a stable bis(amidate) tantalaziridine, via C–H activation of a dimethylamido ligand, is reported. A catalytic intermediate in the proposed mechanism, this tantalaziridine is shown to contain an azametallacyclopropane moiety bound to a d0, Ta(V) centre. The reactivity of this complex with small molecules is also explored.  47  Section 2.2.4 describes the efforts to elucidate the mechanism of the α-alkylation of amines catalyzed by a mono(amidate) tantalum species. Byproduct formation will be discussed as well as the investigation of off-cycle pathways using isotopically labeled substrates. Preliminary kinetic studies were performed to probe the role of the catalyst and substrates in the reaction, resulting in a proposed empirical rate law. The installation of radical probes into precatalysts and substrates to explore the possibility of a one-electron mechanism will be described in Section 2.2.5.  2.2 2.2.1  Results and Discussion Synthesis and Characterization of Mono(amidate) Tantalum Complexes Development of a widely applicable catalyst for the α-alkylation of amines based on  group 5 metals is an area of current academic interest.6 The ability of amidate ligands to generate electropositive metal centres as well as stabilize reactive intermediates makes this class of ligand an obvious starting point.29, 57, 62 To this end, a series of amide proligands with varying degrees of steric bulk at the 2 and 6 position of the phenyl ring bound to nitrogen were synthesized to elucidate the influence of steric bulk on catalyst performance in the α-alkylation of amines. The amides discussed in this chapter (Figure 2.4) were synthesized from their respective primary amines and acid chlorides in the presence of a nitrogen base such as triethylamine or pyridine.  Less bulky amines react more vigorously with acid chlorides,  requiring slow additions and low temperatures.  In contrast, amines with increased steric  demands such as 2,4,6-tri-tert-butylaniline require increased temperatures and longer reaction times for complete conversion to the amide.  48  Figure 2.4 Amide proligands used for synthesizing mono(amidate) tantalum complexes.  The Schafer group has shown protonolysis to be an efficient method of installing amidate ligands with group 3 and 4 metal amido complexes.32, 52-53, 58, 61 However, group 4 complexes are restricted to bis(amidate) species due to ligand disproportionation.89 An analogous protonolysis route was undertaken to synthesize group 5 mono(amidate) complexes. Gratifyingly, reacting equimolar amounts of amide proligand and pentakis(dimethylamido)tantalum in hexanes at room temperature yields the mono(amidate) complexes 1 – 4 in excellent yields (Scheme 2.17). This is presumably due to the increased steric congestion about the tantalum coordination sphere preventing intermolecular interaction.  Scheme 2.17 Protonolysis synthesis of mono(amidate) tantalum complexes 1 – 4.  Slow addition of the ligand is achieved by exploiting the sparing solubility of the amide proligands in hexanes (Scheme 2.17). However, this strategy could not be employed for 5 due to its total insolubility in hexanes. For this reaction, dichloromethane (DCM) was employed as the  49  solvent and the reaction mixture was heated until the ligand was dissolved (Scheme 2.18). Though the ligand is insoluble in hydrocarbon solvents, the resulting complex is highly soluble even in non-polar solvents such as pentane. This leads to a significant decrease in recrystallized yield.  Scheme 2.18 Protonolysis synthesis of mono(amidate) tantalum complex 5.  This series of mono(amidate) tantalum complexes was found to be highly crystalline and the complexes were purified by recrystallization in hexanes or pentane. Each complex was characterized in the solution phase, using NMR techniques, and the solid state, using single crystal X-ray diffraction. Combining these data allows for the evaluation of the amidate ligand’s steric and electronic influences. The solid-state molecular structures (Figure 2.5) for 1 – 4 reveal amidate ligands bound κ2(N,O) to the tantalum centre, while 5 is observed in the κ1(O) configuration (vide infra). The complexes all crystallize in the C1 point group except for 4 which is Cs symmetric due to its mirror plane.  The κ2(N,O) mono(amidate) complexes adopt a pseudo-trigonal bipyramidal  geometry where the amidate ligand is considered to be occupying one coordination site and the Ta–Camidate vector is used for determining bond angles. On average, the bond angles in the plane of the amidate are closer to 120° (trigonal bipyramidal) than 90° (octahedral), largely on account 50  of the tight Oamidate–Ta–Namidate bite angle. Therefore, the amidate ligand occupies an equatorial coordination site along with two dimethylamido ligands. The remaining dimethylamido ligands occupy the two axial positions.  This pseudo-trigonal bipyramidal assignment is further  supported by the observation that the equatorial Namido–Ta– Namido bond angle tends towards 120° as the steric demands of the amidate ligand are alleviated. The dimethylamido ligands exhibit planar nitrogen, with sums of bond angles approximately 360°, suggesting each amido ligand is a monoanionic, 4 electron donor. Therefore, counting the monoanionic, 4 electron amidate ligand, these tantalum mono(amidates) are formally 20 electron complexes.  51  Figure 2.5 ORTEP representations of the solid-state molecular structures for 1, 2, 3, 4 and 5 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  52  Table 2.1 Crystallographic data and relevant metrical parameters for 1 – 5. Simplified illustration showing idealized geometry given for reference.  Space group Ta–Oamidate (Å) Ta–Namidate (Å) Ta–N(trans to Namidate) (Å) Ta–N(trans to Oamidate) (Å) Ta–Naxial1 (Å) Ta–Naxial2 (Å) Camidate–Oamidate (Å) Camidate–Namidate (Å) Oamidate–Ta–Namidate (°) (O–C–N)amidate (°) Nequatorial–Ta–Nequatorial (°)  1 Pī 2.1396(15) 2.3392(17) 1.9931(19) 1.978(2) 2.042(2) 2.0294(19) 1.302(2) 1.298(3) 57.66(6) 112.86(19) 107.62(8)  2 Pī 2.1051(12) 2.4224(14) 1.9978(15) 1.9788(15) 2.0508(15) 2.0422(15) 1.317(2) 1.293(2) 56.99(4) 113.10(15) 103.27(6)  3 P21/n 2.099(3) 2.447(3) 1.996(4) 1.970(3) 2.037(4) 2.050(4) 1.312(5) 1.297(5) 56.90(10) 113.9(3) 101.94(16)  4 Pnma 2.1188(18) 2.451(2) 2.001(2) 1.984(2) 2.0621(19) 2.0621(19) 1.323(3) 1.302(3) 56.85(7) 113.4(2) 101.72(9)  5 Pī 2.0586(15) 1.994(2) 1.319(2) 1.278(2) 127.34(18) 121.64(9) 119.16(9) 119.18(8)  The steric bulk at the 2,6 positions of the N-aryl ring also has a significant effect on the binding mode of the amidate ligand. For instance, the increase in steric bulk from 1 to 2 to 3 corresponds to a lengthening of the Ta–Namidate bond (2.3392(17) Å, 2.4224(14) Å,and 2.447(3) Å respectively). The increase in Ta–Namidate bond length corresponds to a decrease in Ta–Oamidate bond length over the same series: 2.1396(15) Å for 1; 2.1051(12) Å for 2; 2.099(3) Å for 3. This can also be viewed in terms of bond strengths: as the Ta–Namidate bond weakens, the Ta–Oamidate bond becomes stronger in a shift towards a κ1(O) binding mode (vide infra). Solution phase data suggests the mono(amidate) complexes adopt the same geometry in solution as in the solid state. The 1H NMR signals attributed to the amidate ligand are sharp and 53  suggests only one species is present. In contrast, the signals for the methyl groups of the dimethylamido ligands appear as one broad signal between 3.2 and 3.4 ppm. This suggests the axial and equatorial dimethylamido ligands are exchanging on the NMR timescale and therefore cannot be resolved as individual signals. Substituting a phenyl group for the tert-butyl group in the amidate ligand, compound 4, does not change the Ta–Namidate bond length but does lengthen the Ta–Oamidate bond to 2.1188(18) Å. This suggests the amidate ligand is not bound as strongly in 4 as with 3. However, the metrical parameters of the ligand itself are relatively unchanged. For example, even the Camidate– Namidate–Cipso bond angle is within experimental error for complexes 3 and 4 (125.1(3)° and 124.6(2)° respectively), suggesting the change in electronics affects the Ta-amidate interactions but not the ligand geometry.  Interestingly, the Namidate–Ta–Oamidate bite angles are largely  unaffected by steric or electronic perturbations, and are comparable to previously characterized group 4 systems.51 In the case where tBu groups are at the 2,6 positions (5), the nitrogen is too sterically hindered to interact with the metal centre. Interestingly, the ligand adopts a κ1(O) binding mode, binding exclusively through the oxygen which occupies an axial coordination site. Following the observed trend, compound 5 has the shortest Ta–Oamidate bond length at 2.0586(15) Å. This short Ta–Oamidate distance is accompanied by a smaller Camidate–Namidate distance nearing a double bond (1.278(2) Å) and an obtuse Oamidate–Camidate–Namidate bond angle (127.34(18)°) compared to the κ2(N,O) amidates 1 – 4 (113.5(2)° average). In addition, the Camidate–Oamidate–Ta bond angle of 152.33(12)°, together with the aforementioned data, suggests that the amidate ligand in 5 is best described as an iminoalkoxide ligand.  Gratifyingly, the  13  C{1H} NMR shift for the carbon in  the amidate chelate can be used to assign κ1(O) or κ2(N,O) binding modes in solution. While the  54  δC ranges from 177.3-180.4 for the κ2(N,O) binding mode in 1 – 3, a shift of δC = 160.3 for 5 confirms its solution phase κ1(O) coordination. While an increase in steric bulk has little effect on the Ta–Namido contacts in the plane of the amidate chelate, it does cause a lengthening of the axial dimethylamido bonds. Since the phenyl ring of the Namidate substituent is perpendicular to the plane of the N,O chelate, the groups at the 2 and 6 positions are pointing directly towards the axial dimethylamido ligands, thus amplifying the steric interactions. For the mono(amidate) complexes 1 – 4, the axial Ta–Namido bonds are longer (2.047(4) Å average) than those in the equatorial plane of the chelating atoms, with the amido pseudo-trans to Oamidate (1.977(3) Å average) having a shorter Ta–Namido distance than that pseudo-trans to Namidate (1.996(4) Å average).  2.2.2  Catalytic Reactivity of Mono(amidate) Tantalum Complexes Based upon the successful use of Ta(NMe2)5 as a catalyst for the α-alkylation of  amines,39 it was hypothesised that mono(amidate) tantalum complexes could also be used as catalysts for this transformation. This view was enhanced with the discovery that increasing the electrophilicity of the tantalum metal centre, by replacing amido ligands with chlorides, increases catalytic activity.40 The fact that amidate ligands have been shown to support electrophilic early transition metal centres52-53 suggested that mono(amidate) tantalum complexes would be proficient catalysts for the α-alkylation of amines. To this end, compounds 1 – 5 were screened in the α-alkylation of N-methylaniline with 1-octene, a convenient screening reaction for the intermolecular α-alkylation of amines (Table 2.2). These reactions were setup in a nitrogen filled glove-box under rigorously anhydrous conditions as the tantalum amido complexes are sensitive to both H2O and O2.90-91 The precatalyst, N-methylaniline, and 1-octene were dissolved  55  in benzene-d6 and transferred to a Teflon-capped NMR tube. The tube was sealed and brought out of the glove-box and put in a preheated oil bath set to 110° C. The reaction was monitored by 1H NMR spectroscopy by integrating the ortho-protons of N-methylaniline (δH = 6.40) and the product, 6, (δH = 6.48) relative to an internal standard (1,3,5-trimethoxybenzene). Exclusive formation of 6, the branched regioisomer, is observed, consistent with the proposed mechanism (Scheme 1.3) and reports for other group 5 systems.39-40, 42-45, 92-93  Table 2.2 Screening of mono(amidate) tantalum precatalysts for the α-alkylation of Nmethylaniline with 1-octene. Data for Ta(NMe2)539 and [Cl3Ta(NMePh)2]240 is shown for comparative purposes.  Entry  Catalyst  Time (h)  NMR yield (%)  1  1  63  89  2  2  66  93  3  3  63  96  4  4  77  85  5  5  66  16  6  Ta(NMe2)5  > 96  0  [Cl3Ta(NMePh)2]2  24  72  a  7 a  4 mol% catalyst, 90 °C  Gratifyingly, all of the mono(amidate) complexes were found to catalyze the transformation at 110 °C. This is in contrast to the homoleptic dimethylamido complex (Table 56  2.2, entry 6) where temperatures of 130 °C or greater are required for catalysis. The reactivity of the κ2(N,O) mono(amidate) complexes is directly correlated to the steric bulk at the 2,6 positions of the aryl-Namidate moiety. As the steric bulk is increased (H < Me < iPr), so too is the conversion to the desired product (Table 2.2, entries 1-3), with 3 achieving the highest conversion. Increasing the bulk at the 2,6 positions to tBu (5; Table 2.2, entry 5) results in a stark decrease in reactivity with 6 being formed in only 16% conversion. This observation is accompanied by a plethora of new signals in the 1H NMR spectrum, thought to be the result of metal complex decomposition. These observations suggest that κ2(N,O) bonding of mono(amidates) results in more robust complexes than κ1(O) systems. Interestingly, when the electronic nature of the substituent on the Camidate is altered (4; Table 2.2, entry 4), a significant drop in reactivity is observed. Even with extended reaction times, comparable conversions are not obtained.  A thermolysis study of 4 and the tBu  substituted 3 highlights the difference in stability. Heating toluene-d8 solutions containing varying concentrations of 3 to 130 °C for 36 hours (Scheme 2.19) results in two species present in a 4:1 ratio: 3 (major) and the corresponding tantalaziridine (minor). Signals attributable to liberated dimethylamine (HN(CH3)2, δH = 2.21) and the diastereotopic proton signals of a tantalaziridine (Ta–CH2, δH = 2.27 and 2.32) confirm this assignment (Figure 2.6). In fact, after this assignment, it was noted that small amounts of these diagnostic signals were present in spectra collected at room temperature, suggesting facile tantalaziridine formation. These results contrast the findings for chloro-bridged precatalysts as no liberated amine or tantalaziridine complex could be detected at temperatures as high as 150 °C.40 Unfortunately, this transient tantalaziridine could not be isolated even upon the addition of neutral donors such as pyridine or trimethylphosphine oxide.  57  Scheme 2.19 Thermolysis of precatalyst 3.  In contrast, 4 is observed to decompose at temperatures as low as 90 °C.  This  observation is in line with the solid-state data that suggested the amidate ligand is less strongly bound in the case of 4. While dominated by steric effects, the decreased reactivity of 4 also hints at an electronic component to the design of a robust catalyst for the α-alkylation of amines. Based on the results of this initial screen, 3 was identified as the most promising mono(amidate) tantalum precatalyst for the intermolecular α-alkylation of amines.  58  Figure 2.6 Section of the 1H NMR spectra (400 MHz, 25 °C) for the thermolysis of 3. The bottom spectrum was collected before heating; the top spectrum was collected after heating to 130 °C for 36 hours.  2.2.3  Synthesis and Characterization of Bis(amidate) Tantalum Complexes In addition to the series of mono(amidate) complexes, a selection of bis(amidate)  supported tantalum species was synthesized as possible precatalysts for the α-alkylation of amines. By increasing the amide proligand loading to two molar equivalents, a protonolysis reaction with Ta(NMe2)5, analogues to Scheme 2.17, delivers the bis(amidate) ligated species in excellent yields. However, when screened for catalytic activity in the α-alkylation of amines, the bis(amidate) species were found to be significantly less reactive than their mono(amidate) counterparts. This observation is supported by the thorough ligand screening study by Hartwig and Herzon that shows tantalum complexes supported by dianionic bis(amido) and bis(alkoxide) ligands are less reactive than Ta(NMe2)5.40 These data suggest that the coordination geometry of the metal complex greatly impacts reactivity. 59  2.2.3.1  Characterization of a κ1(O), κ2(N,O) Bis(amidate) Tantalum Complex Reacting the least sterically demanding amide proligand in Figure 2.4, N-  phenylpivalamide, with the pentakis(dimethylamido)tantalum precursor in hexanes yields complex 7 in excellent yields (Scheme 2.20).  Scheme 2.20 Protonolysis synthesis of bis(amidate) complex 7.  Complex 7 is purified by recrystallization in hexanes heated to reflux, yielding single crystals suitable for X-ray diffraction upon cooling. In the solid state, this complex exhibits a pseudo-trigonal bipyramidal coordination about Ta, with one amidate ligand coordinated κ1(O) and the other κ2(N,O) (Figure 2.7). Notably, the monodentate κ1(O) bound amidate fragment shows considerable C–N double bond character (Camidate–Namidate 1.273(2) Å) and an obtuse bond angle (Oamidate–Camidate–Namidate 125.81(15)°) compared to the chelating amidate ligand (1.300(2) Å and 112.89(15) respectively).  The amidate bite angle, as well as the various  dimethylamido contacts, are in agreement with the values previously observed for the mono(amidate) tantalum complexes 1 – 3.  60  Figure 2.7 ORTEP representation of the solid-state molecular structure of 7 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  In solution at ambient temperature, a species consistent with the solid-state coordination mode is observed by NMR spectroscopy. Two different amidate environments, the monodentate κ1(O) and bidentate κ2(N,O), are observed with carbonyl resonances at δC = 168.1 and 184.5 respectively. Two broad, overlapping methyl amido resonances are observed in both the 1H (δH = 3.42 and 2.84) and 13C{1H} (δC = 45.2 and 48.7) NMR spectra, consistent with two equatorial and an axial amido ligand slowly exchanging on the NMR time scale. At low temperature (T = -80.9 °C) these broad resonances start to sharpen and eventually 6 distinct (although broad) signals for each of the amido CH3-groups are observed. Figure 2.8 shows the aliphatic region (0.0 – 4.0 ppm) of the temperature dependent 1H NMR spectra of 7 from -80.9 to 101.3 °C which reveal several dynamic processes occurring in the Ta-coordination sphere.  The dynamic  processes observed in the cold regime, -80.9 – -6.5 °C, are totally reversible and unaffected by repeated cooling and warming of the sample within this range.  61  Figure 2.8 Variable temperature (-80.9 – 101.3 °C) 400 MHz 1H NMR study of 7.  At temperatures above 41.3 °C, the amido methyl signals coalesce suggesting fast exchange on the NMR time scale. Further upfield, two distinct amidate ligand tBu signals also coalesce at elevated temperatures (73.3 °C), indicative of a dynamic exchange between the κ1(O) and the κ2(N,O) binding modes (Scheme 2.21). Using a coalescence temperature of 62.6 °C and a maximum peak separation of 274.72 Hz from the slow exchange regime (-80.9 °C), the ΔG‡ for the exchange is estimated to be 9.32 ± 0.02 kcal/mol.  62  Scheme 2.21 Proposed equilibrium between coordination modes of 7 illustrating the dynamic exchange of amidate binding modes observed on the NMR time scale. Labeling of amidate N and O for illustrative purposes only.  At temperatures above 25.3 °C, a doublet corresponding to the methyl groups of neutral dimethylamine appears at δH = 2.21, presumably released due to tantalaziridine formation (A, Scheme 1.3). New signals at δH = 1.03 and 1.11 accompany the dimethylamine signal, suggestive of two additional tBu environments other than the broad signal attributed to 7 (δH = 1.27, T = 101.3 °C). Additional signals at δH = 2.16, 2.26 and 3.22, 3.35 and 3.38 can be assigned as the diastereotopic Ta–CH2 protons and new NCH3 environments generated during tantalaziridine formation. When the sample is kept at a constant temperature of 101.3 °C, numerous amido and tBu CH3 signals emerge that cannot be assigned to 7 or the corresponding tantalaziridine. These signals persist even after the sample has been cooled to ambient temperature, suggesting new species, such as mutli-metallic complexes, are being formed that are not in equilibrium with complex 7. The hemi-lability of amidate ligands is an important property, as it is one of the major differences between 3 and other group 5 catalysts for the α-alkylation of amines.  Ligand  participation in the mechanism cannot be excluded and may explain the vastly larger substrate 63  scope reported for precatalyst 3. Hemi-labile ligands, such as carboxylates, have been shown to actively participate in C–H activation mechanisms in a process known as concerted metalationdeprotonation (CMD).94 It is possible that the hemi-labile nitrogen of the amidate ligand acts as a proton shuttle for either of the protonation steps in the proposed mechanism, lowering the activation barrier. Quantifying the hemi-lability of amidate ligands as well as probing ligand participation in the mechanism would aid in the development of a new generation of catalysts. The fact that both κ1(O) and κ2(N,O) binding modes are observed in 7 suggests that tantalum complexes with more than one κ2(N,O) amidate ligand are disfavoured. However, the crystallographic characterization of a tris(amidate) species disproves this notion. Though 7 is formed as yellow prisms in excellent recrystallized yield, in one particular instance another minor red-orange product was also observed in the recrystallization vessel (Scheme 2.22). The smaller red-orange crystals were manually separated from 7 and found to be suitable for single crystal X-ray diffraction (Figure 2.9).  Scheme 2.22 Protonolysis synthesis of bis(amidate) product 7 and tris(amidate) product 8.  64  Figure 2.9 ORTEP representation of the solid-state molecular structure of 8 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  Interestingly, compound 8 was found to have 3 amidate ligands bound in a κ2(N,O) fashion to the central tantalum atom. In addition, instead of the two expected dimethylamido ligands, assuming a protonolysis reaction with 3 equivalents of amide proligand, a chelating N,N’-dimethylethane-1,2-diamido ligand is bound. The protonated proligand, DMEDA, is a useful organocatalyst95 but no crystallographically characterized tantalum complex supported by the dianionic ligand has been reported.  However, a related monoanionic N,N-chelating  (dimethylaminomethylene)methylamido ligand has previously been observed by Xue and coworkers91, 96-97 as well as the Schafer group.98 If the amidate ligand is once more assumed to occupy one coordination site, 8 can be described as having a pseudo-trigonal bipyramidal geometry, with one amidate ligand trans to the chelating diamido ligand and cis to the two other amidate ligands. The complex contains one mirror plane, along the plane formed by the two N atoms from the diamido ligand, and therefore can be described using the Cs point group.  As with the aforementioned mono(amidate)  complexes 1 – 3, the Ta–Namidate bond is longer than the corresponding Ta–Oamidate bond (2.248(2) Å and 2.192(2) Å respectively) for the amidate trans to the diamido ligand. 65  Interestingly, the other two amidate ligands display the opposite trend, with the Ta–Namidate contacts (2.183(2) Å and 2.190(2) Å) being shorter than their Ta–Oamidate counterparts (2.2197(19) Å and 2.2298(19) Å). A comparable disparity in bond lengths has been observed in other compounds containing multiple κ2(N,O) amidate ligands (vide infra). Unfortunately, solution phase data could not be gathered for 8, due to the inability to completely separate the two crystals in the bulk material. As a result, NMR spectra were dominated by the broad signals observed for 7. For similar reasons, elemental analysis was not attempted. Further attempts to synthesize 8 using 3 equivalents of pivalamide proligand were unsuccessful, even at elevated temperatures.  In all cases, multiple species and unreacted  proligand were present in the 1H NMR spectrum. Reacting 7 with an additional equivalent of proligand produced no reaction as assessed by 1H NMR spectroscopy.  2.2.3.2  Synthesis and Characterization of a Bis(amidate) Tantalaziridine In an effort to hinder the dynamic processes observed in 7 and synthesize a well-defined  bis(amidate) precatalyst, steric bulk can be added to the ortho positions of the amidate Nsubstituent. To this end, Ta(NMe2)5 and 2 equivalents of N-(2,6-dimethylphenyl)pivalamide are reacted to yield a product with magnetically inequivalent amidate ligand resonances (Scheme 2.23).  In this case, in addition to the 2 equivalents of dimethylamine formed during the  protonolysis reaction with the amide proligands, an extra equivalent of HNMe2 is spontaneously eliminated resulting in the formation of an isolable tantalaziridine. Presumably, the increase in steric bulk at the ligand induces the spontaneous room temperature elimination of neutral amine from the metal centre. The resulting complex 9 can be isolated in good yield (84%) with  66  analytically pure samples, suitable for single crystal X-ray analysis, obtained after recrystallization from hexanes.93  Scheme 2.23 Protonolysis synthesis of bis(amidate) complex 9.  The solid-state structure of 9 (Figure 2.10) reveals a C1 symmetric complex with both amidate ligands bound κ2(N,O) to Ta, akin to a tantalaziridine recently reported by Zi and coworkers.42 This is in contrast to the κ1(O), κ2(N,O) binding modes observed in complex 7 despite the added bulk of the N-aryl fragment in 9. Presumably, the decreased steric demand of the metallaziridine moiety, compared to two dimethylamido ligands, allows for both amidate ligands to be bound κ2(N,O).  One amidate ligand shares the equatorial plane with the  tantalaziridine fragment (N1,O1; 14.6° dihedral angle) and the nitrogen contact of the other κ2(N,O) amidate (N2).  The dimethylamido ligand and the remaining oxygen contact (O2)  occupy the axial plane (N3–Ta–O2 angle of 153.17(17)°) of the pseudo-pentagonal bipyramidal geometry. The Ta–N3 bond length of 1.970(5) Å is shorter than those found in compounds 1 – 4 (Table 2.1), perhaps due to the lack of steric hindrance from equatorial dimethylamido ligands. The axial O2 is more strongly bound to the tantalum atom than its equatorial counterpart with bond distances of 2.132(3) and 2.214(4) Å respectively. The longer equatorial Ta–O1 contact is accompanied by the shorter of the two Ta–Namidate bond lengths at 2.179(4) Å (versus 2.245(4) 67  Å), though both are significantly shorter than those of the mono(amidate) series (2.415(3) Å average). As with 8¸ the bonds lengths of the κ2(N,O) amidate contacts appear to depend on the relative location of the other ligands bound to tantalum. Interestingly the amidate bite-angles in the mono(amidate) complexes (56.9 – 57.7°) tend to be slightly more acute when compared to the amidates in the bis(amidate) series (58.3 – 59.3°), likely due to the pronounced spatial demand of the four dimethylamido fragments in the mono-amidate series.  Figure 2.10 ORTEP representation of the solid-state molecular structure of 9 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms and most of the carbon atoms of the ligands omitted for clarity.  Table 2.3 Selected bond lengths (Å) and angles (°) for complex 9. Ta–O1  2.214(4)  C38–N4  1.424(6)  Ta–O2  2.132(3)  O1–Ta–N1  59.35(14)  Ta–N1  2.179(4)  O2–Ta–N2  59.34(12)  Ta–N2  2.245(4)  N3–Ta–O2  153.17(17)  Ta–N3  1.970(5)  68  As discussed in Section 2.1.2, tantalaziridines have two limiting resonance structures: a dianionic, 4 electron donor ligand as an azametallacyclopropane bound to a Ta(V) d0 centre, and a neutral 2 electron donor η2-imine ligand with the d2 Ta(III) centre interacting with the π system of the C=N bond (Figure 2.2). Using metrical parameters, specifically the C–N bond distance, and spectroscopic data, 9 is found to contain an azametallacyclopropane moiety and therefore a Ta(V) d0 centre.  Notably, the cyclometallated fragment has a C38–N4 bond distance of  1.424(6) Å, consistent with a C–N single bond. Planar geometries about the nitrogen atom (∑(bond angles): 359.9(1)°) are also consistent with this assignment. Moreover, the signals corresponding to the diastereotopic protons and carbon atom of the tantalaziridine moiety resonate at H = 2.34 and 2.49 (2JH,H = 3.5 Hz) and C = 59.1 respectively. These values are in the typical range of known pentavalent group 5 metallaziridine complexes and are notably different from those observed for the related η2-imine complexes with average 13C NMR signals of greater than δC = 200.9 To investigate the effectiveness of bis-amidate complexes as precatalysts for the αalkylation of amines, complex 9 was screened in a similar manner to the aforementioned monoamidates (Table 2.2). Despite the fact that the proposed catalytically-relevant metallaziridine moiety is pre-installed, 9 displayed sluggish reactivity compared to 1 – 4, achieving just 71% conversion even with higher temperatures (130 °C) and longer reaction times (168 h). Though initially disappointing, the decreased reactivity of 9 offers the possibility of using it as a model system to access relevant catalytic intermediates, as attempts to isolate the tantalaziridine formed by the thermolysis of 3 were unsuccessful.  69  2.2.3.3  Reactivity of Tantalaziridine 9 with Unsaturated Molecules It has been well documented that tantalaziridines react with unsaturated fragments such  as alkenes, carbonyls, nitriles and isocyanides in a stoichiometric fashion and can yield the corresponding organic insertion products after work-up.9 Surprisingly, exposing 9 to various alkenes, internal alkynes and conjugated ketones does not result in any reactivity, even at temperatures of 90 °C.  In comparison, reaction with terminal alkynes, acetophenone,  benzaldehyde and phenylisocyanate yields complex mixtures of products.  However, when  complex 9 reacts with acetonitrile in benzene at room temperature, the C≡N bond inserts into the Ta–C bond of the metallaziridine creating a 5-membered metallacycle that is ultimately observed after a 1,3-hydride shift to generate 10 (Scheme 2.24).  This 1,3-hydride shift has been  previously observed for nitrile insertions into early transition metal-carbon bonds, both through intra- and intermolecular pathways.68, 99-102 Complex 10 is isolated as a deep-red powder that is recrystallized from refluxing hexanes to give single crystals suitable for X-ray diffraction (Figure 2.11). The solitary olefinic proton of the metallacycle (H = 5.86) is an informative spectroscopic handle. This species can be viewed as a model of the 5-membered metallacycle proposed in the catalytic cycle of the α-alkylation of amines. The acetonitrile inserts into the tantalaziridine with the methyl group pointing away from the metal center, minimizing steric interactions with the auxiliary ligands during the insertion step, consistent with the exclusive regioselectivity observed for branched products in the α-alkylation of amines.  70  Scheme 2.24 Insertion of acetonitrile into Ta–C bond of 9 to generate 10.  Figure 2.11 ORTEP representation of the solid-state molecular structure of 10 drawn at 50% probability for thermal ellipsoids. Most hydrogen atoms omitted for clarity.  Table 2.4 Selected bond lengths (Å) and angles (°) for complex 10. Ta–O1  2.1376(13)  Ta–N4  2.0623(16)  Ta–O2  2.2078(13)  Ta–N5  2.0428(16)  Ta–N1  2.2974(15)  C3–C4  1.345(3)  Ta–N2  2.2223(15)  O1–Ta–N1  58.56(5)  Ta–N3  1.9946(16)  O2–Ta–N2  58.47(5)  71  In the solid state, 10 adopts a distorted trigonal bipyramidal geometry with one amidate ligand in the axial plane and the other in the equatorial. The axial Ta–Namido contact is more in line with the mono(amidate) series at 1.9946(16) Å and the Ta–Oamidate bond lengths (2.1376(13) and 2.2078(13) Å) are shorter than the associated Ta–Namidate values (2.2974(15) and 2.2223(15) Å). By comparison, the nitrogens of the 5-membered metallacycle are more tightly bound to the central Ta atom, with bond distances of 2.0428(16) and 2.0623(16) Å. In agreement with the spectroscopic data, the C–C bond length of 1.345(3) Å for the metallacycle moiety suggests a double bond, consistent with a 1,3-hydride shift. Further proof comes from the summation of angles of 360.0(2)° about the methylated carbon in the ring, confirming a planar sp2 hybridization. Unfortunately, no bis(amido) intermediate model (Scheme 1.3, C) was isolated when 10 was exposed to amines. This is not surprising however as the use of nitriles as substrates for the α-alkylation of amines has neither been reported nor observed. However, the difficulty observed for inserting unsaturated molecules into the tantalaziridine Ta–C bond is consistent with the hypothesis alkene insertion is turnover limiting for the α-alkylation of amines catalyzed by tantalum amidate complexes (vide infra).  2.2.3.4  Attempted Synthesis of an Electronically Modified Bis(amidate) Tantalaziridine In order to elucidate how changes to the electronic structure of the ligand would affect  the resulting complexes, the tBu group in the amide proligand backbone was exchanged for a phenyl group. Interchanging groups in this position maintains the steric bulk of the N-aryl component while modifying the electronic nature of the amidate ligand. Though 3 and 4 were found to be structurally similar, the stark difference in reactivity suggests that there is an  72  electronic component to amidate precatalysts that deserves further investigation. Since the spontaneous, room temperature C–H activation to form 9 is thought to be driven by steric effects, the reaction outlined in Scheme 2.23 was repeated using 2 equivalents of N-(2,6dimethylphenyl)benzamide as the proligand (Scheme 2.25). Unlike 9, the resultant complex, 11, is not soluble in pentane or hexane and the 1H NMR spectrum shows multiple species in solution. Recrystallization from toluene gave yellow crystals suitable for X-ray diffraction (Figure 2.12).  Scheme 2.25 Protonolysis synthesis of bis(amidate) complex 11.  73  Figure 2.12 ORTEP representation of the solid-state molecular structure of 11 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  Table 2.5 Selected bond lengths (Å) and angles (°) for complex 11. Ta–O1  2.1518(17)  Ta–N5  1.993(2)  Ta–O2  2.0290(18)  C2–N2  1.276(3)  Ta–N1  2.330(2)  O1–Ta–N1  58.22(7)  Ta–N3  1.977(2)  O2–C2–N2  126.2(2)  Ta–N4  1.965(2)  Structurally, 11 resembles the κ1(O), κ2(N,O) geometry of 7 more than that of the tantalaziridine-containing 9. An equatorial amidate ligand binds in the expected κ2(N,O) fashion while the other amidate ligand is bound κ1(O) in the axial position. As with 7, the monodentate κ1(O) amidate fragment displays an obtuse bond angle (O2–C2–N2 = 126.2(2)°) and a C–N bond length suggestive of a double bond (C2–N2 = 1.276(3) Å). In fact, as all the angles defining the trigonal bipyramidal geometry are within 3° of each other, 11 can be considered isostructural to 7. 74  These species also portray similar solution phase behaviour, both having broad, illdefined 1H NMR signals.  However, spectroscopy reveals multiple species in solutions  containing 11, even with analytically pure samples. No variable temperature NMR study was carried out with 11, as the lack of diagnostic 1H NMR signals would provide no further insight into the hemi-labile nature of amidate ligands. This compound highlights the notion that small changes in the electronic components of the amide proligand can result in vastly different reactivity for the complex.  Probing the Mechanism of Mono(amidate) Tantalum Catalyzed α-Alkylation of  2.2.4 Amines  The proposed mechanism put forth by Nugent and coworkers (Scheme 1.3) for the αalkylation of amines is rooted in deuterium labeling studies that suggested the formation of a metallaziridine intermediate.8 This initial study led Hartwig and Herzon to investigate aspects of the proposed mechanism, including regioselectivity and metallacycle formation, using deuterium scrambling and 1H NMR studies.40 However, these experiments were qualitative in nature and did not provide a kinetic profile.  The first quantitative studies were performed for the  intramolecular α-alkylation of aminoalkenes catalyzed by Ti(NMe2)4 but could not be applied to intermolecular systems.103 Recently, Hultzsch and Reznichenko disclosed a thorough mechanistic investigation for Ta and Nb catalysts supported by silyl-substituted bis(naphtholate) deuterium labeling as well as kinetic studies.44  ligands by means of  The rate was found to have a first order  dependence on both catalyst and amine, with the order in alkene depending on steric bulk of the  75  substrate (zero order for 1-octene; first order for vinylcyclohexane). Based on these data, either alkene insertion or β-hydrogen abstraction has been proposed to be the turnover limiting step. In an effort to better elucidate the mechanism of the α-alkylation of amines catalyzed by tantalum mono(amidate) complexes, a series of stoichiometric and kinetic investigations were performed.  2.2.4.1  Byproduct Formation During the course of these investigations, pathways off the catalytic cycle that produce  byproducts 12 and 13 (Scheme 2.26) were identified. These byproducts arise from the αalkylation of dimethylamine liberated from the catalyst with 1-octene. Diagnostic signals cannot be distinguished in the 1H NMR spectrum due to overlap with 1-octene and product 6, making in situ monitoring difficult. However, 13 is produced in quantities large enough to be isolated by chromatography. As a result of this byproduct formation, quantitative monitoring of alkene consumption using 1H NMR spectroscopy as it relates to formation of product 6 is impossible, limiting the possibility of a full kinetic profile.  Scheme 2.26 Byproducts 12 and 13 formed from the α-alkylation of 1-octene and dimethylamine liberated from the catalyst.  Notably, when 3 is reacted with stoichiometric amounts of 1-octene (1 equivalent for each dimethylamido ligand) in benzene over 7 days at 110 °C, both the branched (12) and linear 76  (14) mono-alkylated byproducts are isolated after tosylation (Scheme 2.27). This is the first example of a linear product generated with an alkyl alkene substrate catalyzed by a group 5 complex. However, as with the aforementioned titanium systems, regioselectivity is an issue with 12 and 14 being formed in approximately equal amounts.  Interestingly, no signals  attributable to 13 were present in the GC/MS trace.  Scheme 2.27 Reaction of 3 with 1-ocetene to form both branched (12) and linear (14) monoalkylated α-alkylation of amines byproducts.  Though a quantitative mechanistic profile is hampered by the formation of these byproducts, valuable information can be gleaned from other mechanistic probes including isotopic labeling studies, reaction kinetics and theoretical modeling. The information gathered herein will be useful in designing further generations of tantalum-based catalysts for the αalkylation of amines.  2.2.4.2  Isotopic Labeling Studies Early research into transition-metal catalyzed hydroaminoalkylation has used  deuterium-labeled substrates to provide insight into reaction kinetics via isotope effects as well as identify side-reactions such as reversible tantalaziridine formation (Scheme 2.28, D) and ortho-metalation of an aryl substrate (E).8, 39 77  Scheme 2.28 Proposed mechanism for tantalum catalyzed α-alkylation of amines showing reactions off the catalytic pathway that have been previously studied using deuterium labeling.  To investigate amido ligand exchange in Ta-amidate species (Scheme 2.28, A–D), 3 was exposed to excess N-methylaniline-d3 in a Teflon-sealed NMR tube. Complex 3 eliminates detectable amounts of dimethylamine at room temperature over a 24 h period, suggesting that ligand exchange and/or tantalaziridine formation is facile for these systems, even at room temperature. Gentle heating to 65 °C results in substantial H-incorporation into the methyl group  78  of the labeled amine (Scheme 2.29), indicating reversible tantalaziridine formation at subcatalytic temperatures. In addition to the scrambling of isotopes at the methyl position, the integration of the signal corresponding to the ortho protons on the aniline decrease slightly, showing that ortho-metalation is a competing side reaction that also occurs at mild temperatures. However, even after extended reaction times, no deuterium incorporation into the ligand is observed.  To date, none of these complexes have shown catalytic reactivity at these low  temperatures.  Scheme 2.29 Incorporation of H into the labeled methyl group of N-methylaniline-d3.  To this end, N-methylaniline-d3 can be reacted with 1-octene in the presence of group 5 precatalysts to yield the deuterated product 6-d (Table 2.6). When Ta(NMe2)5 is used as the precatalyst, only 45% of the deuterium label remains α to N with 46% deuterium incorporation into the ortho positions.39  In contrast, when [Cl3Ta(NMePh)2]2 or silylated binaphtholate  niobium complexes are employed, >90% of the methylene group remains labeled.40,  44  This  suggests that reversible tantalaziridine formation and ortho-metalation are more facile in the homologous system whereas [Cl3Ta(NMePh)2]2 has a strong preference for alkene insertion (pathway B) over reversible tantalaziridine formation (pathway D). When the Ta amidate 1 is subjected to these reaction conditions, the deuterated product 6-d is isolated in excellent yield  79  (93%) with 60% of the deuterium label remaining in the methylene group α to N and 23% Dincorporation observed in the ortho positions.  Table 2.6 Comparison of deuterium incorporation or retention in 6-d for group 5 catalysts.  Entry  Catalyst  ortho  methylene  methyl  1  Ta(NMe2)5  45%  46%  12%  2  [Cl3Ta(NMePh)2]2  -  92%  -  3  (Si-binaphth)Nb(NMe2)3(HNMe2)  -  95%  32%  4  3  23%  60%  16%  Comparing the relative initial rates (up to 20% consumption of amine) of the reaction of 1-octene with N-methylaniline-d3 and its unlabeled congener offers a method of probing the βhydrogen abstraction step in the catalytic cycle (Scheme 2.28, C–A). With 10 mol% of 3 at 130 °C, no significant kinetic isotope effect (KIE) is observed (kH/kD = 1.4(1), Figure 2.13). However, scrambling of the N–H protons into the labeled methyl group via tantalaziridine formation could negate any kinetic isotope effect, replacing C–D with C–H bonds. In fact, the rate of hydrogen incorporation into the CD3 moiety was found to be comparable to the rate of catalysis at 110 °C, with no detectable scrambling into the ortho positions. In order to minimize loss of the deuterium label, N-methylaniline-d3 was deuterated at the N position to generate N-methylaniline-d4. This substrate design also allows the opportunity to investigate the protonolysis step of the mechanism (Scheme 2.28, B–C).  Subjecting N-  80  methylaniline-d4 to the aforementioned reaction conditions results in a kinetic isotope effect of 1.5(1) (Figure 2.13), within error of the result with N-methylaniline-d3. Therefore, no primary kinetic isotope effect is observed for the α-alkylation of amines catalyzed by 3. This further corroborates a mechanism where neither tantalaziridine formation by breaking the C–H bond α to N nor protonolysis of the 5-membered metallacycle by an amine substrate is the turnover limiting step. Interestingly, the silyl-substituted binaphtholate niobium system developed by Hultzsch and coworkers was found to have a primary kinetic isotope effect of 1.6(1) with N-deuterated Nmethylaniline.44 Assignment of a primary KIE is in line with the first order dependence on amine concentration found for the niobium catalyst. Notably, no kinetic isotope effect was found for N-methylaniline-d3 (kH/kD = 1.08(7)).  Figure 2.13 Consumption of amine as a function of time for the α-alkylation of N-methylaniline and is -d3 and -d4 deuterated analogues with 1-octene catalyzed by 3. Solid lines depict the leastsquares fits (R2 = 0.971, 0.982, 0.980) to the data points.  81  2.2.4.3  Reaction Kinetics In addition to the labeling experiments, a preliminary kinetic study using precatalyst 3, N-  methylaniline and 1-octene in toluene-d8 at 130 °C was performed in Teflon-cap sealed NMR tubes with monitoring by 1H NMR spectroscopy. As with the catalytic screening, the reaction was followed by integrating the ortho-proton signals of N-methylaniline relative to an internal standard (1,3,5-trimethoxybenzene). Unfortunately, heating the samples inside the spectrometer was impossible due to equipment limitations. Therefore, the NMR tubes were heated in an external oil bath and spectra were collected at hourly intervals as the catalyst was found to become less active with frequent heating and cooling cycles, presumably due to catalytically inactive multi-metallic species. Firstly, using pseudo first-order conditions (10-fold excess of 1-octene and 10 mol% catalyst loading), a linear dependence is observed over two half-lives (75% consumption) of Nmethylaniline before a sharp decrease in rate (Figure 2.14). This decrease in rate may be attributable to product inhibition and/or low substrate concentration at this stage of the reaction. This observation suggests a zero order dependence on amine concentration over 2 half-lives and is in good agreement with the kinetic isotope experiments.  82  Figure 2.14 Consumption of amine as a function of time for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Solid line depicts the least-squares fit (R2 = 0.994) to the data points.  Based on the zero-order behaviour of the amine substrate, an experiment was designed to probe the early stage of the α-alkylation of amines at intervals less than one hour. To this end, 5 identical reaction mixtures were loaded into Teflon capped NMR tubes that were placed in a preheated 130 °C oil bath. Each quarter-hour, one tube would be removed and the 1H NMR spectrum collected. This tube would not be returned to the oil bath, therefore avoiding the decrease in catalytic performance from repeated heating and cooling. The resultant zero-order plot (Figure 2.15) shows the expected linear behaviour, suggesting no induction period.  83  Figure 2.15 Consumption of amine as a function of time for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Solid line depicts the least-squares fit (R2 = 0.991) to the data points.  Next, the reaction was carried out with various catalyst loadings (1, 1.5, 2, 2.5, 3, 4, 5 and 7.5 mol%, 4.3 – 32.2 mmol L-1), using a 1:1.5 ratio of N-methylaniline to 1-octene in a constant volume of toluene-d8. Figure 2.16 shows linear, zero-order depletion of N-methylaniline over the concentration range of the catalyst 3. However, when the observed initial rates (up to 30% consumption of N-methylaniline) derived from the linear models are plotted as a function of catalyst concentration, two different regimes are observed over the concentration range (Figure 2.17).  84  Figure 2.16 Consumption of amine as a function of time for the α-alkylation of N-methylaniline with 1-octene catalyzed by varying concentrations of 3. Solid lines depict the least-squares fit to the data points.  Figure 2.17 Observed initial rates of consumption of N-methylaniline as a function of catalyst concentration of 3 (ccatalyst). Solid line depicts the least-squares fit (R2 = 0.947) to the data points.  85  Analysis of the data obtained with up to 5 mol% catalyst loading revealed a linear correlation confirming the anticipated first order behaviour.44 However, increasing the loading above this threshold, up to 20 mol% (85.9 mmol L-1), does not increase the rate of reaction, indicative of the formation of dormant states and/or unfavourable equilibria off the catalytic cycle, such as multi-metallic species, as recently proposed for catalytically active homoleptic titanium catalyst Ti(NMe2)4.103 With 3, it is proposed that these catalytically inactive species are formed at catalyst loadings above 5 mol%.  To confirm this hypothesis, a series of catalytic reactions was  performed with the relative molar ratios of N-methylaniline, 1-octene and 3 (1:1.5:0.05) held constant while the absolute concentration of these species in 0.5 mL of toluene-d8 was varied. This was achieved via 4 serial dilutions of a 1 mL standard solution that contained 2 mmol Nmethylaniline (2.0 mol L-1), 3 mmol 1-octene (3.0 mol L-1), and 0.1 mmol 3 (0.1 mol L-1). Each successive dilution resulted in a reaction mixture half as concentrated as the previous one, yielding, for example, initial amine concentrations of 2, 1, 0.5, 0.25 and 0.125 mol L-1. Interestingly, the initial rates (up to 30%) of N-methylaniline consumption are found to be indistinguishable for all concentrations measured, within two standard deviations, (Table 2.7). This suggests that increases in catalyst concentration relative to the substrates, rather than the absolute concentration of the species present in the reaction mixture, are the determining factor for the observed saturation kinetics.  86  Table 2.7 Comparison of the initial rates of consumption of N-methylaniline to the absolute initial concentration of species present in the reaction mixture. Initial concentrations of Nmethylaniline are given as representative examples. Initial N-methylaniline Concentration (mol L-1)  Normalized Consumption of N-methylaniline at 8 h  Rate of N-methylaniline Consumption (mol L-1 h-1)  0.125  0.69  0.039  0.250  0.66  0.042  0.500  0.69  0.039  1.000  0.71  0.037  2.000  0.69  0.038  Standard Deviation  0.02  0.002  It should be noted that monitoring the catalyst during the reaction is non-trivial and attempts to isolate tantalum species after completion have been unsuccessful. However, further support for inactive tantalum species was found in the mass spectrum of a completed reaction mixture initially containing 10 mol% 3. In addition to the expected organic products, a number of signals between 700 and 1000 m/z were present in the spectrum. These large masses are suggestive of uncharacterized tantalum-based mono or dimeric species formed during the reaction. Notably, no signals attributable to 3 were present. Regrettably, pseudo-first order conditions could not be employed for elucidation of the alkene reaction order. The large excess of N-methylaniline causes an overlap in substrate and product signals (δH = 6.40 and 6.48 respectively) making quantitative analysis impossible. Alternatively, monitoring of alkene consumption is complicated by the formation of byproducts 12 and 13 and thus unreliable. Therefore, in order to probe reaction order in alkene, a series of experiments was conducted in which the initial concentration of alkene was varied while catalyst loading, amine concentration and reaction volume was held constant. Under these conditions, a 87  positive, linear relationship between both the initial rate of consumption of N-methylaniline and the initial rate of formation of 6 is observed (Figure 2.18).  Though byproduct formation  continues to plague the quantitative analysis of this data, as noted by the different rates of substrate consumption and product formation, it can be stated that the rate of the α-alkylation of amines catalyzed by mono(amidate) 3 is dependent on the initial concentration of alkene.  Figure 2.18 Observed initial rates of consumption of N-methylaniline and initial rates of formation of 6 as a function of initial alkene concentration. Solid and dashed lines depict the least-squares fits (R2 = 0.985 and 0.993 respectively) to the data points.  These combined preliminary kinetic data suggest an empirical rate law at low catalyst loadings that is zero order in amine, first order in catalyst and has a non-zero dependence on alkene concentration.  This assertion is in good agreement with the observed difficulty of  insertion of unsaturated molecules into the Ta–C tantalaziridine bond of 9 as well as the lack of primary kinetic isotope effects observed with 3 for both protonolysis and β-hydrogen abstraction  88  steps. Isotopic labeling experiments have also demonstrated that the turnover limiting step does not involve tantalaziridine formation, consistent with the proposed rate law.  2.2.5  Probing a One Electron Mechanism for the α-Alkylation of Amines The mechanism proposed in Scheme 1.3 is comprised of steps involving the movement of  two electrons at a time and therefore describes an ionic mechanism. To date, the possibility of a mechanism for the α-alkylation of amines involving one electron, or radical, processes has not been explored in the literature. However, the redox couple between d0 Ta(V) and d1 Ta(IV) has been found to occur between 0.31 and 0.74 V in traditional organic solvents using electrochemical methods.104-105 Synthetically, Ta(V) centres can be reduced to d1 species with reagents as innocuous as n-butyllithium.106 Based on this ease of reduction, the possibility of radical participation from a Ta(IV) centre during catalysis is plausible and should be considered. Though the absence of ambient light, which can initiate radical processes, does not affect the rate of the α-alkylation of amines catalyzed by mono(amidate) tantalum complexes, further experimental evidence for radical behaviour was pursued. Traditional radical traps such as TEMPO ((2,2,6,6-tetramethylpiperidin-1-yl)oxyl) were discounted because of the oxophilicity observed for early transition metals such as tantalum. The aim is to observe radical behaviour without inducing one electron processes with additives. To this end, a cyclopropyl moiety was built into a pair of mono(amidate) precatalysts as well as an alkene substrate in order to probe the possibility of one electron processes in the α-alkylation of amines. The radically induced ring opening of cyclopropyl rings (Scheme 2.30) has been used extensively to study radical mechanisms in a variety of systems.107  89  Scheme 2.30 Radically induced ring opening of cyclopropyl ring.  2.2.5.1  Cyclopropyl-based Mono(amidate) Precatalysts The modular synthesis of amidate proligand offers two different sites for installation of  functional groups: the substituent bound to the nitrogen or to the carbonyl carbon. Based on the assumption that the unpaired electron density from the Ta(IV) d1 centre could be delocalized across the 4-membered metallacycle formed during κ2(N,O) amidate binding, the cyclopropyl moiety was installed at the carbonyl carbon. Though nitrogen-containing ring-opening radical probes exist,107 systems that are exclusively carbon based are more versatile and were selected as a starting point.108 Based on the success of both 2 and 3 as precatalysts for the α-alkylation of amines, bulk at the 2,6 positions of the N-aryl substituent was maintained at dimethyl and diisopropyl.  Using the commercially available cyclopropanecarbonyl chloride and the  appropriate 2,6-disubstituted aniline, the cyclopropyl-based amide proligands were synthesized in excellent yields as outlined in Scheme 2.31.  Scheme 2.31 Synthesis of cyclopropyl-based amide proligands.  90  The corresponding complexes, 15 and 16 for the dimethyl and diisopropyl species respectively, were synthesized via protonolysis in the same fashion as the aforementioned mono(amidate) species (Scheme 2.17). Recrystallization from refluxing hexanes yielded yellow prisms in good to excellent yields (91% for 15, 84% for 16). X-ray diffraction studies (Figure 2.19) revealed pseudo-trigonal bipyramidal geometries isostructural to those of complexes 1 – 4. Interestingly, while the Ta–Oamidate bonds lengths are identical, within error, the less sterically demanding 15 has a slightly longer Ta–Namidate contact than 16 (2.4107(14) and 2.3894(10) Å respectively). As with 1 – 4, the axial amido bonds are longer than their equatorial counterparts. For both 15 and 16, the cyclopropyl ring is facing away from the N-aryl substituent, minimizing steric interactions.  Figure 2.19 ORTEP representations of the solid-state molecular structures of 15 (left) and 16 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  Both cyclopropyl-based complexes are competent precatalysts for the α-alkylation of amines. When 5 mol% of catalyst is employed for the α-alkylation of N-methylaniline with 1octene in toluene-d8 at 130 °C for 24 hours, both 15 and 16 achieve complete conversion to the  91  expected product, 6. These results are in good agreement with the catalytic activity of tBu-based analogues 2 and 3.  2.2.5.2  Substrate Containing an Ultrafast Radical Probe Since the data obtained using precatalysts containing a radical indicator was not  conclusive, a cyclopropyl based substrate was synthesized. The substrate, 17, is based on a radical clock designed by Baines and coworkers as an ultrafast mechanistic probe.109 Ring opening is observed when unpaired electron density is localized on the terminal sp2 carbon to give a linear chain capped by the phenyl group. The resultant radical is located α to the phenyl group which stabilizes the carbon-centred radical and acts as a driving force for the ring-opening event.110-111 Assuming a stepwise radical process for the insertion of the alkene into the Ta–C bond of the tantalaziridine, instead of the expected 5-membered metallacycle, one of two 8membered rings will be formed (Scheme 2.32). These intermediates will ultimately generate a linear α-alkylation product that can be spectroscopically distinguished from the two electron product containing the intact cyclopropyl ring. However, a concerted radical insertion would be indistinguishable from the two electron process, both generating a 5-membered metallacycle.  92  Scheme 2.32 Formation of 8-membered metallacycles from the stepwise radical insertion of a cyclopropyl-based alkene into a tantalaziridine.  The substrate was synthesized from commercially available starting materials using traditional methods (Scheme 2.33).112 Initially, trans-2-phenylcyclopropanecarboxylic acid was reduced using lithium aluminum hydride to afford the corresponding alcohol. The terminal alcohol was oxidize to an aldehyde, via the Swern oxidation, and subsequently reacted with methyl triphenylphosphonium bromide under Wittig conditions to yield the 17 as a yellow oil. After purification by column chromatography, the resultant clear oil was diluted with toluene-d8 and dried over molecular sieves for 24 hours. This solution was then degassed with three freeze/pump/thaw cycles before being brought into the glove box.  Scheme 2.33 Synthesis of 17 as a substrate for the α-alkylation of amines.  93  Following purification, 17 was used for a series of α-alkylations using N-methylaniline as the amine substrate (Scheme 2.34). A variety of tantalum amidate precatalysts was used (2, 3, 9, 15) in order to confirm consistency of mechanism across a variety of catalysts. In addition, a control experiment without any catalyst was run concurrently to assess the behaviour of the substrates at catalytic temperatures.  After 24 hours at 130 °C in toluene-d8, complete  consumption of 17 is observed for the reactions catalyzed by 2, 3, and 15. In line with previous observations, 9 is a less active catalyst with only 40% consumption observed. Heating to 130 °C has no effect on the substrates in the control experiment based on 1H NMR spectroscopy.  Scheme 2.34 α-Alkylation of N-methylaniline with cyclopropyl-based alkene substrate 17.  Based on the 1H NMR spectra recorded after 24 hours, there are no protons associated with alkene functional groups present in solution. Whereas the 17 has 3 distinct multiplets (δH = 4.88, 5.02, 5.35) assignable to alkene protons, there are no signals in the region between 3.5 and 6.0 ppm after 24 hours (Figure 2.20). In addition, signals attributable to the protons on the cyclopropyl ring of 17 (δH = 0.86, 0.96, 1.53, 1.70) are present after the alkene substrate is consumed. The four distinct signals are shifted slightly upfield to δH = 0.62, 0.72, 1.37, and 1.53 respectively and retain the diagnostic coupling patterns associated with cyclopropyl moieties.  94  Figure 2.20 1H NMR spectra (300 MHz, 25 °C) of the α-alkylation of N-methylaniline with cyclopropyl-based alkene 17 catalyzed by precatalyst 3. Bottom spectrum corresponds to time zero; top spectrum corresponds to heating at 130 °C for 24 hours.  It is possible that the linear α-alkylation product formed through radical coupling could itself act as an alkene substrate, eliminating any alkene proton signals in the above spectrum. However, no signal attributable to an increased mass was found in a GC/MS trace of the completed reaction mixture. This result suggests that the α-alkylation of amines catalyzed by tantalum amidate complexes is not going through a step-wise, one electron mechanism.  2.3  Conclusions A series of mono(amidate) tetrakis(dimethylamido) tantalum complexes with varying  degrees of steric bulk at the 2 and 6 positions of the N-phenyl substituent of the amidate ligand  95  have been synthesized and characterized both in solution and the solid state. Complexes 1 – 4 crystalize in pseudo-trigonal bipyramidal geometries, with the amidate ligand bound in a κ2(N,O) fashion. While the equatorial dimethylamido ligands are largely unaffected by the increasing bulk (from H in 1 to iPr for 3), the Ta–Oamidate and Ta–Namidate bond lengths systematically decrease and increase respectively. Substituting the tBu group (3) for a Ph moiety (4) in the amidate backbone results in a lengthening of the Ta–Oamidate bond but has little effect on other metrical parameters. Increasing the bulk to tBu in the case of 5 results in a change in amidate binding mode to κ1(O). The series of mono(amidate) complexes were screened as precatalysts for the αalkylation of amines. All species were capable of affecting the transformation at 110 °C, with 3 achieving the best conversion in the shortest amount of time. A thermolysis study of 3 revealed the formation of a tantalaziridine at catalytic temperatures. The catalysts with κ2(N,O) bound amidate ligands performed according to steric bulk, with an increase in bulk corresponding to higher reactivity. The κ1(O) bound 5 was found to be the least effective catalyst, suggesting κ2(N,O) binding is required. A significant decrease in reactivity was observed for 4, possibly due to instability at high temperatures. In collaborative work, Dr. Patrick Eisenberger demonstrated that 3 was compatible with a wide variety of alkene and amine substrates (including internal alkenes, dienes, silyl-protected alcohols, piperidine, and para-methoxy-N-methylaniline, a primary amine precursor) with exclusive regioselectivity and excellent diastereoselectivity.92-93  Building upon these  observations of remarkable selectivities, Dr. Rashidat Ayinla developed a biphenyl-based bis(amidate) tantalum complex for the enantioselective α-alkylation of amines. This complex is the first example of an enantioselective precatalyst and is capable of α-alkylating N-aryl amines  96  with terminal alkenes and norbornene in good to excellent yields with moderate enantioselectivities.93 In addition to the mono(amidate) species, a number of bis(amidate) tantalum complexes were also synthesized. While these complexes were not as effective as precatalysts for the αalkylation of amines, they presented interesting venues for investigating the fundamental properties of amidate ligands or as model systems for exploring catalytic intermediates. For instance, the unique κ1(O), κ2(N,O) binding arrangement found in 7 offered the opportunity to study the hemi-lability of amidate ligands. Using variable temperature 1H NMR spectroscopy, the exchange of amidate binding modes as well as tantalaziridine formation was explored. The notion of hemi-lability for amidate ligands also opens the possibility of a concerted metalationdeprotonation mechanism, an avenue yet to be explored for the α-alkylation of amines. Increasing the steric bulk of the amidate ligands results in the spontaneous formation of a tantalaziridine, previously only observed as minor products in solutions of the aforementioned compounds.  9 is a stable bis(amidate) tantalaziridine generated via C–H activation of a  dimethylamido ligand at room temperature. Based on solution phase and solid state data, 9 is a d0 Ta(V) complex containing an azametallacyclopropane moiety. The sluggish reactivity of 9 hinted at the ability to isolate other intermediates further along the catalytic cycle. Though alkenes were found to be unreactive, a model 5-membered metallacycle, 10, can be produced by reacting 9 with acetonitrile. The difficulty of inserting small molecules into the Ta–C bond suggests that insertion may be the turnover limiting step in the catalytic cycle. In an effort to deepen the understanding of the mechanism for the α-alkylation of amines, a series of stoichiometric and kinetic investigations were performed. Non-productive off-cycle pathways were identified using deuterated N-methylaniline. Isotopic labeling studies showed  97  tantalaziridine formation to be rapid and reversible, suggesting C–H activation is not the turnover limiting step. Kinetic investigations showed the reaction to be zero order in amine, supported by the lack of a primary isotope effect. The reaction was found to follow first order behaviour at low catalyst loadings before experiencing saturation kinetics possible due to multi-metallic species. Though byproduct formation prohibits a quantitative reaction order for alkene, an increase in initial alkene concentration corresponds to an increase in initial rate. Together, these data suggest that alkene insertion is the turnover limiting step for the α-alkylation of amines catalyzed by 3. In an effort to probe the possibility of a radical mechanism for the α-alkylation of amines, amide proligands with cyclopropyl moieties were synthesized. The resulting complexes, 15 and 16, displayed comparable catalytic reactivity to their tBu counterparts, supporting the notion of a two electron mechanism.  However, the information available from these experiments was  limited, leading to the design of a cyclopropyl-based ligand, 17. The product of the α-alkylation of N-methylaniline with 17 catalyzed by a variety of tantalum amidate catalysts suggests an ionic mechanism, eliminating the possibility of a radical stepwise insertion.  98  2.4 2.4.1  Experimental Materials and Methods All preparative scale reactions were conducted in oven dried (160 °C) glassware with  magnetic stirring using Schlenk-line techniques or a glove box under an atmosphere of dry dinitrogen. Experiments on NMR tube scale were carried out in Teflon cap sealed NMR tubes (5 mm). Toluene, benzene, hexanes, pentane, dichloromethane (DCM), and tetrahydrofuran (THF) were purified by passage over an activated aluminum oxide column and degassed prior to use. benzene-d6 and toluene-d8 were dried over 4 Å molecular sieves and degassed by 3 freezepump-thaw cycles.  Solvents for chromatography were used as received from commercial  sources and were at least of ACS reagent grade. Silica gel G60 (70 – 230 mesh) and F60 (230 – 400 mesh) was purchased from Silicycle. Reagents for amide synthesis were used as received from Aldrich without further purification. The amides were synthesized from the corresponding commercial acid chlorides and amines and rigorously dried by heating to 80 °C under vacuum unless otherwise noted. All commercial amines and olefins for catalytic reactions were distilled under reduced pressure from CaH2 and degassed by 3 freeze-pump-thaw cycles or sublimed in the case of solids. Ta(NMe2)5 was purchased from Strem and used as received. The spectral data for 6 and 17 are consistent with literature values.39, 112 NMR spectra were recorded on Bruker Avance 300 (1H: 300, 400 (1H: 400,  13  C: 100) or Bruker Avance 600 (1H: 600  13  13  C: 75), Bruker Avance  C: 150) instruments operating at the  denoted spectrometer frequency given in mega Hertz (MHz) for the specified nucleus. The samples were measured as solutions in the stated solvent at ambient temperature in non-spinning mode if not mentioned otherwise. To specify the signal multiplicity, the following abbreviations  99  are used: s = singlet, d = doublet, t = triplet, q = quartet, qu = quintet, sept = septet, oct = octet, and m = multiplet; br. indicates a broad resonance. Shifts δ are reported in parts per million (ppm) relative to tetramethylsilane (TMS) as an external standard for 1H- and 13C NMR spectra and calibrated against the solvent residual peak.113 Coupling constants J are given in Hertz (Hz). GC-MS measurements were performed on an Agilent Technologies GC 6890N/ MS 5973N equipped with an Agilent Technologies HP-5HS column (length: 30 m, 0.25 mm inner diameter, 0.25 μm coating thickness) coupled to a quadrupole mass filter. Helium was used as the carrier gas with a constant flow of 1.2 mL/min. Mass-spectra (MS) and elemental analyses (EA) were measured by the mass spectrometry and microanalysis service at University of British Columbia, UBC Vancouver. Mass spectra were measured on a Kratos MS-50. Fragment signals are given in mass per charge number (m/z). Elemental analyses were performed on a Carlo Erba Elemental Analyzer EA 1108. The content of the specified element is expressed in percent (%). Single Xray structure determinations were performed at the Department of Chemistry, University of British Columbia by Mr. Jacky Yim, Mr. Neal Yonson or Mr. Scott Ryken using either a Bruker APEX or APEX DUO with a molybdenum radiation source (MoKα, λ = 0.71073 Å) under a continuous flow of nitrogen (T = 100(2) °C).  2.4.2  Synthesis and Characterization of Tantalum Amidate Complexes  (N-Phenylpivalamidate)tetrakis(dimethylamido)tantalum (1).  N-  Phenylpivalamide (0.254 g, 2.00 mmol) and Ta(NMe2)5 (0.810 g, 2.02 mmol) were suspended in hexanes (5 mL) and the mixture was allowed to stir overnight (15 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal  100  amount of hot hexanes. The yellow crystals were collected and identified as the title compound. Yield: 0.999 g (1.87 mmol, 93%).  1  H NMR (benzene-d6, 400 MHz) δ 1.12 (s, 9H, C(CH3)3),  3.40 (br. s, 24H, N(CH3)2), 6.89 – 6.93 (m, 1H, CHarom.), 6.99 – 7.02 (m, 2H, CHarom.), 7.09 – 7.13 (m, 2H, CHarom.); 13C{1H} NMR (benzene-d6, 100 MHz) δ 29.7 (C(CH3)3), 42.1 (C(CH3)3), 46.9 (br. s, N(CH3)2), 124.3, 126.2, 128.9 (CHarom.), 147.0 (Carom.N), 180.4 (C=O). MS(EI): m/z 533 (M+), 489 (M+ – NMe2); Anal. Calcd. (%) for C19H38N5OTa: C, 42.78; H, 7.18; N, 13.13; Found: C, 42.86; H, 7.16; N, 13.01. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  (N-[2,6Dimethylphenyl]pivalamidate)tetrakis(dimethylamido)tantalum (2). N-(2,6-Dimethylphenyl)pivalamide  (0.410 g,  2.00 mmol)  and  Ta(NMe2)5 (0.809 g, 2.01 mmol) were suspended in hexanes (5 mL) and the mixture was allowed to stir overnight (15 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal amount hot hexanes. The yellow crystals were collected and identified as the title compound. Yield: 1.064 g (1.90 mmol, 95%).  1  H NMR (benzene-d6, 300 MHz) δ 1.03 (s, 9H, C(CH3)3), 2.24 (s, 6H,  CCH3), 3.25 (br. s, 24H, N(CH3)2), 6.85 – 6.94 (m, 3H, CHarom.);  13  C{1H} NMR (benzene-d6,  75 MHz) δ 17.9 (CCH3), 28.2 (C(CH3)3), 41.4 (C(CH3)2), 46.5 (br. s, N(CH3)2), 124.0, 127.6 (CHarom.), 131.7 (CCH3), 143.9 (CN), 177.3 (C=O). MS(EI): m/z 517 (M+ – NMe2); Anal. Calcd. (%) for C21H42N5OTa: C, 44.92; H, 7.54; N, 12.47; Found: C, 43.21; H, 7.38; N, 13.39. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  101  (N-[2,6Diisopropylphenyl]pivalamidate)tetrakis(dimethylamido)tantalum (3).  N-(2,6-Diisopropylphenyl)pivalamide (0.785 g, 3.01 mmol) and  Ta(NMe2)5 (1.208 g, 3.01 mmol) were suspended in hexanes (10 mL) and the mixture was allowed to stir overnight (24 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal amount of hot hexanes. The yellow crystals were collected and identified as the title compound. Yield: 1.698 g (2.75 mmol, 91%). 3  1  H NMR (benzene-d6, 400 MHz) δ 1.10 (s, 9H, C(CH3)3), 1.28 (d,  JH,H = 6.7 Hz, 6H, CH(CH3)2), 1.36 (d, 3JH,H = 7.1 Hz, 6H, CH(CH3)2), 3.31 (br. s, 24H,  N(CH3)2), 3.51 (sept, 3JH,H = 6.7 Hz, 2H, CH(CH3)2), 7.05 (m, 3H, CHarom.);  13  C{1H} NMR  (benzene-d6, 100 MHz) δ 24.9, 26.6 (CH(CH3)2), 27.3 (CH(CH3)2), 29.7 (C(CH3)3), 42.4 (C(CH3)3), 47.7 (br. s, N(CH3)2), 123.9, 125.8 (CHarom.), 141.3 (CN), 143.0 (CCH), 177.7 (C=O). MS(EI): m/z 573 (M+ – NMe2); Anal. Calcd. for C25H50N5OTa: C, 48.61; H, 8.16; N, 11.34; Found: C, 47.75; H, 8.05; N, 9.98. Reliable elemental analysis could not be obtained for this compound.  It is suspected this is due to partial tantalaziridine formation during sample  preparation. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  (N-[2,6Diisopropylphenyl]benzamidate)tetrakis(dimethylamido)tantalum (4).  N-(2,6-Diisopropylphenyl)benzamide (0.563 g, 2.00 mmol) and  Ta(NMe2)5 (0.809 g, 2.01 mmol) were suspended in hexanes (5 mL) and the mixture was allowed to stir overnight (26 h) to give a yellow suspension. All volatiles were  102  removed in vacuo and the remaining solid was recrystallized from the minimal amount of hot hexanes. The yellow crystals were collected and identified as the title compound. Yield: 1.196 g (1.88 mmol, 93%). 1H NMR (benzene-d6, 400 MHz) δ 1.06 (d, 3JH,H = 6.8 Hz, 6H, CH(CH3)2), 1.40 (d, 3JH,H = 6.8 Hz, 6H, CH(CH3)2), 3.31 (br. s, 24H, N(CH3)2), 3.65 (sept, 3JH,H = 6.8 Hz, 2H, CH(CH3)2), 6.94 – 7.00 (m, 3H, CHarom.), 7.26 (m, 3H, CHarom.), 7.68 (d, 3JH,H = 7.6 Hz, 2H, CHarom.); 13C{1H} NMR (benzene-d6, 100 MHz) δ 24.7, 24.8 (CH(CH3)2), 27.3 (CH(CH3)2), 47.0 (br. s, N(CH3)2), 124.2, 125.8, 127.9, 130.0, 130.7 (CHarom.), 133.8 (CC(O)N), 141.4 (CN), 142.4 (CCH), 167.3 (C=O).  MS(EI): m/z 637 (M+), 593 (M+ – NMe2); Anal. Calcd. (%) for  C27H46N5OTa: C, 50.86; H, 7.27; N, 10.98; Found: C, 50.97; H, 7.16; N, 10.62. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  N-(2,4,6-tri-tert-butylphenyl)benzamide.  2,4,6-Tri-tert-butylaniline  (1.00 g, 3.8 mmol) was dissolved in benzoyl chloride (8 mL, 68.9 mmol) and pyridine (0.62 mL, 7.6 mmol).  The mixture was  heated to 100°C and allowed to stir overnight (18h). The resulting solution was diluted with hexanes and filtered. The filtered solid was dried under high vacuum. The filtrate was washed sequentially with 1M NaOH, 1M HCl, water and saturated NaCl solution. The organic layer was separated from the aqueous layer and the solvent was removed under vacuum. Yield: 1.21 g (3.3 mmol, 87%).  1  H NMR (chloroform-d, 300 MHz) δ 1.35 (s, 9H, C(CH3)3), 1.42 (s, 18H,  C(CH3)3), 7.45 (br. s, 1H, NH), 7.47 (s, 2H, CHarom), 7.56 (m, 3H, CHarom.), 7.94 (dd, 2H, CHarom.);  13  C{1H} NMR (chloroform-d, 100 MHz) δ 31.4, 32.0 (C(CH3)3), 36.4 (C(CH3)3),  123.3, 127.2 (CHarom.), 130.0 (Carom.N), 131.6 (Carom.CO), 135.2, 148.2 (Carom.C(CH3)3), 167.4  103  (C=O). MS(EI): m/z 365 (M+); Anal. Calcd. (%) for C25H35NO: C, 82.14; H, 9.65; N, 3.83; Found: C, 81.88; H, 9.63; N, 3.98.  (N-[2,4,6-tri-tertbutylphenyl]benzamidate)tetrakis(dimethylamido)tantalum (5).  N-(2,4,6-Tri-tert-butylphenyl)benzamide  (0.228 g,  0.62 mmol) and Ta(NMe2)5 (0.250 g, 0.62 mmol) were suspended in dichloromethane (5 mL) and the mixture was heated until a clear yellow solution was present. The solution was allowed to stir for 30 minutes before all volatiles were removed in vacuo and the remaining solid was recrystallized from pentane at –30 °C.  The yellow crystals were  collected and identified as the title compound. Yield: 0.102 g (0.14 mmol, 23%).  1  H NMR  (benzene-d6, 600 MHz) δ 1.10 (s, 9H, C(CH3)3), 1.32 (s, 18H, C(CH3)3), 2.62 (br. s, 24H, N(CH3)2), 6.78 (t, 1H, CHarom), 6.81 (s, 2H, CHarom.), 6.89 (t, 2H, CHarom.), 7.69 (br. d, 2H, CHarom.);  13  C{1H} NMR (benzene-d6, 150 MHz) δ 32.3, 32.5 (C(CH3)3), 35.0, 36.9 (C(CH3)3),  45.7 (N(CH3)2), 121.7, 128.5, 128.7 (CHarom.), 128.9 (Carom.CO), 129.1 (CHarom.), 139.9, 142.3 (Carom.C(CH3)3), 147.4 (Carom.N), 160.3 (C=O). MS(ESI): m/z 720.8 (M+ – H); Anal. Calcd. (%) for C33H58N5OTa: C, 54.91; H, 8.10; N, 9.70; Found: C, 55.06; H, 8.01; N, 9.78. Single crystal X-ray quality samples were obtained by recrystallization from pentane at –30 °C.  Bis(N-phenylpivalamidate)tris(dimethylamido)tantalum (7). NPhenylpivalamide (0.222 g, 1.25 mmol) and Ta(NMe2)5 (0.25 g, 0.625 mmol) were suspended in hexanes (5 mL) and the mixture was allowed to stir overnight (15 h) to give a clear yellow solution.  104  All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal amount of hot hexanes. The yellow crystals were collected and identified as the title compound. Yield: 0.399g (0.60 mmol, 96%). 1H NMR (benzene-d6, 600 MHz) δ 1.06 (br. s, 9H, C(CH3)3), 1.66 (br. s, 9H, C(CH3)3), 2.84 (br. s, 6H, N(CH3)2), 3.42 (br. s, 12H, N(CH3)2), 6.83 (m, 1H, CHarom.), 6.90 (m, 1H, CHarom.), 7.00 (m, 2H, CHarom.), 7.08 (m, 4H, CHarom.), 7.12 (m, 2H, CHarom.);  13  C{1H} NMR (benzene-d6, 150 MHz) δ 29.1, 29.9 (C(CH3)3), 40.3, 42.1 (C(CH3)3),  45.2 (br. s, N(CH3)2), 48.7 (br. s, N(CH3)2), 121.4, 123.4, 125.4, 126.4, 128.3, 128.7, 128.9 (CHarom.), 145.6 (Carom.N),151.2 (Carom.N), 168.6 (C=O), 184.9 (C=O). MS(EI): m/z 621 (M+ – NMe2); Anal. Calcd. (%) for C28H46N5O2Ta: C, 50.52; H, 6.97; N, 10.52; Found: C, 50.64; H, 6.90; N, 10.44. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  The Gibbs energy of activation (ΔG‡) for the dynamic exchange of amidate binding modes studied using the variable temperature 1H NMR study (toluene-d8, 400 MHz) was estimated using the Eyring equation.114 The rate of exchange at the coalescence temperature is calculated based on the maximum peak separation in the cold regime where no exchange is observed: kc = π(Δν/√2) Substituting this expression into the Eyring equation gives the following equation: ΔG‡(Tc) = 0.00457Tc[9.972 + log10 (Tc/Δv)] Assuming a coalescence temperature (Tc) of 62.6 °C (335.8 K) and a maximum peak separation (Δν) of 274.72 Hz gives a free energy estimation of 9.32 ± 0.02 kcal/mol.  105  Tris(N-phenylpivalamidate)(N,N’-dimethylethane-1,2diamido)tantalum (8). Single crystal X-ray quality samples were obtained from a reaction mixture containing mostly 7 after recrystallization from hot hexanes. Deep-red crystals were manually separated from the yellow prisms of 7 and submitted for X-ray analysis. Spectroscopic NMR data is not available due to contamination with 7. Due to the small amount of material collected (approximately 5 mg) and the inability to purify the material, MS and EA data are also unavailable. Subsequent attempts to synthesize 8 were unsuccessful.  Bis(N-[2,6-dimethylphenyl]pivalamidate)(dimethylamido)Nmethyltantalaaziridine (9). N-(2,6-dimethylphenyl)pivalamide (0.350 g, 1.71 mmol) was suspended in hexanes (2 mL). Ta(NMe2)5 (0.342 g, 0.853 mmol) was added as a solid and the solution was allowed to stir overnight. The solvent was removed in vacuo and the resulting solid was recrystallized from the minimal amount of hot hexanes. The light yellow crystals were collected and identified as the title compound. Yield: 0.487g (0.72 mmol, 84%).  1  H NMR  (benzene-d6, 400 MHz) δ 1.05 (s, 9H, C(CH3)3), 1.08 (s, 9H, C(CH3)3), 2.17 (s, 3H, CCH3), 2.34 (d, 2JH,H = 3.5 Hz, 1H, TaCH2), 2.41 (s, 3H, CCH3), 2.49 (d, 2JH,H = 3.5 Hz, 1H, TaCH2), 2.56 (s, 3H, CCH3), 2.87 (s, 3H, CCH3), 3.07 (br. s, 6H, N(CH3)2), 3.16 (s, 3H, NCH3), 6.94 (br. m, 4H, CHarom.), 7.00 (d, 3JH,H = 7.0 Hz, 1H, CHarom.), 7.06 (d, 3JH,H = 7.0 Hz, 1H, CHarom.); 13  C{1H} NMR (benzene-d6, 100 MHz) δ 18.4, 19.2, 19.6, 20.1 (CCH3), 27.5, 27.6 (C(CH3)3),  41.7, 41.9 (C(CH3)2), 44.8 (NCH3), 45.9 (br. s, N(CH3)2), 59.1 (TaCH2), 124.7, 125.7, 127.7, 128.3, 128.3, 128.4 (CHarom.), 133.0, 133.4, 133.8, 134.2 (Carom.CH3), 144.0, 144.3 (Carom.N),  106  182.3, 192.1 (C=O). MS(EI): m/z 676 (M+); Anal. Calcd. (%) for C30H47N4O2Ta: C, 53.25; H, 7.00; N, 8.28; Found: C, 53.00; H, 6.81; N, 8.16. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  Bis(N-[2,6-dimethylphenyl]pivalamidate)(dimethylamido)[N1methylprop-1-ene-1,2-diamido(2-)-κN,κN1])tantalum (10). Bis(N-[2,6dimethylphenyl]pivalamidate)(dimethylamido)-N-methyltantalaaziridine (9, 0.569 g, 0.84 mmol) was dissolved in benzene (3 mL) and dry acetonitrile (52 μL, 1.00 mmol) was added. The mixture was allowed to stir at ambient temperature for 24 h to give a deep- red solution. All volatiles were removed in vacuo and the remaining red powder was identified as the title compound. Yield: 0.602 g (0.84 mmol, 100%).  1  H NMR (benzene-d6, 400 MHz) δ 1.03 (s, 9H, C(CH3)3), 1.10 (s, 9H,  C(CH3)3), 1.50 (s, 3H, Carom.CH3), 1.83 (s, 3H, CH3C=CH), 2.10 (s, 3H, Carom.CH3), 2.16 (s, 3H, Carom.CH3), 2.42 (s, 3H, Carom.CH3), 3.38 (s, 3H, N(CH3)2), 3.89 (s, 3H, N(CH3)2), 4.26 (s, 3H, C=CHNCH3), 5.86 (d, 3JH,H = 1.6 Hz, 1H, C=CHN), 6.76 (d, 3JH,H = 7.2 Hz, 1H, CHarom.), 6.84 (d, 3JH,H = 7.4 Hz, 1H, CHarom.), 6.89 – 7.97 (m, 4H, CHarom.), 7.03 (br. s, 1H, NH); 13  C{1H} NMR (benzene-d6, 100 MHz) δ 16.4 (CH3C=CH), 18.7, 18.9, 19.1 20.8 (Carom.CH3),  28.0, 28.5 (C(CH3)3), 41.8, 42.1 (C(CH3)3), 43.8, 45.1 (N(CH3)2), 47.7 (C=CHNCH3), 125.0 (CHarom.), 126.0 (C=CHN), 126.6, 127.7, 128.1, 128.4, 129.2 (CHarom.), 130.0 (C=CHN), 133.4, 134.8, 134.9, 136.0 (Carom.CH3), 140.6, 145.0 (Carom.N), 180.7, 191.1 (C=O). MS(EI): m/z 717 (M+); Anal. Calcd. (%) for C32H50N5O2Ta: C, 53.55; H, 7.02; N, 9.76; Found: C, 53.66; H, 7.04; N, 9.37. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  107  Bis(N-[2,6Dimethylphenyl]benzamidate)tris(dimethylamido)tantalum (11). N-(2,6-dimethylphenyl)benzamide (0.264 g, 1.25 mmol) and Ta(NMe2)5 (0.235 g, 0.59 mmol) were suspended in hexanes (5 mL) and the mixture was allowed to stir overnight (15 h) to give a clear yellow solution.  All volatiles were  removed in vacuo and the remaining solid was recrystallized from the minimal amount of hot toluene. The yellow crystals were collected and identified as the title compound. Yield: 0.330g (0.43 mmol, 74%).  1  H NMR spectroscopy showed multiple species in solution which, upon  confirmation of sample purity, are assumed to be constitutional isomers. MS(EI): m/z 716 (M+ – NMe2); Anal. Calcd. (%) for C36H46N5O2Ta: C, 56.76; H, 6.09; N, 9.19; Found: C, 56.82; H, 6.10; N, 9.22. Single crystal X-ray quality samples were obtained by recrystallization from hot toluene.  N-(2,6-dimethylphenyl)cyclopropanecarboxamide.  2,6-dimethylaniline  (2.36 mL, 19.1 mmol) was dissolved in DCM (25 mL) and triethylamine (8.00 mL,  57.4  mmol).  The  mixture  was  cooled  to  -78  °C  and  cyclopropanecarbonyl chloride (1.74 mL, 19.1 mmol) was added dropwise to the solution. The reaction mixture was allowed to warm and stir overnight (18h). The solution was washed sequentially with 1M NaOH, 1M HCl, water and saturated NaCl solution. The organic layer was separated from the aqueous layer and the solvent was dried with MgSO4, filtered and removed under vacuum.  The resultant solid was purified by washing with hexanes.  Yield: 3.07 g  (16.2 mmol, 85%). Two sets of signals (2:1 ratio) are observed due to the hindered rotation  108  about the amide bond. In all cases the major signal is reported first.  1  H NMR (chloroform-d,  600 MHz) δ 0.76/0.67 (m, 2H, CH2), 0.98/1.03 (m, 2H, CH2), 1.58/1.22 (m, 1H, CH), 2.17/2.32 (s, 6H, CH3), 7.01/7.14 (m, 3H, CHarom.), 7.41/7.09 (br. s, 1H, NH); 13C{1H} NMR (chloroformd, 151 MHz) δ 7.1/8.1, 14.3/10.2 (CH2), 18.2/18.5 (CH), 126.8 (Carom.N), 127.8/128.4, 134.1/134.7 (CHarom.), 135.5/137.0 (Carom.CH3), 172.1/176.0 (C=O). MS(EI): m/z 189 (M+); Anal. Calcd. (%) for C12H15NO: C, 76.16; H, 7.99; N, 7.40; Found: C, 75.90; H, 7.95; N, 7.37.  N-(2,6-diisopropylphenyl)cyclopropanecarboxamide. Same procedure as above except 2,6-diisopropylaniline (3.61 mL, 19.1 mmol) was used as the amine. Yield: 4.26 g (17.4 mmol, 91%). Two sets of signals (4:3 ratio) are observed due to the hindered rotation about the amide bond. If two signals are reported, the major signal is reported first. 1H NMR (chloroform-d, 400 MHz) δ 0.68/0.85 (m, 2H, CH2), 1.07 (m, 2H, CH2), 1.21, 1.26, 1.19 (d, 2JH,H = 6.85 Hz, 12H, CHCH3), 1.23/1.26 (m, 1H, CH(CH2)2), 3.31/3.11 (sep, 2JH,H = 6.85 Hz, 1H, CHCH3) 6.80/6.74 (br. s, 1H, NH), 7.21/7.16 (s, 1H, CHarom.), 7.23/7.18 (s, 1H, CHarom.), 7.35/7.33 (s, 1H, CHarom.);  13  C{1H} NMR (chloroform-d,  151 MHz) δ 8.5/7.1, 10.7/14.4 (CH2), 23.5 (br. CH(CH2)2), 24.0/23.1 (CH(CH3)2), 28.4/28.6 (CH(CH3)2), 123.7/123.2 (CHarom.), 128.7/127.9 (CHarom.), 131.9/131.5 (Carom.N), 147.3/146.3 (Carom.CH), 176.6/172.9 (C=O). MS(EI): m/z 245 (M+); Anal. Calcd. (%) for C16H23NO: C, 78.32; H, 9.45; N, 5.71; Found: C, 78.27; H, 9.42; N, 5.82.  109  (N-[2,6Dimethylphenyl]cyclopropanecarboxamidate)tetrakis(dimethylami do)tantalum (15).  N-(2,6-dimethylphenyl)cyclopropanecarboxamide  (0.118 g, 0.62 mmol) and Ta(NMe2)5 (0.250 g, 0.62 mmol) were suspended in hexanes (4 mL) and the mixture was allowed to stir overnight (15 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal amount of hot hexanes. The yellow crystals were collected and identified as the title compound. Yield: 0.308g (0.57 mmol, 91%). 1H NMR (benzene-d6, 400 MHz) δ 0.35 (m, 2H, CH2), 1.08 (m, 1H, CH), 1.17 (m, 2H, CH2), 2.44 (s, 6H, CH3), 3.41 (br. s, 24H, N(CH3)2), 7.09 (t, 1H, CHarom.), 7.20 (d, 2H, CHarom.); 13C{1H} NMR (benzene-d6, 100 MHz) δ 7.4 (CH2), 11.8 (CH), 18.6 (CCH3), 47.2 (N(CH3)2), 125.0, 128.7 (Carom.H), 133.5 (Carom.CH3), 144.0 (CN), 175.7 (C=O). MS(EI): m/z 501 (M+ – NMe2); Anal. Calcd. (%) for C20H38N5OTa: C, 44.04; H, 7.02; N, 12.84; Found: C, 44.38; H, 7.16; N, 12.85. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  (N-[2,6Diisopropylphenyl]cyclopropanecarboxamidate)tetrakis(dimethyla mido)tantalum  (16).  N-(2,6-  diisopropylphenyl)cyclopropanecarboxamide (0.153 g, 0.62 mmol) and Ta(NMe2)5 (0.250 g, 0.62 mmol) were suspended in hexanes (5 mL) and the mixture was allowed to stir overnight (16 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal amount of hot hexanes. The yellow crystals were collected and identified as the title compound. Yield: 0.316g (0.53 mmol,  110  84%). 1H NMR (benzene-d6, 400 MHz) δ 0.19 (m, 2H, CH2), 0.81 (m, 1H, CH(CH2)2), 1.04 (m, 2H, CH2), 1.27 (d, 2JH,H = 6.83 Hz, 6H, CHCH3), 1.33 (d, 2JH,H = 6.83 Hz, 6H, CHCH3), 3.27 (br. s, 24H, N(CH3)2), 3.47 (sep, 2JH,H = 6.83 Hz, 1H, CHCH3), 7.13 (m, 2H, CHarom.), 7.17 (s, 1H, CHarom.); 13C{1H} NMR (benzene-d6, 100 MHz) δ 8.6 (CH2), 12.8 (CH), 24.9, 26.0 (CH(CH3)2), 27.5 (CH(CH3)2), 47.2 (br. N(CH3)2), 126.0, 124.1 (CHarom.), 141.2 (CN), 144.3 (CCH), 176.3 (C=O). MS(EI): m/z 557 (M+ – NMe2); Anal. Calcd. (%) for C24H46N5OTa: C, 47.91; H, 7.71; N, 11.64; Found: C, 48.26; H, 7.94; N, 11.56. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  2.4.3  General Procedure for Screening of Mono(amidate) Precatalysts In a nitrogen filled glove box, the appropriate tantalum precatalyst (1 – 5; 0.025 mmol, 5  mol%) was weighed into a small vial and subsequently dissolved in 500 mg of toluene-d8. 1,3,5trimethoxybenzene (13.9 mg, 0.083 mmol) was then weighed into another small vial and dissolved in the same solution containing the tantalum precatalyst before being transferred to an NMR tube equipped with a Teflon cap. N-methylaniline (54 µL, 0.498 mmol) and 1-octene (118 µL, 0.752 mmol) were then added sequentially by means of a μL-pipette. The NMR tube was closed, shaken and removed from the glove box. Following the recording of a 1H NMR spectrum, the NMR tube was placed in a preheated, 110 °C thermostated oil bath.  The  conversion was periodically monitored by 1H NMR until the maximum conversion had been achieved.  111  2.4.4  Isotopic Labeling Studies  Isotope exchange in the absence of alkene. In a nitrogen filled glove box, a solution of 3 and 1,3,5-trimethoxybenzene (3.5 mg, 0.0208 mmol) in toluene-d8 (500 mg) was transferred to a NMR tube equipped with a Teflon cap. N-methylaniline-d3 (56 µL, 0.5 mmol) was added by means of a μL-pipette. The NMR tube was closed, shaken and removed from the glove box. Following the recording of a 1H NMR spectrum, the NMR tube remained at ambient temperature for 20 hours when another 1H NMR spectrum was recorded. The NMR tube was then placed in a preheated, 65 °C thermostated oil bath and 1H NMR spectra were recorded after 24 and 55 hours.  Isotope exchange in the catalytic process. In a nitrogen filled glove box, a solution of 3 (15.4 mg, 0.025 mmol, 5 mol%) in toluene (600 μL) was transferred to an NMR tube equipped with a Teflon cap. 1-octene (117.7 μL, 0.75 mmol) followed by N-methylaniline-d3 (55.7 μL, 0.5 mmol) were added by means of a μL-pipette. The NMR tube was closed, shaken and removed from the glove box and placed in a preheated, 130 °C thermostated oil bath for 48 hours. The reaction mixture was poured into a vial and all volatiles were removed under reduced pressure. Flash chromatography (silica gel F60, hexanes/EtOAc/NEt3 100:2:1) gave 6-d as a colorless oil. Yield: 0.097 g (0.47 mmol, 93%).  1  H and 2H NMR spectra match the previously  reported compound.39  2.4.5  Kinetic Investigations  General procedure for kinetic runs. In a nitrogen filled glove box, standard solutions of 3 and 1,3,5-trimethoxybenzene in toluene-d8 was transferred to a NMR tube equipped with a Teflon  112  cap. Appropriate amounts of N-methylaniline and 1-octene were added by means of a μL-pipette (standard solutions where permitted). Toluene-d8 was added until the final volume of deuterated solvent reached 600 μL. The NMR tube was closed, shaken and removed from the glove box and a t0 1H NMR spectrum was recorded. The NMR tube was then placed in a preheated, 130 °C thermostated oil bath. The sample was then removed from the oil bath and a spectrum was collected every hour for 10 hours.  7.0  6.5  6.0 Chemical Shift (ppm)  5.5  5.0  4.5  Figure 2.21 Typical kinetics experiment monitoring the conversion of N-methylaniline (δH = 6.37) and 1-octene (δH = 5.00, 5.78) to product 6 (δH = 6.46) catalyzed by 3 with integrations relative to 1,3,5-trimethoxybenzene (δH = 6.14). Time zero spectrum is shown at the bottom with each subsequent spectrum taken at 1 hour intervals up to 10 hours.  113  Amine concentration was quantified by integration of the N-methylaniline ortho-protons relative to internal standard (1,3,5-trimethoxybenzene). To normalize the data collected between different runs, concentrations are plotted as c(t)/c(t0). All errors on linear corrections were estimated from the standard error of the regression analysis performed using the Data Analysis Toolpack in Microsoft Excel 2010 unless otherwise noted. Error bars on raw kinetic data are derived from the standard deviation of duplicate or triplicate experiments. Initially, kinetic runs were performed in triplicate for two data points in order to ensure a well-behaved system with reproducible results, as judged by the standard deviation in c(t)/c(t0). If the standard deviation was less than or equal to 5% of the average value between triplicate runs the system was deemed to be well behaved and duplicate runs were used for the remainder of the data points collected.  114  Chapter 3: Computational Modeling of the Catalytic Cycle for the αAlkylation of Amines  3.1  Introduction The modeling of catalytic cycles with computational chemistry is becoming increasingly  powerful, accurate and common.115-116  Using quantum chemical calculations, valuable  information about a catalytic system can be accessed that is exceedingly difficult or impossible to observe with traditional experimental methods.  Perhaps most importantly, these  computational techniques allow the visualization of transition states and intermediates which are invaluable tools for catalyst development. In addition to molecular geometries, thermodynamic energies can be calculated for each structure to give a simplified potential energy surface (PES) for the catalytic cycle. Visualization of electronic properties of optimized species, including molecular orbitals, is also useful for interpreting reactivity trends and designing further generations of catalysts.  3.1.1  Computational Investigations Involving Amidate Species Previously, the Schafer group has performed in-house ground state optimizations to  evaluate relative energies and explore the electronic interactions within discrete complexes. For example, optimization of all possible orientations of titanium bis(amidate) bis(amido) species is consistent with the lowest energy geometric isomer being experimentally isolated in the solid state (Figure 3.1).51 The same study also showed that the occupied molecular orbitals associated with the amidate ligands are low in energy suggesting strong binding to the metal centre. Additionally, the frontier molecular orbitals are consistent with activation of the ancillary amido 115  ligands to form imido species, a proposed intermediate in the hydroamination of alkenes and alkynes. However, no modeling of the catalytic cycle for hydroamination using early transition metal amidate complexes was performed at that time.  Figure 3.1 Possible geometric isomers of a bis(amidate) bis(amido) titanium complex. The highlighted isomer is crystallographically observed and is calculated to have the lowest ground state energy. Relative free energies (ΔG) are reported in kcal/mol.  After the publication of an axially chiral biphenyl-based bis(amidate) zirconium catalyst for the enantioselective hydroamination of aminoalkenes (Figure 3.2), the Schafer group collaborated with theoretician and University of British Columbia professor Dr. Wang and his graduate student, Dr. Stephen Hepperle, to elucidate the origin of the selectivity observed for the catalyst. Using established quantum chemical techniques, the catalytic cycle was modeled using density functional theory (DFT) and a favourable pathway in good agreement with the experimentally observed stereochemistry was charted.117 In this case, all calculations were performed by Dr. Hepperle in consultation with the Schafer group.  116  Figure 3.2 Axially chiral bis(amidate) bis(amido) zirconium precatalyst for the enantioselective hydroamination of aminoalkenes.  3.1.2  Modeling of the Intramolecular α-Alkylation of Amines In addition to the kinetic investigations discussed in Section 2.2.4, Doye and coworkers  have modeled the catalytic cycle for the intramolecular α-alkylation of amines using DFT calculations.103 Using Ti(NMe2)4 as the precatalyst, the catalytic cycle shown in Scheme 3.1 was calculated for a model aminoalkene substrate.  Based on free energies computed for the  intermediates and transition states, β-hydrogen abstraction (C to D, Scheme 3.1) was found to be the turnover limiting step. This is in good agreement with the observed zero order dependence on aminoalkene substrate concentration and the large primary kinetic isotope effect (kH/kD = 7.3) found for substrates deuterated α to the nitrogen atom.  117  Scheme 3.1 Calculated intermediates for the catalytic cycle of the intramolecular α-alkylation of aminoalkenes using Ti(NMe2)4 as the precatalyst.103  Starting from the 5-membered bicyclic metallacycle A, the aminoalkene substrate is found to coordinate to the titanium centre (B) before protonolysis of the Ti–C bond to give 118  bis(amido) intermediate C. Turnover limiting β-hydrogen abstraction yields titanaziridine D with the cyclized product coordinated to the titanium centre. Insertion of the double bond into the Ti–C bond gives intermediate E with the product now bound to the amido nitrogen through an NH–N bridge. The product is replaced by an aminoalkene substrate in intermediate F allowing for coordination to the metal centre and subsequent closing of the catalytic cycle. Based on the success of the aforementioned theoretical studies, and the recent advances in computational power available to chemists, performing the experimental and in silico studies concurrently for a catalytic α-alkylation of amines system was a possibility. To this end, the first computational investigation of a catalytic cycle by a Schafer group member is described herein. At the time of writing, this is the first reported example of a computational investigation of the intermolecular α-alkylation of amines.  3.1.3  Scope of Chapter The focus of this chapter is the computational modeling of the catalytic cycle for the α-  alkylation of amines catalyzed by tantalum amidate complexes.  In addition to gaining  quantitative information about the turnover limiting step of the catalytic cycle, qualitative visualization of transition state geometries greatly aids in the development of improved catalysts. Based on the solid state molecular structure of precatalyst 3 as well as previous mechanistic suggestions in the literature,8,  33-36, 39-40, 44, 103  a more detailed proposal for the catalytic cycle  outlined in Scheme 1.3 is depicted below (Scheme 3.2).  119  Scheme 3.2 Proposed catalytic cycle for the α-alkylation of N-methylaniline with 1-octene catalyzed by complex 3. Generalized transition states as well as intermediates are depicted for the first turnover.  120  At the outset, the C–H activation of a dimethylamido ligand to generate the catalytically active tantalaziridine is investigated (Section 3.2.1). The catalytic cycle based on this calculated geometry is modeled in Section 3.2.2 by locating transition states on the potential energy surface and relaxing them to relevant intermediates. Interesting changes in the binding mode of the amidate ligand will be discussed. Substitution of the model dimethylamine substrate for N-methylaniline will be explored in Section 3.2.3. The steric demand of the phenyl group is found to have profound effects on the geometry of intermediates and transition states alike. This corresponds to a change in the energy landscape of the PES and a different turnover limiting step.  Upon regeneration of the  tantalaziridine, the azametallacyclopropane moiety is found to have the reverse geometry in relation to the amidate ligand, producing a new set of transition states and intermediates. These two pathways are compared and the possibility of a cross-over point is also explored. Computational data investigating the possibility of one-electron processes are presented in Section 3.2.4. By optimizing intermediate and transition states in the triplet state, arguments against the likelihood of radical behaviour are supported by molecular orbital surfaces as well as a comparison of the relative energies to singlet species. Finally, the preliminary efforts to establish a steric parameter for amidate precatalysts as a predictive tool for catalyst development are reported (Section 3.2.5). Using buried volume and solid angle calculations, trends in reactivity are related to calculated steric values for the mono(amidate) complexes described in Chapter 2 and axially chiral tantalum precatalysts reported in the literature.  121  3.2  Results and Discussion In order to further investigate the mechanism for the α-alkylation of amines catalyzed by  mono(amidate) tantalum catalysts, the proposed catalytic cycle (Scheme 1.3) was modeled using quantum chemical calculations, specifically density functional theory.118  All calculations  described herein were performed using the Gaussian suite of programs119-120 and run on the Glacier or Orcinus clusters maintained by WestGrid, a division of ComputeCanada. Based on previous success in the modeling of early transition metal amidate complexes, density functional theory (DFT) was chosen as the modelling method.117 All geometries have been optimized using the B3LYP hybrid functional121-124 due to its ability to accurately predict crucial molecular properties such as bond lengths and vibrational frequencies.125 The double zeta 6-31G Pople basis set was modified to include polarization functions for C, H, O and N to better describe interactions with the tantalum centre. The tantalum centre was modeled using the Los Alamos LANL2DZ basis set with an effective core potential that included f-orbitals.126-127 Calculated bond lengths and angles for the tantalum coordination sphere of 3 were found to be in good agreement with solid-state data derived from crystallographic studies (standard deviations of 0.027Å and 0.62° for bond lengths and angles respectively). All energies herein are reported in units of kilocalories per mole (kcal/mol) and were calculated using MP2, a second-order implementation of Møller-Plesset perturbation theory, with the aforementioned basis set.128  3.2.1  Tantalaziridine Formation The first step in the catalytic cycle is the reversible C–H activation, as shown through  deuterium labelling experiments, of a dimethylamido ligand bound to 3 to form the active tantalaziridine catalyst (Scheme 3.3).  122  Scheme 3.3 Intermediates and transition state for the C–H activation of a dimethylamido ligand of complex 3 to form tantalaziridine I. Relative free energies (ΔG) are reported in kcal/mol.  Starting from the coordinates for the solid-state molecular structure of 3, a transition state geometry (TS(3/I), Figure 3.3) was found for the formation of the tantalaziridine in the equatorial plane of the amidate chelate. The negative vibrational mode associated with the transition state assignment clearly shows the transfer of the proton from the carbon to the nitrogen along a linear path, indicating a concerted bond making / bond breaking step.  Figure 3.3 Optimized geometry for C–H activation transition state TS(3/I). Most hydrogen atoms omitted for clarity. Colouration: C = gray, H = white, N = blue, O = red, Ta = purple 123  Relaxation of the transition state yields a species isostructural with complex 3 (bond lengths involving the Ta centre are within 0.01 Å on average) and a tantalaziridine with a datively bound dimethylamine ligand shielding the reactive Ta–C bond (I, Scheme 3.3). This neutral ligand was removed and the resultant structure was optimized to ensure a stable tantalaziridine is indeed feasible with this geometry. Interestingly, intermediate (II) is found with axial dimethylamido ligands distorted from the ideal linear arrangement (Namido–Ta–Namido = 124.0°) to occupy the space vacated by the expelled dimethylamine (Figure 3.4).  Figure 3.4 Optimized geometries for intermediates I (left) and II (right). Most hydrogen atoms omitted for clarity.  The activation barrier (ΔG‡) for the C–H activation was found to be 34.1 kcal/mol at 110 °C in toluene. Initially, this barrier was thought to be too large for the formation of minor, yet detectable, amounts of tantalaziridine present in room temperature solutions containing 3 (ΔG‡ = 33.3 kcal/mol at 20 °C). However, comparison of the calculated equilibrium constant (Keq) at 20 and 110 °C (1.879 x 10-7 and 3.354 x 10-6 respectively) suggests an increase in the formation of 124  the tantalaziridine product I at higher temperatures.  This result qualitatively supports the  increased ratio of tantalaziridine observed at elevated temperatures during the thermolysis of 3 (Scheme 2.19). In order to verify that TS(3/I) and the resultant tantalaziridine II were located on the lowest energy pathway, alternative transition state geometries were investigated.129 Activating a C–H bond of the other equatorial dimethylamido ligand required a rotation of the Namido–C bonds, resulting in a κ1(O) amidate binding mode and a higher activation barrier (42.3 kcal/mol). Additionally, no transition state for the C–H activation of an axial dimethylamido ligand, orthogonal to the plane of the amidate chelate, could be found. However, relaxation of an insertion transition state investigating the same alternative geometry (vide supra) resulted in a tantalaziridine 9.7 kcal/mol higher in energy than II. This evidence suggests that the path described in Scheme 3.3 is the most favourable C–H activation pathway, justifying the use of II as the starting point for further modeling of the catalytic cycle.  3.2.2 3.2.2.1  Dimethylamine as a Model Substrate Insertion of 1-Octene into Tantalaziridine Rather than immediately replace the methyl group of the azametallacyclopropane moiety  with a phenyl group to simulate the use of N-methylaniline as a substrate, the catalytic cycle was investigated using dimethylamine as a model amine substrate. The reasoning is twofold: one, methyl groups are less computationally expensive therefore requiring less resources and time to complete calculations and two, the formation of byproduct 12 could be investigated. To this end, a transition state geometry for the insertion of 1-octene into the tantalaziridine Ta–C bond was found (TS(III/IV), Scheme 3.4).  125  Scheme 3.4 Intermediates and transition state for the insertion of 1-octene into the Ta–C bond of tantalaziridine II. Relative free energies (ΔG) are reported in kcal/mol.  The negative vibrational mode shows the terminal sp2 carbon of 1-octene moving towards the tantalum centre and adopting a more tetrahedral geometry. The methine carbon of the alkene substrate is travelling in the direction of the azametallacyclopropane carbon, lengthening the double bond. An intrinsic reaction coordinate analysis of the relaxation suggests coordination of the terminal carbon to form a Ta–C bond followed by C–C bond formation to close the 5membered metallacycle. The amidate ligand is bound κ2(N,O) and does not participate in the vibrational mode. In contrast to the geometry of tantalaziridine II, the dimethylamido ligands are closer to an ideal axial position (Figure 3.5).  126  Figure 3.5 Optimized geometry for transition state TS(III/IV). Hydrogen atoms omitted for clarity.  Relaxation of TS(III/IV) yields the expected 5-membered metallacycle IV (Figure 3.6) as well as tantalaziridine II with the alkene substrate approximately 5 Å away from the tantalum centre (intermediate III, Scheme 3.4). This suggests that there is no pre-coordination of the alkene substrate to the tantalum centre to activate the double bond before insertion occurs. Though there is no chemical interaction between the tantalaziridine and the 1-octene, the proximity of the substrate in III results in a 12.3 kcal/mol increase in energy compared to II and 1-octene being at an infinite distance. This difference in energy is proposed to arise because of the difficulty in calculating the entropy of systems with two non-interacting molecules.130  127  Figure 3.6 Optimized geometry for 5-membered metallacycle intermediate IV. Hydrogen atoms omitted for clarity.  An alternative transition state with the insertion occurring perpendicular to the plane of the amidate chelate was also calculated;129 however it was found to be 13.1 kcal/mol higher energy than TS(III/IV), further supporting the notion that reactivity in the same plane as the amidate chelate is the lowest energy pathway along the PES.  3.2.2.2  Protonolysis of 5-Membered Metallacycle by N-Methylaniline Having established that IV, with the reactive Ta–C bond in the equatorial plane along  with the amidate chelate, is the lowest energy 5-membered metallacycle along the PES, the subsequent protonolysis transition state was based on this geometry (Scheme 3.5).  128  Scheme 3.5 Intermediates and transition state for the protonolysis of the Ta–C bond of metallacycle IV by N-methylaniline. Relative free energies (ΔG) are reported in kcal/mol.  Gratifyingly, TS(V/VI) was found to occur in the equatorial plane with the amidate bound in a κ2(N,O) fashion (Figure 3.7). The negative vibrational mode shows a proton transfer along a straight vector from the N-methylaniline substrate to the carbon previously bound to tantalum.  Figure 3.7 Optimized geometry for transition state TS(V/VI). Most hydrogen atoms omitted for clarity. 129  The “backwards” relaxation of this transition state, intermediate V, gives a 5-membered metallacycle identical to IV with an N-methylaniline molecule 4.8 Å from the tantalum centre. As with the insertion of 1-octene, pre-coordination of the amine substrate to the metal centre is not required to activate the N–H bond. This is in contrast to the model calculated by Doye and coworkers for the intramolecular α-alkylation of amines where the aminoalkene substrate does coordinate to the titanium centre before the proton transfer occurs.103  Interestingly, the  intermediate found further along the reaction pathway (VI) is considerably lower in energy than V (Scheme 3.5) and is trending towards a κ1(O) binding mode with a Ta–Namidate bond length of 3.11 Å compared to the 2.43 Å observed for κ2(N,O) species V.  3.2.2.3  β-Hydrogen Abstraction to Release Product and Reform Tantalaziridine In order to facilitate the β-hydrogen abstraction to release product 12 and reform the  tantalaziridine, the CH3 group of the N-methylanilido ligand must be rotated through the equatorial plane.  Intermediate VII, from the relaxation of TS(VII/VIII), represents the  minimum found for this geometry (Scheme 3.6).  130  Scheme 3.6 Intermediates and transition state for the β-hydrogen abstraction of a methyl C–H bond of bis(amido) VII to release amine 12. Relative free energies (ΔG) are reported in kcal/mol.  Notably, the 115° rotation of the amido methyl group results in a κ1(O) binding mode for the amidate ligand with the complex adopting a more idealized trigonal bipyramidal geometry (Figure 3.8). This increased Ta–Namidate bond length (3.44 Å) and Ta–Oamidate–Camidate bond angle (140.1°) is accompanied by a 5.4 kcal/mol increase in energy. Though κ1(O) species were previously found to be higher in energy for less sterically crowded complexes, no intermediate was found with a κ2(N,O) bound amidate ligand. This observation supports the proposal that the decreased reaction temperatures and increased substrate scope observed for 3 are attributable, at least in part, to the hemi-labile nature of amidate ligands.  131  Figure 3.8 Optimized geometries for bis(amido) intermediates VI (left) and VII (right). Hydrogen atoms and select carbon atoms form the alkyl chain omitted for clarity. Simplified representations (above) are giving for reference.  The transition state based on the aforementioned geometry (TS(VII/VIII), Figure 3.9) is found to have a negative vibrational mode that depicts a β-hydrogen abstraction from the methyl group of the N-methylanilido ligand derived from the N-methylaniline substrate to release product 12. Though the initial input for TS(VII/VIII) had the amidate ligand bound in a κ1(O) fashion (based on the geometry of intermediate VII), the transition state features a κ2(N,O) amidate chelate in the plane of the proton transfer. This result showcases the ability of the  132  computational algorithms to optimize to the lowest energy geometry without being restricted to the κ1(O) or κ2(N,O) amidate binding mode of the input geometry.  Figure 3.9 Optimized geometry for transition state TS(VII/VIII). Most hydrogen atoms omitted for clarity.  Interestingly, as with I, the neutral amine remains coordinated to the tantalum centre upon relaxation from the transition state (VIII, Scheme 3.6). Removal of the amine donor gives an aryl-substituted tantalaziridine (IX) with the azametallapropane moiety twisted approximately 30° out of the equatorial plane and the Oamidate cis to the Naziridine, akin to the intermediate calculated for tantalaziridine formation with a κ1(O) bound amidate ligand.129  This is the  opposite relative geometry observed in II (Namidate cis to the Naziridine) and will be further discussed in Section 3.2.3.  3.2.2.4  Potential Energy Surface Comparing the energies of the calculated intermediates and transition states relative to  tantalaziridine II gives the potential energy surface presented in Figure 3.10. The highest energy point along the catalytic cycle is transition state TS(VII/VIII), suggesting that β-hydrogen 133  abstraction is the turnover limiting step for the α-alkylation of dimethylamine with 1-octene. Though in agreement with the computational model put forth by Doye and coworkers for their intramolecular Ti system,103  the experimental evidence for intermolecular α-alkylation of  amines catalyzed by 3 suggests that alkene insertion is turnover limiting. However, TS(III/IV) is found to be a low energy transition state and have the lowest activation barrier (ΔG‡ = 26.7 kcal/mol). Nevertheless, Figure 3.10 represents the PES using dimethylamine as a model substrate. Substituting N-methylaniline as the amine substrate alters both the steric and electronic properties of the various intermediates and transition states, possibly yielding calculated results more in line with experimental observations.  134  Figure 3.10 Potential energy surface (PES) for the α-alkylation of dimethylamine with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol.  135  3.2.3 3.2.3.1  Potential Energy Surface using N-Methylaniline as a Substrate Namidate cis to the Naziridine Investigations involving N-methylaniline as a substrate were based upon the calculated  species optimized with dimethylamine. Interestingly, the amidate ligand accommodates for the extra bulk of the substrate (vide infra) leaving the transition states and intermediates relatively unchanged from the dimethylamine model. For example, replacing the methyl group bound to the nitrogen of the tantalaziridine in TS(III/IV) with a phenyl group led to a transition state (TS(XI/XII), Figure 3.11) isostructural with TS(III/IV) which was relaxed to give the anticipated 5-membered metallacycle (XII) and a tantalaziridine paired with a non-coordinating 1-octene molecule (XI).  The tantalaziridine X, optimized upon removal of the alkene, is  structurally identical to II, ignoring the exchange of the nitrogen-bound substituent.  136  Figure 3.11 Optimized geometries for transition states TS(XI/XII) (left), TS(XIII/XIV) (middle), and TS(XV/XVI) (right) viewed along the axial plane. Most hydrogen atoms and select carbon atoms form the alkyl chain omitted for clarity.  Interestingly, the Ta–Namidate bond in metallacycle XII is noticeably lengthened (3.02 Å) compared to the less hindered VI (2.44 Å). With the phenyl group occupying the same plane as the amidate chelate, adjacent to the sterically demanding Namidate substituent, the ligand assumes an asymmetric κ1(O) binding mode to alleviate steric crowding in the equatorial plane. This notion is further evidenced by the inability to locate geometries containing κ2(N,O) bound amidate ligands for both protonolysis (TS(XIII/XIV)) and β-hydrogen abstraction transition states (TS(XV/XVI)) shown in Figure 3.11. In these cases the Namidate is not bound to the tantalum atom and the Ta–Oamidate–Camidate bond angle is increased, pointing the bulky N-aryl substituent away from the metal centre.  Relaxation of both these transition states yield  137  intermediates with the amidate ligand bound in a κ1(O) arrangement. This trend is broken with tantalaziridine XVII which adopts a κ2(N,O) amidate binding mode, presumably to stabilize the strained 3-membered metallacycle.  The fact that a sterically congested transition state  containing the amidate bound κ1(O) can relax to an intermediate stabilized by a κ2(N,O) chelating amidate is further evidence of the advantageous hemi-lability of this class of ligands. Despite the variation in amidate binding mode, the use of N-methylaniline as the amine substrate does not drastically change the calculated structures of the transition states or intermediates from those discussed in the dimethylamine model. With only three tantalum– ligand contacts in the equatorial plane, the 5 coordinate species adopt a more idealized trigonal bipyramidal geometry when N-methylaniline is used as a substrate. Though similar in geometry, the PES for the N-methylaniline pathway is significantly different from using dimethylamine as a substrate (Figure 3.12). Compared to the dimethylamine model (Figure 3.10), small variations in relative energy are observed for TS(XI/XII) and related intermediates with more pronounced increases observed for XIII and TS(XIII/XIV) (6.8 and 5.5 kcal/mol respectively). However, the most relevant change is the decrease in relative energy calculated for the β-hydrogen abstraction transition state TS(XV/XVI). At 40.0 kcal/mol, TS(XV/XVI) is significantly lower in energy than the protonolysis transition state TS(XIII/XIV) (47.7 kcal/mol) indicating that protonolysis is the turnover limiting step when N-methylaniline is used as the amine substrate. Although this result is still not aligned with the experimental evidence, it is interesting to observe how changes in the substrates can affect the nature of the potential energy surface.  138  Figure 3.12 Potential energy surface (PES) for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol.  139  3.2.3.2  Oamidate cis to the Naziridine In line with observations for the dimethylamine system, a reversal of relative geometry is  observed for the azametallapropane moiety with respect to the amidate chelate in tantalaziridine XVII. Whereas the Namidate is cis to the Naziridine for the tantalaziridine X at the beginning of the catalytic cycle, the reverse is found for XVII (Oamidate is cis to the Naziridine, Figure 3.13). Notably, the azacyclopropane ring of XVII is twisted out of the equatorial plane approximately 30° and the Ta–Namidate contact (2.65 Å) is longer than that found for X (2.24 Å) to allow for the axial dimethylamido ligand to move towards the equatorial plane.  This out-of-plane  metallacycle geometry is similar to that observed with tantalaziridine IX.  Figure 3.13 Optimized geometries for tantalaziridines X (left) and XVII (right). Hydrogen atoms omitted for clarity.  Starting from the geometry of XVII, the catalytic cycle was modeled beginning with the insertion of 1-octene into the Ta–C bond of the tantalaziridine. Based on the earlier findings that reactivity in the plane of the amidate ligand is the lowest energy pathway, the dimethylamido ligands were repositioned into the axial plane allowing the alkene molecule to approach equatorially. The corresponding transition state (TS(XVIII/XIX), Figure 3.14) was found to 140  have a negative vibrational mode showing the insertion of the C=C double bond into the Ta–C bond of the tantalaziridine.  Figure 3.14 Optimized geometries for intermediate XVIII (left) and transition state TS(XVIII/XIX) (right). Hydrogen atoms omitted for clarity.  In contrast to XI, the alkene remains coordinated to the tantalum centre upon relaxation of TS(XVIII/XIX) to intermediate XVIII (Figure 3.14). The Ta–Calkene distance of 2.58 Å is just outside the sum of the covalent radii (2.46(8) Å) but well within the van der Waals radii (3.90 Å) of the two atoms, suggestive of interaction between the alkene and the metal centre.131132  This result marks the first calculated structure displaying pre-coordination for either the  alkene or amine substrate. As a result, only a small energy difference (2.6 kcal/mol) between XVIII and TS(XVIII/XIX) is observed. Gratifyingly, an optimized tantalaziridine structurally indistinguishable from XVII is found upon removal of the coordinated 1-octene confirming that XVII and TS(XVIII/XIX) are on the same reaction pathway. Though no longer adjacent to the bulky aryl group of the amidate ligand, the planar phenyl group bound to the Naziridine still occupies a considerable amount of space in the equatorial plane. Consequently, the amidate ligand in the 5-membered metallacycle XIX is bound in a 141  κ1(O) fashion (Ta–Namidate bond length of 3.43 Å), a binding mode is also observed for TS(XX/XXI) (Figure 3.15) and the bis(amido) intermediates XXI and XXII. The κ1(O) binding of the amidate results in a vacant coordination site, allowing the N-methylaniline substrate to bind to the tantalum centre (Ta–Namine = 2.52 Å) before proton transfer takes place (XX, Figure 3.15).  Figure 3.15 Optimized geometries for intermediate XX (left) and transition state TS(XX/XXI) (right). Most hydrogen atoms omitted for clarity.  In contrast to TS(XV/XVI), a κ2(N,O) amidate binding mode is observed for the βhydrogen abstraction transition state TS(XXII/XXIII). Relaxation of TS(XXII/XXIII) gives intermediate XXIII, a tantalaziridine species with a κ2(N,O) bound amidate ligand and an equivalent of amine 6 datively bound to the tantalum centre. Optimization after elimination of the amine product yields tantalaziridine X, with the Namidate cis to the Naziridine, thus resuming the pathway defined by the PES illustrated in Figure 3.12.  The PES describing the pathway  beginning with tantalaziridine XVII, with the Oamidate cis to the Naziridine, is depicted in Figure 3.16.  142  Figure 3.16 Potential energy surface for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol. 143  Despite the pre-coordination of the alkene to the metal centre, the activation barrier (ΔG‡) for TS(XVIII/XIX) (28.3 kcal/mol) is marginally larger than that calculated for TS(XI/XII) (24.6 kcal/mol). However, the barriers for the protonolysis and β-hydrogen abstraction transition states are significantly lower (26.5 and 23.9 kcal/mol respectively) than those calculated for TS(XIII/XIV) (34.5 kcal/mol) and TS(XV/XVI) (38.2 kcal/mol), in better agreement with experimental observations. Rotation of the N-methylanilido group into the equatorial plane (intermediate XXI to XXII) results in a 8.3 kcal/mol decrease in energy in contrast to the 3.7 kcal/mol increase in energy observed for intermediates XIV and XV. Notably, the lowest energy transition state on the PES is TS(XXII/XXIII), describing the β-hydrogen abstraction step. Disappointingly, protonolysis of the 5-membered metallacycle is still found to be the turnover limiting step despite the reversal of relative tantalaziridine geometry. Due to the additional carbon and hydrogen atoms present in calculations using Nmethylaniline instead of dimethylamine as a substrate, the energies calculated for the dimethylamine system (Figure 3.10) cannot be directly compared with those generated using Nmethylaniline as a substrate. However, since the only difference between the two pathways that use N-methylaniline as a substrate, described in Figure 3.12 and Figure 3.16, is a matter of geometry, the absolute energies of the calculated structures can be compared directly. A plot of the two pathways, with energies relative to tantalaziridine X, is shown in Figure 3.17.  144  Figure 3.17 Comparison of the two pathways found for the α-alkylation of N-methylaniline with 1-octene catalyzed by 3. Relative free energies (ΔG) are reported in kcal/mol.  145  Intermediate XVII is found to be significantly higher in energy (8.7 kcal/mol) than tantalaziridine X, supporting the notion that deviation from the equatorial plane results in higher energy species. Compounding this fact with the aforementioned increase in activation barrier results in transition state TS(XVIII/XIX) being 12.5 kcal/mol higher in energy than the corresponding TS(XI/XII). Both pathways contain 5-membered metallacycle that are higher in energy than the intermediate tantalaziridines or bis(amido) species found along the respective pathways. Notably, the absolute energies of the protonolysis transition states, TS(XIII/XIV) and TS(XX/XXI), are almost identical, suggestive of a potential cross-over point between these two pathways (vide infra). However, when the relative energies of each pathway are considered, the barrier calculated for TS(XX/XXI) is 8.1 kcal/mol lower than that for TS(XIII/XIV), suggestive of a more favourable pathway. Relaxation to the bis(amido) intermediates XIV and XXI results in a large decrease in energy, where rotation of the methyl group into the plane of the amidate results either in an increase (XV) or decrease (XXII) in energy.  Interestingly, the activation barrier of 38.2  kcal/mol for TS(XV/XVI) is considerably higher than the 23.9 kcal/mol barrier calculated for TS(XXII/XXIII). These barriers are particularly significant as both the computational study by Doye and coworkers and the kinetic studies of Hultzsch and coworkers predict β-hydrogen abstraction to be the rate limiting step in the catalytic cycle. The relaxation of each of these transition states yields the tantalaziridine for the beginning of the other pathway; the end of one pathway is the start of the other. Interestingly, an apparent 10.2 kcal/mol change in energy is seen for tantalaziridines X and XVII depending on whether they are found at the beginning or the end of the pathways in Figure 3.17. This discrepancy arises from the energy difference between the reactants (1-octene and N-methylaniline) and the α-alkylation product 6.  146  Though neither pathway agrees with the experimentally determined turnover limiting step, it is obvious that small changes in the geometry of the catalyst can largely affect the PES for the catalytic cycle. For instance, reactivity in the plane of the amidate ligand is found to be the lowest energy pathway. Additionally, a reversal of the tantalaziridine geometry relative to the amidate ligand results in the lowering of activation barriers to generate a shallower, and therefore more desirable, potential energy surface.  With the knowledge that these small  variations could cause large changes in the energetics of a system, other pathways were considered to identify rectify the disconnect between the quantum chemical calculations and experimental observations.  3.2.3.3  Alternative Pathways using N-methylaniline as a Substrate Based on the insignificant energy difference between the protonolysis transition states for  the two pathways, the possibility of a cross-over point in the PES was explored. Ideally, this cross-over point would combine the first portion of the pathway beginning with tantalaziridine X (X to XII) and then transition to the low-energy section of the second pathway (XXI to XXIII) to regenerate tantalaziridine X and continue along the cycle. Such a pathway may reduce the energy of the protonolysis transition state making alkene insertion the turnover limiting step, thereby aligning experimental and computational findings. In order for the cross-over to occur, a switch in relative tantalaziridine geometry is necessary after alkene insertion and before protonolysis.  To this end, input structures based on 5-membered metallacycle XIX were  generated by rotating the amidate ligand 180° (isostructural with XII) in 10° increments to search for an intermediate with the amidate ligand orthogonal to the metallacycle. Due to the interaction of the steric bulk of the N-aryl moiety with the axial dimethylamido ligand, the  147  amidate was initially modeled with a κ1(O) binding mode. While rotation less than 80° and greater than 110° optimized to XIX and XII respectively, rotating the amidate ligand 90 or 100° led to a minimum with a pseudo-trigonal bipyramidal geometry (Figure 3.18). In this case however, the equatorial plane is occupied by the Oamidate, a dimethylamido ligand and the carbon contact of the metallacycle, leaving the nitrogen metallacycle contact in the axial plane with the other amido ligand.  Figure 3.18 Optimized geometry for a 5-membered metallacycle intermediate with a κ1(O) amidate ligand orthogonal to the metallacycle. Hydrogen atoms omitted for clarity.  Interestingly, this intermediate is lower in energy than XII (5.5 kcal/mol) and XIX (15.6 kcal/mol) making it the first structure with an out-of-plane amidate ligand to have lower energy than its in-plane equivalents. Based on this promising geometry, a protonolysis transition state is found with the proton transfer taking place orthogonal to the plane of the amidate N,O atoms (Figure 3.19). Unlike the orthogonal intermediate, the dimethylamido ligands occupy both axial positions in the transition state and are aligned with the N,O amidate plane. Relaxation of this transition state gave the intermediate in Figure 3.18 and bis(amido) intermediate XXI, proving the low-energy section of the second pathway (XXI to XXIII) is accessible via this route. 148  Unfortunately, in line with previous observations, the out-of-plane transition state was found to be slightly higher in energy than TS(XIII/XIV) and TS(XX/XXI) by 7.6 and 6.7 kcal/mol respectively.  Figure 3.19 Optimized geometry for a protonolysis transition state with a κ1(O) amidate ligand orthogonal to the N,O amidate plane. Most hydrogen atoms omitted for clarity.  Working backwards from the orthogonal intermediate, a transition state was found modeling the insertion of 1-octene into a tantalaziridine Ta–C bond perpendicular to the plane of the N,O of the amidate ligand (Figure 3.20). Regrettably, the transition state was found to be 7.8 kcal/mol lower in energy than the corresponding protonolysis transition state. Despite having a smaller difference in transition state energies than those found in Figure 3.17, these results suggest protonolysis is the turnover limiting step for this pathway as well.  149  Figure 3.20 Optimized geometry for an alkene insertion transition state with a κ1(O) amidate ligand orthogonal to the N,O amidate plane. Hydrogen atoms omitted for clarity.  Thus far, dimethylamido ligands have been used as ancillary ligands to model each calculated structure. With a catalyst loading of 5 mol%, a large excess of amine substrate is present in solution and in situ amide exchange reactions cannot be ruled out. To this end, both pathways were re-optimized substituting one or both of the dimethylamido ligands for Nmethylanilido ligands derived from the N-methylaniline substrate. Due to the asymmetric nature of the calculated species, substitution of one dimethylamido ligand could result in two different tantalum complexes (top and bottom, Figure 3.21). In spite of the increased steric demands of the arylamido ligands, relative energies for the N-methylanilido substituted species are found to be comparable to their dimethylamido ligated counterparts along both pathways (Figure 3.21) suggesting that the nature of the axial amido ligands has little effect on reactivity in the equatorial plane.  150  Figure 3.21 Potential energy surfaces showing the effects of axial amido substitution. Relative free energies (ΔG) are reported in kcal/mol. Colouration: none = black dash, bottom = red, top = blue, both = green. 151  Unfortunately, the potential energy surfaces presented in this section do not agree with the experimental evidence discussed in Chapter 2.  This could be due to intermolecular  interactions with substrates or a more intricate mechanism where the amidate ligand plays an active role that was not explored in this study.  While these possibilities have yet to be  investigated, the modeling of this catalytic cycle has extended the knowledge of amidate ligands that will be useful in the development of further generations of catalysts for the α-alkylation of amines. Perhaps the most important finding of these computational investigations is the ability of amidate ligands to adopt a κ1(O) binding mode as required to alleviate steric strain or open a coordination site in both intermediates and transition states. This binding mode flexibility could explain the broad substrate scope observed for 3 compared to other α-alkylation of amines catalysts.  3.2.4  Triplet Species as Possible Transition States and Intermediates To complement the experimental investigations in Section 2.2.5, the possibility of a one  electron mechanism was explored using computational methods. Whereas the aforementioned tantalum species were optimized in the singlet state, both tantalaziridine intermediates and alkene insertion transition states were optimized in the triple state to probe for the possibility of radical behaviour. All three tantalaziridines were optimized in this fashion to explore the effects of tantalaziridine substitution and relative geometry on the resultant triplet species. Interestingly, the N-substituent of the tantalaziridine is bent approximately 40° out of the plane of the amidate chelate (tantalaziridine X shown as a representative example in Figure 3.22). This bending is accompanied by a lengthening of the Ta–Caziridine and Ta–Naziridine bonds an average of 0.22 and 0.24 Å respectively. Additionally, the Caziridine–Naziridine bond length is  152  shortened from 1.43 Å to 1.37 Å, suggestive of more double bond character. This notion is supported by the increase in H–Caziridine–H bond angle from 111° to 117°, bringing it closer to 120° expected for sp2 centres. Based on these changes, the tantalaziridine moieties are best described as η2-imines in the triplet state.  Figure 3.22 Optimized geometry of tantalaziridine X in the triplet state. Select carbon atoms from the amidate backbone and all hydrogen atoms have been removed for clarity.  A study of the molecular orbitals shows the highest occupied molecular orbital (HOMO) for each tantalaziridine represents a single electron in a tantalum d-orbital, suggesting that most of the unpaired electron density is located on the d1 Ta(IV) centre (X is shown as a representative example in Figure 3.23). The HOMO-1 is also a singly occupied molecular orbital (SOMO) and shows significant amounts of unpaired electron density on both the Ta–C and Ta–N bond in the tantalaziridine moiety. This would suggest that radical reactivity at these sites is possible. However, radical species are often short lived in solution and thus triplet tantalaziridines are not likely stable intermediates. Support for this view is found in the free energies: the triplet versions of tantalaziridines X and XVII are 77.3 and 68.6 kcal/mol higher in energy than their singlet counterparts! 153  Figure 3.23 Molecular orbital surfaces for the SOMOs (HOMO, top left; HOMO-1, top right) calculated for the triplet version of X. A view without molecular orbitals is given for reference (bottom).  Though triplet intermediates are energetically unfavourable, the possibility of radical character in the shorter-lived transition states must be explored.  As alkene insertion is  considered to be the most likely single electron process (vide supra), transition states TS(III/IV), TS(XI/XIII), and TS(XVIII/XIX) were optimized in the triplet state.  Unlike the  aforementioned tantalaziridines, no significant change in geometry is observed between the singlet and triplet species. Relaxation of the triplet transition states in the singlet state yields the corresponding  intermediates  previously  observed  for  TS(III/IV),  TS(XI/XIII),  and  TS(XVIII/XIX). This result suggests that the triplet transition states are located on the same 154  pathway along the potential energy surface as their singlet counterparts. However, the SOMOs associated with the triplet transition states do not display any unpaired electron density on either the C=C unsaturation or the Ta–C bond (TS(XI/XII) shown as a representative example in Figure 3.24). The absence of unpaired electron density on the reactive sites of the transition state suggests that the insertion of the 1-octene into the tantalaziridine bond is not a radical process.  Figure 3.24 Molecular orbital surfaces for the SOMOs (HOMO, top left; HOMO-1, top right) calculated for the triplet version of TS(XI/XII). A view without molecular orbitals is given for reference (bottom).  155  The calculated energies of the system are in good agreement with this assessment as the triplet species are again found to be vastly higher in energy than the singlet transition states, with differences of 65.8 and 57.0 kcal/mol calculated for TS(XVIII/XIX) and TS(XI/XIII) respectively. An analogous value for TS(III/IV) could not be calculated due to convergence failure for the self-consistent field, regardless of the number of cycles. Upon review of both the experimental and computational data amassed, it can be stated that radical processes in the catalytic cycle of the α-alkylation of amines using precatalyst 3 are not favoured. Though still unable to reconcile the in silico data with the experimental evidence, certain assumptions made during the modeling process must be considered.  For example, each  calculation takes into account only 1 discrete metal complex and associated organic fragment, if applicable. However, a large excess of amine and alkene substrates are present in solution, heightening the possibility of intermolecular processes taking place during the catalytic cycle. Based on the evidence presented herein, changes in the relative geometry of the tantalum transition states and intermediates can cause large variations in free energies. However, due to difficulties monitoring the tantalum species in situ, the structural identity of complexes present in solution cannot be confirmed. While potential energy surfaces were plotted based on complexes with fixed geometry, in solution at catalytic temperatures a number of different molecular orientations are possible. Indeed, the room temperature 1H NMR spectrum of 3 contains only 1 broad signal for the dimethylamido CH3 groups indicative of fast exchange the axial and equatorial positions on the NMR time scale. At higher, catalytically relevant temperatures, these types of dynamic processes should be even more facile. Unfortunately, modeling each possible geometric isomer for the catalytic species along the postulated PES is unfeasible.  156  Another possible avenue of inquiry is the participation of the amidate ligand in the mechanism. Tantalum amidate complexes have been shown to adopt both κ1(O) and κ2(N,O) binding modes both experimentally and computationally with species such as complex 7 displaying dynamic exchange between the two isomers. Computationally optimized complexes have been also shown to adopt the κ1(O) binding mode to relieve steric bulk or make a coordination site available as necessary.  As mentioned in Chapter 2, the possibility of a  mechanism with concerted metalation-deprotonation aided by the Namidate cannot be discounted and remains an active topic of research.  3.2.5  Development of a Steric Parameter for Tantalum Precatalysts The quantitative evaluation of the steric demand of a particular ligand is an exercise that  has been utilized and improved upon for decades. Through synthetic methods, more or less bulk can be applied to a system and corresponding trends in properties can be demonstrated. However, the truly powerful notion of these relationships is the possibility of developing a predictive tool based on reliable, quantitative steric parameters. To this end, a number of methods including the Tolman cone angle and Taft-Dubois steric parameter have been developed.133-134 These techniques have proven to be valuable predictive tools but have been developed with a specific compound or reaction target in mind (PR3 ligands for Tolman cone angles, rates of esterification and hydrolysis of esters for Taft-Dubois steric parameters). Recently, two new techniques have emerged that offer the opportunity to calculate a value describing the steric bulk of ligands bound to metal atoms, namely buried volume and solid angle.135-136 These methods are straightforward to use, accept multiple input formats, and are available to the entire chemical community to use free of charge.137-138  157  Buried volume (%Vbur) is determined by defining the metal atom in the complex as the centre of a sphere. From this point in space, a sphere with a radius defined by the user is drawn and the amount of the sphere’s volume occupied by the ligand is calculated (Figure 3.25). This value is calculated for one ligand and returned as a percentage and can only be calculated for one ligand at a time. A web based implementation of this technique, SambVca, was developed by Cavallo and coworkers.137, 139 Initially developed for and applied to symmetric N-heterocyclic carbene ligands,135,  140-145  the technique has since proven useful for a range of organometallic  ligands.146-149  Figure 3.25 Sphere with radius of arbitrary size drawn from the centre of the tantalum atom in complex 3. The buried volume is the amount of sphere’s volume occupied by the ligand.  The solid angle parameter is determined by treating the central metal atom as a point which emanates light from the centre of a sphere containing the molecule. The ligands will block a certain amount of light from reaching the surface of the sphere and a solid angle value (G) is calculated from this “shadow” (Figure 3.26). A G value is assigned to each ligand (GL) in 158  the molecule as a percentage and a visualization of the “shadows” is also part of the output. These calculations are performed using the downloadable application Solid-G, written and designed by Guzei and Wendt.136-137 G values have shown promise in rationalizing the steric effects of a catalytic system and how they influence the rates of reaction.150-152  Figure 3.26 Solid-G output for complex 3. Shadow for amidate ligand shown in blue. Shadows for dimethylamido ligands shown in green, yellow, purple, and red (not visible).  The buried volume and solid angle have been calculated for each of the ligands presented in Figure 2.5 using coordinates derived from single crystal X-ray analysis in an attempt to relate the steric parameters to the catalytic activity of the respective complex.  While the %Vbur  technique offers more control in evaluating the sterics of the ligand at specific distances from the metal centre, the solid angle calculations offer information about the complex as a whole because of the inclusion of the dimethylamido ligands. While any description of a multidimensional object with one parameter will be inherently flawed, the use of both techniques in combination  159  serves to generate a more accurate model of the sterics effects pertaining to these related systems. As the %Vbur technique was initially developed for N-heterocyclic carbene ligands, the input parameters are based on a single nitrogen atom bound to the metal centre with two carbon atoms bound symmetrically to the nitrogen. However, due to the asymmetric binding of amidate ligands, the centre of the sphere must be mathematically defined using a “dummy atom”.153 Starting from the solid-state molecular structure, the amidate ligand is rotated 180° about the Namido–Ta–Oamidate–Camidate dihedral using a molecular modeling program such as GaussView. The new Cartesian coordinates for the Camidate are recorded and the ligand is returned to its initial geometry. Each atom not associated with the amidate ligand is deleted and the file is saved such that the Cartesian coordinates for each atom are listed. The “dummy atom” is then added manually at the end of the file and used to define the placement of the metal centre. The %Vbur for the amidate ligands found in complexes 1 – 5 were plotted at varying distances from the tantalum centre (Figure 3.27). At distances close to the metal centre, the κ2(N,O) amidate ligands show similar steric demand while 5 occupies much less space due to the κ1(O) binding mode. Further from the metal centre, 1 is significantly less demanding than the ligands substituted at the 2,6-positions of the N-aryl substituent. Interestingly, 2 and 3 are virtually sterically identical below 4 Å, a trend which is reflected in their similar reactivity (Table 2.2).  Substituting a planar phenyl group (4) for the tBu fragment on the amidate  backbone results in a substantial decrease in sterics %Vbur at distance greater than 3.5 Å.  160  Figure 3.27 Buried volumes (%Vbur) calculated for mono(amidate) tantalum precatalysts.  Unlike the %Vbur parameter, the Solid-G program evaluates all the ligands in the coordination sphere, generating other useful steric information. While GL describes the solid angle of each individual ligand, GTa describes the percentage of the tantalum atom’s surface shielded by the ligated atoms from which the available space in the coordination sphere may be calculated (Gspace). A plot of these two parameters is shown in Figure 3.28. As expected, the GL increases as the substituents at the 2,6-positions of the amidate ligand increase in steric bulk. This is in line with catalytic screening results, except for the κ1(O) bound 5. However, this does not translate directly to a decrease in space in the tantalum coordination sphere, probably due to the steric bulk imposed by the four dimethylamido ligands. The Gspace parameter is also in good agreement with the observed reactivity of the precatalysts with the exception of 4, though this increase in Gspace may be a factor in the lack of thermal stability observed.  161  Figure 3.28 Solid-G parameters GL and Gspace calculated for mono(amidate) tantalum precatalysts.  Though no ideal value for buried volume or solid angle could be extrapolated from the mono(amidate) tantalum data, trends relating the steric parameters to catalytic reactivity were investigated. From the experimental data it is clear that a κ2(N,O) bound amidate is necessary as well as bulk at the 2,6-position of the N-aryl substituent. With these trends in mind, the steric parameter evaluation was expanded to ligands used for the enantioselective α-alkylation of amines. The set of crystallographically characterized complexes with reported catalytic activity data is larger than that of the previously discussed mono(amidate) complexes. This study focuses solely on reactivity and does not consider enantioselectivities. Figure 3.29 depicts the axially chiral bis(amidate) complexes developed in the Schafer group (Dr. Rashidat Ayinla and Neal Yonson) and by Zi and coworkers,43 as well as a binaphtholate-ligated precatalyst synthesized by Hultzsch and coworkers,45 all of which were  162  evaluated using SambVca and Solid-G starting from their solid state molecular structures. For ease of interpretation, the complexes are segregated into groups based on laboratory and listed in order of decreasing reactivity (from left to right). Turnover frequencies and enantioselectivities are also reported for the α-alkylation of N-methylaniline by 1-octene in Table 3.1. At this time these ligands are being evaluated for reactivity only, not enantioselectivities.  Figure 3.29 Axially chiral tantalum precatalysts for the enantioselective α-alkylation of amines.  163  Table 3.1 Reactivity data for a series of axially chiral tantalum precatalysts.  Entry Precatalyst  Yield (%)  Turnover Frequency (h-1)  ee (%)  1  18  48  86  0.179  44  2  19  48  79  0.165  5  3  20  72  80  0.111  21  4  21  72  62  0.086  5  5  22  38  28  0.074  1  6  23  44  11  0.025  n/a  7a  24  120  86  0.143  63  8a  25  120  79  0.132  70  9a  26  120  76  0.127  60  12  79  1.317  49  a,b  10 a  Time (h)  27 b  5 mol% catalyst. 150 °C  A plot of the %Vbur values calculated for each complex is shown in Figure 3.30.129 The most reactive bis(amidate) complex, 18, occupies the least amount of space close to the metal centre, consistent with the κ2(O,O) binding mode observed. Complex 18 also possesses the only amidate ligand that shows an increase in %Vbur further from the tantalum atom. This can be rationalized from the added bulk of the iPr groups which will occupy more space than the methyl groups of 19 and 20. For comparison, the geometry for the κ2(O,O) isomer of 20 (20-κ2), observed by Dr. Rashidat Ayinla using variable temperature NMR spectroscopy,98 was optimized using the aforementioned quantum chemical methods based on the crystallographic data for 18. An optimized geometry was located and the %Vbur values presented herein are based on these calculated coordinates. The data is comparable to 18 from 2.5 – 3.1 Å but as 18 starts to 164  increase in steric bulk, 20-κ2 continues to decrease until 3.7 Å. This result suggests that while too much steric hindrance around the metal centre retards catalytic activity, some degree of bulk promotes efficient catalysis.  Figure 3.30 Buried volumes (%Vbur) calculated for axially chiral tantalum precatalysts.  Ligands 19, 20, 21, and 22 seem to follow a general trend as well: the bulkier ligands form more reactive hydroaminoalkylation catalysts.129 Though the steric parameters of 23 are similar to those of 19, the reactivity is much lower. This suggests that although sterics are a dominant factor for determining catalytic efficiency, electronic effects perturb reactivity.40 The precatalysts synthesized by Zi and coworkers (24 – 26) vary primarily in the substitution of the biaryl backbone and therefore display similar steric properties at distances up to 3.5 Å, reflected in the similar observed reactivities. For this reason, only data for the most reactive complex, 24, 165  is plotted. Interestingly, the SiMe3-substituted binapthalate system 27 shows a novel steric pattern with only small deviations (range for 2.5 – 4.0 Å: 41.2 – 41.6%) in the %Vbur as the distance from the metal is increased. Unfortunately, no reactivity data is available at 130 °C making direct comparison to the bis(amidate) species difficult. Investigating the GL and Gspace parameters generated for the enantioselective precatalysts (Figure 3.31), 18 is found to be a notable outlier of the broader trends. The GL for 19, 20, 21, and 22 decreases as reactivity of the corresponding complex decreases.129  However, 18  (45.43%) has a lower GL than 19 (46.27%) but larger than 20 (45.13%). While these differences may seem insignificant, a recent report noted a 143 fold increase in reaction rate with a GL difference of only 1.6%.151 Complex 18 has the most available Gspace, probably owing to the κ2(O,O) binding of the bis(amidate) ligand. This notion is supported by the calculated geometry of 20-κ2, whose Gspace is larger (16.15%) than that for 18 (15.80%) presumably due to the less sterically demanding methyl groups at the 2,6-positions. Generally, an increase in available Gspace correlates to a decrease in reactivity. It is proposed that steric shielding of the metal centre (small Gspace values) prevents the formation of catalytically inactive species, such as multimetallic complexes, formed via intermolecular pathways. Clearly, 18 has achieved a balance between sterically congesting the metal centre and imposing enough bulk for efficient reactivity.  166  Figure 3.31 Solid-G parameters GL and Gspace calculated for axially chiral tantalum precatalysts.  As with the %Vbur data, the brominated biphenyl complex 23 does not follow the observed trends in reactivity. When the data for complexes 24 – 26 are plotted the opposite trend emerges: an increase in GL value corresponds to a less efficient catalyst.129 Notably, the κ2(O,O) nature of the binaphtholate complex 27 results in similar data to 20-κ2, namely small GL value (43.12%) and large Gspace (17.10%). As previously mentioned, there are a number of issues with performing analyses using calculated steric data. Specifically for buried volume and solid angles, the input geometry is generated from solid-state data and catalytic reactions are run in the solution phase. However, the purpose of this exercise is to correlate trends in reactivity to steric parameters and use those conclusions to design new catalysts. Unfortunately, the steric parameters cannot currently be applied to the observed enantioselectivities but elucidating new trends from the data is a current avenue of research. Coupled with the ability to design metal complexes in silico, calculating 167  steric parameters could be a powerful tool in the design of tantalum precatalysts for the αalkylation of amines.  3.3  Conclusions A catalytic cycle based on the proposed mechanism for the α-alkylation of amines has  been modeled using quantum chemical calculations. The C–H activation of a dimethylamido ligand to form the catalytically active tantalaziridine was found to occur most favourably in the plane of the amidate chelate. Alternative routes with a κ1(O) bound amidate ligand or activation of an axial dimethylamido were discounted due to higher energy transition states and intermediates. From this tantalaziridine geometry, transition states modeling the insertion of 1octene into the Ta–C bond, protonolysis by an amine substrate, β-hydrogen abstraction to release product 12, and reformation of the catalyst were optimized. Relaxation of these transition states gave chemically relevant intermediates and the potential energy surface was plotted (Figure 3.10), revealing β-hydrogen abstraction to be the turnover limiting step. As with tantalaziridine formation, reactivity in the equatorial plane was identified as the lowest energy pathway. Based on the transition states found for the aforementioned model system, the cycle was re-optimized with a phenyl group as the N-substituent on the tantalaziridine to simulate the use of N-methylaniline as the amine substrate. Interestingly, the amidate ligand was found to adopt a κ1(O) binding mode as various stages of the catalytic cycle, presumably due to the increased steric demands of the phenyl group. Plotting of the PES (Figure 3.12) showed protonolysis to be the turnover limiting step. Notably, the intermediate regenerated at the end of the cycle was found to have the reverse geometry of the κ2(N,O) amidate ligand relative to the tantalaziridine moiety. Starting from this tantalaziridine, another pathway along the PES was calculated (Figure  168  3.16). Though a number of geometric and energetic differences exist between the two pathways, protonolysis remains the turnover limiting step. In an attempt to reconcile experimental and computational data, a cross-over point between the two pathways was sought. Though initially promising, this route proved to be unfavourable due to higher energy transition states. Replacing the axial dimethylamido ligands with N-methylanilido ligands was found to have little effect on the PES. In order to supplement the experimental data of Section 2.2.5, the possibility of radical processes in the catalytic cycle was explored computationally. While triplet tantalaziridines show unpaired electron density on the azametallacyclopropane moiety, the calculated energies suggested they are much less stable than their singlet counterparts. The same trend is observed for alkene insertion transition states optimized in the triplet state, with calculated energy differences of greater than 57 kcal/mol. Though relaxation to relevant singlet intermediates is observed, no unpaired electron density is observed on the alkene or the tantalum-carbon bond contradicting the likelihood of radical reactivity. Preliminary efforts were made to establish a steric parameter for amidate ligands and correlate the values to catalytic activity. An effective method for generating %Vbur input files for asymmetric amidate ligands was developed and applied to a series of tantalum amidate complexes. The experimentally derived notion that having κ2(N,O) bound amidate ligands with substantial steric bulk is necessary for effective catalysts was supported in the trends observed for the %Vbur, GL, and Gspace values calculated for mono(amidate) precatalysts. These values were also calculated for a series of axially chiral bis(amidate) catalysts that displayed a different set of trends with significant differences observed for κ2(N,O) and κ3(N,O,O) complexes with identical ligands. Unfortunately, the equilibrium between the κ1(O) and κ2(N,O) binding modes  169  observed in the solution phase cannot be factored in to the calculation of these steric parameters, limiting the value of these approaches. Nevertheless, this marks the first application of this type of steric analysis to asymmetric, bidentate amidate ligands and is an important stepping stone towards using predictive tools in catalyst design.  170  3.4 3.4.1  Experimental Computational Methods Geometric optimization of transition states and local minima were confirmed using  vibrational analysis: transition states having one imaginary (negative) vibrational mode and minima having only positive frequencies.  Intermediates along the catalytic pathway were  located by relaxing the transition state in either direction along the PES and confirmed using intrinsic reaction coordinate (IRC) analysis. The initial geometries for the arbitrarily named “forward” and “reverse” relaxed intermediates are generated by multiplying the Cartesian displacement values of the imaginary vibrational mode by a scale factor and then adding them to the transition state geometries. These calculations are performed under the default conditions in the Gaussian program (298.15 K, 1 atm). No solvent models are added at this time (vide infra). While DFT is a fast and accurate method of optimizing molecular geometries, ground state energies calculated using DFT have been shown to be less accurate than those calculated with ab initio methods such as Møller-Plesset perturbation theory. Therefore, for this study all calculations at single points along the PES (describing either intermediates or transition states) were performed using MP2, a second-order implementation of Møller-Plesset perturbation theory,128 and the aforementioned 6-31G(d,p) basis set. At this stage, an integral equation formalization polarizable continuum model (IEFPCM) is introduced to mimic toluene solvation based on an empirically derived dielectric constant. No discrete solvent molecules are added to the model. Optimizations are not performed with PCM solvation due to the increase in model complexity and therefore computational resources required. In order to show that gas-phase optimization has a negligible effect on the calculated catalytic cycle, single point MP2 calculations were performed without the solvent model added. Non-solvated transition states  171  and intermediates were found to be higher in energy by 1.21 – 2.59 kcal/mol, suggesting no significant solvation effects are observed. This is also confirmed experimentally, as replacing toluene with the more polar trifluorotoluene results in no change in reaction rate. The free energies reported herein are the summation of the MP2 single point energy and the thermal correction to Gibbs free energy calculated during vibrational analysis.  The  temperature dependant nature of this value necessitates frequency calculations to be performed on optimized structures at 110 °C to obtain a more realistic ΔG value. All ball-and-stick models of calculated structures presented in this chapter have been produced using GaussView5.154 Molecular orbital surfaces are generated from natural bond orbital (NBO) analysis,155 plotted at an isovalue of 0.05.  172  Chapter 4: Tethered Amidate and Ureate Ligands with a Neutral Donor Bridge  4.1 4.1.1  Introduction Tethered Diamido Ligands with Neutral Donors Diamido ligands with neutral donors built into the aryl tether were initially developed by  Schrock and coworkers to support discrete early transition metal catalysts for the polymerization of terminal olefins (left, Figure 4.1).156-161 Titanium and zirconium complexes supported by this class of ligand with R = tBu-d6 were found to be highly active as polymerization catalysts as well as more robust than the established metallocene-based systems.159  Figure 4.1 Diamido ligands with an aryl tether featuring a neutral oxygen donor.  Notably, the steric bulk of the R group has been found to affect the binding geometry of group 4 systems. Whereas the R groups of the diamido ligand are pointing towards the reactive R2 ligands in the meridional isomer (mer, Scheme 4.1), maximizing steric interactions with the ancillary ligands, the facial isomer (fac, Scheme 4.1) has the bulky R groups pulled away from 173  the metal centre, and thus the reative ligands, alleviating the unfavourable steric pressure on the system. As a result, the tBu-d6 substituted ligand adopts a facial binding mode while the less bulky iPr group are found to crystalize in a meridional fashion.160 However, Danopoulos and coworkers found that a secondary aryl tether could force sterically bulky diamido ligands to adopt the planar mer isomer when installed on titanium (right, Figure 4.1).162  Scheme 4.1 Two possible isomers for group 4 metal complexes supported by diamido ligands with an aryl tether featuring a neutral oxygen donor.  This disparity in binding mode corresponds to a change in reactivity: while the fac-bound zirconium system (R = tBu-d6) is found to catalyze the living polymerization of 1-hexene, the mer isomer (R = iPr) produces only oligomers under the same conditions.160 However, even at temperatures as low as -80 °C, the auxiliary ligands appear equivalent in the NMR spectra, suggesting rapid interconversion between the two isomers in solution.159 When the group 4 metal centres are exchanged for a larger yttrium atom, the tBu-d6-substituted diamido ligand crystalizes in a mer fashion, suggesting this isomer is preferred when sterically accessible.161 Schrock and coworkers have also synthesized diamido ligands using sulfur as the neutral donor in the aryl tether (Figure 4.2).163 When installed on zirconium, these ligands adopt the fac 174  binding mode in the solid state and do not rapidly isomerize to the mer binding mode in solution. Substituting the hard oxygen atom in the diamido tether for a softer sulfur donor makes the ligand more attractive for coordination to late transition metals, exemplified by the ruthenium complexes synthesized by Hidai and coworkers as precatalysts for the hydrogenation of benzonitrile.164 Additionally, Deng and coworkers have shown that the bulky mesityl substituted diamido ligand supports high spin iron(II) complexes.165  In line with the aforementioned  zirconium systems, the sulfur bridged diamido ligands adopt the fac binding mode in the solid state for both Ru and Fe complexes.  Figure 4.2 Diamido proligands with an aryl tether featuring a neutral sulfur donor.  Using the least sterically encumbered ligand in Figure 4.2 (R = H), Otero and coworkers have synthesized a number of tantalum complexes that crystallize in a fac κ3(N,N,S) arrangement.166 Interestingly, reaction of tert-butylisocyanide with a dihydride tantalum species supported by the sulfur bridged diamido ligand led to the double insertion into the Ta–H bonds to yield a tantalaziridine (Scheme 4.2). Though no solid state data is available, the spectroscopic  175  signals associated with the protons (δH = 1.72, 1.82) and carbon (δC = 65.7) of the metallacycle suggest a tantalaziridine moiety rather than an η2-imine.  Scheme 4.2 Synthesis of a tantalaziridine supported by a diamido ligand with a neutral sulfur donor.166  The Fryzuk group has enjoyed success in the field of dinitrogen activation using a similar diamido framework. Initial studies featured a macrocyclic ligand (left, Figure 4.3) which was shown to support a zirconium dinitrogen complex that could protonate the dinitrogen unit upon exposure to H2.167 However, the ligand was found to participate in unwanted side reactions,168 including silicon migration to the N2 unit,169 leading to the development of the aryl bridged ligand in Figure 4.3 (right).170  Figure 4.3 Diamine proligands featuring a neutral phosphorus donor.  176  When coordinated to zirconium, the diamido ligand adopts a fac binding mode (Figure 4.4),170 in line with observations for the zirconium systems synthesized by Schrock and coworkers (vide supra). Gratifyingly, reduction of a dichloro zirconium species supported by the diamido ligand in the presence of N2 results in a side-on bound dinitrogen unit bridging two Zr centres (Figure 4.4).171 Like the aforementioned macrocyclic system, exposure of a solution of the N2 complex to H2 results in protonation of the dinitrogen moiety as well as the installation of a bridging hydride (Figure 4.5). Notably, the phosphorus remains bound to the zirconium centre in both solution and the solid state for all synthesized compounds.170-171  Figure 4.4 Synthesis of a side-on bound dinitrogen unit bridging a zirconium dimer supported by a diamido ligand with a neutral phosphorus donor.171  177  Figure 4.5 Protonation of a side-on bridged dinitrogen moiety in a zirconium dimer supported by a diamido ligand with a neutral phosphorus donor.171  Diamido ligands with neutral donors built into the aryl tether have been shown to support a number of early and late transition metal complexes for both stoichiometric and catalytic reactivity. Interestingly, the Schafer group has not explored the use of tethers bearing a donor atom for bis(amidate) ligands. However, a recently reported alkyl-tethered bis(ureate) zirconium complex has been shown to be a broadly applicable hydroamination catalyst, while untethered bis(ureate) complexes show reduced reactivity.172 These observations suggest that a tether may dramatically impact catalytic activity. In an effort to build upon the success of the bis(ureate) zirconium precatalyst and to further explore the effects of the tether on catalytic systems, the monoanionic amido donors of the aryl-bridged framework were replaced with N,O-chelating functional groups, namely amides and ureas.  Installation of these ligands on zirconium and tantalum via protonolysis with  homoleptic dimethylamido precursors (Scheme 4.3) yields precatalysts similar to those discussed for the α-alkylation of amines and hydroamination of alkenes.172  However, this series of  178  precatalysts possess a neutral donor, either O or S, in the tether that can donate additional electron density to the metal centre as required. This additional electron density may further stabilize the electropositive early transition metal centres, enabling interesting reactivity.  Scheme 4.3 Synthesis of Zr and Ta complexes supported by a tethered bis(amidate) or bis(ureate) ligand with a neutral chalcogen donor.  4.1.2  Scope of Chapter This chapter focuses on a series of zirconium and tantalum complexes supported by  tethered bis(amidate) or bis(ureate) ligands that contain a neutral chalcogen donor in the backbone. The synthesis of the diamide (Section 4.2.1) and diurea proligands (Section 4.2.2) will be described as well as the synthesis of the resultant Zr and Ta complexes. The solid state and solution phase behaviour of the bis(amidate) complexes will be discussed in Section 4.2.1, including the need for additional donor ligands for the Zr complexes and the dynamic behaviour of the Ta species. Section 4.2.2 is focused on the bis(ureate) complexes with comparisons drawn between the chalcogen-tethered Zr species and the aforementioned alkyl-bridged bis(ureate) hydroamination precatalyst. Initial investigations into using electrochemical analyses to characterize these air- and moisture-sensitive early transition metal complexes will be explored in Section 4.2.3. 179  Section 4.2.4 describes the screening of the complexes as catalysts for the α-alkylation of amines as well as intramolecular hydroamination. Differences in activity between Zr and Ta species for the various reactions will be highlighted in addition to the influence of the tethered ligand.  4.2 4.2.1  Results and Discussion Tethered Bis(amidate) Complexes of Zr and Ta The general synthesis of the tethered diamide proligands is shown in Scheme 4.4. The  oxygen bridged dianiline is prepared by reducing the corresponding dinitro compound which itself is synthesized via nucleophilic aromatic substitution.173-174 The dianiline quickly oxidizes back to the dinitro compound, making storage of this synthon impractical.  As with the  proligands in Chapter 2, the dianiline reacts with two equivalents of the appropriate acid chloride to install the amide functionality. Interestingly, the commercially available sulfur congener of the easily oxidized dinitro compound is considerably more stable.  180  Scheme 4.4 Synthesis of diamide proligands.  After initial studies with the oxy-bridged system, the tBu substituted diamide was found to be the most promising based on the solubility of the proligand and the corresponding complex. As a result, only the tBu version of the thio-bridged diamide was synthesized. In solution, both ligands display C2 symmetry as evidenced by the single set of 1H NMR signals observed for both the tert-butyl groups and the aromatic protons. Despite the possibility of a delocalized π network between the bridged aryl groups, both ligands experience an out of plane twisting in the solid state (Figure 4.6), possibly due to crystal packing interactions. The metrical parameters for the bridged ligands are comparable except for the C–E bond length, with the C–O bond distances (1.386(3) Å average) being significantly shorter than the C–S contacts (1.786(1) Å average) as anticipated. This increase in bond length is accompanied by a more acute C–E–C bong angle for the thio-bridged diamide (102.26(6)°) compared to the oxygen congener (117.23(16)°). The longer C–S bond could offer more flexibility in the ligand and therefore different binding modes for amidate coordination. 181  Figure 4.6 ORTEP representations of the solid-state molecular structures for oxy- and thiobridged diamide proligands drawn at 50% probability for thermal ellipsoids. Only one part of a disordered tBu group for the oxy-bridged species is shown for clarity. Hydrogen atoms omitted for clarity.  When the oxy-bridged proligand is reacted with the homoleptic Zr(NMe2)4 in benzene-d6, clean conversion to one symmetric product is observed in the 1H NMR spectrum.175 Only one set of signals is observed for both the tBu (δH = 1.00) and aryl protons (δH = 6.77, 6.86, 6.99). In addition to the signal assigned to the dimethylamido protons at δH = 3.18, a broad signal attributed to a coordinated dimethylamine molecule at δH = 2.21 is observed.56 This suggests that a coordination site is being occupied by a neutral amine donor rather than the oxygen atom of the ligand tether. Validation of this hypothesis was found in the solid state data for the related complexes 28 and 29 (Figure 4.7).  These isostructural complexes are synthesized in the presence of  pyridine (Scheme 4.5), a more rigid donor than dimethylamine, which results in highly crystalline products.  The κ4(N,N,O,O) bis(amidate) complexes adopt a pseudo-trigonal  bipyramidal geometry with the amidate chelates occupying one equatorial and one axial coordination site.  This is in contrast to a previously reported κ4(N,N,O,O) axially chiral 182  bis(amidate) Zr complex where both amidate contacts are found in the equatorial plane.56 Interestingly, the Zr–Oamidate bond lengths for 28 and 29 (2.2444(15) Å average) are significantly shorter than the reported tethered bis(amidate) zirconium species (2.2846(37) Å average) while the Zr–Namidate contacts (2.3242(17) Å average for 28 and 29) are not statistically different.56 As predicted from the 1H NMR spectrum, there is no interaction between the zirconium atom and the chalcogen donor in the ligand (Zr–O ≈ 3.813; Zr–S ≈ 3.962 Å).  Scheme 4.5 Protonolysis synthesis of bis(amidate) zirconium complexes 28 and 29.  Figure 4.7 ORTEP representation of the solid state molecular structure of 28 (left) and 29 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  183  Table 4.1 Selected bond lengths (Å) and angles (°) for complexes 28 and 29. 28  29  Zr–O1  2.2531(15)  2.2512(13)  Zr–O2  2.2415(14)  2.2320(13)  Zr–N1  2.3194(17)  2.3179(15)  Zr–N2  2.3242(17)  2.3354(15)  Zr–N3  2.1006(18)  2.0911(16)  Zr–N4  2.0822(18)  2.0894(16)  Zr–N5  2.4780(18)  2.4789(16)  O1–Zr–N1  56.71(6)  57.03(5)  O2–Zr–N2  57.32(6)  57.32(5)  The bridged bis(amidate) ligands can also be installed on tantalum via protonolysis as outlined in Scheme 4.6. After a few minutes, a white precipitate is observed in the yellow solution of the reaction mixture. This solid is found to be insoluble in hydrocarbon solvents and is filtered off before recrystallization of the soluble material. Crystals suitable for study by X-ray diffraction are grown from a saturated hexanes solution at -35 °C and the resultant solid state molecular structure is shown in Figure 4.8. The insoluble nature of the byproduct is suggestive of aggregate species bridged by one or more amidate ligands.  Scheme 4.6 Protonolysis synthesis of bis(amidate) zirconium complexes 30 and 31.  184  Figure 4.8 ORTEP representation of the solid-state molecular structure of 30 (left) and 31 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  Table 4.2 Selected bond lengths (Å) and angles (°) for complexes 30 and 31. 30  31  κ2(N,O)  2.1355(17)  2.1439(12)  κ1(O)  2.0012(18)  1.9947(11)  Ta–N  κ2(N,O)  2.309(2)  2.3517(14)  C–N  κ1(O)  1.277(3)  1.279(2)  Ta–Nequitorial  N trans  2.006(2)  1.9967(14)  O trans  1.970(2)  1.9719(14)  Ta–Naxial  1.999(2)  1.9988(15)  O–Ta–N  58.35(7)  57.88(5)  Ta–O  Like their Zr congeners, both complexes adopt a pseudo-trigonal bipyramidal geometry. However, here the bis(amidate) ligand is bound κ3(N,O,O) in both 30 and 31, with the chelating amidate in the equatorial plane and the κ1(O) contact in the axial plane. In line with the observations for the untethered bis(amidate) complex 7, the Ta–Oamidate distance for the κ1(O)  185  bound amidate (1.998(2) Å average) is noticeably shorter than its κ2(N,O) counterpart (2.140(2) Å average). The equatorial dimethylamido ligands are also asymmetrically bound, with the Ta– Namido trans to the Namidate being longer (2.001(2) Å average) than the amido trans to the Oamidate (1.971(2) Å average).  Other metrical parameters for 30 and 31 are consistent with other  κ3(N,O,O) tantalum bis(amidate) structures.98 In contrast to the tantalaziridine species in Section 4.1.1, no interaction is observed between the tantalum centre and the oxygen or sulfur donor in the solid state (Ta–O/S ≈ 4.000 and 4.388 Å respectively). Broad signals in the room temperature 1H NMR spectrum for 30 (Figure 4.9) suggest dynamic exchange processes like those observed for untethered bis(amidate) complex 7. Additionally, two sets of signals are observed for the tBu groups indicative of different magnetic environments for the individual binding modes. Upon heating to 40 °C, coalescence of these two distinct sets of tBu signals for the κ1(O) (δH = 1.50, 1.67) and κ2(N,O) (δH = 1.11, 0.85) bound amidate units is observed at δH = 1.63 and 1.10 respectively.  Interestingly, at higher  temperatures (80 °C) only one broad signal is observed for the tBu groups (δH = 1.25) suggesting rapid interconversion of the two binding modes on the NMR timescale.  As with 7, the  coalescence of the dimethylamido methyl groups (δH = 3.25) is observed at approximately 40 °C. Holding the solution temperature at 90 °C for multiple hours results in the formation of signals assignable to liberated dimethylamine as well as diastereotopic tantalaziridine protons. These signals persist in the sample after cooling and suggest that 30 might be a suitable precatalyst for the α-alkylation of amines (vide infra). Though the 1H spectrum of 31 collected at ambient temperature is also broad, only one signal is observed for the dimethylamido ligands (δH = 3.19) and each tBu environment (δH = 1.71 and 1.09). In contrast to the aforementioned  186  diamido systems of Schrock and coworkers, this result implies reduced barriers to exchange in 31 perhaps due to the increased flexibility of the ligand.  Figure 4.9 Variable temperature (25 – 90 °C) 400 MHz 1H NMR study of 30.  4.2.2  Tethered Bis(ureate) Complexes of Zr and Ta Encouraged by the success of the tethered bis(ureate) zirconium hydroamination  precatalyst introduced in Section 4.1.1, two diurea proligands were synthesized according to  187  literature procedure.60  Reaction of the oxy- or thio-dianiline starting material with phenyl  chloroformate gives the intermediate dicarbamate which, upon exposure to diisopropylamine, is converted to the desired diurea product in good yields (Scheme 4.7).  Scheme 4.7 Synthesis of diurea proligands.  In line with observations made for the diamide proligands, the diurea compounds are twisted about the chalcogen bridge in the solid state (Figure 4.10). As expected, the C–S bond lengths (1.7798(15) Å average) are considerably longer than the corresponding C–O distances (1.4018(16) Å average) and the C–E–C bond angles are comparable to the diamide values. In line with the diamide proligands, the diurea species also adopt a C2 symmetric geometry in solution as evidenced by the presence of a single methine signal in the 1H NMR spectrum.  188  Figure 4.10 ORTEP representations of the solid-state molecular structures for oxy- and thiobridged diurea proligands drawn at 50% probability for thermal ellipsoids. Only one part of a disordered N(iPr)2 group for the thio-bridged species is shown for clarity. Hydrogen atoms omitted for clarity.  Following literature precedence,172 the oxy-bridged diurea proligand was reacted with Zr(NMe2)4 in pentane at reflux to yield the κ3(N,O,O) bis(amidate) complex 32•HNMe2 in 80% yield (Scheme 4.8). The pseudo-trigonal bipyramidal geometry of 32•HNMe2 features both κ1(O) and κ2(N,O) ureate contacts in the equatorial plane with the neutral dimethylamine donor occupying an axial position (Figure 4.11). The presence of dimethylamine in the axial position is in good agreement with the solid-state molecular structure of the related bis(ureate) zirconium hydroamination precatalyst.172 A broad signal in the 1H NMR spectrum at δH = 2.21 suggests the amine remains coordinated in solution as well. Unfortunately, the presence of the amine donor negates the need for electron density from the oxy-bridge resulting in a Zr–O distance of approximately 3.724 Å.  189  Scheme 4.8 Protonolysis synthesis of bis(ureate) zirconium complex 32•HNMe2.  Figure 4.11 ORTEP representation of the solid-state molecular structures for 32•HNMe2 drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  In an attempt to encourage a dative interaction between the linking oxygen atom and the zirconium centre of 32•HNMe2, the dimethylamine donor was removed in vacuo. In contrast to dimer formation as observed with the alkyl-bridged species,176 abstraction of the amine donor here yields monometallic complex 32 supported by a κ4(N,N,O,O) bound bis(ureate) ligand (Figure 4.12). This complex is structurally similar to the thio-bridged variant 33 which cannot be synthesized in pentane due to low solubility of the ligand. Synthesis of these two species in toluene (Scheme 4.9) necessitates removal of the solvent under reduced pressure. 190  Scheme 4.9 Protonolysis synthesis of bis(ureate) zirconium complexes 32 and 33.  Figure 4.12 ORTEP representation of the solid-state molecular structure of 32 (left) and 33 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  191  Table 4.3 Selected bond lengths (Å) and angles (°) for complexes 32 and 33. 32  33  Zr–O1  2.1949(15)  2.1749(12)  Zr–O2  2.1855(16)  2.1863(12)  Zr–N1  2.3129(16)  2.3156(14)  Zr–N2  2.2529(18)  2.3229(13)  Zr–N5  2.0314(18)  2.0525(14)  Zr–N6  2.0466(17)  2.0532(14)  O1–Zr–N1  57.68(6)  58.81(4)  O2–Zr–N2  59.20(6)  58.20(4)  With an average angle of 109.3° between the atoms bound to the metal centre, and no obviously defined planes, 32 and 33 are best described as tetrahedral complexes with each κ2(N,O) ureate unit occupying one coordination site. The oxy-bridged species 32 has slightly longer Zr–Oureate bonds (2.190(2) Å average) compared to complex 33 (2.181(1) Å average), corresponding to shorter Zr–Nureate contacts (2.283(2) and 2.319(1) Å average respectively). In contrast, the alkyl-bridged ureate precatalyst is found to have more symmetric N,O-chelation to Zr, with a smaller discrepancy (0.043(2) Å average) between Zr–Oureate and Zr–Nureate bond lengths. Bis(ureate) complexes 32 and 33 have shorter average Zr–Oureate (2.1854(2) Å) and Zr– Nureate (2.3011(2) Å) contacts than the comparable bis(amidate) species 28 and 29 (2.2444(2) and 2.3242(2) Å), putting the κ2(N,O) chelate closer to the metal centre. These metrical parameters are in line with observations made for complexes containing comparable alkyl-bridged ligands.177 The Zr–Namido distances for 32 and 33 are uniformly shorter (2.046(2) Å average) than those for the related alkyl-bridged species (2.114(2) Å average).  The principal difference  between the chalcogen- and alkyl-bridged complexes in the solid phase is the in-plane binding of 192  the bis(ureate) ligand. Even with the additional flexibility offered by the sulfur linker compared to its oxygen congener, neither species allows both κ2(N,O) ureate units to bind to the zirconium in the equatorial plane as found with the alkyl bridge. This attribute may significantly impact the catalytic activity of the complex (vide infra). Disappointingly, removal of the dimethylamine donor does not induce orbital interaction with the tether donor as evidenced by Zr–O and Zr–S distances of approximately 3.836 and 3.823 Å respectively. The solution phase data for 32 and 33 is characterized by broad signals for the isopropyl methyl protons in the 1H NMR spectra (0.5 – 1.5 ppm for 32; 0.0 – 2.0 ppm for 33). Due to the observed broadening and high multiplicity, the methine C–H signals cannot be distinguished from the baseline in either spectrum. In contrast, well-resolved signals are observed for both the aryl protons of the bis(ureate) ligand and the methyl protons of the dimethylamido ligands. As observed for the tethered bis(amidate) systems, bis(ureate) tantalum species can also be synthesized via protonolysis. Unlike their bis(amidate) counterparts, no precipitate is formed under the reaction conditions outlined in Scheme 4.10, leading to higher isolated yields. Recrystallization from pentane at -35 °C yields single crystals suitable for X-ray diffraction studies. Interestingly, oxy-bridged 34 is found to have a κ3(N,O,O) bound bis(ureate) ligand while the more flexible thio-bridged bis(ureate) ligand of 35 binds in a κ4(N,N,O,O) fashion (Figure 4.13). The smallest M–E distance observed in the series of tethered bis(amidate) and bis(ureate) complexes is found for 34 with a Ta–O distance of approximately 2.873 Å. At this distance, electron donation to the metal centre cannot be ruled out.  193  Scheme 4.10 Protonolysis synthesis of bis(ureate) tantalum complexes 34 and 35.  Figure 4.13 ORTEP representation of the solid-state molecular structure of 34 (left) and 35 (right) drawn at 50% probability for thermal ellipsoids. Hydrogen atoms omitted for clarity.  194  Table 4.4 Selected bond lengths (Å) and angles (°) for complexes 30 and 31. 30  31  Ta–O1  2.257(2)  2.2968(18)  Ta–O2  1.999(2)  2.2353(17)  Ta–N1  2.237(2)  2.2567(19)  Ta–N2  n/a  2.244(2)  N trans  1.992(3)  1.993(2)  O trans  1.986(3)  2.019(2)  Ta–Naxial  1.978(3)  2.018(2)  O1–Ta–N1  57.88(8)  57.48(7)  O2–Ta–N2  n/a  57.74(7)  Ta–Nequitorial  Both complexes crystalize in a distorted-trigonal bipyramidal geometry with a κ2(N,O) ureate chelate and two dimethylamido ligands in the equatorial plane. The Ta–Namido bond length for the dimethylamido ligand trans to the ureate chelate is significantly shorter for 34 (1.986(3) Å) than for 35 (2.019(2) Å), a trend that is also observed for the axial dimethylamido ligands (1.978(3) and 2.018(2) Å respectively). Notably, the bond lengths for the dimethylamido ligands cis to the κ2(N,O) ureate unit, and therefore trans to the chalcogen tether atom, are identical within experimental error (1.992(3) Å). This suggests that there is no trans influence from the tether oxygen atom in 34. In contrast to the bis(ureate) zirconium species, 34 has shorter Ta–Oureate (2.257(2) Å) and Ta–Nureate (2.237(2) Å) distances than 35 (2.297(2) and 2.257(2) Å respectively) for the equatorial κ2(N,O) ureate. Overall, ureate chelates are more symmetrically bound to the metal centre for tantalum species 34 and 35 compared to their zirconium counterparts. Notably, the M–Namido distances are shorter for the tantalum ureate complexes than the zirconium species, attributable to the smaller ionic radius of Ta(V) compared to Zr(IV).178  195  In line with the solution phase data for 30 and the zirconium bis(ureate) species, the signals in the room temperature 1H NMR spectrum for 34 are extremely broad, even in the aromatic region. Cooling of an analytically pure sample to -70 °C reveals a number of magnetic environments for various protons in the alkyl region (0.0 – 5.0 ppm), suggestive of multiple isomers in solution (Figure 4.14). Heating the same sample to 80 °C resolves the signals enough to elucidate the expected doublet for the isopropyl methyl groups, the singlet observed for dimethylamido methyl groups, the multiplet for the isopropyl methine groups as well as a set of aromatic protons with various coupling patterns, suggesting facile exchange of environments at elevated temperatures.  196  Figure 4.14 Variable temperature 400 MHz 1H NMR study of 34. Select spectra taken at -70, 25 and 80 °C shown.179  Akin to 30, signals indicative of tantalaziridine formation and release of dimethylamine are visible (δH = 2.24 and 2.20 respectively) after heating of the sample for multiple hours. The formation is not reversible as the diagnostic signals are still present upon cooling of the sample to room temperature. Unlike the variable temperature studies for the κ3(N,O,O) 30, no distinct signals can be assigned to either the κ1(O) or κ2(N,O) ureate units. Though 35 has better  197  resolved 1H NMR signals at room temperature, especially for the aromatic protons, the number of signals in the alkyl region clearly indicates multiple isomers in solution. In an effort to further characterize this novel series of tethered better bis(amidate) and bis(ureate) complexes, electrochemical analyses were performed. These results will aid in the elucidation of the effects that amidate and ureate ligands have on the electronic nature of a metal complex, as well as highlighting the fundamental differences between Zr and Ta.  4.2.3  Electrochemical Analysis of Tethered Bis(amidate) and Bis(ureate) Complexes As discussed in Section 4.1.1, the Fryzuk group has had excellent success reducing early  transition metals to generate dinitrogen complexes.168, 171, 180-188 However, Dr. Patricia HorrilloMartinez, a joint post-doctoral fellow in the Schafer and Fryzuk groups, found that tantalum amidate complexes are unstable when exposed to chemical reductants such as potassium graphite (KC8) and permethylcobaltocene (Cp2*Co).189  Treatment of pentamethylcyclopentadienyl  amidate tantalum chloride complexes with either reductant results in the extrusion of the oxygen atom from the amidate ligand to generate a tantalum oxo complex and an η2-iminoacyl ligand (Scheme 4.4). This is in contrast to the comparable early metal κ2(N,N) amidinate systems developed by Sita and coworkers that form bimetallic complexes with bridging dinitrogen moieties when reduced with KC8 in a nitrogen atmosphere.190-192 The extrusion of oxygen from the amidate ligand highlights the oxophilicity of tantalum and suggests that N,O-chelating ligands are not compatible with harsh reducing agents.  Unfortunately, the corresponding  trimethyl amidate tantalum complex shows no reactivity with hydrogen gas.188  198  Scheme 4.11 Reduction of tantalum amidate complexes to yield η2-iminoacyl ligated tantalum oxo species.  Though the reduction of zirconium and tantalum species by chemical reductants has been well documented by the Fryzuk group, these compounds are rarely characterized using electrochemical methods, specifically cyclic voltammetry.193 Cyclic voltammetry, or CV, is an electrochemical technique that involves applying a potential to a conductive solution containing an analyte and measuring the resultant current in the solution as a function of the applied potential.194 By scanning a range of potentials, reduction(s) and oxidation(s) of the analyte are observed as “peaks” in the voltammogram, offering a method of quantitatively characterizing a specific analyte in an electrochemical fashion.195 CV is useful for organometallic complexes as it quantifies potentials that redox events occur at, facilitating the choice of an appropriate chemical reductant/oxidant.  Additionally, the stability of the complex under reducing or  oxidative conditions can be evaluated as well as the reversibility of a redox process. Traditionally CV is performed under ambient conditions, yet the sensitivity of early transition metal complexes to H2O and O2 render these species incompatible with these experimental conditions. However, using purified solvents and electrolytes, electrochemical analyses such as CV can be performed under rigorously air- and moisture-free conditions.196 As a result, electrochemical analysis has become a powerful tool for the characterization of sensitive 199  organometallic complexes and the elucidation of catalytic mechanisms related to these species.197-198 The focus of the analyses herein is the determination of the reduction potential of the Zr and Ta centres and how it is affected by the ligand, specifically the difference between amidate- and ureate-supported complexes. To this end, a selection of complexes containing tethered bis(amidate) or bis(ureate) ligands was characterized using cyclic voltammetry. tetrabutylammonium  hexafluorophosphate  (TBAPF6)  Using THF as the solvent and as  the  electrolyte,  the  voltammograms were collected under an argon atmosphere to prevent reaction with N2.  cyclic The  working (platinum button) and counter (glassy carbon) electrodes were polished using alumina powder before each set of experiments while residual organic materials were removed from the reference electrode (silver wire) by heating with a butane torch. All data presented herein are referenced to ferrocene (Fc0/+). Full data sets are available in Appendix D. The voltammogram for complex 28 (Figure 4.15) shows two irreversible oxidations at 0.18 and -1.09 V that are persistent through a range of scan rates. These oxidations are presumed to occur at the ligand as the zirconium centre is already in a d0 electronic state. This hypothesis is supported by the presence of oxidative events in the voltammogram of the diamide proligand.199  200  Figure 4.15 Cyclic voltammogram for 28 at a scan rate of 100 mV/s (0.005 mol L-1 in THF, 0.1 mol L-1 TBAPF6). Electrodes as follows: working, Pt button; counter, glassy carbon; reference, Ag wire.  In contrast to the Zr system, reversible reductive events are observed for the corresponding tantalum species. Figure 4.16 shows the voltammograms for 30 and 31 with the reversible events occurring at -2.87 V for both species. This suggests that the chalcogen donor of the ligand does not significantly influence the electronic properties of the complex. These events are tentatively assigned to the reduction of the tantalum centre from Ta(V) to Ta(IV) based on calculated frontier orbitals.199 These values are 1.6 – 2.0 V more negative than other reported reduction potentials,104-105,  200  however, comparable complexes and experimental  conditions were not found in the literature which could explain the large discrepancy. After the set of experiments was complete, a filmy precipitate was found to have formed in the electrochemical cell.  It is proposed this material is the result of reduction induced  decomposition.  201  Figure 4.16 Cyclic voltammograms for 30 and 31 at a scan rate of 50 mV/s (0.005 mol L-1 in THF, 0.1 mol L-1 TBAPF6). Electrodes as follows: working, Pt button; counter, glassy carbon; reference, Ag wire.  The oxy-bridged bis(ureate) tantalum species 34 has a similar reductive event at -3.08 V that appears to be less reversible than the reductions for the bis(amidate) species based on qualitative line-shape analysis (Figure 4.17). This reduction potential is more negative than for the bis(amidate) species (ΔV = 0.21 for both 30 and 31) suggesting tantalum is more difficult to reduce when supported by bis(ureate) ligands. This result supports the notion that ureate ligands promote the formation of more electrophilic metal centres.177 Interestingly, there are also nonreversible oxidative events present near the positive edge of the solvent window (0.16 and 0.41 V).  In an effort to resolve these oxidative events, the electrolyte was changed to  tetrabutylammonium tetraphenylborate (TBABPh4). Unfortunately, the change in electrolyte caused a loss of detail in the oxidative region and the events were no longer visible at any scan rate (Figure 4.17). However, the quasi-reversible reduction (-3.10 V) is still present. Notably,  202  qualitatively less precipitate was observed after these analyses when compared to the bis(amidate) species.  Figure 4.17 Cyclic voltammograms for 34 (0.005 mol L-1 in THF) using 0.1 mol L-1 TBAPF6 (left, scan rate = 100 mV/s) and 0.1 mol L-1 TBABPh4 (right, scan rate = 500 mV/s) as the electrolyte. Electrodes as follows: working, Pt button; counter, glassy carbon; reference, Ag wire.  At present, the Schafer and Fryzuk groups are in the early stages of using electrochemical methods to characterize and investigate metal complexes.  However, the air-sensitive  experimental set-up now available to both groups will allow these techniques to become commonplace. As more complexes are subjected to electrochemical analysis, trends in the electronic nature of these early transition metal complexes will be revealed. The knowledge of the reduction potentials of metal centres is invaluable to the design of systems for the activation of dinitrogen. The selection of an appropriate chemical reductant for a system is easily achieved  203  with a quantitative measure of the reduction potential which may avoid harsh reaction conditions. Additionally, the stability of the complex under redox conditions can aid in the design of further generations of supporting ligands.  4.2.4  Catalytic Activity of Tethered Bis(amidate) and Bis(ureate) Complexes In light of the observations of tantalaziridine formation at high temperatures for 30 and  34, as well as the excellent reactivity observed for the alkyl-bridged bis(ureate) zirconium species, complexes 28 – 35 were investigated as precatalysts for the intermolecular α-alkylation of amines as well as the intramolecular hydroamination of aminoalkenes. Screening for the intermolecular α-alkylation of amines was performed using N-methylaniline and 1-octene as substrates with complex 3 acting as a benchmark for catalytic activity.39, 43, 45, 93 Intramolecular hydroamination of both primary and secondary aminoalkenes was attempted in order to compare reactivity to the alkyl-bridged bis(ureate) precatalyst.172  The zirconium precatalysts were  prepared in situ to avoid the competitive binding from the pyridine donor found in the crystalline complexes 28 and 29. The catalytic results are summarized in Table 4.5 along with the specific reaction conditions.  204  Table 4.5 Catalytic screening of complexes 28 – 35 for the intermolecular α-alkylation of amines (α-alk, top) and intramolecular hydroamination of primary (1° HA, middle) and secondary amines (2° HA, bottom). Conversions determined by 1H NMR spectroscopy using 1,3,5trimethoxybenzene as an internal standard.  Catalyst  Conversion (%) α-alk  1° HA  2° HA  28  <2  > 98  <2  29  <2  > 98  <2  30  34  3  4  31  28  7  3  32  <2  > 98  10  33  <2  > 98  6  34  15  15  8  35  <2  5  12  In line with the literature overview presented in Chapter 1, the zirconium complexes are ineffective as precatalysts for the intermolecular α-alkylation of amines. After 24 hours at 130 °C, no signals attributable to product 6 are found in the 1H NMR spectrum. In contrast, the 205  tantalum species are found to be competent precatalysts for the reaction, in good agreement with tantalaziridine formation at elevated temperatures. Like the bis(amidate) complex 9, 30 and 31 form less active catalysts (34 and 28% conversion respectively) than the κ2(N,O) mono(amidate) complexes screened in Table 2.2.  However, based on turnover frequency, the tethered  bis(amidate) species (average TOF = 0.258 h-1) are significantly more active than the untethered bis(amidate) tantalaziridine (TOF = 0.042 h-1).93 For reference, full conversion is achieved when complex 3 is used as the precatalyst with the above conditions. The κ3(N,O,O) bis(ureate) tantalum complex 34 also catalyzes the α-alkylation of N-methylaniline with 1-octene albeit in low conversion (15%). Interestingly, no reactivity is observed with the thio-bridged κ4(N,N,O,O) species 35. The zirconium species perform much better as hydroamination precatalysts as both the bis(amidate) and bis(ureate) complexes achieve full conversion with the primary aminoalkene substrate after the 2 hour time period. However, bis(amidate) complexes 28 and 29 show no reactivity for the secondary amine substrate under the described reaction conditions.  This  reactivity profile is similar to the homoleptic Zr(NMe2)4, suggesting an imido-based mechanism for 28 and 29.201 In contrast, the bis(ureate) zirconium precatalysts 32 and 33 do show some conversions with the secondary aminoalkene substrate (10 and 6% respectively), suggesting a mechanism more akin to that observed for the alkyl-bridged species.176, 202 The low reactivity of these complexes might be attributed to the non-planar geometry of the κ4(N,N,O,O) bis(ureate) ligand, as the secondary amine substrate is completely converted under these conditions with the in-plane alkyl-bridged complex. Less impressive results for hydroamination are obtained with the tantalum precatalysts. Though trace amounts of secondary aminoalkene conversion are detected with the κ3(N,O,O)  206  bis(amidate) complexes 30 and 31 (4 and 3% respectively), less than 10% conversion is observed for the primary amine substrate after 2 hours. Extending the reaction time to 18 hours increases the conversions to 9 and 14% respectively, well below the reactivity observed for tetrakis(dimethylamido) zirconium. Notably, the bis(ureate) tantalum complexes catalyze the intramolecular hydroamination of both primary and secondary aminoalkenes to a small degree. Oxy-bridged 34 outperforms its thio-bridge counterpart 31 for primary amines (15 and 5% conversion respectively), with the κ4(N,N,O,O) species 31 displaying higher reactivity for the secondary substrate (8 and 12% conversions). Unlike their bis(amidate) equivalents, increasing the reaction time for primary aminoalkene hydroamination does not result in an increase in conversion. Overall, tantalum precatalysts performed better for the α-alkylation of amines while zirconium complexes are more effective for hydroamination catalysis.  The bis(amidate)  tantalum species achieved higher conversions for the α-alkylation of amines than their bis(ureate) counterparts but were not as active as mono(amidate) 3. In comparison, the silylatedbinaphtholate niobium system developed by Hultzsch and coworkers is significantly more reactive at 150 °C. However, the tethered tantalum species reported herein display exclusive regioselectivity, a feat thus far unmatched by any group 4 system. In contrast, the bis(ureate) tantalum species are more active as precatalysts for hydroamination. All the tethered zirconium complexes screened were found to catalyze the hydroamination of a primary aminoalkene substrate while bis(ureate) precatalysts 32 and 33 are compatible, to a small degree, with a secondary aminoalkene. Though the conversions are poor, this type of reactivity is interesting as hydroamination with secondary amines is rare for group 4 complexes. In fact, the precatalysts found in the literature are sensitive cationic species, unlike  207  the robust complexes described herein, or display modest reactivities like those observed for the tethered bis(ureate) species.33, 203-206 However, none of the chalcogen-bridged complexes were found to be as reactive as the versatile alkyl-bridged precatalyst discussed in Section 4.1.1.  4.3  Conclusions A series of Zr and Ta complexes supported by bis(amidate) and bis(ureate) ligands with a  neutral chalcogen donor in the tether have been synthesized and characterized in the solution phase and the solid state. The bis(amidate) zirconium complexes 28 and 29 require an amine donor to fill the coordination sphere (HNMe2 in solution, pyridine in the solid state) and display a κ4(N,N,O,O) bis(amidate) binding mode. The Ta congeners 30 and 31 adopt a κ3(N,O,O) binding mode and have broad 1H NMR signals at ambient temperatures. At higher temperatures the signals coalesce, suggesting dynamic ligand behaviour on the NMR timescale. None of the data collected for the bis(amidate) species suggests there is electron donation from the neutral donor to the metal centre in solution or the solid state. Bis(ureate) species 32 and 33 are found to have a κ3(N,O,O) binding mode in the solid state with the ureate moieties cis to one another. This is in contrast to the in-plane binding of an alkyl-bridge bis(ureate) zirconium complex found to be a broadly applicable hydroamination catalyst. Interestingly, complex 34 is found to have a κ3(N,O,O) bis(ureate) binding mode that permits the oxygen donor in the tether to be approximately 2.873 Å from the tantalum centre, the shortest M–E distance observed herein. The more flexible sulfur-bridged ligand of 35 binds κ4(N,N,O,O), forcing the sulfur atom more than 3 Å from the metal centre.  1  H NMR  spectroscopy data for the bis(ureate) tantalum species suggests multiple isomers in solution at ambient temperatures with time-averaged signals observed at increased temperatures.  208  Using newly acquired electrochemical equipment, the bis(amidate) and bis(ureate) complexes were subjected to electrochemical analysis, specifically cyclic voltammetry, under air- and moisture-free conditions. Non-reversible oxidations are found for 28 while a reversible reduction assigned to the Ta(V) / Ta(IV) couple is observed for the tantalum complexes 30 and 31. Complex 34 produces a more complex voltammogram with a quasi-reversible reduction at Ta and non-reversible oxidative events near the edge of the spectral window. Replacing the TBAPF6 electrolyte with TBABPh4 causes a loss of resolution, completely obscuring the oxidative events. Though these early results are qualitative in nature, trends in electrochemical data could prove useful in the design of future complexes for accessing reduced metal centres appropriate for redox chemistry. Complexes 28 – 35 were also screened as catalysts for the α-alkylation of amines as well as the intramolecular hydroamination of primary and secondary aminoalkenes. None of the Zr species are found to catalyze the α-alkylation of amines, in line with literature precedence. In contrast, the Ta species do affect reactivity, with the bis(amidate) species outperforming the bis(ureate) complexes. However, no precatalyst is more active than the benchmark complex 3, achieving complete conversion in the time allotted. The zirconium species are found to be more effective for hydroamination than the Ta precatalysts, with complete conversion of the primary amine substrate observed for each Zr complex. However, neither the bis(amidate) nor the bis(ureate) Zr species can convert more than 10% of the secondary amine substrate whereas complete conversion is observed under the same conditions for the alkyl-bridged bis(ureate) precatalyst. Notably, the bis(ureate) Ta species are significantly more active for hydroamination than the bis(amidate) precatalysts but still only manage to achieve conversions as high as 15%.  209  4.4 4.4.1  Experimental Materials and Methods General manipulations and techniques are as outlined in Section 2.4.1 and also include  the following addition: All electrochemical measurements were collected using a CH Instruments 630D Potentiostat using proprietary CH Instruments software. The potentiostat was connected to the electrodes using shielded cable that interfaced with a BNC junction built into the glove box. The electrodes were modified to include a pin-type junction for ease of handling inside the glove box. Further information can be found in Section 4.4.4.  4.4.2  Synthesis and Characterization of Proligands and Complexes  General procedure for the synthesis of diamide proligands. Unless otherwise noted, all proligands were prepared using the following general procedure and purified by recrystallization from hexanes and ethyl acetate. All compounds were isolated as colourless crystalline solids. The synthesis of N,N’-(2,2’-oxybis(2,1-phenylene))bis(tert-butylamide) is given as an example: 2,2’-oxydianiline (2.09 g, 10.4 mmol) was dissolved in dichloromethane (DCM) (35 mL). Triethylamine (4.3 mL, 30.9 mmol) was added in one portion and the solution was cooled to -78 °C. Trimethylacetyl chloride (2.6 mL, 21.1 mmol) was added dropwise and the solution was allowed to warm to room temperature overnight. A precipitate formed and was dissolved in DCM (25 mL) and 1M aqueous NaOH solution (50 mL) and was added to the solution. The organic portion was washed with 50 mL portions of 1M aqueous HCl solution (3x), water and brine and dried with MgSO4. The solution containing the remaining product was run through a  210  short silica gel plug using DCM. The solvent was removed by rotary evaporation and the resulting solid was washed with hexanes.  N,N’-(2,2’-oxybis(2,1-phenylene))bis(tert-butylamide). Yield: 3.29 g (86%). 1H NMR (chloroform-d, 600 MHz) δ 1.22 (s, 18H, C(CH3)3), 6.83 (dd, 3JH,H = 8.19 Hz, 3JH,H = 1.02 Hz, 2H, CHarom.), 7.03 (td, 3JH,H = 8.70 Hz, 3JH,H = 1.54 Hz, 2H, CHarom.), 7.13, (td, 3JH,H = 7.68 Hz, 3JH,H = 1.02 Hz, 2H, CHarom.), 7.96 (br s, 2H, NH), 8.27 (d, 3JH,H = 8.19 Hz, 2H, CHarom.);  13  C{1H} NMR (chloroform-d, 150  MHz) δ 27.12 (C(CH3)3), 39.51 (C(CH3)3), 117.2, 122.0, 124.3, 124.3 (CHarom.), 129.0 (CN), 145.2 (Carom.OC), 176.4 (C=O); MS(ESI) m/z 369 (M+ + H), 391 (M+ + Na); Anal. Calcd. (%) for C22H28N2O3: C, 71.71; H, 7.66; N, 7.60. Found: C, 71.50; H, 7.56; N, 7.60. Single crystal Xray quality samples were obtained by recrystallization from ethanol at 4 °C. Disordered tBu group was modeled using two unique parts.  N,N’-(2,2’-oxybis(2,1-phenylene))dibenzamide.  2,2’-oxydianiline  (3.03 g, 15.1 mmol) was dissolved in DCM (75 mL). Triethylamine (6.3 mL, 45.3 mmol) was added in one portion and the solution was cooled to -78 °C before benzoyl chloride (4.4 mL, 37.8 mmol) was introduced. Yield: 5.12 g (83%). 1H NMR (chloroform-d, 400 MHz) δ 6.96 (d, 3JH,H = 8.22 Hz, 2H, CHarom.), 7.12 (td, 3  JH,H = 5.87 Hz, 3JH,H = 1.56 Hz, 2H, CHarom.), 7.22 (td, 3JH,H = 7.83 Hz, 3JH,H = 1.17 Hz, 2H,  CHarom.), 7.43 (t, 3JH,H = 7.43 Hz, 4H, CHarom.), 7.52 (td, 3JH,H = 7.43 Hz, 3JH,H = 1.17 Hz, 2H, CHarom.), 7.79 (d, 3JH,H = 8.22 Hz, 4H, CHarom.), 8.40 (d, 3JH,H = 8.22 Hz, 2H, CHarom.) 8.51 (br s, 2H, NH);  13  C{1H} NMR (chloroform-d, 150 MHz) δ 117.7, 122.6, 124.4, 125.0, 127.0, 128.4  211  (CHarom.), 129.0 (CN), 131.7 (CHarom.), 134.1 (Carom.CON), 146.1 (Carom.OC), 165.5 (C=O); MS(ESI) m/z 409 (M+ + H), 431 (M+ + Na); Anal. Calcd. (%) for C26H20N2O3: C, 76.45; H, 4.94; N, 6.86. Found: C, 76.40; H, 4.90; N, 6.81.  N,N’-(2,2’-oxybis(2,1-phenylene))bis(4-methylbenzamide). 2,2’-oxydianiline (3.54 g, 17.7 mmol) was dissolved in DCM (75 mL).  Triethylamine (6.5 mL, 46.9 mmol) was added in one  portion and the solution was cooled to -78 °C before 4-methylbenzoyl chloride (4.7 mL, 35.9 mmol) was introduced. Yield: 6.57 (85%). 1H NMR (chloroform-d, 600 MHz) δ 2.37 (s, 6H, CCH3), 6.93 (d, 2H, 3JH,H = 8.70 Hz, CHarom.), 7.10 (t, 2H, 3JH,H = 7.17 Hz, CHarom.), 7.17 (m, 6H, CHarom.), 7.69 (d, 4H, 3JH,H = 8.19 Hz, CHarom.), 8.33 (d, 2H, 3JH,H = 8.19 Hz, CHarom.), 8.68 (br s, 2H, NH);  13  C{1H} NMR (chloroform-d, 150 MHz) δ 21.4 (CCH3), 117.8, 122.7, 124.5, 125.0,  127.2, 129.2 (CHarom.), 129.3 (CN), 131.5 (Carom.CON), 142.3 (CCH3), 146.2 (Carom.OC), 165.7 (C=O); MS(ESI) m/z 459 (M+ + Na); Anal. Calcd. (%) for C28H24N2O3: C, 77.04; H, 5.54; N, 6.42. Found: C, 77.30; H, 5.49; N, 6.38.  N,N’-(2,2’-oxybis(2,1-phenylene))bis(2,4,6trimethylbenzamide).  This compound was synthesized by a  modification of the general procedure outlined above.  2,2’-  oxydianiline (1.00 g, 5.00 mmol) was dissolved in 1,2-dichloroethane (50 mL) and stirred vigorously. Pyridine (1.2 mL, 15.00 mmol) and 2,4,6-trimethylbenzoyl chloride (2.1 mL, 12.5 mmol) were added individually and the mixture was heated at reflux overnight. The aqueous workup and purification followed the previously discussed general procedure. Yield: 1.94 g  212  (79%).  1  H NMR (chloroform-d, 400 MHz) δ 2.20 (s, 12H, ortho-CCH3), 2.30 (s, 6H, para-  CCH3), 6.84 (s, 4H, CHarom.), 6.89 (dd, 2H, 3JH,H = 8.22 Hz, 3JH,H = 1.17 Hz, CHarom.), 7.12 (td, 2H, 3JH,H = 7.83 Hz, 3JH,H = 1.56 Hz, CHarom.), 7.22 (td, 2H, 3JH,H = 7.43 Hz, 3JH,H = 1.17 Hz, CHarom.), 8.14 (br s, 2H, NH), 8.41 (d, 2H, 3JH,H = 8.22 Hz, CHarom.); 13C{1H} NMR (chloroformd, 100 MHz) δ 19.0 (ortho-CCH3), 21.0 (para-CCH3), 117.9, 122.2, 124.7, 125.0, 128.2 (CHarom.), 129.2 (CN), 134.2 (ortho-CCH3), 134.6 (Carom.CON), 138.8 (para-CCH3), 145.9 (Carom.OC), 168.7 (C=O); MS(ESI) m/z 515 (M+ + Na); Anal. Calcd. (%) for C32H32N2O3: C, 78.02; H, 6.55; N, 5.69. Found: C, 77.82; H, 6.45; N, 5.68.  N,N’-(2,2’-thiobis(2,1-phenylene))bis(tert-butylamide).  2,2’-  thiodianiline (2.00 g, 9.24 mmol) was dissolved in DCM (25 mL). Triethylamine (3.9 mL, 27.7 mmol) was added in one portion and the solution was cooled to -78 °C before pivoyl chloride (2.8 mL, 23.1 mmol) was introduced. Yield: 3.03 g (85%). 3  1  H NMR (chloroform-d, 300 MHz) δ 1.24 (s, 18H, C(CH3)3), 7.02 (td,  JH,H = 7.42 Hz, 3JH,H = 1.10 Hz, 2H, CHarom.), 7.12 (dd, 3JH,H = 7.97 Hz, 3JH,H = 1.65 Hz, 2H,  CHarom.), 7.32, (td, 3JH,H = 7.15 Hz, 3JH,H = 1.37 Hz, 2H, CHarom.), 8.07 (br s, 2H, NH), 8.26 (dd, 3  JH,H = 7.15 Hz, 3JH,H = 1.10 Hz, 2H, CHarom.);  13  C{1H} NMR (chloroform-d, 75 MHz) δ 27.4  (C(CH3)3), 39.9 (C(CH3)3), 122.2, 122.2, 124.9, 129.2 (CHarom.), 131.7 (CN), 137.5 (Carom.OC), 176.6 (C=O); MS(ESI) m/z 385 (M+ + H), 407 (M+ + Na); Anal. Calcd. (%) for C22H28N2O2S: C, 68.72; H, 7.34; N, 7.29. Found: C, 68.72; H, 7.38; N, 7.29. Single crystal X-ray quality samples were obtained by recrystallization from hexanes and ethyl acetate.  213  N,N’-(2,2’-oxybis(2,1-phenylene))bis(tertbutylamidate)bis(dimethylamido)(pyridine)zirconium (28). oxybis(2,1-phenylene))bis(tert-butylamide)  (0.276  g,  N,N’-(2,2’-  0.748  mmol),  Zr(NMe2)4 (0.200 g, 0.748 mmol) and pyridine (60 μL, 0.748 mmol) were suspended in toluene (3 mL) and the mixture was allowed to stir overnight (18 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was washed thrice with hexanes and then recrystallized from the minimal amount of hot hexanes. The colourless crystals were collected and identified as the title compound. Yield: 0.265 g (0.426 mmol, 57%). 1  H NMR (benzene-d6, 600 MHz) δ 0.99 (s, 18H, C(CH3)3), 3.20 (s, 12H, N(CH3)2), 6.67 (m, 2H,  CHarom.), 6.77 (m, 4H, py–CHarom.), 6.89 (d, 3JH,H = 7.2 Hz, 2H, CHarom.), 6.97 (m, 1H, py– CHarom.), 7.01 (d, 3JH,H = 7.2 Hz, 2H, CHarom.), 8.62 (m, 2H, CHarom.); 13C{1H} NMR (benzene-d6, 150 MHz) δ 28.7 (CCH3), 41.2 (C(CH3)2), 43.6 (br. s, N(CH3)2), 122.5, 123.8, 124.5, 125.0, 127.3, 128.7, 136.0 (CHarom.), 139.8 (Carom.OC), 150.5 (CN), 151.3 (C=O); MS(EI): m/z 500 (M+ – pyridine, – NMe2); Anal. Calcd. (%) for C31H43N5O3Zr: C, 59.58; H, 6.94; N, 11.21; Found: C, 59.84; H, 6.84; N, 10.99. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  N,N’-(2,2’-thiobis(2,1-phenylene))bis(tertbutylamidate)bis(dimethylamido)(pyridine)zirconium (29).  N,N’-(2,2’-  thiobis(2,1-phenylene))bis(tert-butylamide) (0.288 g, 0.75 mmol), Zr(NMe2)4 (0.200 g, 0.75 mmol) and pyridine (60 μL, 0.75 mmol) were suspended in toluene (3 mL) and the mixture was allowed to stir overnight (18 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was washed with hexanes  214  and then recrystallized from the minimal amount of hexanes at -35 °C. The colourless crystals were collected and identified as the title compound. 1  Yield: 0.239 g (0.375 mmol, 50%).  H NMR (benzene-d6, 600 MHz) δ 0.92 (s, 18H, C(CH3)3), 3.21 (s, 12H, N(CH3)2), 6.71 (m, 4H,  CHarom.), 6.88 (m, 4H, CHarom.), 6.99 (br. s, 1H, CHarom.), 7.55 (d, 3JH,H = 7.7 Hz, 2H, CHarom.), 8.62 (br. s, 2H, CHarom.); 13C{1H} NMR (benzene-d6, 150 MHz) δ 28.7 (CCH3), 41.1 (C(CH3)2), 44.1 (br. s, N(CH3)2), 123.9, 124.8, 126.5, 128.7, 129.1, 129.2, (CHarom.), 136.0 (Carom.OC), 149.5 (CN), 150.5 (C=O); MS(EI): m/z 516 (M+ – pyridine, – NMe2); Anal. Calcd. (%) for C31H43N5O2SZr: C, 58.09; H, 6.76; N, 10.93; Found: C, 58.28; H, 6.48; N, 10.74. Single crystal X-ray quality samples were obtained by recrystallization from hexanes at -35 °C.  N,N’-(2,2’-oxybis(2,1-phenylene))bis(tertbutylamidate)tris(dimethylamido)tantalum  (30).  N,N’-(2,2’-  oxybis(2,1-phenylene))bis(tert-butylamide) (0.184 g, 0.498 mmol) and Ta(NMe2)5 (0.200 g, 0.498 mmol) were suspended in toluene (3 mL) and the mixture was allowed to stir overnight (18 h) to give a white precipitate suspended in a yellow solution. The solution was filtered through a Celite plug and washed thrice with hexanes. All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal amount of hexanes at -35 °C. The yellow crystals were collected and identified as the title compound. Due to dynamic exchange processes involving the bis(amidate) ligand, NMR data is reported at 90 °C. At this temperature the appropriate signals have coalesced but appear broad. This broadening reduces the intensity of the signals such that no signals for quaternary carbons are present in the 13C NMR spectrum. Yield: 0.203 g (0.299 mmol, 60%). 1H NMR (benzene-d6, 400 MHz, 90 °C) δ 1.28 (br. s, 18H, C(CH3)3), 3.30 (br. s, 18H, N(CH3)2), 6.75, 6.86, 7.02 (br.  215  m, 8H, CHarom.);  13  C{1H} NMR (benzene-d6, 100 MHz, 90 °C) δ 28.9 (br. S, CCH3), 43.6 (br.,  N(CH3)2); MS(EI): m/z 635 (M+ – NMe2); Anal. Calcd. (%) for C28H44N5O3Ta: C, 49.48; H, 6.53; N, 10.30; Found: C, 49.66; H, 6.61; N, 10.56. Single crystal X-ray quality samples were obtained by recrystallization from hexanes at -35 °C.  N,N’-(2,2’-thiobis(2,1-phenylene))bis(tertbutylamidate)tris(dimethylamido)tantalum  (31).  N,N’-(2,2’-  thiobis(2,1-phenylene))bis(tert-butylamide) (0.192 g, 0.498 mmol) and Ta(NMe2)5 (0.200 g, 0.498 mmol) were suspended in toluene (3 mL) and the mixture was allowed to stir overnight (16 h) to give a white precipitate suspended in a yellow solution. The solution was filtered through a Celite plug and washed thrice with hexanes. All volatiles were removed in vacuo and the remaining solid was recrystallized from the minimal amount of hexanes at -35 °C. The yellow crystals were collected and identified as the title compound. Yield: 0.207 g (0.299 mmol, 60%).  1  H NMR (benzene-d6, 400 MHz) δ 1.08, 1.71  (br. s, 18H, C(CH3)3), 3.19 (br. s, 18H, N(CH3)2), 6.61, 6.69, 6.95, 7.23, 7.72 (br. m, 8H, CHarom.); 13C{1H} NMR (benzene-d6, 100 MHz) δ 28.6, 29.8 (br., CCH3), 39.3, 41.7 (br., CCH3) 47.5 (br., N(CH3)2), 121.6, 122.2, 125.6, 128.9, 131.0, 140.7, 150.0, 152.1 (Carom.), 169.4, 181.6 (C=O); MS(EI): m/z 651 (M+ – NMe2); Anal. Calcd. (%) for C28H44N5O2STa: C, 48.34; H, 6.37; N, 10.07; Found: C, 47.99; H, 6.34; N, 10.40. Single crystal X-ray quality samples were obtained by recrystallization from hexanes at -35 °C.  216  1,1’-(oxybis(2,1-phenylene))bis(3,3-diisopropylurea). This compound was synthesized according to modified literature procedure.60  1  Overall yield: 12.46 g (82%).  H NMR  (chloroform-d, 400 MHz) δ 1.22 (d, 3JH,H = 7.0 Hz, 24H, CH(CH3)2), 3.97 (m, 4H, CH(CH3)2), 6.77 (d, 3JH,H = 7.9 Hz, 2H, CHarom.), 6.87 (br s, 2H, NH), 6.90 (m, 2H, CHarom.), 7.10 (t, 2H, 3  JH,H = 7.61 Hz, CHarom.), 8.29 (d, 2H, 3JH,H = 8.22 Hz, CHarom);  13  C{1H} NMR (chloroform-d,  100 MHz) δ 21.1 (CH(CH3)2), 45.0 (CH(CH3)2), 117.2, 120.4, 122.1, 124.6 (CHarom.), 131.1 (CN), 144.3 (Carom.OC), 154.0 (C=O); MS(ESI) m/z 454 (M+), 477 (M+ + Na); Anal. Calcd. (%) for C26H38N4O3: C, 68.69; H, 8.43; N, 12.32. Found: C, 68.36; H, 8.29; N, 12.34. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  1,1’-(thiobis(2,1-phenylene))bis(3,3-diisopropylurea). This compound was synthesized according to modified literature procedure.60 Overall yield: 4.98 g (79%). 1H NMR (chloroform-d, 400 MHz) δ 1.31 (d, 3JH,H = 7.0 Hz, 24H, CH(CH3)2), 3.94 (m, 4H, CH(CH3)2), 6.96 (td, 3JH,H = 7.9 Hz, 3JH,H = 1.5 Hz, 2H, CHarom.), 7.01 (br s, 2H, NH), 7.12 (dd, 2H, 3JH,H = 7.6 Hz, 3JH,H = 1.2 Hz, CHarom.), 7.31 (td, 3JH,H = 7.0 Hz, 3JH,H = 1.5 Hz, 2H, CHarom.), 8.22 (dd, 2H, 3JH,H = 8.2 Hz, 3JH,H = 0.9 Hz, CHarom.);  13  C{1H} NMR (chloroform-d, 100 MHz) δ 21.1  (CH(CH3)2), 45.6 (CH(CH3)2), 120.5, 121.3, 123.1, 129.0 (CHarom.), 131.5 (CN), 139.3 (Carom.OC), 153.9 (C=O); MS(ESI) m/z 471 (M+ + H); Anal. Calcd. (%) for C26H38N4O2S: C, 66.35; H, 8.14; N, 11.90. Found: C, 66.21; H, 8.05; N, 11.83. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes. Disordered N(iPr)2 group was modeled using two unique parts.  217  1,1’-(oxybis(2,1-phenylene))bis(3,3diisopropylureate)bis(dimethylamido)zirconium  1,1’-  (32).  (oxybis(2,1-phenylene))bis(3,3-diisopropylurea) (0.297 g, 0.654 mmol) and Zr(NMe2)4 (0.175 g, 0.654 mmol) were suspended in toluene (3 mL) and the mixture was allowed to stir overnight (18 h) to give a clear solution. All volatiles were removed in vacuo and the remaining solid was washed thrice with hexanes and then recrystallized from the minimal amount of hot hexanes. The colourless crystals were collected and identified as the title compound. Yield: 0.145 g (0.229 mmol, 35%). 1H NMR (benzene-d6, 400 MHz) δ 0.97 (br. s, 24H, CH(CH3)2), 3.27 (s, 12H, N(CH3)2)), 6.72 (t, 3JH,H = 7.6 Hz, 2H, CHarom.), 6.83 (t, 3JH,H = 7.6 Hz, 2H, CHarom.), 6.97 (d, 3JH,H = 7.6 Hz, 2H, CHarom.), 7.03 (t, 2H, 3  JH,H = 7.6 Hz, CHarom);  13  C{1H} NMR (benzene-d6, 100 MHz) δ 21.3 (br., CH(CH3)2), 42.3  (N(CH3)2), 47.1 (br., CH(CH3)2), 121.2, 123.4, 124.5, 126.1 (CHarom.), 141.6 (CN), 152.4 (Carom.OC), 166.4 (C=O); MS(ESI) m/z 586 (M+ – NMe2); Anal. Calcd. (%) for C30H48N6O3Zr: C, 57.02; H, 7.66; N, 13.30. Found: C, 57.24; H, 8.06; N, 13.57. Single crystal X-ray quality samples were obtained by recrystallization from hot hexanes.  1,1’-(thiobis(2,1-phenylene))bis(3,3diisopropylureate)bis(dimethylamido)zirconium (thiobis(2,1-phenylene))bis(3,3-diisopropylurea)  1,1’-  (33).  (0.264  g,  0.56  mmol) and Zr(NMe2)4 (0.150 g, 0.56 mmol) were suspended in toluene (3 mL) and the mixture was allowed to stir overnight (14 h) to give a clear solution. All volatiles were removed in vacuo and the remaining solid was washed thrice with hexanes and then recrystallized from the minimal amount of pentane at -35 °C. The colourless crystals were collected and identified as  218  the title compound. Yield: 0.171 g (0.25 mmol, 44%). 1H NMR (toluene-d8, 400 MHz) δ 1.27 (br. s, 24H, CH(CH3)2), 3.19 (br. s, 4H, CH(CH3)2), 3.27 (s, 12H, N(CH3)2)), 6.72 (m, 2H, CHarom.), 6.98 (d, 3JH,H = 4.0 Hz, 4H, CHarom.), 7.54 (d, 3JH,H = 7.6 Hz, 2H, CHarom.);  13  C{1H}  NMR (toluene-d8, 100 MHz) δ 20.8 (br., CH(CH3)2), 42.3 (N(CH3)2), 46.9 (br., CH(CH3)2), 122.9, 128.7, 130.0, 134.2 (CHarom.), 137.4 (CN), 151.4 (Carom.OC), 166.2 (C=O); MS(ESI) m/z 602 (M+ – NMe2); Anal. Calcd. (%) for C30H48N6O2SZr: C, 55.60; H, 7.47; N, 12.97. Found: C, 55.95; H, 7.69; N, 13.25. Single crystal X-ray quality samples were obtained by recrystallization from pentane at -35 °C. 1,1’-(oxybis(2,1-phenylene))bis(3,3diisopropylureate)tris(dimethylamido)tantalum  (34).  1,1’-  (oxybis(2,1-phenylene))bis(3,3-diisopropylurea) (0.227 g, 0.498 mmol) and Ta(NMe2)5 (0.200 g, 0.498 mmol) were suspended in toluene (4 mL) and the mixture was allowed to stir overnight (15 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was washed with hexanes and then recrystallized from the minimal amount of pentane at -35 °C. The yellow crystals were collected and identified as the title compound. Yield: 0.332 g (0.433 mmol, 87%). 1H NMR (toluene-d8, 400 MHz, 80 °C) δ 1.20 (d, 24H, 3JH,H = 6.7 Hz, CH(CH3)2), 3.45 (s, 12H, N(CH3)2)), 3.66 (m, 4H, CH(CH3)2), 6.69 (t, 3JH,H = 7.0 Hz, 2H, CHarom.), 6.84 (d, 3JH,H = 7.6 Hz, 2H, CHarom.), 6.88 (t, 3JH,H = 7.6 Hz, 2H, CHarom.), 7.05 (d, 2H, 3JH,H = 7.6 Hz, CHarom); 13C{1H} NMR (toluene-d8, 100 MHz, 80 °C) δ 21.2 (CH(CH3)2), 45.9 (N(CH3)2), 46.4 (br., CH(CH3)2), 122.7 (CHarom.), 137.0 (CN); MS(ESI) m/z 721 (M+ – NMe2); Anal. Calcd. (%) for C32H54N7O3Ta: C, 50.19; H, 7.11; N, 12.80. Found: C, 50.27; H, 7.01; N, 12.67. Single crystal X-ray quality samples were obtained by recrystallization from pentane at -35 °C. 219  Figure 4.18 Variable temperature 400 MHz 1H NMR study of 34.  1,1’-(thiobis(2,1-phenylene))bis(3,3diisopropylureate)tris(dimethylamido)tantalum  (35).  1,1’-  (thiobis(2,1-phenylene))bis(3,3-diisopropylurea) (0.235 g, 0.498 mmol) and Ta(NMe2)5 (0.200 g, 0.498 mmol) were suspended in toluene (3 mL) and the mixture was allowed to stir overnight (20 h) to give a clear yellow solution. All volatiles were removed in vacuo and the remaining solid was washed thrice with hexanes and then recrystallized from the minimal amount of pentane at -35 °C. The yellow crystals were collected and identified as the title compound. Yield: 0.272 g (0.393 mmol, 79%).  1  H NMR  (benzene-d6, 400 MHz) δ 1.20 (br. s, 24H, CH(CH3)2), 3.12 (br. s, 4H, CH(CH3)2), 3.64 (br. s, 12H, N(CH3)2)), 6.65 (t, 3JH,H = 7.2 Hz, 2H, CHarom.), 6.90 (m, 4H, CHarom.), 7.61 (d, 3JH,H = 7.5 Hz, 2H, CHarom); 13C{1H} NMR (benzene-d6, 100 MHz) δ 20.9, 22.5 (br., CH(CH3)2), 47.2 (br., 220  CH(CH3)2), 49.8 (N(CH3)2), 121.8, 123.6, 127.6, 128.7 (CHarom.), 135.5 (br., CN), 150.1 (Carom.OC), 164.7 (br., C=O); MS(ESI) m/z 693 (M+ – 2x NMe2); Anal. Calcd. (%) for C32H54N7O2STa: C, 49.16; H, 6.96; N, 12.54. Found: C, 49.48; H, 6.83; N, 12.81. Single crystal X-ray quality samples were obtained by recrystallization from pentane at -35 °C. Disordered pentane molecule was removed from the unit cell using the SQUEEZE routine.  4.4.3  General Procedures for Catalytic Screening  The α-alkylation of amines. In a nitrogen filled glove box, Zr(NMe2)4 (6.7 mg, 0.025mmol, 5 mol%) and the appropriate diamide/diurea proligand (0.025mmol, 5 mol%) or tantalum precatalyst 30, 31, 34, 35 (0.025 mmol, 5 mol%) was weighed into a small vial and subsequently dissolved in 433.6 μL of toluene-d8 before being transferred to an NMR tube equipped with a Teflon cap. N-methylaniline (54.0 µL, 0.498 mmol), 1-octene (118.0 µL, 0.752 mmol), and a standard solution containing 1,3,5-trimethoxybenzene in toluene-d8 (66.4 µL, 1.25 mol L-1, 0.083 mmol) were then added sequentially by means of a μL-pipette. The NMR tube was closed, shaken and removed from the glove box. Following the recording of a 1H NMR spectrum, the NMR tube was placed in a preheated, 130 °C thermostated oil bath. The NMR tube was removed from the oil bath after 24 hours and another 1H NMR spectrum was recorded.  Hydroamination of aminoalkenes.  In a nitrogen filled glove box, the appropriate  diamide/diurea proligand (0.015mmol, 10 mol%) or tantalum precatalyst 30, 31, 34, 35 (0.015 mmol, 10 mol%) was weighed into a small vial and subsequently dissolved in 300.0 μL of toluene-d8 before being transferred to an NMR tube equipped with a Teflon cap. Standard solutions containing 1,3,5-trimethoxybenzene (40.0 µL, 1.25 mol L-1, 0.083 mmol), the  221  aminoalkene substrate (100.0 µL, 1.5 mol L-1, 0.15 mmol), and Zr(NMe2)4 (if appropriate, 100.0 µL, 0.15 mol L-1, 0.015 mmol, 10 mol%) in toluene-d8 were then added sequentially by means of a μL-pipette. The corresponding amount of toluene-d8 required for a total volume of 600 µL (60 µL for Zr systems, 160 µL for Ta systems) was added before the NMR tube was closed, shaken and removed from the glove box. Following the recording of a 1H NMR spectrum, the NMR tube was placed in a preheated, 100 °C thermostated oil bath. The NMR tube was removed from the oil bath after 2 hours and another 1H NMR spectrum was recorded. For the secondary aminoalkene substrate, the NMR tube was reintroduced to the oil bath for an additional 16 hours before a final 1H NMR spectrum was recorded.  4.4.4  General Procedure for Electrochemical Analysis All electrodes were fitted with a b14 septum with a 5/16” opening cut by a cork borer.  Before being brought into an argon filled glove box, the working (platinum button, 2 mm diameter) and counter (glassy carbon, 3 mm diameter) electrodes were polished with 0.3, 0.1 and 0.05 μm powder polish (alumina slurry) sequentially, sonicated for 10 minutes in deionized water, and rinsed with acetone. The reference electrode (99.9% silver wire, 0.25 mm diameter) was exposed to the flame of a butane torch and wiped with a fine polish pad in order to remove the residual organic matter before being brought into the glove box with the other electrodes. The analyte, either TBAPF6 (387 mg, 1 mmol) or TBABPh4 (562 mg, 1 mmol), was dissolved in 10 mL of dry THF (0.1 mol L-1) that had been sparged with argon and transferred to a 25 mL 3necked round bottom flask with b14 female joints. The electrodes were connected to the appropriate leads from the potentiostat and inserted into the necks of the flask with the counter electrode installed in the central position. Once the solution had settled, a cyclic voltammogram  222  (100 mV/s, -2.8 to 1.5 V, negative scan starting at 0 V) was collected to act as the baseline for the experiment. Upon successful completion of the baseline the analyte (0.05 mmol, 0.005 mol L-1) was dissolved in the electrolyte/THF solution and added to the flask and further CV experiments were conducted. Between each run, the flask was agitated to aid with diffusion and if a decrease in signal intensity was observed, the electrodes were wiped with a KimWipe. After completion of the set of experiments, a single crystal of ferrocene (Fc) was added to the solution and a cyclic voltammogram (100 mV/s, -2.8 to 1.5 V, negative scan starting at 0 V) was collected for reference.  223  Chapter 5: Summary and Future Directions  5.1  Summary The work contained in this thesis is largely focused on amidate complexes of tantalum  and their role as catalysts for the α-alkylation of amines. The mechanism of this reaction was explored with isotopic labeling studies and kinetic investigations as well as stoichiometric experiments with bis(amidate) model compounds. Based on these findings, transition states and intermediates along the catalytic cycle were modeled using quantum chemical calculations. Additionally, a series of tethered bis(amidate) and bis(ureate) ligands were installed on Ta and Zr and the resulting complexes were characterized using cyclic voltammetry and screened as precatalysts for the α-alkylation of amines and hydroamination. In Chapter 2, a series of mono(amidate) tantalum complexes was synthesized and screened as precatalysts for the α-alkylation of amines. The catalytic activity of the precatalysts was found to be related to the steric bulk of the amidate ligand with the sterically encumbered κ2(N,O) mono(amidate) complex 3 identified as the most active precatalyst. In contrast, complex 5, which contains a bulky, κ1(O) bound amidate ligand, was found to be the least active catalyst which was proposed to be a result of the κ1(O) binding mode. Varying the electronic nature of the amidate ligand gives rise to comparable solid-state geometry and metrical parameters but drastically influences catalytic activity, as evidenced by the significant decrease in catalyst performance observed for complex 4, supported by a benzamidate ligand, when compared to the pivalamidate supported 3. Bis(amidate) complexes of tantalum were also synthesized and although these species are not effective precatalysts for the α-alkylation of amines, they do offer the opportunity to  224  investigate fundamental properties of amidate ligands and isolate model catalytic intermediates. The bis(amidate) complex 7, featuring both κ1(O) and κ2(N,O) binding modes for the amidate ligands, was subjected to a variable temperature 1H NMR spectroscopy study that explored the hemi-lability of amidate ligands as well as tantalaziridine formation at elevated temperatures. The assignment of tantalaziridine formation with amidate complexes was supported by the isolation and characterization of a bis(amidate) tantalaziridine formed via C–H activation at ambient temperatures. Using solid-state and solution phase data, complex 9 was confirmed as an azametallacyclopropane moiety bound to a d0 Ta(V) centre. Interestingly, the Ta–C bond was found to be unreactive towards stoichiometric alkene insertion but did react with acetonitrile to generate a representative 5-membered metallacycle, complex 10. Chapter 2 also describes a set of experiments aimed at elucidating the mechanism for the α-alkylation of amines using both stoichiometric and kinetic investigations. Isotopic labelling studies revealed non-productive off-cycle pathways and showed that complex 3 could undergo rapid and reversible tantalaziridine formation under mild conditions.  From kinetic  investigations, the rate of reaction was shown to be independent of amine concentration but is influenced by catalyst loading, though saturation kinetics are observed at loading above 5 mol%. Though the quantitative monitoring of alkene consumption is complicated by the formation of byproducts, an increase in initial alkene concentration correlates to an increase in the rate of product formation. Based on these mechanistic investigations, and the aforementioned model systems, the turnover limiting step for the α-alkylation of amines catalyzed by mono(amidate) tantalum complexes is hypothesised to be alkene insertion into the Ta–C bond of a tantalaziridine.  225  To investigate the possibility of a radical mechanism for the α-alkylation of amines, cyclopropyl moieties were incorporated into ligand backbones and an alkene substrate. While experiments with the precatalysts bearing the cyclopropyl groups were consistent with an ionic mechanism, catalysis performed with the cyclopropyl alkene substrate definitively eliminated the possibility of a radical stepwise insertion. In order to supplement the experimental mechanistic work done in Chapter 2, the proposed catalytic cycle was modeled using quantum chemical calculations in Chapter 3. Initially performed using dimethylamine as a model substrate, a complete set of transition states and minima were calculated using complex 3 as the precatalyst. Throughout the investigation, reactivity in the plane of the amidate chelate was found to be the lowest energy pathway. Though the majority of calculated structures have a κ2(N,O) bound amidate ligand, the ligand was shown to adopt a κ1(O) binding mode when additional space was required in the tantalum coordination sphere. This notion was confirmed when dimethylamine was replaced by the bulkier N-methylaniline, with multiple transition states and intermediates calculated as κ1(O) amidate species. The potential energy surface generated for the use of N-methylaniline as a substrate suggests that protonolysis is the turnover limiting step, not alkene insertion as indicated by experimental results. The regeneration of the active species to close the calculated catalytic cycle results in a reversal of tantalaziridine geometry relative to the κ2(N,O) amidate ligand.  Although the  transition states and intermediates calculated for this new pathway differ geometrically and energetically, protonolysis is maintained as the turnover limiting step. Other routes to align the computed findings with experimental data, including cross-over pathways and axial ligand  226  exchange, proved to be fruitless. However, calculated molecular orbitals and energetic barriers agree with the experimental evidence that alkene insertion is not a radical process. In addition to quantum chemical calculations, preliminary efforts to establish a steric parameter that correlates to catalyst activity were described in Chapter 3. Upon development of a method for generating asymmetric amidate ligand input files, the buried volume (%Vbur) and solid angle (GL, Gspace) steric parameters were calculated for mono(amidate) complexes 1 – 5 as well as a series of axially chiral bis(amidate) precatalysts. Unfortunately, no universal trends were identified when correlating the parameters to catalytic activity, perhaps due to the difficulties in accurately describing solution phase behaviour in these systems. Chapter 4 describes attempts to build upon the success of early transition metal complexes supported by diamido ligands bearing a neutral donor in the tether by replacing the amido contacts with amidate N,O chelates.  A series of diamide proligands containing a  chalcogen donor were synthesized and installed on zirconium and tantalum.  Bis(amidate)  ligands supporting the metal complexes adopt different binding modes (κ4(N,N,O,O) for Zr, κ3(N,O,O) for Ta) with dynamic behaviour of the ligand observed in solution at ambient temperatures for the tantalum species. Diurea proligands based on the same aryl backbone were also synthesized and installed on Zr and Ta. The bis(ureate) ligands of the Zr complexes adopts a κ4(N,N,O,O) binding mode with the N,O chelate units cis to one another, a similar binding mode observed for the more flexible thio-bridged bis(ureate) bound to tantalum. In contrast, the oxy-bridged bis(ureate) Ta species is bound κ3(N,O,O) allowing the chalcogen donor in the tether closer to the metal centre than for any other complex. In addition to solid-state and spectroscopic characterization, the bis(amidate) and bis(ureate) species were analysed using electrochemical methods, specifically cyclic  227  voltammetry. Non-reversible oxidative events are found for a bis(amidate) Zr complex while the corresponding tantalum species undergo reversible reductions, assigned to the Ta(V) / Ta(IV) couple. Quasi-reversible reductive events are observed for a bis(ureate) Ta complex under the same conditions. The bis(amidate) and bis(ureate) complexes were also subjected to screening as precatalysts for the α-alkylation of amines as well as the hydroamination of aminoalkenes. The zirconium species showed no reactivity for the α-alkylation of amines while the tantalum precatalysts were significantly less reactive than benchmark complex 3. The trend is reversed for hydroamination, with full conversion of the primary aminoalkene substrate observed for the Zr precatalysts. The bis(ureate) ligated complexes display higher activity than the bis(amidate) species though conversions are low compared to the alkyl-tethered bis(ureate) zirconium benchmark. This decrease in reactivity is proposed to be a result of the non-planar binding geometry of the κ4(N,N,O,O) bis(ureate) ligand which is more sterically encumbering compared to the in-plane bound alkyl-tethered ligand. These results suggest that the mere presence of a tether in the ligand framework does not positively affect catalysis and that the nature of the tether influences the binding geometry of the ligand.  5.2 5.2.1  Future Directions In Situ Catalyst Monitoring The precatalysts presented in this thesis have been rigorously characterized in the solid  state as well as in solution. However, characterizing catalytic intermediates during the course of a reaction is difficult. Since the reaction progress is monitored by 1H NMR spectroscopy, this would be a convenient method for also observing the various catalytic species present in  228  solution. Regrettably, the catalyst is found in low concentrations relative to the substrates in solution resulting in obscured signals. Also, the wide range of isomers possible for amidate catalysts makes conclusive characterization even more challenging. However, by isotopically enriching the amidate ligand with an NMR active nucleus, the only signals present in the spectrum would be associated with the catalytic species in solution. To this end, the possibility of synthesizing  15  N-enriched amide proligands was initially investigated by Alexandru  Vlasceanu, an undergraduate researcher supervised by the author.207 As isotopically enriched compounds are expensive, a synthetic path was developed that uses labeled ammonium chloride (15NH4Cl) as the  15  N source (Scheme 5.1). Early in the  investigations, it was discovered that the proposed route was not compatible with 2,6disubstituted benzoic acids (vide infra). However, the route did prove useful for the synthesis of 15  N-aniline and subsequently 15N-phenylpivalamide (Scheme 5.2).  Scheme 5.1 Proposed synthetic route for 15N-enriched amide proligands.  229  Scheme 5.2 Synthesis of isotopically enriched 15N-phenylpivalamide.  Starting from commercially available benzoyl chloride, reaction with well under basic conditions to generate  15  15  NH4Cl proceeds  N-benzamide in 77% yield. The species is found to  undergo a Hoffman rearrangement using the mild oxidant (diacetoxy)iodobenzene (DIB) to give a carbamate which is cleaved under basic conditions yielding the  15  N-aniline.208 Subsequent  reaction with pivaloyl chloride completes the unoptimized synthesis of the proligand in 39% overall yield. One signal is present in the  15  N{1H} NMR spectrum at δN = -253.1 while the  proton coupled spectrum shows a coupling constant of 89.3 Hz. A corresponding splitting is found in the 1H spectrum as two signals for the N–H proton are found (δH = 7.26, 7.48) with a one bond J-coupling of 88.8 Hz. The enriched ligand can be installed on tantalum via protonolysis with Ta(NMe2)5 akin to complexes 1 (Scheme 2.17) and 7 (Scheme 2.20). The identities of confirmed by comparison of 1H and  13  15  N-1 and  15  N-7 were  C NMR data to their unlabeled congeners before  undergoing further investigation. As expected, the 15N NMR spectrum for 15N-1 shows only one signal that is not split in the proton coupled spectrum. Located at δN = -180.3, the signal for the κ2(N,O) bound amidate is shifted significantly downfield compared to the proligand. Interestingly, an almost identical shift (δN = -181.2) is observed in the  15  N NMR spectrum of 230  15  N-7 (Figure 5.1) along with an additional signal, attributable to the κ1(O) ligand, at δN = -134.8.  This suggests that the κ1(O) and κ2(N,O) binding environments are clearly distinguishable using 15  N NMR spectroscopy.  Figure 5.1 15N spectrum (40 MHz, 25 °C) of 15N-7.  Notably, the signals in  15  N NMR spectrum of  15  N-7 collected at room temperature are  sharp and well defined compared to the broad signals observed using 1H NMR spectroscopy. Preliminary variable temperature  15  N NMR spectroscopy studies do show a broadening of  signals at elevated temperatures but no coalescence is observed at this time. As evidenced by the signal to noise ratio, the  15  N NMR spectroscopy is considerably less sensitive than its 1H  equivalent. Despite this shortcoming, directly probing the binding mode of an amidate ligand is of great interest and to this end a more general method of synthesizing  15  N-labeled amide  proligands was investigated. 231  Based on a recent report, it was found that 2,6-disubstituted benzoic acids could be reacted directly with  15  NH4Cl in the solid state to yield a primary amide intermediate (Scheme  5.1).209 Though initially promising, this modified route was abandoned when attempts to induce a Hoffman rearrangement with a variety of oxidants proved unsuccessful. A new route was formulated with the  15  N label installed via cross-coupling with a boronic acid using Cu2O as a  catalyst (Scheme 5.3).210 Unfortunately, the installation of the  15  N label was never realized but  the route remains a viable method of synthesizing isotopically labeled amide proligands and is a promising direction for future research.  Scheme 5.3 Proposed synthesis of 2,6-disubstituted 15N-enriched amide proligands. Mesitylbromide used as an example starting material.  5.2.2  Alternate Mechanistic Pathways As discussed in Chapter 3, the potential energy surfaces for the calculated transition  states and intermediates of the catalytic α-alkylation of amines do not agree with experimental results: whereas stoichiometric and kinetic data predict alkene insertion to be the turnover limiting step, quantum chemical calculations suggest protonolysis to be turnover limiting. Though significant effort was made to find the lowest energy pathway, it is not possible to model every alternative pathway that may align experimental and computational findings. To this end,  232  two of the most promising alternative pathways that lower the energy of the protonolysis transition state are highlighted here as possibilities for future investigations. The first pathway probes the possibility of productive interaction between species on the catalytic cycle and additional equivalents of amine substrate found in solution. The starting point of this investigation would be a transition state combining the alkene insertion and metallacycle protonolysis steps (Scheme 5.4), akin to the species proposed for intramolecular hydroamination catalyzed by a bis(ureate) zirconium complex.176, 202 This concerted insertionprotonolysis transition state was initially proposed by Hultzsch and coworkers based on isotopic labeling studies.44 In the proposed structure, an additional equivalent of amine substrate is datively bound to the Ta centre in a highly-ordered chair-like transition state. The proton from the amine substrate is protonating the terminal carbon in the alkene substrate, which is simultaneously forming a C–C bond with the tantalaziridine carbon.  Scheme 5.4 Proposed alternative protonolysis transition state for the α-alkylation of amines involving the participation of an additional equivalent of amine substrate.  The amidate ligand is expected to adopt a κ1(O) binding mode, as seen for both aforementioned protonolysis transition states using N-methylaniline as a substrate. Relaxation of  233  the transition state would give a bis(amido) species that is poised for β-hydrogen abstraction to regenerate the active tantalaziridine catalyst, isostructural to the intermediate found for the previously calculated PES.  This theory is supported by the observation of calculated  intermediates with datively bound amines, including tantalaziridines (vide supra). The second alternative pathway that deserves further investigation involves a κ1(O) amidate ligand acting as a proton shuttle to reduce the energy of the protonolysis transition state. This mode of ligand participation is similar to the concerted metalation-deprotonation (CMD) mechanism established for late transition metal mechanisms.94  However, instead of the  participating ligand sequestering the proton, the imine moiety of the κ1(O) ligand delivers the proton to the reactive Ta–C bond (Scheme 5.5). Upon delivery, the amidate ligand binds to the metal centre in a κ2(N,O) fashion to stabilize the positive charge throughout the chelate before attack by the negatively charged N-methylphenyl amido. Notably, in order for this pathway to be viable, the tantalaziridine must possess a geometry such that the Oamidate cis to the Naziridine.  234  Scheme 5.5 Proposed alternative protonolysis pathway for the α-alkylation of amines involving a κ1(O) amidate ligand as a proton shuttle.  5.3  Concluding Remarks The findings within this thesis have highlighted some important overarching notions  about amidates as ligands as well as design features of catalysts for the α-alkylation of amines. From the series of tantalum amidate complexes screened in Chapter 2, the most active catalysts are supported by one amidate ligand bound κ2(N,O) in the solid state and solution phase. Bis(amidate) species, including the tethered species discussed in Chapter 4, were found to be significantly less reactive than mono(amidate) complexes while the κ1(O) binding mode observed in the solid-state molecular structure of 5 correlates to poor catalytic performance and non-productive reactivity. Although a lack of steric bulk at the 2,6-positions of the amidate aryl group decreases catalytic activity, too much bulk causes a shift to the κ1(O) binding mode which is far more detrimental to catalytic performance.  235  That being said, from the expanded substrate scope of 3 it is clear that amidate ligands offer an advantage over the other supporting ligands reported for α-alkylation of amines catalysts.92-93 As seen in the computational investigations in Chapter 3, amidate ligands have the ability to adopt a κ1(O) binding mode when necessary and then revert to the more robust κ2(N,O) when needed. Of the precatalysts reported for the α-alkylation of amines, this ligand hemilability is unique to mono(amidate) precatalysts and may be the key to explaining their broad substrate scope. However, this hemi-lability is also a burden when attempting to develop a predictive tool based on steric parameters as it is difficult to quantify the equilibrium between binding modes in solution. In conclusion, the research presented herein will aid in the development of further generations of catalysts supported by amidate ligands. From a deeper fundamental knowledge of amidate ligand behaviour to the introduction of new techniques to the arsenal of complex characterization, this work will enable others to take meaningful steps towards an ideal catalyst for the α-alkylation of amines.  236  References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.  Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 259. Organization for Economic Co-operation and Development Environmental Outlook for the Chemicals Industry; 2001. Ananikov, V. P.; Tanaka, M., Hydrofunctionalization. Springer: Heidelberg New York Dordrecht London, 2013; Vol. 43, p 367. Nugent, T. C., Chiral Amine Synthesis. Methods, Developments and Applications. February 2010 ed.; Wiley-VCH: Weinheim, Germany, 2010; Vol. I, p 494. Anastas, P. T.; Warner, J. C., Green Chemistry: Theory and Practice. Oxford University Press: New York, 1998. Roesky, P. W. Angew. Chem., Int. Ed. 2009, 48, 4892. Clerici, M. G.; Maspero, F. Synthesis 1980, 305. Nugent, W. A.; Ovenall, D. W.; Holmes, S. J. Organometallics 1983, 2, 161. Lauzon, J. M. P.; Schafer, L. L. Dalton Trans. 2012, 41, 11539. Cummings, S. A. T., Jon A.; Norton, Jack R. Top. Organomet. Chem. 2005, 10, 1. Buchwald, S. L.; Watson, B. T.; Wannamaker, M. W.; Dewan, J. C. J. Am. Chem. Soc. 1989, 111, 4486. Coles, N.; Whitby, R. J.; Blagg, J. Synlett 1992, 143. Coles, N.; Whitby, R. J.; Blagg, J. Synlett 1990, 271. Nugent, W. A.; Thorn, D. L.; Harlow, R. L. J. Am. Chem. Soc. 1987, 109, 2788. Nugent, W. A.; Calabrese, J. C. J. Am. Chem. Soc. 1984, 106, 6422. Hart, D. W.; Blackburn, T. F.; Schwartz, J. J. Am. Chem. Soc. 1975, 97, 679. Grossman, R. B.; Davis, W. M.; Buchwald, S. L. J. Am. Chem. Soc. 1991, 113, 2321. Barluenga, J.; Sanz, R.; Fañanás, F. J. J. Org. Chem. 1997, 62, 5953. Harris, M. C. J.; Whitby, R. J. Tetrahedron Lett. 1995, 36, 4287. Gately, D. A.; Norton, J. R.; Goodson, P. A. J. Am. Chem. Soc. 1995, 117, 986. Tunge, J. A.; Czerwinski, C. J.; Gately, D. A.; Norton, J. R. Organometallics 2001, 20, 254. Jensen, M.; Livinghouse, T. J. Am. Chem. Soc. 1989, 111, 4495. Buchwald, S. L.; Wannamaker, M. W.; Watson, B. T. J. Am. Chem. Soc. 1989, 111, 776. Honda, T.; Mori, M. Organometallics 1996, 15, 5464. Honda, T.; Mori, M. J. Org. Chem. 1996, 61, 1196. Dekura, F.; Honda, T.; Mori, M. Chem. Lett. 1997, 26, 825. Bexrud, J. A.; Eisenberger, P.; Leitch, D. C.; Payne, P. R.; Schafer, L. L. J. Am. Chem. Soc. 2009, 131, 2116. Zhai, H.; Borzenko, A.; Lau, Y. Y.; Ahn, S. H.; Schafer, L. L. Angew. Chem., Int. Ed. 2012, 51, 12219. Zhang, Z.; Leitch, D. C.; Lu, M.; Patrick, B. O.; Schafer, L. L. Chem. Eur. J. 2007, 13, 2012. Ayinla, R. O.; Schafer, L. L. Inorg. Chim. Acta 2006, 359, 3097. Lee, A. V.; Schafer, L. L. Synlett 2006, 18, 2973. Zhang, Z.; Schafer, L. L. Org. Lett. 2003, 5, 4733. Müller, C.; Saak, W.; Doye, S. Eur. J. Org. Chem. 2008, 2008, 2731. Kubiak, R.; Prochnow, I.; Doye, S. Angew. Chem., Int. Ed. 2009, 48, 1153. 237  35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68.  Prochnow, I.; Kubiak, R.; Frey, O. N.; Beckhaus, R.; Doye, S. ChemCatChem 2009, 1, 162. Kubiak, R.; Prochnow, I.; Doye, S. Angew. Chem., Int. Ed. 2010, 49, 2626. Jasper, D. S., Wolfgang; Doye, S. Synlett 2012, 23, 2098. Dörfler, J.; Doye, S. Angew. Chem., Int. Ed. 2013, 52, 1806. Herzon, S. B.; Hartwig, J. F. J. Am. Chem. Soc. 2007, 129, 6690. Herzon, S. B.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 14940. See Supporting Information for Herzon, S. B. H., J. F. J. Am. Chem. Soc. 2008, 130, 14940. Zhang, F.; Song, H.; Zi, G. Dalton Trans. 2011, 40, 1547. Zi, G.; Zhang, F.; Song, H. Chem. Commum. 2010, 46, 6296. Reznichenko, A. L.; Hultzsch, K. C. J. Am. Chem. Soc. 2012, 134, 3300. Reznichenko, A. L.; Emge, T. J.; Audorsch, S.; Klauber, E. G.; Hultzsch, K. C.; Schmidt, B. Organometallics 2011, 30, 921. Jun, C.-H. Chem. Commum. 1998, 1405. Reger, D. L.; Garza, D. G.; Baxter, J. C. Organometallics 1990, 9, 873. Chatani, N.; Asaumi, T.; Yorimitsu, S.; Ikeda, T.; Kakiuchi, F.; Murai, S. J. Am. Chem. Soc. 2001, 123, 10935. Pan, S.; Endo, K.; Shibata, T. Org. Lett. 2011, 13, 4692. Lee, A. V.; Schafer, L. L. Eur. J. Inorg. Chem. 2007, 2243. Thomson, R. K.; Zahariev, F. E.; Zhang, Z.; Patrick, B. O.; Wang, Y. A.; Schafer, L. L. Inorg. Chem. 2005, 44, 8680. Stanlake, L. J. E.; Beard, J. D.; Schafer, L. L. Inorg. Chem. 2008, 47, 8062. Stanlake, L. J. E.; Schafer, L. L. Organometallics 2009, 28, 3990. Thomson, J. A.; Schafer, L. L. Dalton Trans. 2012, 41, 7897. Lee, A. V.; Sajitz, M.; Schafer, L. L. Synthesis 2009, 97. Wood, M. C.; Leitch, D. C.; Yeung, C. S.; Kozak, J. A.; Schafer, L. L. Angew. Chem., Int. Ed. 2007, 46, 354. Thomson, R. K.; Bexrud, J. A.; Schafer, L. L. Organometallics 2006, 25, 4069. Li, C.; Thomson, R. K.; Gillon, B.; Patrick, B. O.; Schafer, L. L. Chem. Commum. 2003, 2462. Walsh, P. J.; Baranger, A. M.; Bergman, R. G. J. Am. Chem. Soc. 1992, 114, 1708. Leitch, D. C.; Beard, J. D.; Thomson, R. K.; Wright, V. A.; Patrick, B. O.; Schafer, L. L. Eur. J. Inorg. Chem. 2009, 2009, 2691. Thomson, R. K.; Patrick, B. O.; Schafer, L. L. Can. J. Chem. 2005, 83, 1037. Thomson, R. K.; Schafer, L. L. Organometallics 2010, 29, 3546. Gott, A. L.; Clarke, A. J.; Clarkson, G. J.; Scott, P. Organometallics 2007, 26, 1729. Durfee, L. D.; Rothwell, I. P. Chem. Rev. 1988, 88, 1059. Castellano, B.; Solari, E.; Floriani, C.; Re, N.; Chiesi-Villa, A.; Rizzoli, C. Chem. Eur. J. 1999, 5, 722. Fandos, R.; Hernandez, C.; Otero, A.; Rodriguez, A.; Ruiz, M. J.; Terreros, P. Eur. J. Inorg. Chem. 2003, 493. Fandos, R.; López-Solera, I.; Otero, A.; Rodríguez, A.; Ruiz, M. J.; Terreros, P. Organometallics 2004, 23, 5030. Strickler, J. R.; Bruck, M. A.; Wigley, D. E. J. Am. Chem. Soc. 1990, 112, 2814.  238  69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96.  Smith, D. P.; Strickler, J. R.; Gray, S. D.; Bruck, M. A.; Holmes, R. S.; Wigley, D. E. Organometallics 1992, 11, 1275. Fandos, R.; Fern ndez-Gallardo, .; L pez-Solera, M. a. I.; Otero, A.; Rodr guez, A.; Ruiz, M. a. J.; Terreros, P. Organometallics 2008, 27, 4803. Bonanno, J. B.; Veige, A. S.; Wolczanski, P. T.; Lobkovsky, E. B. Inorg. Chim. Acta 2003, 345, 173. Neithamer, D. R.; Parkanyi, L.; Mitchell, J. F.; Wolczanski, P. T. J. Am. Chem. Soc. 1988, 110, 4421. Covert, K. J.; Neithamer, D. R.; Zonnevylle, M. C.; LaPointe, R. E.; Schaller, C. P.; Wolczanski, P. T. Inorg. Chem. 1991, 30, 2494. Rankin, M. A.; Cummins, C. C. J. Am. Chem. Soc. 2010, 132, 10021. Takahashi, Y.; Onoyama, N.; Ishikawa, Y.; Motojima, S.; Sugiyama, K. Chem. Lett. 1978, 525. Airoldi, C.; Bradley, D. C.; Vuru, G. Transition Met. Chem. 1979, 4, 64. Scoles, L.; Ruppa, K. B. P.; Gambarotta, S. J. Am. Chem. Soc. 1996, 118, 2529. Mayer, J. M.; Curtis, C. J.; Bercaw, J. E. J. Am. Chem. Soc. 1983, 105, 2651. Bazinet, P.; Yap, G. P. A.; Richeson, D. S. Organometallics 2001, 20, 4129. Spencer, L. P.; Beddie, C.; Hall, M. B.; Fryzuk, M. D. J. Am. Chem. Soc. 2006, 128, 12531. Boncella, J. M.; Cajigal, M. L.; Abboud, K. A. Organometallics 1996, 15, 1905. Fandos, R.; Hernández, C.; López-Solera, I.; Otero, A.; Rodríguez, A.; Ruiz, M. J.; Terreros, P. Organometallics 2000, 19, 5318. Galakhov, M. V.; Gomez, M.; Jimenez, G.; Pellinghelli, M. A.; Royo, P.; Tiripicchio, A. Organometallics 1994, 13, 1564. Galakhov, M. V.; Gómez, M.; Gómez-Sal, P.; Velasco, P. Organometallics 2005, 24, 848. Galakhov, M. V.; Gomez, M.; Jimenez, G.; Royo, P.; Pellinghelli, M. A.; Tiripicchio, A. Organometallics 1995, 14, 2843. Matsuo, Y.; Mashima, K.; Tani, K. Organometallics 2002, 21, 138. Takai, K.; Ishiyama, T.; Yasue, H.; Nobunaka, T.; Itoh, M.; Oshiki, T.; Mashima, K.; Tani, K. Organometallics 1998, 17, 5128. Mayer, J. M. Ph.D. Thesis. California Institute of Technology, Pasadena, California, USA, 1983. Thomson, R. K. Ph.D. Thesis. University of British Columbia, Vancouver, British Columbia, Canada, 2008. Bradley, D. C.; Thomas, I. M. Can. J. Chem. 1962, 40, 449. Chen, S.-J.; Zhang, X.-H.; Yu, X.; Qiu, H.; Yap, G. P. A.; Guzei, I. A.; Lin, Z.; Wu, Y.D.; Xue, Z.-L. J. Am. Chem. Soc. 2007, 129, 14408. Eisenberger, P.; Schafer, L. L. Pure Appl. Chem. 2010, 82, 1503. Eisenberger, P.; Ayinla, R. O.; Lauzon, J. M. P.; Schafer, L. L. Angew. Chem., Int. Ed. 2009, 48, 8361. Lapointe, D.; Fagnou, K. Chem. Lett. 2010, 39, 1118. Liu, W.; Cao, H.; Zhang, H.; Zhang, H.; Chung, K. H.; He, C.; Wang, H.; Kwong, F. Y.; Lei, A. J. Am. Chem. Soc. 2010, 132, 16737. Qiu, H.; Chen, S.-J.; Wang, C.-S.; Wu, Y.-D.; Guzei, I. A.; Chen, X.-T.; Xue, Z.-L. Inorg. Chem. 2009, 48, 3073. 239  97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114.  115. 116. 117. 118. 119.  120.  Zhang, X.-H.; Chen, S.-J.; Cai, H.; Im, H.-J.; Chen, T.; Yu, X.; Chen, X.; Lin, Z.; Wu, Y.-D.; Xue, Z.-L. Organometallics 2008, 27, 1338. Ayinla, R. O. Ph.D. Thesis. University of British Columbia, Vancouver, British Columbia, Canada, 2011. Barroso, S.; Cui, J.; Carretas, J. M.; Cruz, A.; Santos, I. C.; Duarte, M. T.; Telo, J. P.; Marques, N.; Martins, A. M. Organometallics 2009, 28, 3449. Suzuki, E.; Komuro, T.; Okazaki, M.; Tobita, H. Organometallics 2007, 26, 4379. Stroot, J.; Saak, W.; Haase, D.; Beckhaus, R. Z. Anorg. Allg. Chem. 2002, 628, 755. Cohen, S. A.; Bercaw, J. E. Organometallics 1985, 4, 1006. Prochnow, I.; Zark, P.; Müller, T.; Doye, S. Angew. Chem., Int. Ed. 2011, 50, 6401. Tau, P.; Nyokong, T. J. Porphyrins Phthalocyanines 2006, 10, 69. Arney, D. J.; Bruck, M. A.; Wigley, D. E. Organometallics 1991, 10, 3947. Epshteyn, A.; Zavalij, P. Y.; Sita, L. R. J. Am. Chem. Soc. 2006, 128, 16052. Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317. Newcomb, M.; Toy, P. H. Acc. Chem. Res. 2000, 33, 449. Milnes, K. K.; Gottschling, S. E.; Baines, K. M. Org. Biomol. Chem. 2004, 2, 3530. Newcomb, M.; Johnson, C. C.; Manek, M. B.; Varick, T. R. J. Am. Chem. Soc. 1992, 114, 10915. Le Tadic-Biadatti, M.-H.; Newcomb, M. J. Chem. Soc., Perkin Trans. 2 1996, 0, 1467. Theberge, C. R.; Verbicky, C. A.; Zercher, C. K. J. Org. Chem. 1996, 61, 8792. Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997, 62, 7512. Shoemaker, R. Determining the Energy of Activation Parameters from Dynamic NMR Experiments. http://chemnmr.colorado.edu/manuals/DNMR_Calculations.pdf (accessed January 12, 2013). Houk, K. N.; Cheong, P. H.-Y. Nature 2008, 455, 309. Fey, N. Dalton Trans. 2010, 39, 296. Hepperle, S. S. Ph.D. Thesis. University of British Columbia, Vancouver, British Columbia, Canada, 2011. Detailed information regarding methodology can be found in Section 3.4. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision C.01, 2009. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. 240  121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146.  E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Laham, A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision E.01, 2003. Becke, A. D. J. Chem. Phys. 1993, 98, 5648. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785. Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200. Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623. Perdew, J. P.; Ernzerhof, M.; Burke, K. J. Chem. Phys. 1996, 105, 9982. Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270. Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299. Møller, C.; Plesset, M. S. Phys. Rev. 1934, 46, 618. See Appendix C for details. Watson, L.; Eisenstein, O. J. Chem. Educ. 2002, 79, 1269. Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.; Echeverria, J.; Cremades, E.; Barragan, F.; Alvarez, S. Dalton Trans. 2008, 0, 2832. Batsanov, S. S. Inorganic Materials 2001, 37, 871. Tolman, C. A. Chem. Rev. 1977, 77, 313. Taft, R. W. J. Am. Chem. Soc. 1952, 74, 3120. Hillier, A. C.; Sommer, W. J.; Yong, B. S.; Petersen, J. L.; Cavallo, L.; Nolan, S. P. Organometallics 2003, 22, 4322. Guzei, I. A.; Wendt, M. Dalton Trans. 2006, 3991. Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.; Scarano, V.; Cavallo, L. SambVca @ MoLNaC. https://www.molnac.unisa.it/OMtools/sambvca.php. Guzei, I. A.; Wendt, M. Solid-G, 2005. Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.; Scarano, V.; Cavallo, L. Eur. J. Inorg. Chem. 2009, 2009, 1759. Kelly Iii, R. A.; Clavier, H.; Giudice, S.; Scott, N. M.; Stevens, E. D.; Bordner, J.; Samardjiev, I.; Hoff, C. D.; Cavallo, L.; Nolan, S. P. Organometallics 2007, 27, 202. Viciu, M. S.; Navarro, O.; Germaneau, R. F.; Kelly, R. A.; Sommer, W.; Marion, N.; Stevens, E. D.; Cavallo, L.; Nolan, S. P. Organometallics 2004, 23, 1629. Clavier, H.; Correa, A.; Cavallo, L.; Escudero-Adán, E. C.; Benet-Buchholz, J.; Slawin, A. M. Z.; Nolan, S. P. Eur. J. Inorg. Chem. 2009, 2009, 1767. Bazinet, P.; Ong, T.-G.; O'Brien, J. S.; Lavoie, N.; Bell, E.; Yap, G. P. A.; Korobkov, I.; Richeson, D. S. Organometallics 2007, 26, 2885. Dorta, R.; Stevens, E. D.; Scott, N. M.; Costabile, C.; Cavallo, L.; Hoff, C. D.; Nolan, S. P. J. Am. Chem. Soc. 2005, 127, 2485. Gaillard, S.; Bantreil, X.; Slawin, A. M. Z.; Nolan, S. P. Dalton Trans. 2009, 6967. W rtz, S.; Lohre, C.; Fr hlich, R.; Bergander, K.; Glorius, F. J. Am. Chem. Soc. 2009, 131, 8344. 241  147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178.  Clavier, H.; Nolan, S. P. Chem. Commum. 2010, 46, 841. Mandell, C. L.; Kleinbach, S. S.; Dougherty, W. G.; Kassel, W. S.; Nataro, C. Inorg. Chem. 2010, 49, 9718. DiFranco, S. A.; Maciulis, N. A.; Staples, R. J.; Batrice, R. J.; Odom, A. L. Inorg. Chem. 2011, 51, 1187. Dible, B. R.; Cowley, R. E.; Holland, P. L. Organometallics 2011, 30, 5123. Cowley, R. E.; Holland, P. L. Inorg. Chem. 2012, 51, 8352. Skvortsov, G. G.; Yakovenko, M. V.; Fukin, G. K.; Cherkasov, A. V.; Trifonov, A. A. Russ. Chem. Bull. 2007, 56, 1742. See Supporting Information for DiFranco, S. A.; Maciulis, N. A.; Staples, R. J.; Batrice, R. J.; Odom, A. L. Inorg. Chem. 2011, 51, 1187. Dennington, R.; Keith, T.; Millam, J. GaussView, Version 5, 2009. Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F. NBO Version 3.1. Baumann, R.; Davis, W. M.; Schrock, R. R. J. Am. Chem. Soc. 1997, 119, 3830. Liang, L.-C.; Schrock, R. R.; Davis, W. M. Organometallics 2000, 19, 2526. Schrock, R. R.; Liang, L.-C.; Baumann, R.; Davis, W. M. J. Organomet. Chem. 1999, 591, 163. Schrock, R. R.; Baumann, R.; Reid, S. M.; Goodman, J. T.; Stumpf, R.; Davis, W. M. Organometallics 1999, 18, 3649. Baumann, R.; Stumpf, R.; Davis, W. M.; Liang, L.-C.; Schrock, R. R. J. Am. Chem. Soc. 1999, 121, 7822. Graf, D. D.; Davis, W. M.; Schrock, R. R. Organometallics 1998, 17, 5820. Porter, R. M.; Danopoulos, A. A. Polyhedron 2006, 25, 859. Graf, D. D.; Schrock, R. R.; Davis, W. M.; Stumpf, R. Organometallics 1999, 18, 843. Takemoto, S.; Kawamura, H.; Yamada, Y.; Okada, T.; Ono, A.; Yoshikawa, E.; Mizobe, Y.; Hidai, M. Organometallics 2002, 21, 3897. Xiao, J.; Deng, L. Organometallics 2012, 31, 428. Fandos, R.; Fern ndez-Gallardo, .; Otero, A.; Rodr guez, A.; Ruiz, M. a. . Organometallics 2011, 30, 1551. Fryzuk, M. D.; Love, J. B.; Rettig, S. J.; Young, V. G. Science 1997, 275, 1445. Fryzuk, M. D. Acc. Chem. Res. 2008, 42, 127. Fryzuk, M. D.; MacKay, B. A.; Johnson, S. A.; Patrick, B. O. Angew. Chem., Int. Ed. 2002, 41, 3709. MacLachlan, E. A.; Fryzuk, M. D. Organometallics 2005, 24, 1112. MacLachlan, E. A.; Hess, F. M.; Patrick, B. O.; Fryzuk, M. D. J. Am. Chem. Soc. 2007, 129, 10895. Leitch, D. C.; Payne, P. R.; Dunbar, C. R.; Schafer, L. L. J. Am. Chem. Soc. 2009, 131, 18246. Wilshire, J. Aust. J. Chem. 1988, 41, 995. Ashton, P. R.; Hörner, B.; Kocian, O.; Menzer, S.; White, A. J. P.; Stoddart, J. F.; Williams, D. J. Synthesis 1996, 1996, 930. See Appendix B for spectrum. Leitch, D. C.; Platel, R. H.; Schafer, L. L. J. Am. Chem. Soc. 2011, 133, 15453. Leitch, D. C. Ph.D. Thesis. University of British Columbia, Vancouver, British Columbia, Canada, 2010. Shannon, R. Acta Crystallogr., Sect. A: Found. Crystallogr. 1976, 32, 751. 242  179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210.  See Section 4.4 for spectra at additional temperatures. Zhu, T.; Wambach, T. C.; Fryzuk, M. D. Inorg. Chem. 2011, 50, 11212. Morello, L.; João Ferreira, M.; Patrick, B. O.; Fryzuk, M. D. Inorg. Chem. 2008, 47, 1319. Fryzuk, M. D.; Jafarpour, L.; Rettig, S. J. Organometallics 1999, 18, 4050. Cohen, J. D.; Fryzuk, M. D.; Loehr, T. M.; Mylvaganam, M.; Rettig, S. J. Inorg. Chem. 1998, 37, 112. Fryzuk, M. D.; Mylvaganam, M.; Zawarotko, M. J.; MacGillivray, L. R. Polyhedron 1996, 15, 689. Fryzuk, M. D.; Mylvaganam, M.; Zaworotko, M. J.; MacGillivray, L. R. J. Am. Chem. Soc. 1993, 115, 10360. Fryzuk, M. D.; Haddad, T. S.; Mylvaganam, M.; McConville, D. H.; Rettig, S. J. J. Am. Chem. Soc. 1993, 115, 2782. Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. J. Am. Chem. Soc. 1990, 112, 8185. Fryzuk, M. D.; Johnson, S. A.; Patrick, B. O.; Albinati, A.; Mason, S. A.; Koetzle, T. F. J. Am. Chem. Soc. 2001, 123, 3960. Horrillo-Martinez, P.; Patrick, B. O.; Schafer, L. L.; Fryzuk, M. D. Dalton Trans. 2012, 41, 1609. Fontaine, P. P.; Yonke, B. L.; Zavalij, P. Y.; Sita, L. R. J. Am. Chem. Soc. 2010, 132, 12273. Hirotsu, M.; Fontaine, P. P.; Zavalij, P. Y.; Sita, L. R. J. Am. Chem. Soc. 2007, 129, 12690. Hirotsu, M.; Fontaine, P. P.; Epshteyn, A.; Sita, L. R. J. Am. Chem. Soc. 2007, 129, 9284. Fryzuk, M. D.; Johnson, S. A.; Rettig, S. J. Organometallics 2000, 19, 3931. Mabbott, G. A. J. Chem. Educ. 1983, 60, 697. Kissinger, P. T.; Heineman, W. R. J. Chem. Educ. 1983, 60, 702. See Section 4.4.4 for details. Geiger, W. E. Organometallics 2007, 26, 5738. Jutand, A. Chem. Rev. 2008, 108, 2300. See Appendix D for details. Chauke, V. P. Ph.D. Thesis. Rhodes University, Grahamstown, South Africa, 2011. McGrane, P. L.; Jensen, M.; Livinghouse, T. J. Am. Chem. Soc. 1992, 114, 5459. Tobisch, S. Inorg. Chem. 2012, 51, 3786. Majumder, S.; Odom, A. L. Organometallics 2008, 27, 1174. Stubbert, B. D.; Marks, T. J. J. Am. Chem. Soc. 2007, 129, 6149. Gribkov, D. V.; Hultzsch, K. C. Angew. Chem., Int. Ed. 2004, 43, 5542. Knight, P. D.; Munslow, I.; O'Shaughnessy, P. N.; Scott, P. Chem. Commum. 2004, 0, 894. Vlasceanu, A. Undergraduate Thesis. University of British Columbia, Vancouver, British Columbia, Canada, 2011. Moriarty, R. M.; Chany, C. J.; Vaid, R. K.; Prakash, O.; Tuladhar, S. M. J. Org. Chem. 1993, 58, 2478. Khalafi-Nezhad, A.; Parhami, A.; Zare, A.; Moosavi Zare, A. R. JICS 2008, 5, 413. Rao, H.; Fu, H.; Jiang, Y.; Zhao, Y. Angew. Chem., Int. Ed. 2009, 48, 1114.  243  Appendices  Appendix A X-Ray Crystallographic Data Table A.1 Crystallographic parameters for mono(amidate) complexes (Chapter 2). 1  2  3  4  5  formula  C19H38N5OTa  C21H42N5OTa  C25H50N5OTa  C27H46N5OTa  C33H58N5OTa  Fw  533.49  561.55  617.65  637.64  crystal size (mm)  0.10 x 0.10 x 0.10  0.30 x 0.20 x 0.10  0.20 x 0.20 x 0.10  0.40 x 0.30 x 0.20  0.25 x 0.20 x 0.10  colour, habit  yellow, prism  yellow, prism  yellow, prism  yellow, prism  yellow, prism  cell setting  triclinic  triclinic  monoclinic  orthorhombic  triclinic  space group  Pī  Pī  P21/n  Pnma  Pī  a (Å)  9.0186(1)  9.0567(3)  9.8185(3)  23.416(3)  8.4753(1)  b (Å)  9.5412(1)  9.5752(3)  16.6259(5)  13.3876(16)  11.8023(2)  c (Å)  15.3639(2)  16.5314(5)  17.5923(5)  9.6348(12)  18.7984(2)  α (°)  81.2757(6)  94.7636(7)  90  90  96.3177(6)  β (°)  85.3360(5)  97.7344(6)  95.5120  90  102.1258(6)  γ (°)  62.9467(5)  116.8754(6)  90  90  100.9086(6)  V (Å3)  1163.67(5)  1250.22(12)  2858.52(26)  3020.4(6)  1783.08(5)  2  2  4  4  2  ρcalcd (g cm )  1.523  1.492  1.435  1.402  1.344  F(000)  536  568  1264  1296  744  μ (Mo Kα) (mm-1)  4.739  4.415  3.869  3.664  3.112  2θmax (°)  55.10  55.10  55.02  64.96  60.24  total no. of reflns no. of unique reflns I = 2σ(I) no. of variables  19268  20151  26506  21912  36770  5323  5763  6559  5632  10304  246  266  304  185  378  R1 (all data)  0.0171  0.0159  0.0296  0.0287  0.0257  wR2 (all data)  0.0410  0.0354  0.0482  0.0590  0.0586  R1 (I > 2σ(I))  0.0159  0.0146  0.0219  0.0185  0.0233  wR2 (I > 2σ(I))  0.0405  0.0349  0.0458  0.0451  0.0575  goodness of fit  1.144  1.068  1.034  1.143  1.118  Z -1  721.79  244  15  16  formula  C20H38N5OTa  C24H46N5OTa  Fw  545.50  601.61  crystal size (mm)  0.60 x 0.40 x 0.30  0.40 x 0.40 x 0.20  colour, habit  yellow, prism  yellow, prism  cell setting  monoclinic  monoclinic  space group  P21/n  P21/n  a (Å)  10.9701(10)  14.9765(2)  b (Å)  14.3785(13)  11.2993(1)  c (Å)  14.7030(14)  16.0705(2)  α (°)  90  90  β (°)  95.363(4)  95.263(5)  γ (°)  90  90  2309.0(5)  2708.03(7)  4  4  ρcalcd (g cm )  1.569  1.476  F(000)  1096  1224  μ (Mo Kα) (mm-1)  4.779  4.082  2θmax (°)  60.14  60.32  total no. of reflns no. of unique reflns I = 2σ(I) no. of variables  46048  55178  6763  7966  254  292  R1 (all data)  0.0166  0.0148  wR2 (all data)  0.0398  0.0333  R1 (I > 2σ(I))  0.0153  0.0129  wR2 (I > 2σ(I))  0.0393  0.0321  goodness of fit  1.171  1.111  3  V (Å ) Z -1  245  Table A.2 Crystallographic parameters for bis and tris(amidate) complexes (Chapter 2). 7 formula  8  9  10  11  C28H46N5O2Ta C37H52N5O3Ta C30H47N4O2Ta C32H50N5O2Ta C36H46N5O2Ta  Fw  665.65  795.79  676.67  717.72  761.73  crystal size (mm)  1.00 x 0.60 x 0.40  0.38 x 0.22 x 0.22  0.30 x 0.20 x 0.10  0.40 x 0.40 x 0.40  0.20 x 0.20 x 0.10  colour, habit  yellow, prism  orange, prism  yellow, prism  red, prism  yellow, prism  cell setting  triclinic  orthorhombic  monoclinic  monoclinic  monoclinic  space group  Pī  Pbca  P21/n  P21/c  P21/n  a (Å)  11.0813(8)  14.4075(4)  10.2760(2)  19.0960(2)  14.4420(16)  b (Å)  11.2182(7)  20.8963(6)  19.7276(5)  8.8234(1)  13.7445(16)  c (Å)  14.3989(10)  24.4035(7)  15.8409(4)  19.8973(3)  17.0767(19)  α (°)  74.381(3)  90  90  90  90  β (°)  82.122(3)  90  102.7330(10)  106.780(5)  101.228(2)  γ (°)  60.520(2)  90  90  90  90  1500.65(18)  7347.00(52)  3132.31(19)  3209.78(11)  3324.8(3)  2  8  4  4  4  ρcalcd (g cm )  1.473  1.439  1.435  1.485  1.522  F(000)  676  3248  1376  1464  1544  μ (Mo Kα) (mm-1)  3.694  3.033  3.540  3.460  3.345  2θmax (°)  60.54  54.98  67.08  60.24  55.9  total no. of reflns no. of unique reflns I = 2σ(I) no. of variables  29737  36854  31036  34597  53995  8395  8418  10345  9451  7965  337  426  334  383  407  R1 (all data)  0.0194  0.0464  0.0618  0.0222  0.0263  wR2 (all data)  0.0453  0.0531  0.1378  0.0455  0.0507  R1 (I > 2σ(I))  0.0176  0.0244  0.0479  0.0205  0.0211  wR2 (I > 2σ(I))  0.0444  0.0466  0.1267  0.0450  0.0480  goodness of fit  1.131  1.011  1.039  1.246  1.155  3  V (Å ) Z -1  246  Table A.3 Crystallographic parameters for diamide and diurea proligands (Chapter 4). N,N’-(2,2’-oxybis(2,1phenylene))bis(tert-butylamide)  N,N’-(2,2’-thiobis(2,1phenylene))bis(tert-butylamide)  formula  C22H28N2O3  C22H28N2O2S  Fw  368.46  384.52  crystal size (mm)  0.34 x 0.12 x 0.08  0.24 x 0.22 x 0.18  colour, habit  colourless, prism  colourless, prism  cell setting  monoclinic  orthorhombic  space group  P21/n  Pbca  a (Å)  14.6313(5)  18.749(4)  b (Å)  8.7128(3)  10.5989(3)  c (Å)  16.8936(5)  20.9873(5)  α (°)  90  90  β (°)  107.576(2)  90  γ (°)  90  90  V (Å )  2053.05(15)  4170.5(3)  Z  4  8  ρcalcd (g cm-1)  1.192  1.225  792  1648  μ (Mo Kα) (mm )  0.079  0.174  2θmax (°)  56.14  59.92  total no. of reflns no. of unique reflns I = 2σ(I) no. of variables  27732  24223  4935  6010  258  258  R1 (all data)  0.1065  0.0756  wR2 (all data)  0.1757  0.1108  R1 (I > 2σ(I))  0.0598  0.0461  wR2 (I > 2σ(I))  0.1417  0.0991  goodness of fit  1.026  1.023  3  F(000) -1  247  1,1’-(oxybis(2,1phenylene))bis(3,3diisopropylurea)  1,1’-(thiobis(2,1phenylene))bis(3,3diisopropylurea)  formula  C26H38N4O3  C26H38N4O2S  Fw  454.60  470.66  crystal size (mm)  0.16 x 0.16 x 0.15  0.45 x 0.42 x 0.31  colour, habit  colourless, prism  colourless, prism  cell setting  monoclinic  orthorhombic  space group  C2/c  P212121  a (Å)  24.3934(17)  11.4401(6)  b (Å)  14.2979(10)  14.1467(8)  c (Å)  15.4783(10)  15.9802(9)  α (°)  90  90  β (°)  106.3685(9)  90  γ (°)  90  90  V (Å3)  5180(1)  2586.2(4)  8  4  1.166  1.209  1968  1016  μ (Mo Kα) (mm )  0.077  0.154  2θmax (°)  59.20  60.12  total no. of reflns no. of unique reflns I = 2σ(I) no. of variables  27959  17054  7209  7194  314  422  R1 (all data)  0.0819  0.0508  wR2 (all data)  0.1202  0.0881  R1 (I > 2σ(I))  0.0470  0.0392  wR2 (I > 2σ(I))  0.1040  0.0826  goodness of fit  1.016  1.043  Z ρcalcd (g cm ) -1  F(000) -1  248  Table A.4 Crystallographic parameters for tethered bis(amidate) complexes (Chapter 4). 28  29  30  31  formula  C31H43N5O3Zr  C31H43N5O2SZr  C28H44N5O3Ta  C28H44N5O3STa  Fw  624.92  640.98  679.63  695.69  crystal size (mm)  0.34 x 0.28 x 0.21  0.31 x 0.23 x 0.19  0.59 x 0.30 x 0.26  0.40 x 0.40 x 0.40  colour, habit  colourless, prism  colourless, prism  yellow, prism  yellow, prism  cell setting  monoclinic  triclinic  monoclinic  orthorhombic  space group  P21/n  Pī  P21/c  P212121  a (Å)  15.4558(6)  10.0142(16)  10.2195(4)  11.6511(6)  b (Å)  11.4937(4)  11.9935(19)  18.7977(7)  12.9814(7)  c (Å)  17.5330(7)  15.140(2)  15.9056(6)  19.5667(10)  α (°)  90  98.252(3)  90  90  β (°)  99.210(2)  100.240(2)  102.131(2)  90  γ (°)  90  93.922(3)  90  90  V (Å3)  3074.5(2)  1762.7(8)  2987.3(3)  2959.4(5)  4  2  4  4  1.350  1.208  1.511  1.561  1312  672  1376  1408  μ (Mo Kα) (mm )  0.397  0.403  3.715  3.818  2θmax (°)  55.34  58.36  60.18  60.18  total no. of reflns no. of unique reflns I = 2σ(I) no. of variables  54049  34080  52208  19904  7133  9637  8730  8517  371  371  346  347  R1 (all data)  0.0493  0.0499  0.0244  0.0141  wR2 (all data)  0.0884  0.0976  0.0802  0.0302  R1 (I > 2σ(I))  0.0347  0.0377  0.0181  0.0135  wR2 (I > 2σ(I))  0.0799  0.0920  0.0573  0.0300  goodness of fit  1.029  1.055  1.350  1.057  Z ρcalcd (g cm ) -1  F(000) -1  249  Table A.5 Crystallographic parameters for tethered bis(ureate) complexes (Chapter 4). 32•HNMe2  32  33  34  35  formula  C32H55N7O3Zr  C30H48N6O3Zr  C30H48N6O2SZr  C32H54N7O3Ta  C32H54N7O2STa  Fw  677.05  631.96  648.02  765.77  781.83  crystal size (mm)  0.40 x 0.40 x 0.20  0.38 x 0.21 x 0.18  0.34 x 0.28 x 0.24  0.60 x 0.50 x 0.45  0.25 x 0.23 x 0.10  colour, habit  colourless, prism  colourless, prism  colourless, prism  yellow, prism  yellow, prism  cell setting  triclinic  monoclinic  monoclinic  monoclinic  monoclinic  space group  Pī  C2/c  P21/n  P21/c  C2/c  a (Å)  10.863(4)  19.439(2)  13.708(2)  15.5867(7)  32.379(3)  b (Å)  12.107(5)  9.0266(10)  12.8834(19)  10.0693(5)  15.3502(15)  c (Å)  13.975(6)  38.255(4)  19.530(3)  22.5554(11)  19.7503(19)  α (°)  90.769(17)  90  90  90  90  β (°)  107.503(15)  102.962(4)  101.967(3)  99.982(3)  118.501(2)  γ (°)  97.568(16)  90  90  90  90  3  1734.9(17)  6541(2)  3374(1)  3486.4(4)  8626(2)  2  8  4  4  8  ρcalcd (g cm )  1.210  1.283  1.511  1.459  1.204  F(000) μ (Mo Kα) (mm-1) 2θmax (°) total no. of reflns no. of unique reflns I = 2σ(I) no. of variables R1 (all data)  668  2672  1376  1568  3200  0.353  0.374  0.422  3.194  2.628  60.12  59.86  60.14  60.38  60.32  37283  24785  70153  67931  87769  9993  8173  9851  10190  12723  402  435  373  402  402  0.0304  0.0597  0.0460  0.0540  0.0464  wR2 (all data)  0.0670  0.0852  0.0838  0.0703  0.0709  R1 (I > 2σ(I)) wR2 (I > 2σ(I)) goodness of fit  0.0253  0.0372  0.0326  0.0273  0.0289  0.0645  0.0737  0.0756  0.0585  0.0671  1.032  1.029  1.066  1.060  1.048  V (Å ) Z -1  250  Appendix B Selected NMR Spectra  3.44  7.03 7.01 6.91  7.16 7.11  1.13  1, 1H NMR (benzene-d6, 400 MHz)  1.94 2.00 0.95 7.5  7.0  24.08 6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  9.00 3.0  2.5  2.0  1.5  1.0  0.5  0  C{1H} NMR (benzene-d6, 100 MHz)  180  160  140  120  100 80 Chemical Shift (ppm)  60  42.12  47.07  180.35  147.01  124.32  128.39 128.23 126.15  128.94  29.64  13  40  20  0  251  1.55  7.45  7.55 7.53  7.95 7.93  7.47 7.27  1.35  1.42  N-(2,4,6-tri-tert-butylphenyl)benzamide, 1H NMR (chloroform-d, 400 MHz)  2.00 3.77 1.95 0.72 8.0  7.5  21.66 9.52  7.0  6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  2.5  2.0  1.5  1.0  0.5  20  10  0  C{1H} NMR (chloroform-d, 100 MHz)  160  150  140  130  120  36.38  135.24 131.62 128.95  148.22  167.36 170  127.15 123.26  31.45  32.08  76.68  77.32 77.00  13  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  0  252  1.75 8.0  1.80 2.52  7.5  7.0  24.00  6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  2.5  100 90 80 70 Chemical Shift (ppm)  60  50  0.53  1.08  1.28  7.69  6.90 6.89 6.81 6.80 6.78  7.16  2.62  1.10  1.32  5, 1H NMR (benzene-d6, 600 MHz)  19.54 9.64 2.0  1.5  1.0  0.5  0  20  10  0  C{1H} NMR (benzene-d6, 150 MHz)  160  150  140  130  120  32.47 36.87  121.70  129.06 128.85  142.29 139.85  147.42  160.32  45.71  32.26  128.56 128.22  128.39  13  110  40  30  253  2.23 3.94 1.99 1.29 0.98 7.5  7.0  12.88  6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  5.29 3.0  9.00 2.5  2.0  1.01 0.89  1.17 1.13  2.20 2.19  2.97 2.84  4.00  7.01 7.00 6.90 6.84 6.83  3.42  7.08  7.16  1.66  1.06  7, 1H NMR (benzene-d6, 600 MHz)  11.25  1.5  1.0  0.5  0  C{1H} NMR (benzene-d6, 150 MHz)  180  160  140  120  100 80 Chemical Shift (ppm)  60  40  28.52  48.71 45.20 42.13 40.29  128.94 126.39 125.38 123.37 121.41  145.59  151.17  168.58  184.92  29.88 29.11  128.56  128.39  128.23  13  20  0  254  2.89  6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  2.89 2.96 2.81 2.85 2.89 3.0  2.5  2.0  18.00 0.5  0  0.48  7.0  0.91  20.79 19.12 18.92 18.72 16.37  7.5  1.52  3.40  2.88  5.89 5.88 5.88  6.86 6.84 6.78  6.91 6.89  6.98  7.16 7.16  0.86 1.15 2.73 0.99 0.84  2.19 2.11 1.86  3.91  2.86  2.43  4.28  1.03  1.11  10, 1H NMR (benzene-d6, 400 MHz)  1.5  1.0  C{1H} NMR (benzene-d6, 100 MHz)  180  160  140  120  100 Chemical Shift (ppm)  47.73 45.11 43.79 42.08 41.79  129.20 127.72 126.55 126.03 124.99  140.63 136.01 134.82  180.78  191.11  27.97  28.50  128.63 128.39 128.16  13  80  60  40  20  0  255  10 9 8 7 6 5 Chemical Shift (ppm) 4 3  0.87  2.49  2.11  2.25  2.59  6.83 6.83  3.22  6.85  2.21 2.14 2.19  2.74  6.99 6.93 6.92  0.30  7.76 7.74 7.72 7.71  2.77  8.84  3.37  7.16  11, 1H NMR (benzene-d6, 400 MHz)  2 1 0  256  0.99 0.98 0.77 0.76 0.68 0.67 0.75 0.67 0.75 0.66  7.00  1.94 1.59 1.59 1.58 1.57 1.23 1.22 1.04 1.03  7.41  7.16  7.14  7.01  2.32  2.17  N-(2,6-dimethylphenyl)cyclopropanecarboxamide, 1H NMR (chloroform-d, 600 MHz)  1.00 1.49 0.47 2.97 8.0  7.5  7.0  6.5  2.96 6.25 6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  2.5  1.03 0.47 1.02 2.06 2.12 1.03  2.0  1.5  1.0  0.5  0  C{1H} NMR (chloroform-d, 150 MHz)  180  170  160  150  140  10.17  18.52 134.10  175.98 172.10  137.00  135.46  18.24  14.29  128.37 126.83  8.13  127.85  7.13  13  130  120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  10  0  257  7.5  7.0  6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  1.01 0.97 0.75 0.64 0.63 1.57 1.56  1.35 1.00  1.93  3.32 3.31 3.30 3.29 3.10 3.09 3.08  0.65 0.76 2.37 0.96  7.55 7.33  7.25 7.23  7.21 7.20 7.14 7.13  1.25 1.17  1.19  1.18  N-(2,6-diisopropylphenyl)cyclopropanecarboxamide, 1H NMR (chloroform-d, 600 MHz)  0.10 0.60 14.92 2.56 1.04 1.37 2.5  2.0  1.5  1.0  0.5  0  C{1H} NMR (chloroform-d, 150 MHz)  180  170  160  150  140  14.42  131.94  176.62 172.90  147.31 146.33  127.93  10.70  128.70  28.56  123.15  7.06  23.09  123.70  28.40  24.00  8.47  13  130  120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  10  0  258  1.19 1.18 1.17 1.17 1.16 1.16 1.09 1.08 0.37 0.36 0.35 0.34 0.33  7.26 7.21 7.19 7.11 7.09 7.07  2.44  3.41  15, 1H NMR (benzene-d6, 400 MHz)  1.84 0.94 7.5  7.0  23.27 6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  6.00 3.0  1.99 0.98  2.5  2.0  1.5  1.0  2.02 0.5  0  C{1H} NMR (benzene-d6, 100 MHz)  180  170  160  150  140  130  11.79  18.56 47.22  133.45  175.69  144.03  124.96  7.43  128.63 128.39 128.16  13  120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  10  0  259  7.0  2.43 22.08 6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  100 90 80 Chemical Shift (ppm)  70  0.19 0.19 0.17 0.20  0.81  0.82  3.44  0.68 1.97 7.5  1.04 1.04 1.03 1.02  3.49 3.47  7.18 7.17 7.14 7.13  7.16  3.27  1.33 1.32 1.28 1.26  16, 1H NMR (benzene-d6, 400 MHz)  6.18 6.25 2.06 1.31 2.5  2.0  1.5  1.0  2.00  0.5  0  170  160  150  140  47.15  12.77 8.59  126.03  144.30 141.18  176.29 180  27.53 25.98 24.87  124.09  128.15  C{1H} NMR (benzene-d6, 100 MHz) 128.62  13  130  120  110  60  50  40  30  20  10  0  260  1.94 1.92 8.5  8.0  0.08  7.01  7.15  8.28 8.27 7.96  7.13 7.03 6.83 6.82  1.22  N,N’-(2,2’-oxybis(2,1-phenylene))bis(tert-butylamide), 1H NMR (chloroform-d, 600 MHz)  2.00 1.93 1.92 7.5  7.0  18.00  6.5  6.0  5.5  5.0 4.5 4.0 Chemical Shift (ppm)  3.5  3.0  2.5  2.0  70  60  50  40  1.5  1.0  0.5  20  10  0  C{1H} NMR (chloroform-d, 150 MHz)  180  170  160  150  140  130  39.51  77.00 76.79 76.58  128.95  145.23  176.42  124.31 124.28 121.99 117.21  27.12  13  120  110  100 90 80 Chemical Shift (ppm)  30  0  261  1.91 1.80 8.5  8.0  7.11 7.11 7.05 7.05 7.03 7.02  7.32  8.27 8.27 8.25 8.24 8.07  1.24  N,N’-(2,2’-thiobis(2,1-phenylene))bis(tert-butylamide), 1H NMR (chloroform-d, 300 MHz)  2.04 1.89 1.96 7.5  7.0  18.00 6.5  6.0  5.5  5.0 4.5 4.0 Chemical Shift (ppm)  3.5  3.0  2.5  2.0  1.5  1.0  0.5  0  C{1H} NMR (chloroform-d, 75 MHz)  170  160  150  140  77.42 77.00 76.58  137.52  176.57 180  39.92  131.69 129.22 124.94 122.24  27.35  13  130  120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  10  0  262  1.96 2.44 4.12 7.5  7.0  6.5  12.01 6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  4.52 2.5  1.12 0.96 0.84  1.39 1.34 1.29 1.22  2.22  3.24 3.22  6.99 6.85 6.78 6.77 6.76 6.74  7.16  3.18  1.00  N,N’-(2,2’-oxybis(2,1-phenylene))bis(tertbutylamidate)bis(dimethylamido)(dimethylamine)zirconium, 1 H NMR (benzene-d6, 400 MHz)  18.73 2.0  1.5  1.0  0.5  0  263  1.74 9.0  8.5  1.08 0.97 0.94 0.89  8.62 8.61  7.16 7.02 7.00 6.90 6.89 6.79 6.77 6.76 6.75 6.67  3.20  0.99  28, 1H NMR (benzene-d6, 600 MHz)  3.07 2.56 4.31 1.99 8.0  7.5  7.0  6.5  12.00 6.0  5.5  5.0 4.5 4.0 Chemical Shift (ppm)  3.5  18.84  3.0  2.5  2.0  1.5  1.0  0.5  0  C{1H} NMR (benzene-d6, 150 MHz)  180  160  140  41.20  128.68  139.79 135.95  187.23  151.30  150.50  127.30 124.47 122.51  43.55  128.54  128.39  28.68  128.22  13  120  100 80 Chemical Shift (ppm)  60  40  20  0  264  8.0  7.5  7.0  6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  2.5  1.09 1.08  1.64 1.31 1.22 1.21  2.21 2.19 2.18 2.11  3.42  6.79 6.61  7.72  7.23 7.13  6.95  1.71  7.16  3.19  31, 1H NMR (benzene-d6, 400 MHz)  2.0  1.5  1.0  0.5  0  C{1H} NMR (benzene-d6, 100 MHz)  180  160  140  120  28.56 41.70 39.31  125.58 122.16 121.63  130.99  140.74  152.13 149.99  169.39  181.61  29.78  47.50  128.18  13  100 80 Chemical Shift (ppm)  60  40  20  0  265  1  H NMR (chloroform-d, 400 MHz)  1.83 8.5  4.00 3.99 3.97 3.95 3.94  6.91 6.89 6.78 6.76  8.30 8.28  7.12 7.10  1.21  1.22  1,1’-(oxybis(2,1-phenylene))bis(3,3-diisopropylurea),  2.01 4.01 1.91 8.0  7.5  7.0  6.5  4.00 6.0  5.5  5.0 4.5 4.0 Chemical Shift (ppm)  24.98 3.5  3.0  2.5  2.0  1.5  1.0  0.5  0  C{1H} NMR (chloroform-d, 100 MHz)  150  140  130  77.32 77.00 76.68  124.59 122.10 120.41 117.15  131.09  144.25  154.03  45.04  21.14  13  120  110  100  90 80 70 Chemical Shift (ppm)  60  50  40  30  20  10  0  266  1  H NMR (chloroform-d, 400 MHz)  8.23 8.21 8.21 7.33 7.32 7.31 7.29 7.13 7.13 7.11 7.01 6.97 6.96  3.97 3.96 3.94 3.92 3.90  1.31  1.32  1,1’-(thiobis(2,1-phenylene))bis(3,3-diisopropylurea),  1.85 8.5  2.04 1.93 1.92 2.03  8.0  7.5  7.0  6.5  4.00 6.0  5.5  5.0 4.5 4.0 Chemical Shift (ppm)  24.62 3.5  3.0  2.5  2.0  1.5  1.0  0.5  0  C{1H} NMR (chloroform-d, 100 MHz)  140  120  45.58  130  77.32 77.00 76.68  123.06 121.33 120.52  150  131.55 128.96  139.32  153.91  21.14  13  110  100  90 80 70 Chemical Shift (ppm)  60  50  40  30  20  10  0  267  4.00 2.27 1.58 7.5  7.0  6.5  1.23 1.03 0.97 0.95 0.90 0.88 0.87  3.09 2.98  3.34  7.16 7.04 7.04 7.02 7.02 6.99 6.98 6.97 6.97 6.83 6.72  3.27  32, 1H NMR (benzene-d6, 400 MHz)  13.59 6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  24.22  3.0  2.5  2.0  1.5  1.0  0.5  0  10  0  C{1H} NMR (benzene-d6, 100 MHz)  126.09 124.51 123.40 121.17  141.59  166.42  152.42  42.31  128.63 128.39 128.15  13  21.34  47.07  170  160  150  140  130  120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  268  2.00  4.24 2.02  7.5  7.0  1.37 1.33 1.32 1.30 1.28 1.28 1.27 1.25 0.96 0.94 0.92 0.32  2.16 2.15  3.24 3.19  7.16 7.04 6.74 6.73 6.72 6.71 6.70  7.55 7.53  6.98 6.97  3.27  33, 1H NMR (toluene-d8, 400 MHz)  16.14 4.49 6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  34.99  3.0  2.5  2.0  1.5  1.0  0.5  0  10  0  20.40  134.20 129.97  151.36  166.15  20.78 20.01  122.89  124.85  42.34  20.59 20.21  127.67  129.07 128.83  128.17 127.93 127.69  C{1H} NMR (toluene-d8, 100 MHz) 137.44  13  46.89  170  160  150  140  130  120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  269  7.5  7.0  6.5  6.0  5.5  5.0  4.5 4.0 3.5 Chemical Shift (ppm)  3.0  2.5  0.26  0.88  2.12  3.66  7.06 7.04 6.89 6.88 6.86 6.86 6.85 6.83 6.71 6.69  1.21 1.19  3.45  34, 1H NMR (toluene-d8, 400 MHz, 80 °C)  2.0  1.5  1.0  0.5  0  C{1H} NMR (toluene-d8, 100 MHz, 80 °C) 21.17 19.61 19.80  127.20 122.66  19.22  46.37  45.88  19.42  124.37  128.58 128.34  127.68 127.44  137.04  13  180  170  160  150  140  130  120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  10  0  270  2.00 8.0  7.5  0.88 17.09  6.5  6.0  5.5  5.0  4.66  4.5 4.0 3.5 Chemical Shift (ppm)  0.95 0.89 0.86  0.76  1.38  3.25 3.12 3.02  4.93 2.30 7.0  1.25 1.24  1.22 1.21 1.20 1.19  6.67 6.65 6.64 6.63  6.94  6.92  7.62 7.60  7.16  6.90 6.90  3.64  35, 1H NMR (benzene-d6, 400 MHz)  21.22  3.0  2.5  2.0  1.5  1.0  0.5  0  10  0  C{1H} NMR (benzene-d6, 100 MHz)  180  170  160  150  140  130  47.15 46.75 46.06 39.30 120  110  100 90 80 Chemical Shift (ppm)  70  60  50  40  30  20  14.63  23.07 22.49 20.93  121.84  123.62  138.13 135.52 129.18  164.68  150.12  128.74 127.64  49.84  128.63 128.39 128.15  13  271  Appendix C Supplemental Computational Data C.1  Alternate Mechanistic Pathways for the α-Alkylation of Amines In order to verify that TS(3/I) and the resultant tantalaziridine II were located on the  lowest energy pathway, alternate transition state geometries were investigated. Activating a C–H bond of the other equatorial dimethylamido ligand required a rotation of the Namido–C bonds, resulting in a κ1(O) amidate binding mode in the optimized transition state geometry (Figure C.1).  Figure C.1 Optimized geometry for an alternate transition state for the formation of a tantalaziridine. Most hydrogen atoms omitted for clarity.  Relaxation of this transition state yields a κ1(O) isomer of 3 and a tantalaziridine with a bound neutral dimethylamine akin to I (Scheme C.1). Interestingly, the azametallacyclopropane moiety is twisted slightly out-of-plane (22.4°) with respect to the amidate chelate (Figure C.2). Additionally, the Namido–Ta–Namido bond angle (132.9°) for the axial ligands is decreased due to the lack of Ta–Namidate bond in the κ1(O) binding geometry. When the neutral amine ligand is removed, the amidate ligand adopts more of a κ2(N,O) binding geometry with a Ta–Namidate bond  272  length of 2.60 Å. This is accompanied by an additional twisting of the tantalaziridine moiety 8.3° out of the plane of the amidate chelate and further attenuation of the Namido–Ta–Namido bond angle (115.9°). Interestingly, the calculated barrier of 42.3 kcal/mol is larger than that of TS(3/I), suggesting a less-favourable pathway. This notion is further supported by the fact that each intermediate and transition state is higher in energy then its κ2(N,O) counterpart (Scheme C.1).  Scheme C.1 Intermediates and transitions state for the C–H activation of an alternative dimethylamido ligand. Free energies (ΔG) are reported in kcal/mol and are calculated relative to the κ1(O) species depicted (top numbers) and to the corresponding κ2(N,O) species (bottom numbers).  273  Figure C.2 Optimized geometries for the intermediates depicted in Scheme C.1. Most hydrogen atoms omitted for clarity.  Alternatively, a pathway resulting in a tantalaziridine orthogonal to the plane of the amidate chelate was sought. Unfortunately, no transition state was found for the C–H activation of an axial dimethylamido ligand.  However, relaxation of an insertion transition state  investigating the same alternative geometry (vide infra) led to a κ2(N,O) tantalaziridine in which the azametallacyclopropane moiety is perpendicular to the plane of the amidate ligand (Figure C.3). In addition to the absence of a suitable transition state geometry, this precatalyst is 9.7 kcal/mol higher in energy than II. This evidence suggests that the path described in Section 3.2.1 is the most favourable C–H activation pathway, justifying the use of II as the starting point for further modeling of the catalytic cycle.  274  Figure C.3 Optimized geometry for a tantalaziridine with the azametallacyclopropane moiety perpendicular to the plane of the amidate ligand. Hydrogen atoms omitted for clarity.  Though species II was found to be the lowest energy tantalaziridine, a transition state for the insertion of 1-octene into a tantalaziridine perpendicular to the plane of the amidate chelate was found (Figure C.4). The negative vibrational mode found corresponds to the same atomic motion observed in TS(III/IV), confirming the correct bond-making/bond-breaking process. Relaxation of the transition state yields intermediates that are 11.8 and 12.0 kcal/mol higher in energy than the in-plane equivalents III and IV. This is in good agreement with the finding that the transition state is 13.1 kcal/mol higher energy than TS(III/IV), further supporting the notion that reactivity in the same plane as the amidate chelate is the lowest energy pathway along the PES.  275  Figure C.4 Optimized geometry for a transition state with the insertion of 1-octene into the azametallacyclopropane moiety perpendicular to the plane of the amidate ligand. Hydrogen atoms omitted for clarity.  C.2  Supplemental Data for Steric Parameter Calculations  Figure C.5 Buried volumes (%Vbur) calculated for axially chiral tantalum precatalysts.  276  Figure C.6 Solid-G parameters GL and Gspace calculated for axially chiral tantalum precatalysts.  277  Appendix D Supplemental Electrochemical Data All cyclic voltammograms presented herein, unless otherwise noted, were collected from a 10 mL THF solution containing the analyte (0.005 mol L-1) and the electrolyte TBAPF6 (0.1 mol L-1) at the scan rates listed in the legend. Each individual scan started at 0 V towards negative potentials.  Figure D.7 Cyclic voltammograms for N,N’-(2,2’-oxybis(2,1-phenylene))bis(tert-butylamide).  278  Figure D.8 Cyclic voltammograms for 28.  Figure D.9 Cyclic voltammograms for 30.  279  Figure D.10 Cyclic voltammograms for 31.  Figure D.11 Molecular orbital surfaces for the LUMOs calculated for 30 (left) and 31 (right).  280  Figure D.12 Cyclic voltammograms for 1,1’-(oxybis(2,1-phenylene))bis(3,3-diisopropylurea). Scans start from -2.8 V towards positive potentials.  During  the  electrochemical  analysis  of  1,1’-(oxybis(2,1-phenylene))bis(3,3-  diisopropylurea) it was found that an oxidation is required before the observed reaction can take place. A single scan starting from 0 V scanning through negative potentials first revealed no reduction event. However, when multiple scans are performed using the same scan method, reduction events are observed on every scan after the initial pass.  281  Figure D.13 Cyclic voltammogram for 34.  Figure D.14 Cyclic voltammogram for 34 using TBABPh4 as the electrolyte.  282  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

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.24.1-0073860/manifest

Comment

Related Items