SYNTHESIS, STRUCTURE, AND REACTIVITY OF ... - Open Collections [PDF]

SYNTHESIS, STRUCTURE, AND REACTIVITY OF EARLY TRANSITION METAL PRECATALYSTS BEARING (N,O)-CHELATING LIGANDS

by Philippa Robyn Payne

B.Sc., University of Ottawa, 2008

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry)

THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)

December 2013

© Philippa Robyn Payne, 2013

ABSTRACT

The synthesis, structure, and reactivity of early transition metal complexes containing (N,O)-chelating ancillary ligands are described. The ligands investigated include ureates, pyridonates, amidates, and sulfonamidates. These related ligands generate four-membered metallacycles when bound to the metal center in a κ2-(N,O) fashion. The zirconium and tantalum complexes have been examined in terms of their activity and selectivity as precatalyst systems for hydroamination or hydroaminoalkylation. A chiral cyclic ureate ligand has been synthesized from enantiopure L-valine for application in zirconium-catalyzed asymmetric hydroamination of aminoalkenes. Chiral zirconium complexes, prepared in situ from two equivalents of the urea proligand and tetrakis(dimethylamido) zirconium, promote the formation of pyrrolidines and piperidines in up to 12% ee. Isolation of an asymmetric bimetallic zirconium complex containing three bridging ureate ligands confirms that ligand redistribution occurs in solution and is most likely responsible for the low enantioselectivities. Mechanistic investigations focusing on the hydroaminoalkylation reactivity promoted by a bis(pyridonate) bis(dimethylamido) zirconium precatalyst expose a complex catalytic system in solution. Stoichiometric investigations reveal the formation of polymetallic complexes upon addition of primary amines. The kinetic and stoichiometric investigations are most consistent with a bimetallic catalytically active species. A series of mono(amidate) tantalum amido complexes with varying steric and electronic properties

have

been

synthesized

via

protonolysis.

Solid-state

and

solution-phase

characterization indicate that the amidate substituents influence the observed binding mode of ii

the ligand. Salt metathesis and protonolysis routes to the synthesis of mixed tantalum chloro amidate complexes are investigated. Sulfonamide proligands react with pentakis(dimethylamido) tantalum to generate well-defined monomeric complexes containing a κ2-(N,O) bound sulfonamidate. The hemilabile (N,O)-chelating amidate ligands, which generate four-membered metallacycles, are the most active of the precatalysts examined for the intermolecular hydroaminoalkylation of terminal olefins with secondary amines. The substrate scope of a mono(amidate) tetrakis(dimethylamido) tantalum complex has been examined for the α-alkylation of unprotected piperidine, piperazine, and azepane Nheterocyclic amines. The lack of reactivity with pyrrolidine substrates is examined by quantum chemical calculations and isotopic labeling studies. Two (N,O)-chelating ureate ligands are also successful ancillary ligands for this transformation and, with a C1-symmetric chiral ureate complex, enantioselective α-alkylation of piperidine is observed.

iii

PREFACE

Parts of the research conducted for this thesis were carried out collaboratively with other members of the Schafer research group. I, in consultation with my supervisor Dr. Laurel Schafer, designed and performed all of the experiments described herein except in the following instances. The initial concept and synthetic approach to the synthesis of the cyclic urea proligand was designed by Dr. David C. Leitch (Chapter 2). Compound 16 was synthesized and characterized through X-ray diffraction by Dr. Patrick Eisenberger (Chapter 3). The amidate ligands 22 and 23 (Chapter 4) and the related tantalum complexes were synthesized and characterized by Benedict J. Barron, an undergraduate researcher, under my supervision. The Xray diffraction data for the crystalline mixture of compounds 29 and 30 were refined by Dr. Nicholas C. Payne. The DFT calculations were performed by Dr. J.M. Lauzon (Chapter 5). The N-heterocyclic substrate screening was performed in collaboration with Dr. Patrick Eisenberger (Table 5.1, Chapter 5). Jacky Yim performed the synthesis and characterization for compound 47 (Table 5.1, entry 3). The solid-state molecular data presented herein was collected by Neal Yonson, Jacky Yim, or Scott Ryken while I performed the final refinements. The following publications have been reported based on this work. Any work from these papers that I did not directly carry out, with the exceptions listed previously, does not appear in this thesis or is appropriately referenced. I wrote these manuscripts with editorial assistance from the co-authors listed, except for the Tetrahedron where I was involved with the editing but not the initial draft. 

Payne, P. R.; Bexrud, J. A.; Leitch, D. C.; Schafer, L. L. Can. J. Chem. 2011, 89, 1222. (Chapter 2) iv



Garcia, P.; Payne, P. R.; Chong, E.; Webster, R. L.; Barron, B. J.; Behrle, A. C.; Schmidt, J. A. R.; Schafer, L. L. Tetrahedron 2013, 69, 5737. (Chapter 4)



Payne, P. R.; Garcia, P.; Eisenberger, P.; Yim, J. C.-H.; Schafer, L. L. Org. Lett. 2013, 15, 2182. (Chapter 5)

v

TABLE OF CONTENTS

ABSTRACT ................................................................................................................................... ii PREFACE ..................................................................................................................................... iv TABLE OF CONTENTS ............................................................................................................ vi LIST OF TABLES ....................................................................................................................... xi LIST OF FIGURES ................................................................................................................... xiii LIST OF SCHEMES ................................................................................................................. xix LIST OF ABBREVIATIONS AND ACRONYMS .................................................................xxv ACKNOWLEDGEMENTS ......................................................................................................xxx DEDICATION.......................................................................................................................... xxxi CHAPTER 1: Synthesis, structure, and reactivity of early transition metal precatalysts bearing (N,O)-chelating ligands. ...................................................................................................1 1.1

Introduction ..................................................................................................................... 1

1.2

Mono anionic, monodentate ligands based on N- and O-donors. ................................... 5

1.2.1

Multidentate, monoanionic ligands of (O,O)-, (N,N)-, and (N,O)-compounds .......... 6

1.2.2

Amidate complexes for hydroamination and hydroaminoalkylation........................ 11

1.2.3

Ureate complexes for hydroamination ...................................................................... 18

1.2.4

Pyridonate complexes for hydroamination and hydroaminoalkylation .................... 22

1.2.5

Sulfonamidate complexes for hydroamination ......................................................... 25

1.3

Scope of thesis .............................................................................................................. 28

CHAPTER 2: C1-symmetric ureate complexes of zirconium for the asymmetric hydroamination of unactivated aminoalkenes ..........................................................................31 vi

2.1

Introduction ................................................................................................................... 31

2.1.1

Asymmetric hydroamination of unactivated olefins ................................................. 31

2.1.2

Rare-earth metal systems .......................................................................................... 33

2.1.3

Group 4 metal systems .............................................................................................. 34

2.1.4

Expansion of substrate scope using a tethered bis(ureate) zirconium catalyst ......... 36

2.1.5

Scope of chapter ........................................................................................................ 37

2.2

Results and discussion .................................................................................................. 38

2.2.1

Ligand and complex .................................................................................................. 38

2.2.2

Intramolecular hydroamination................................................................................. 41

2.2.3

Isolated bimetallic complex ...................................................................................... 43

2.3

Conclusions ................................................................................................................... 47

2.4

Experimental ................................................................................................................. 48

2.4.1

General methods ....................................................................................................... 48

2.4.2

Materials ................................................................................................................... 49

2.4.3

General experimental procedure ............................................................................... 50

2.4.4

Compound synthesis and characterization ................................................................ 50

CHAPTER 3: Mechanistic investigations of a bis(pyridonate) zirconium complex for intramolecular hydroaminoalkylation .......................................................................................53 3.1

Introduction ................................................................................................................... 53

3.1.1

Titanium-based precatalysts...................................................................................... 53

3.1.2

A zirconium pyridonate precatalyst .......................................................................... 58

3.1.3

Scope of chapter ........................................................................................................ 61

3.2

Results and discussion .................................................................................................. 61 vii

3.2.1

Solution-phase behaviour of the precatalyst ............................................................. 61

3.2.2

Stoichiometric reactivity ........................................................................................... 65

3.2.3

Kinetic analysis of intramolecular hydroaminoalkylation ........................................ 71

3.3

Conclusions and mechanistic proposal ......................................................................... 77

3.4

Experimental ................................................................................................................. 79

3.4.1

Materials ................................................................................................................... 79

3.4.2

General experimental procedures ............................................................................. 79

3.4.3

Synthesis and characterization .................................................................................. 81

CHAPTER 4: Synthesis, structure, and reactivity of new tantalum complexes ....................84 4.1

Introduction ................................................................................................................... 84

4.1.1

Chloro ancillary ligands ............................................................................................ 84

4.1.2

(N,O)- and (O,O)-chelating ancillary ligands ........................................................... 85

4.1.3

Scope of chapter ........................................................................................................ 88 Results and discussion .................................................................................................. 89

4.2 4.2.1

Tantalum precatalysts with amidate ligand modifications........................................ 89

4.2.2

Hydroaminoalkylation reactivity of modified tantalum amidate complexes ............ 99

4.2.3

Mixed chloro and amidate tantalum complexes ..................................................... 102

4.2.4

Tantalum complexes supported by sulfonamidate ligands ..................................... 117

4.3

Conclusions ................................................................................................................. 125

4.4

Experimental ............................................................................................................... 127

4.4.1

Materials ................................................................................................................. 127

4.4.2

General experimental procedures ........................................................................... 127

4.4.3

Synthesis and characterization ................................................................................ 129 viii

4.4.3.1

Proligands ....................................................................................................... 129

4.4.3.2

Tantalum complexes ....................................................................................... 135

CHAPTER 5: Synthesis of α-alkylated N-heterocycles via hydroaminoalkylation .............144 5.1

Introduction ................................................................................................................. 144

5.1.1

Functionalized N-heterocycles ................................................................................ 144

5.1.2

Stoichiometric α-lithiation and functionalization ................................................... 145

5.1.3

Oxidative functionalization of α-C–H bonds .......................................................... 148

5.1.4

Directed transition metal-catalyzed C–H activation ............................................... 151

5.1.5

Hydroaminoalkylation of N-heterocycles with early transition metals .................. 153

5.1.6

Scope of chapter ...................................................................................................... 155

5.2

Results and discussion ................................................................................................ 156

5.2.1

The hydroaminoalkylation of N-heterocyclic substrates ........................................ 156

5.2.2

Urea proligands for the hydroaminoalkylation of N-heterocycles .......................... 163

5.2.3

Attempted α-alkylation with pyrrolidine and indoline ........................................... 164

5.2.4

Computational modeling of the catalytic cycles for the α-alkylation of amines .... 167

5.2.4.1

Previous investigation into the hydroaminoalkylation of acyclic amines ...... 167

5.2.4.2

Computational investigations with N-heterocyclic substrates ........................ 169

5.2.4.3

Deuterium labeling.......................................................................................... 175

5.3

Conclusions ................................................................................................................. 180

5.4

Experimental ............................................................................................................... 182

5.4.1

Materials ................................................................................................................. 182

5.4.2

General experimental procedures ........................................................................... 182

5.4.3

Synthesis and characterization ................................................................................ 183 ix

5.4.4

Attempts towards the synthesis of bicyclic and tricyclic compounds .................... 196

5.4.5

Deuterium labeling experiments ............................................................................. 197

5.4.6

Computational methods .......................................................................................... 197

CHAPTER 6: Summary, conclusions, and future directions ................................................199 6.1

Summary and conclusions .......................................................................................... 199

6.2

Future directions ......................................................................................................... 201

6.2.1

Enantioselective hydroaminoalkylation with simple chiral ligands ....................... 201

6.2.2

Reactivity with pyrrolidines .................................................................................... 202

6.2.3

Pyridonate complexes of tantalum .......................................................................... 204

6.3

Concluding remarks .................................................................................................... 207

REFERENCES ...........................................................................................................................209 APPENDICES ............................................................................................................................222 Appendix A X-ray crystallographic data ................................................................................ 222 Appendix B Selected NMR spectra ........................................................................................ 228 Appendix C SFC analysis for determination of enantiomeric excess..................................... 261

x

LIST OF TABLES

Table 2.1 Relevant bond lengths (Å) and angles (°) for 8. .......................................................... 40 Table 2.2 Reactivity studies using proligand 8 for intramolecular hydroamination with alkenes. ....................................................................................................................................................... 42 Table 2.3 Relevant bond lengths (Å) and angles (º) for complex 10. .......................................... 45 Table 3.1 Relevant bond lengths (Å) and angles (º) for complex 16. .......................................... 67 Table 3.2 DOSY experiments of 5 and the reaction mixture of 5 with 1-bultylamine. ............... 70 Table 4.1 Selected bond lengths (Å) and angles (º) for complex 27. ........................................... 94 Table 4.2 Relevant bond lengths (Å) and angles (º) for complex 28. .......................................... 95 Table 4.3 Screening of new tantalum precatalysts for the hydroaminoalkylation of Nmethylaniline and p-methoxy-N-methylaniline with 1-octene. .................................................. 101 Table 4.4 Relevant bond lengths (Å) and angles (º) for complexes 38 and 39. ......................... 109 Table 4.5 Relevant bond lengths (Å) and angles (º) of 42. ........................................................ 116 Table 4.6 Relevant bond lengths (Å) and angles (º) for complexes 43, 44, and 45. .................. 123 Table 5.1 Hydroaminoalkylation of saturated N-heterocycles. .................................................. 157 Table A.1 Crystallographic parameters for the cyclic urea proligand and the bimetallic ureate and pyridonate zirconium complexes (Chapter 2). ..................................................................... 222 Table A.2 Crystallographic parameters for the bridging imido pyridonate complex (Chapter 3). ..................................................................................................................................................... 223 Table A.3 Crystallographic parameters for the mono(amidate) tetrakis(dimethylamido) tantalum complexes (Chapter 4). ............................................................................................................... 224

xi

Table A.4 Crystallographic parameters for the mixed chloro amidate tantalum complexes (Chapter 4). ................................................................................................................................. 225 Table A.5 Crystallographic parameters for the mono(sulfonamidate) tetrakis(dimethylamido) tantalum complexes (Chapter 4). ................................................................................................ 226 Table A.6 Crystallographic parameters for the homoleptic indolinyl tantalum complex (Chapter 5). ................................................................................................................................................ 227

xii

LIST OF FIGURES

Figure 1.1 Representative Nobel prize winning catalytic systems based on early, mid, and late transition metals. ............................................................................................................................. 2 Figure 1.2 Representative metallocene complexes for the selective synthesis of polypropenes with diverse tacticities.48-51 ............................................................................................................. 4 Figure 1.3 Monodentate, monoanionic ligands based upon oxygen or nitrogen donors. .............. 5 Figure 1.4 Bidentate, monoanionic ligands with (O,O)- or (N,N)-donors. .................................... 7 Figure 1.5 Fluxional behaviour of a metal complex containing a hemilabile bidentate ligand. .... 7 Figure 1.6 Bidentate, monoanionic ligands with (N,O)-donors. .................................................... 8 Figure 1.7 Protonolysis (top) and salt metathesis (bottom) synthetic strategies for the formation of early transition metal complexes and potential coordination modes of (N,O)-ligands. ............. 9 Figure 1.8 Catalytic hydroamination (A – D) and hydroaminoalkylation (E,F). The bonds broken are indicated in red and those formed in blue. .................................................................. 10 Figure 1.9 Early transition metal complexes containing (N,O)-ligand(s) synthesized via protonolysis with organic amide proligands. ................................................................................ 12 Figure 1.10 Isocyanate insertion into a metal-amido bond to generate ureate complexes. ......... 19 Figure 1.11 The tautomeric equilibrium between 2-hydroxypyridine (2-pyridinol) and 2pyridone. ....................................................................................................................................... 22 Figure 1.12 Bis(pyridonate) bis(dimethylamido) zirconium complex 5 that promotes both hydroamination and hydroaminoalkylation. ................................................................................. 24 Figure 1.13 Representative early transition metal systems containing sulfonamidate and sulfamide ligands. Mes = mesityl. ................................................................................................ 26 xiii

Figure 2.1 Representative late transition metal, alkaline metal, and Brønsted acid based catalytic systems. ......................................................................................................................................... 32 Figure 2.2 Chiral lanthanide precatalysts for asymmetric hydroamination. ................................ 33 Figure 2.3 Cationic and neutral zirconium systems for the asymmetric intramolecular hydroamination of aminoalkenes. ................................................................................................. 35 Figure 2.4 Untethered chiral amidate ligands for intramolecular asymmetric hydroamination. . 36 Figure 2.5 Tethered bis(ureate) zirconium hydroamination precatalyst 4. .................................. 37 Figure 2.6 ORTEP depiction of the solid-state molecular structure of proligand 8. The ellipsoids are plotted at 50% probability and the majority of the hydrogen atoms are omitted for clarity. The hydrogens displayed (N–H) have been located from unassigned electron density and their positions refined. ........................................................................................................................... 39 Figure 2.7 Proximity to the zirconium center of steric bulk and source of chirality in the cyclic ureate compared with the amidate ligand. .................................................................................... 43 Figure 2.8 ORTEP depiction of the solid-state molecular structure of complex 10. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. In the simplified structure (right) the methyl groups of dimethylamido ligands (N7-11) and the cyclohexyl groups of the ureate ligands ((N1, N3, N5) are removed for clarity......................................................... 44 Figure 2.9 ORTEP depiction of the solid-state molecular structure of complex 11. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. .......................... 47 Figure 3.1 Titanium precatalysts containing bidentate ligands for the hydroaminoalkylation of secondary amines. ......................................................................................................................... 58 Figure 3.2 Variable temperature 1H NMR spectra of the aryl region of complex 5 in d8-toluene. ....................................................................................................................................................... 63 xiv

Figure 3.3 ORTEP representation of the solid-state molecular structure of complex 16, with a simplified core structure shown on the right. Symmetry equivalent atoms (i) were generated with the symmetry operation (−x+1, −y, −z+1). The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. ........................................................................................ 67 Figure 3.4 Variable 1H NMR spectroscopy of the reaction of 5 with 1.5 equivalents of benzylamine in d6-benzene. .......................................................................................................... 69 Figure 3.5 Plot of consumption of 19 (c/c0) and ln(c/c0) as a function of time (min). The solid trendline depicts the least-squares fit of the data points. .............................................................. 73 Figure 3.6 Primary kinetic isotope effect observed for the cyclization of 19 compared with αdeuterated substrate 20. ................................................................................................................. 74 Figure 3.7 Observed rates of consumption of 19 as a function of catalyst concentration. .......... 76 Figure 3.8 Eyring plot in the temperature range of 85 – 110 °C and the relevant activation parameters. Error on activation parameters estimated from regression analysis. ......................... 77 Figure 4.1 Tantalum precatalysts for the α-alkylation of amines supported by (N,O)- and (O,O)chelating ligands. .......................................................................................................................... 87 Figure 4.2 New amide proligands (top) based upon the amidate ligand of 2 and potential alternative coordination geometries (bottom). .............................................................................. 90 Figure 4.3 ORTEP representation of the solid-state molecular structure of complex 27. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. .......... 93 Figure 4.4 ORTEP representation of the solid-state molecular structure of complex 28. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. .......... 95 Figure 4.5 ORTEP depiction of solid-state molecular structure of the crystalline material obtained using proligand 22. The asymmetric unit cell contains two unique tantalum complexes, xv

29 and 30. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. ........................................................................................................................................... 97 Figure 4.6 Key potential intermediates proposed by Porco and co-workers for the formation of amides using a dimeric zirconium alkoxide complex via an intramolecular transformation.288 .. 98 Figure 4.7 ORTEP depiction of the solid-state molecular structures of complexes 38 and 39. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. ........ 108 Figure 4.8 1H NMR spectrum of complex 40 in d6-benzene. .................................................... 113 Figure 4.9 ORTEP representations of the solid-state molecular structure of previously synthesized complex 42.297 The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity.................................................................................................................. 115 Figure 4.10 Phosphoramidate tantalum alkyl precatalyst 21 and the pKas of a series of amide, phosphoramide and sulfonamide compounds.300 ........................................................................ 117 Figure 4.11 Early transition metal hydroaminoalkylation precatalysts supported by sulfur containing ligands. ...................................................................................................................... 118 Figure 4.12 ORTEP depictions of solid-state molecular structures of sulfonamidate complexes 43, 44, and 45. The ellipsoids are plotted at 50% probability and hydrogen atoms are omitted for clarity. ......................................................................................................................................... 120 Figure 4.13 Proposed interconversion between κ1- and κ2-binding modes in solution. ............ 121 Figure 5.1 Natural products and pharmaceuticals containing an α-alkyl/aryl N-heterocycle. .. 144 Figure 5.2 ORTEP representation of the solid-state molecular structure of compound 47. The ellipsoids are plotted at 50% probability and the majority of the hydrogen atoms are omitted for clarity. ......................................................................................................................................... 160

xvi

Figure 5.3 ORTEP depiction of the solid-state molecular structure of complex 51. The asymmetric unit contains a disordered indoline molecule which has been successfully modeled. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. . 165 Figure 5.4 Optimized geometry for transition state TS(III/IV) for the pyrrolidine and piperidine substrates. The majority of the hydrogen atoms are omitted for clarity. .................................... 172 Figure 5.5 Optimized geometry for the bis(amido) intermediates IIIpyrr and IIIpip (top) and the metallaziridine intermediates IVpyrr and IVpip containing the neutral-bound α-alkylated product (bottom). The majority of the hydrogen atoms are omitted for clarity. ...................................... 173 Figure 5.6 Optimized geometry for the tantalaziridine intermediates Vpyrr and Vpip. The hydrogen atoms are omitted for clarity. ...................................................................................... 174 Figure 5.7 Potential energy surface for the formation of the tantalaziridine intermediates V from the bis(amido) tantalum complexes III....................................................................................... 175 Figure 6.1 Optimized geometries for the tantalaziridines with pyrrolidine (left) and piperidine (right). ......................................................................................................................................... 203 Figure 6.2 Potential new precatalysts for the hydroaminoalkylation of N-heterocycles such as pyrrolidine. ORTEP depiction of solid-state molecular structure of complex 42. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. ........................ 204 Figure 6.3 Preliminary synthesis and stacked 1H NMR spectra for the proligand and the crude mono and bis(pyridonate) complexes 52 and 53 in d6-benzene. ................................................ 205 Figure 6.4 ORTEP depiction of solid-state molecular structure of complex 54. The three equatorial pyridonate ligands are highly disordered, both orientations are shown. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. ........................ 206

xvii

Figure 6.5 Potential commercially available pyridone proligands (A) and pyridone precursor (B) as well as a reported synthetic route to highly substituted pyridones (C).410 ............................. 207

xviii

LIST OF SCHEMES

Scheme 1.1 Regioselective hydroamination of terminal alkynes promoted by precatalyst 1. ..... 14 Scheme 1.2 Intramolecular hydroamination of primary aminoalkenes with neutral bis(amidate) zirconium complexes.131,134........................................................................................................... 15 Scheme 1.3 Simplified [2+2] cycloaddition mechanism for the bis(amidate) titanium-catalyzed intermolecular

hydroamination

of

alkynes

and

the

bis(amidate)

zirconium-catalyzed

intramolecular cyclization of aminoalkenes. [Zr] and [Ti] = bis(amidate) zirconium and titanium, respectively. .................................................................................................................................. 16 Scheme 1.4 Intramolecular hydroaminoalkylation of unactivated olefins with secondary amines catalyzed by mono and bis(amidate) tantalum amido complexes. ............................................... 17 Scheme 1.5 Proposed catalytic cycle for the hydroaminoalkylation of secondary amines with tantalum precatalysts. .................................................................................................................... 18 Scheme 1.6 Synthesis of bis(ureate) alkyl, chloro, or amido complexes of group 4 metals. DME = 1,2-dimethoxyethane. ................................................................................................................ 20 Scheme 1.7 Proposed catalytic cycle for hydroamination using precatalyst 4 in which the key step is a proton-assisted σ-bond insertion. [Zr] = tethered bis(ureate) ......................................... 22 Scheme 1.8 Proposed mechanism for hydroaminoalkylation promoted by complex 5 via a bridging zirconaziridine. [Zr] = bis(pyridonate). .......................................................................... 25 Scheme 2.1 Hydroamination of alkenes using a chiral catalyst to generate enantioenriched alkylamines. .................................................................................................................................. 31

xix

Scheme 2.2 Modular route for the synthesis of urea proligand 8. Boc = tert-butoxycarbonyl, HOBt

=

hydroxybenzotriazole,

DCC

=

N,N'-dicyclohexylcarbodiimide,

DIPEA

=

diisopropylethylamine, TFA = trifluoroacetic acid. ..................................................................... 39 Scheme 2.3 Precatalyst formation by the protonolysis reaction of Zr(NMe2)4 with 8. ................ 40 Scheme 3.1 Inter- and intramolecular hydroaminoalkylation of amines. .................................... 53 Scheme 3.2 Select titanium precatalysts for the intramolecular hydroaminoalkylation of primary and secondary aminoalkenes. a10 mol% [Ti]. .............................................................................. 54 Scheme 3.3 Simplified mechanism of titanium-catalyzed intramolecular hydroaminoalkylation. ....................................................................................................................................................... 56 Scheme 3.4 Intermolecular hydroaminoalkylation of 1-octene promoted by complexes 12, 13 and 14. ........................................................................................................................................... 57 Scheme 3.5 Intramolecular hydroaminoalkylation with zirconium pyridonate precatalyst 5. ..... 59 Scheme 3.6 Proposed mechanism for intramolecular hydroaminoalkylation with precatalyst 5. [Zr] = bis(pyridonate). .................................................................................................................. 60 Scheme 3.7 Synthesis of bis(pyridonate) bis(dimethylamido) zirconium and ORTEP representation of the solid-state molecular structure of complex 5.159 The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. ................................................ 62 Scheme 3.8 Reported routes for the synthesis of monomeric metallaziridines of zirconium and tantalum......................................................................................................................................... 64 Scheme 3.9 Synthesis of dimeric imido zirconium complex 16 via aminolysis of complex 5. ... 66 Scheme 3.10 Synthesis of bimetallic titanaziridine 17................................................................. 68 Scheme 3.11 Synthesis of kinetic substrate 19 and isotopically labeled 20. ................................ 72 Scheme 4.1 Intermolecular hydroaminoalkylation of secondary amine substrates. .................... 84 xx

Scheme 4.2 The effect of chloro ancillary ligands in tantalum complexes for the intermolecular hydroaminoalkylation of 1-octene with of N-methylaniline. ........................................................ 85 Scheme 4.3 New amide proligands 22 and 23 containing modified nitrogen substituents. ......... 91 Scheme 4.4 Synthesis of proligands 24, 25, and 26 bearing pendant methoxy or tertiary amine donors............................................................................................................................................ 92 Scheme 4.5 Protonolysis methodology using amide proligands and Ta(NMe2)5. ....................... 92 Scheme 4.6 In situ substitution of the ester moiety of the proligand to generate tantalum complex 29................................................................................................................................................... 96 Scheme 4.7 Substitution at the ester moiety of the proligand to generate tantalum complex 29. 98 Scheme 4.8 Salt metathesis (top) and protonolysis (bottom) approaches to new organometallic compounds and potential mixed chloro amido tantalum precursors 32, 33, 35, and 36 (right). 103 Scheme 4.9 The syntheses of mixed dichloro, trichloro, and tetrachloro tantalum amido starting materials. ..................................................................................................................................... 104 Scheme 4.10 Reaction of pentachloro tantalum starting material with the amidate sodium salt. ..................................................................................................................................................... 105 Scheme 4.11 Non-productive ligand substitution reaction of the dimethylamido ligands of 2. 106 Scheme 4.12 Transformations investigated using via one-step protonolysis (top) and salt metathesis (bottom) procedures from mixed dichloro and trichloro amido tantalum starting materials. ..................................................................................................................................... 107 Scheme 4.13 Previously synthesized bis(amidate) tantalum complexes.146............................... 110 Scheme 4.14 Synthetic methodologies that result in the formation of complex 40. .................. 111 Scheme 4.15 Unexpected reactivity observed with aryloxo ligands. ......................................... 114 Scheme 4.16 Synthesis of 42 via salt metathesis of the tantalum dichloro precursor 33. .......... 114 xxi

Scheme 4.17 Hydroaminoalkylation reactivity of precatalyst 42 using N-methylaniline and pmethoxy-N-methylaniline and 1-octene. ..................................................................................... 116 Scheme 4.18 Sulfonamide proligands with varied N-substituents. ............................................ 119 Scheme 4.19 Synthesis of mono(sulfonamidate) tetrakis(dimethylamido) tantalum complexes 43 – 46.............................................................................................................................................. 119 Scheme

5.1

Multiple-step

organolithium-mediated

α-deprotonation/functionalization

methodology. .............................................................................................................................. 146 Scheme

5.2

Organolithium-mediated

α-deprotonation-transmetallation-Negishi

coupling

strategy. ....................................................................................................................................... 147 Scheme 5.3 Ruthenium catalyzed α-cyanation of tertiary amines. ............................................ 148 Scheme 5.4 Copper-catalyzed cross-dehydrogenative coupling reactions of tertiary amines. .. 148 Scheme 5.5 α-Functionalization via an iminium cation. ............................................................ 149 Scheme 5.6 Photoredox approach for the α-functionalization of tertiary amines via an iminium ion (top) or an α-amino radical cation (bottom). ........................................................................ 151 Scheme 5.7 Directed ruthenium-catalyzed α-arylation of N-substituted heterocycles. ............. 152 Scheme 5.8 Directed ruthenium-catalyzed α-alkylation of N-substituted heterocycles. ............ 153 Scheme 5.9 Intermolecular hydroaminoalkylation of alkenes with secondary amines.............. 154 Scheme 5.10 Hydroaminoalkylation of piperidine with mono(amidate) tetrakis(dimethylamido) tantalum precatalyst 2. ................................................................................................................ 155 Scheme 5.11 Postulated mechanism for the tantalum-catalyzed hydroaminoalkylation of Nheterocycles................................................................................................................................. 159 Scheme 5.12 Proposed synthesis of bicyclic and tricyclic compounds following α-alkylation of an N-heterocyclic amine. ............................................................................................................ 160 xxii

Scheme 5.13 α-Alkylation and ring-closure methodology using N-arylalkyl amines (top)284 and attempts at the analogous methodology with 1,2,3,4-tetrahydroquinoline. ................................ 161 Scheme 5.14 Hydroaminoalkylation with olefin substrate containing a protected carbonyl moiety. ........................................................................................................................................ 162 Scheme 5.15 Attempted hydroaminoalkylation-hydroamination procedure catalyzed by tantalum and zirconium precatalysts, respectively. ................................................................................... 163 Scheme 5.16 α-Alkylation of piperidine catalyzed by in situ generated tantalum ureate complexes. .................................................................................................................................. 164 Scheme 5.17 Homoleptic tantalum species 51 produced from reaction of 2 with indoline. ...... 165 Scheme 5.18 No hydroaminoalkylation reactivity observed with pyrrolidine. .......................... 166 Scheme 5.19 Idealized intermediates and transition states for the catalytic α-alkylation of dimethylamine with 1-octene, followed by the coordination/activation of an N-methylaniline substrate. Relative free energies (∆G) are reported in kcal/mol.146 ............................................ 168 Scheme 5.20 Simplified catalyst activation via hydroaminoalkylation of a dimethylamido ligand (I – III, left) and proposed mechanistic cycle with N-heterocyclic substrates (IV – VI, right). 170 Scheme 5.21 Summary of calculated intermediates and the transition state for the C–H activation of pyrrolidine (pyrr) and piperidine (pip) to form tantalaziridines V. Relative free energies (∆G) are reported in kcal/mol. ............................................................................................................. 171 Scheme 5.22 Deuterium scrambling experiment reported with dimethylamine (top) and proposed experiment with N-heterocyclic amines (bottom)....................................................................... 176 Scheme 5.23 Sites of deuterium incorporation for piperidine and pyrrolidine substrates. ........ 177 Scheme 5.24 Calculated intermediates leading to α- or β-arylation reported by Baudoin and coworkers.394................................................................................................................................... 178 xxiii

Scheme 5.25 Potential mechanism for the β-deuteration observed and the optimized geometry for Vpyrr with the shortest dimethylamido–β-hydrogen distance indicated. The majority of the hydrogen atoms are omitted for clarity. ...................................................................................... 179 Scheme 6.1 Urea proligands incorporating chirality in the carbon backbone (top) or the amino groups (bottom) for tantalum-catalyzed asymmetric hydroaminoalkylation.............................. 202

xxiv

LIST OF ABBREVIATIONS AND ACRONYMS

Å

angstrom (10-10 m)

Ac

acetyl

anal.

analysis

Ar

aryl

B3LYP

Becke 3-parameter Lee-Yang-Par functional

Bhyd

benzhydryl

Bn

benzyl

Boc

tert-butoxycarbonyl

bpy

2,2'-bipyridine

br

broad

n

Bu, sBu, tBu

normal butyl, sec-butyl, tert-butyl

C

Celsius

c

concentration

calcd.

calculated

Cp

cyclopentadienyl

Cp*

pentamethylcyclopentadienyl

Cy

cyclohexyl

°

degree

d

deuterium

d

doublet

δ

chemical shift xxv

DBU

1,8-diazabicycloundec-7-ene

DCC

N,N'-dicyclohexylcarbodiimide

DCM

dichloromethane

DEPT

distortionless enhancement by polarization transfer

DFT

density functional theory

∆G‡

Gibbs free energy of activation

∆H‡

enthalpy of activation

∆S‡

entropy of activation

DIPEA

diisopropylethylamine

DMA

dimethylacetamide

DME

1,2-dimethoxyethane

DMSO

dimethylsulfoxide

DOSY

diffusion ordered spectroscopy

d.r.

diastereomeric ratio

E+

electrophile

ee

enantiomeric excess

EI

electron impact

Et

ethyl

e.u.

entropy units (cal K-1 mol-1)

EWG

electron withdrawing group

fac

facial

gem

geminal

h

hours xxvi

HMDS

hexamethyldisilazane (N(SiMe3)2)

HOBt

hydroxybenzotriazole

HRMS

high resolution mass spectrometry

Hz

Hertz

Ind

indenyl

i

Pr

isopropyl

J

coupling constant

κ

denticity

k

rate constant

K

kelvin

Ka

acid dissociation constant

kB

Boltzmann’s constant

kcal

kilocalorie

KIE

kinetic isotope effect

L

supporting ligand

Ln

ligand set

Ln

rare-earth metal

m

multiplet

M

metal, mol/L

M+

molecular ion

MAO

methylaluminoxane

μ

bridging ligand

Me

methyl xxvii

Mes

mesityl

min

minute

mol

mole

mol%

mole percent

MP2

second order Møller-Plesset perturbation theory

MS

mass spectrometry

Ms

mesyl, methanesulfonyl

m/z

mass-to-charge ratio

Naphth

naphthyl

NMR

nuclear magnetic resonance

Nu

nucleophile

ORTEP

Oak Ridge thermal ellipsoid plot

p

-log (as in pKa)

[P]

photocatalyst

Ph

phenyl

pip

piperidine

PMP

p-methoxyphenyl

ppm

parts per million

pyrr

pyrrolidine

py

pyridine

[O]

oxidant

q

quartet

R

organic substituent xxviii

rac

racemic

s

singlet, second

sept

septet

SET

single-electron transfer

t

Bu3SiO–

silox

Σ

summation

t

triplet

T

temperature

TBAF

tetrabutylammonium fluoride

TBS

tert-butyldimethylsilyl

tert

tertiary

TFA

trifluoroacetic acid

THF

tetrahydrofuran

TMS

trimethylsilyl

tol

toluene

Ts

tosyl, para-toluenesulfonyl

X

halide

xxix

ACKNOWLEDGEMENTS

I sincerely thank my supervisor Dr. Laurel Schafer for the guidance, inspiration, encouragement, and unwavering support she provided throughout my graduate studies. Thank you for giving me so many opportunities to grow as both a scientist and an individual. You are a phenomenal teacher and mentor. Thank you to my committee members, Dr. Glenn Sammis and Dr. Michael Fryzuk, for their valuable input and thorough critique of this thesis prior to submission. Thank you to Ken Love for his tireless help with the GC-MS, Dr. Brian Patrick for his assistance with X-ray crystallography, Maria Ezhova and the NMR staff, and Marshall Lapawa for help with the mass spectrometry. I would like to acknowledge the following funding agencies that have supported my research: the University of British Columbia, the Natural Sciences and Engineering Research Council of Canada, and Boehringer Ingelheim Ltd. My heartfelt thanks to all past and present members of the Schafer group for creating such a supportive and fun lab environment. I have learnt so much from you all. Special thanks to Dr. J.M. Lauzon, Dr. Erin Morgan, and Dr. Désirée Sauer for their input and guidance during the assembly of this thesis. Thank you to my proofreaders: Dr. J.M. Lauzon, Dr. Erin Morgan, Scott Ryken, Mitch Perry, and Jacky Yim. Thank you to Helen Huang-Hobbs, my thesis writing partner. Thank you to Peter Christensen for his unending encouragement throughout the thesis writing process. Thanks to my siblings, Nicki and Alex, who are always available for pick-me-up phone calls. Much love to my parents, Nick and Jan, who truly understand what it takes to write a thesis. I am deeply grateful for your unfailing love and support. xxx

DEDICATION

This thesis is dedicated to my parents,

Dr. Gillian Janet Ann Payne

and

Dr. Nicholas Charles Payne

xxxi

CHAPTER 1: Synthesis, structure, and reactivity of early transition metal precatalysts bearing (N,O)-chelating ligands.

1.1

Introduction The first Nobel Prize in inorganic chemistry was awarded one hundred years ago, in

1913, to Alfred Werner for his work defining the basics of coordination chemistry.1 Currently, the application of coordination and organometallic complexes spans many aspects of science and technology and their importance cannot be overemphasized.2-7 Arguably, the most significant application of metal complexes, in industry and academia, is the field of heterogeneous and homogeneous catalysis.8-13 The use of a catalyst allows for milder reaction conditions and the reduction or elimination of waste resulting in more sustainable and environmentally benign chemical processes.14,15 Homogeneous catalysis, in which the catalyst and the reactants share one phase, affords active molecular species that can be identified and monitored spectroscopically. Furthermore, a well-characterized catalytic system may allow for a mechanistic understanding of the performance of the catalyst at the molecular level. Such insight can be instrumental in the rational design of catalytic systems with precise reactivity and selectivity. One of the major applications of homogeneous catalysis is in the synthesis of commodity chemicals. Notable large scale industrial processes include the Monsanto process for the production of acetic acid through the carbonylation of methanol,16-19 and hydroformylation to generate aldehydes and value-added chemicals from raw petrochemical materials.16,20-22 Homogeneous catalysis is also critical for the synthesis of intermediates and products in the fine chemical, agrochemical, and pharmaceutical industries.11,13 Research in the field of 1

homogeneous catalysis involves both the development of new catalytic systems to improve existing methodologies as well as the discovery of new catalytic reactions. The importance of these advances has been recognized repeatedly; three Nobel Prizes in the last decade have been awarded to four catalytic reactions: asymmetric hydrogenation and oxidation (Knowles,23 Noyori,24 Sharpless,25 2001), metathesis (Chauvin,26 Grubbs,27 Schrock,28 2005), and crosscoupling reactions (Heck,29,30 Negishi,31 Suzuki,32 2010).

iPr

Ar2 Cl H2 P N Ru P Ar2 Cl N H2

iPr

Ti(O Pr)4, EtO2C HO

CO2Et OH

F3C F3C

Cy3P

Cl

CF3

Schrock

Sharpless

Ph Ru

Ph O

F3C Noyori

Cl

Mo

O

N

N

N

i

Grubbs

Figure 1.1 Representative Nobel prize winning catalytic systems based on early, mid, and late transition metals.

Catalyst systems have been developed using metals from across the periodic table. However, early transition metal systems, which contain the d-block elements of group 3 (scandium, yttrium), group 4 (titanium, zirconium, hafnium), and group 5 (vanadium, niobium, tantalum), are especially appealing as they are inexpensive, naturally abundant, and biologically benign metals particularly compared with the late transition metals.33,34 The properties and reactivity of the resultant metal catalysts are as dependent on the supporting ligand environment as they are on the identity of the metal itself. The ancillary ligands influence physical properties of the metal complex (stability, solubility), the reactivity and, therefore, the required reaction conditions (pressure, temperature, time), and the selectivity imparted by the catalyst.

2

A key example of the application of early transition metal catalysts, and the remarkable impact of the supporting ligands, are the group 4 Ziegler-Natta catalysts35-38 for the controlled polymerization of ethylene and propylene to generate α-polymers for plastics manufacturing, a multi-billion dollar industry. In the early 1950’s, Ziegler and Natta showed that, under mild reaction conditions using titanium tetrachloride in the presence of alkylaluminium activators, the polymerization of ethylene39 and propylene40 could be achieved. The prochiral propene monomers can give rise to a variety of polymers containing different tacticities (Figure 1.2); control of the regio- and stereoselectivity of the polymerization can, therefore, result in a variety of polymeric microstructures, each with different physical and mechanical properties. However, the initial TiCl4/AlEt3 catalyst system is ill-defined. This results in limited control over polymer structure and molecular weights and hampers a thorough mechanistic understanding. The development of the well-defined, homogeneous single-site group 4 metallocene catalysts (Figure 1.2) activated by co-catalysts such as methylaluminoxane (MAO), represents a significant advancement to the field.37,38,41-44 These complexes are soluble in hydrocarbon solvents and impart control over the polydispersity and the microstructure of the polymer by the judicious choice of the tunable ancillary ligands.42,45 Their well-defined active site and solubility means their reactivity can be intimately studied and this thorough mechanistic understanding has been used as a guide for on-going catalyst design efforts.46,47

3

2n

[Zr] MAO

n

Naphth

[Zr] =

Si Zr Cl Cl

C2v-symmetric

Si Zr Cl Cl

Ph Zr Cl Cl

Naphth C2-symmetric

Cs-symmetric

n

n

n

Atactic

Ph

Isotactic

Syndiotactic

Cl

Zr

Cl

C1-symmetric

n

Hemiisotactic

Figure 1.2 Representative metallocene complexes for the selective synthesis of polypropenes with diverse tacticities.48-51

These catalyst design efforts and the wide-scale application of group 4 metallocenes in the synthesis of polyolefins is an illustrative example of the impact ancillary ligands can have on reactivity, selectivity, stability, and solubility of metal complexes. While the cyclopentadienyl (Cp) ligand motif and its analogues continue to find diverse use in both catalytic and stoichiometric synthetic transformations,52 limitations, such as synthetic challenges, have precipitated the departure from this scaffold to other ligand architectures.53-56 Ligand scaffolds that are synthesized in a modular manner in minimal steps are particularly attractive, as this allows for rapid tuning and optimization of a particular metal-ligand combination to suit the transformation, substrate combination, or selectivity required. The majority of non-Cp ligands for early transition metals are based upon oxygen- or nitrogen-donors, as these hard, multipleelectron donor atoms are effective at stabilizing the high oxidation states of the hard, Lewis-

4

acidic, oxophilic metals. An overview of relevant monoanionic non-Cp ligands will be discussed in the following sections.

1.2

Mono anionic, monodentate ligands based on N- and O-donors. Common Cp-analogues are the monoanionic ligands based on nitrogen and oxygen

(Figure 1.3), such as the oxygen-donor alkoxo ligands.57 The alkoxo ligands appear simplistic, however their use is not without challenges; in particular, the steric bulk of the ligand is further removed from the metal centre and therefore these ligands are less sterically bulky than the Cp ligands. The complexes formed are coordinatively unsaturated and the alkoxo ligands are able to bridge metal atoms. These characteristics often result in the formation of bridged bimetallic species or aggregates that can access low energy pathways for ligand exchange and redistribution reactions that hamper controlled catalytic applications. One effective strategy to minimize dimerization and aggregation is to increase the steric bulk of the ligand to access well-define monomeric complexes.58-62 Highly sterically demanding alkoxides and phenoxides (Figure 1.3) have been shown to support d0 alkylidene and alkylidyne complexes for olefin and acetylene metathesis reactions.63 R1

O R1

O

O R3

R2

alkoxide

R2

R1

phenoxide

R1

Si

R3

R2

siloxide

R2

N

R1

amido

Figure 1.3 Monodentate, monoanionic ligands based upon oxygen or nitrogen donors.

An alternative strategy are the siloxo ligands (Figure 1.3) pioneered by Wolczanski and co-workers.60,64,65 The presence of the silicon in the backbone decreases the electron-donating 5

ability of the siloxides compared with the alkoxides and results in a more electrophilic metal centre.65,66 These ligands have been applied to generate a variety of early transition metal complexes with rich and varied chemistry. The tBu3SiO– ligand (termed “silox”) supports reduced low-coordinate, and highly reactive early transition metal centres in the MIII oxidation state (M = Ti, V, Ta, Nb, Mo), which have shown the notable reactivity in C–H bond activation of hydrocarbons.64 The simple amido ligands (Figure 1.3) offer an advantage over the alkoxo, phenoxo, and siloxo systems because of the potential for two substituents at the nitrogen.56,67-69 Homoleptic amido systems are extensively applied in catalysis, including Ti(NMe2)470,71 that is a precatalyst for hydroamination. Homoleptic amido metal complexes are also useful starting materials for the preparation of new coordination complexes (vide infra). Importantly, the homoleptic dimethylamido complexes of titanium, zirconium, hafnium, tantalum, and niobium are commercially available.

1.2.1

Multidentate, monoanionic ligands of (O,O)-, (N,N)-, and (N,O)-compounds Polydentate ligand architectures containing multiple N- or O-donors are often applied to

generate well-defined monomeric complexes. These compounds exhibit improved kinetic and thermodynamic stability because of the chelate effect.72 The more complex ligand scaffold also contains additional sites for ligand modification and optimization; alterations to the rigidity or flexibility of the tether, the distance between donor atoms and therefore different chelate ring sizes, the orientation of the donors, and the identity of the chelating atoms all influence the structure and reactivity of the resulting complex. Bidentate ligands incorporating two identical donors have been extensively investigated and have many applications in homogeneous catalysis 6

including their role as ancillary ligands in post-metallocene catalysts for olefin and ring-opening polymerization reactions,54,73 hydroamination,74-76 and hydroaminoalkylation.77,78 Representative ligands of this type include the four-membered chelating amidinate,79-81 guanidinate,74,79,82,83 and 2-aminopyridonate75,78,84 ligands, as well as the five- and six-membered chelating aminotropionate,69,92 β-diketiminate,85,86 and β-diketonate87-91 ligands (Figure 1.4).68,76 R1 N

R3

R1 N

R2

2

R2

(R )2N

N R1 amidinate

N R1 guanidinate

R1 N

N N

R2 R1

2-aminopyridonate

R3

N

N

R1

N R1

aminotroponiminate

R3

O

O

R1 R1

R1

R2

R2

β-diketiminate

β-diketonate

Figure 1.4 Bidentate, monoanionic ligands with (O,O)- or (N,N)-donors.

Chelating ligands that can bind to a metal centre through two different donor atoms are another class of bidentate ligands.92 This ligand scaffold is especially appealing because the electronic asymmetry induced by the presence of two different donor atoms can promote ligand hemilability (Figure 1.5).93-95 Hemilabile ligands display dynamic coordination behaviour of the more weakly bound donor allowing for the intermittent generation of vacant sites. Z Y

Z

[M]

Y [M]

Figure 1.5 Fluxional behaviour of a metal complex containing a hemilabile bidentate ligand.

The fluxional behaviour of the hemilabile ligands is particularly attractive for catalytic systems; the ligand adds stability to the metal catalyst while retaining the potential for the open coordination sites required for desirable reactivity. The ligand can also assist in the displacement of coordinated products to regenerate the active catalyst. Ligand hemilability is governed by 7

subtle energy differences and is still difficult to control. However, it provides yet another facet of ligand design that can be tuned and optimized and is, therefore, a useful approach to generating new catalytic systems.96-98 One class of hemilabile ligands ideally suited to the coordination of early transition metal complexes is (N,O)-chelating ligands (Figure 1.6). The amidate,99 ureate,100,101 pyridonate,102 and sulfonamidate103 ligands, when bound in a κ2-(N,O) fashion, result in four-membered chelates with tight bite angles. O R2

O

O NH R1

amidate

R2

2

(R )2N ureate

N R1

O O N

S

R2 R1

2-pyridonate

N

R1 sulfonamidate

Figure 1.6 Bidentate, monoanionic ligands with (N,O)-donors.

The (N,O)-proligands are easily synthesized organic compounds that contain numerous sites of potential modification and optimization (eg. R1 and R2, Figure 1.6). Their modular syntheses allow for straightforward structure-reactivity studies and adjustment of ligand steric and electronic parameters to influence catalytic behaviour. The neutral proligands contain protons of sufficiently acidity for the protonolysis of metal alkyl or amido ligands, which can be used to advantage for the one-step synthesis of metal complexes (Figure 1.7, top). The byproducts of the protonolysis approach are volatile alkanes or amines that can be easily removed from the reaction mixture under reduced pressure, simplifying the workup and purification procedures. Another commonly used synthetic strategy involves deprotonation of the proligand by alkali metal bases to generate a ligand salt that can then react with metal halide precursors via salt metathesis (Figure 1.7, bottom). The resultant metal complexes can contain the ligands 8

bound in a variety of fashions. The most common of these binding modes include monodentate, O- or N-bound, chelating (N,O)-bound, and a bridging motif involving multiple metal centres (Figure 1.7). O [M] O MRx NH

- HR R = alkyl, amido

O Na N

MClx

- NaCl

R (N,O-ligand)MRx-1

N R

R

R

[M] N R

N [M] R Monometallic O [M]

(N,O-ligand)MClx-1

O

O R

O

[M]

R [M] N [M] N R R Bimetallic

Figure 1.7 Protonolysis (top) and salt metathesis (bottom) synthetic strategies for the formation of early transition metal complexes and potential coordination modes of (N,O)-ligands.

The amidate, ureate, and pyridonate ligands have been extensively applied to generate early transition metal complexes for applications in catalysis.99,104-112 These include, but are not limited to, catalytic transamidation,108 the amidation of aldehydes,113 and the polymerization of ethylene,105 lactams,104 ε-caprolactone114 and rac-lactide.115 The most extensive application of these complexes is as precatalysts for two complementary atom-economic catalytic syntheses of amines: hydroamination111,116,117 and hydroaminoalkylation112,118 (Figure 1.8).

9

Alkyne hydroamination H R3 catalyst N 2 1 A R R R3

R3

H N

R1

R3

R3

R3

R1

R2

H N

( )n

catalyst B

H

( )n R1

R2 R3

Alkene hydramination

R1

R3 R2

R3

catalyst C

∗

H R1

N

R3

Alkene hydroaminoalkylation R1

H N

H

R2 R2

R3

R3

H

R3 catalyst E

R1

H N R2

R1

R2

R1

∗

H N

catalyst D

∗

H R1

H ( )n R2

( )n

N

R2

H

R3

R3 R2

( )n R2

H N

R2

R3

R3 H N

N

catalyst F

R1

H N

R3

∗

( )n R2

Figure 1.8 Catalytic hydroamination (A – D) and hydroaminoalkylation (E,F). The bonds broken are indicated in red and those formed in blue.

Hydroamination119,120 and hydroaminoalkylation121 are two hydrofunctionalization reactions consisting of the formal addition of an N–H or C–H bond across a C–C unsaturation. These are attractive methodologies for the synthesis of higher value amine products with increased molecular complexity from readily available starting materials. The amine compounds produced are relevant for a variety of applications including pharmaceutical drugs, agrochemicals, and natural product synthesis. These methodologies are both 100% atomeconomic, in that all of the atoms in the reagents are retained in the product, eliminating the production of wasteful by-products. The push towards more environmentally benign methodologies and the inherent economic advantages means that technologies such as these are attractive alternatives to established stoichiometric procedures. The kinetic and thermodynamic

10

factors of these reactions, as well the potential for a variety of regio- and stereoisomeric products necessitate the mediation of highly active and selective catalysts. The development of a general catalyst system, capable of promoting hydroamination reactivity with unactivated alkenes and alkynes, primary and secondary amines, in both an intraand intermolecular fashion is an important goal. Transition metal catalysts from across the periodic table have been investigated for this transformation.122,123 Early transition metal catalysts124 are of particular interest due to their high reactivities, with minimized air- and moisture-sensitivity compared with the rare-earth metal systems, and lower cost and toxicity compared with the late transition metal catalysts. The (N,O)-ligands in Figure 1.6 have been studied in precatalysts for hydroamination and hydroaminoalkylation methodologies that display promising substrate scope and reactivity. The following sections will provide an overview of the coordination chemistry of early transition metals with amidate, ureate, pyridonate, and sulfonamidate ligands in the context of their application in the catalytic synthesis of amines.

1.2.2

Amidate complexes for hydroamination and hydroaminoalkylation A modular class of easily synthesized (N,O)-ligands are the amidates, which are effective

in supporting a range of early transition metal complexes.99 A salt metathesis synthetic route has been used with limited success to generate mixed amidate chloro complexes; however this methodology can result in intractable mixtures of products, including bridged dimeric or illdefined multi-metallic species.99,101,114 On the other hand, protonolysis is a reliable route to a variety of group 3, 4, and 5 metal amidate complexes (Figure 1.9).

11

O n

MRx NH

-n HR

n=3

n=4

Group 3

O

R

Y N Ar

Group 4

R1 N Ar

Y(N(SiMe3)2) R N Ar

N Ar

THF

O M(NR2)2 O

N Ar

O

2

Y(N(SiMe3)2)2 N Ar

THF

2

R

Ti

M

O

R

NR2

Ph O

THF

3

O N Ar

n=1

n=2

O

Not applicable

(N,O-ligand)nMRx-n

M = Y (x = 3); Zr, Ti, Hf (x = 4); Ta (x = 5) R = NMe2, NEt2, N(SiMe3)2, Bn

N R

Ph

Ligand redistribution

Ph

4

M = Zr, Hf

O

O

Ph

Zr(NMe2) N Ar

R

M(NR2)2 N Ar

3

2

M = Ti, Zr, Hf

O Group 5

No reported attempts of n = 3 − 5

R

O

Ta(NMe2)3 N Ar

R

O N Ar

R

Ta(NMe2)4 N Ar

Figure 1.9 Early transition metal complexes containing (N,O)-ligand(s) synthesized via protonolysis with organic amide proligands.

In most cases, adjusting the ligand to metal stoichiometry allows for the selective generation of targeted early transition metal complexes (Figure 1.9). Group 3 mono-, bis-, and tris(amidate) yttrium complexes have all been characterized and the amidate ligands are bound in a κ2-(N,O) chelating motif with a neutrally bound tetrahydrofuran (THF) solvent molecule in the coordination sphere. Of these three classes of yttrium amidate complexes, the tris(amidate) 12

complexes are the most active as precatalysts for the amidation of aldehydes113 and as initiators for the polymerization of ε-caprolactone to generate high molecular weight polymers.114 Monoand bis(amidate) yttrium complexes are active precatalysts for intramolecular aminoalkene hydroamination (Figure 1.8, D).117 Highly sterically crowded group 4 metal complexes bearing three or four amidate ligands can be synthesized via protonolysis (Figure 1.9).125 The tris(amidate) complexes display different coordination modes dependent on the size of the metal centre; the complex containing the smaller titanium metal, less able to accommodate expanded coordination numbers and high steric crowding, contains two κ2 and one κ1(O)-bound amidate. The larger zirconium and hafnium metal centres are capable of accommodating three, and even four κ2-bound amidates, resulting in complexes with expanded coordination numbers. These complexes, however, do not contain the two reactive ligands required for application as hydroamination or hydroaminoalkylation precatalysts. The bis(amidate) bis(amido) complexes of titanium and zirconium (Figure 1.9) are a broadly applicable class of precatalysts for hydroamination. The titanium complexes are the most active for the hydroamination of alkynes126 and complex 1 has been identified as a regioselective catalyst for the intramolecular126 and intermolecular hydroamination of alkynes,127,128 to generate the aldimine products selectively (Scheme 1.1). The titanium catalyst functions well with both aryl- and alkylamines, displays good tolerance to esters, silyl-protected alcohols, and aryl halide functional groups, and promotes excellent regioselectivity. The reactive aldimine products can be further elaborated using one-pot procedures to synthesize aldehydes,143,144 substituted amines,143,144 α-cyano amines,129 α-amino acids,129 piperazines and morpholines,130 and tetrahydroisoquinoline and benzoquinolizine alkaloids.128 13

R1

NH2

R2

5 mol% 1 24 h, 65 °C

R1HN

NHR1 R2

+

R2 Not observed

O Ph iPr

Ti(NEt2)2 N 2

R 1N

NR1

1

iPr

R2 +

R2 Not observed

Scheme 1.1 Regioselective hydroamination of terminal alkynes promoted by precatalyst 1.

Group 4 bis(amidate) bis(amido) complexes have also been identified as precatalysts for the more challenging hydroamination of alkenes. The majority of investigations in this field focus on the intramolecular cyclization of aminoalkenes with zirconium based catalysts.131 Neutral group 4 bis(amidate) zirconium amido or imido complexes are efficient precatalysts for the intramolecular cyclization of primary amines to form pyrrolidine and piperidine products (Scheme 1.2). The monomeric imido complex can be generated by reaction of the bis(amido) complex with 2,6-dimethylaniline and trapped with triphenylphosphine oxide.131 The bis(amido) and imido complexes show comparable half-lives for the cyclization reactions, which implies both precatalysts share a common catalytically active species. The asymmetric version of this transformation, producing enantioenriched α-chiral amines, is an attractive goal. Following a report by Bergman and co-workers of neutral bis(amido) zirconium precatalysts displaying ee values of up to 80%,132 Schafer and co-workers described the use of neutral biaryl bis(amidate) zirconium complexes for this transformation that are proficient at the cyclization of aminoalkenes with enantiomeric excesses up to 93%.48,49,133 Research efforts concerning the asymmetric hydroamination of unactivated olefins are covered in more depth in Section 2.1.

14

R1 R2 ( )n

NH2

∗

10 mol% [Zr] 110 °C

H N

( )n R1

R2

R1 = Ph, Me R2 = Ph, Me, H n = 1,2

Ar O [Zr] =

O

Ph iPr

Zr(NMe2)2 N

N

O PPH3

N

Zr

Ph iPr

O

N

N

2

2

iPr

iPr

Zr(NMe2)2HNMe2 O

Ar Ar = 2,4,6-trimethylphenyl

Scheme 1.2 Intramolecular hydroamination of primary aminoalkenes with neutral bis(amidate) zirconium complexes.131,134

The majority of group 4 catalyzed hydroamination reactivity has been proposed to occur via a [2+2] cycloaddition mechanism involving a catalytically active metal-imido species (Scheme 1.3).131,135-140 Mechanistic investigations for the bis(amidate) catalysts are consistent with this proposal, supported by the lack of reactivity observed with secondary amine substrates.131,139-142 The reaction, therefore, involves initial formation of the zirconium-imido species, followed by [2+2] cycloaddition with the C–C unsaturation. The orientation of the alkyne or alkene during the cycloaddition step determines the regioselectivity of the reaction. Successive protonation of the metallacycle and the new amine release the product and regenerates the catalytically active imido species.

15

[Ti] NH

NMe2

NMe2

[Zr]

NMe2

NMe2

R R

H2NR1

2 H 2N

H 2 HNMe2

2 HNMe2

R2

R [Ti]

NH2

N

R2

R1

H N H

H

R

R2

N

R

R

R R

NHR1 [Ti]

[Zr]

[Ti]

NR1

R2

H

NR1

H 2N R

H

R2

[Zr]

N R

H2NR1

Scheme 1.3 Simplified [2+2] cycloaddition mechanism for the bis(amidate) titanium-catalyzed intermolecular hydroamination of alkynes and the bis(amidate) zirconium-catalyzed intramolecular cyclization of aminoalkenes. [Zr] and [Ti] = bis(amidate) zirconium and titanium, respectively.

These precatalysts require two or more reactive ligands to allow for the open two coordination sites required for productive reactivity. Mono(amidate) complexes of group 4 metals would also fulfill this requirement; however, all attempted syntheses of these complexes have not been successful and ligand redistribution is observed resulting in the formation of the bis(amidate) bis(amido) and the homoleptic metal amido complexes. Group 5 mono(amidate) complexes have been successfully prepared and, along with bis(amidate) tris(dimethylamido) tantalum complexes, are active precatalysts for hydroaminoalkylation.118 In particular, mono(amidate) tantalum complex 2 is a broadly applicable precatalyst for the intermolecular αalkylation of secondary amines (Scheme 1.4).118

16

R

R4

H 1N

H R3

R2

Ta(NMe2)4 N

[Ta] =

iPr

H

N ∗ R1

∗

R3

R2 Ar N

O tBu

5 − 10 mol% [Ta] 130 °C

R4

H

iPr

O O Ta(NMe2)3 N Ar

Mes N N

O Ta(NMe2)3 O Mes

Ar = 2,6-dimethylphenyl

2, Schafer

Schafer

3, Zi

Scheme 1.4 Intramolecular hydroaminoalkylation of unactivated olefins with secondary amines catalyzed by mono and bis(amidate) tantalum amido complexes.

Complex 2 remains the most broadly applicable catalyst for this reaction, catalyzing the α-alkylation of N-arylalkylamines, controlled mono-alkylation with dienes, and displays tolerance of oxygen-containing substrates.118 Most importantly, this remains the only catalytic system capable of the α-alkylation of N-heterocyclic amine substrates, which are important structural motifs for the pharmaceutical industry. In all cases thus far this precatalyst shows regioselective hydroaminoalkylation to generate the branched product, and excellent diastereoselectivity when applicable (Scheme 1.4). The axially chiral biaryl-based bis(amidate) ligands support catalytic systems, such as 3 (Scheme 1.4), which are generally less reactive than their mono(amidate) counterparts112,118,143 but promote enantioselective hydroaminoalkylation with ee’s reported of up to 93%.112 Mechanistic investigations indicate that the tantalum precatalysts all react by the formation of a catalytically active tantalaziridine following α-C–H activation (Scheme 1.5).118,144,145 The incoming olefin undergoes insertion into the reactive Ta–C bond, setting both 17

the diastereoselectivity and regioselectivity of the product and generating the five-membered metallacycle. Protonation of the metallacycle by an incoming amine substrate and subsequent αC–H activation results in the regeneration of the tantalaziridine and release of the α-alkylated product. Extensive computational experiments are consistent with this proposed mechanism.146 [Ta] HN

NMe2 NMe2

R1 2 HNMe2 [Ta]

R1

H N

R2

A

N [Ta] R1

N R1

R2

R2

R1

[Ta] N

N

R2

R1 B

C HN

R1

Scheme 1.5 Proposed catalytic cycle for the hydroaminoalkylation of secondary amines with tantalum precatalysts.

1.2.3

Ureate complexes for hydroamination The ureate ligand class, closely related to the amidates, contains an amino substituent in

the backbone that is capable of engaging in resonance. This alters the electronic properties of the system, which improves the electron-donating abilities of these ligand via π-donation of the nitrogen lone-pair. Initial studies into the preparation of early transition metal ureate complexes used an isocyanate insertion approach to generate the ureate complexes from metal amido 18

species (Figure 1.10).147,148 However, this is a laborious approach, where modular changes to the resultant metal complexes require a variety of metal amido and isocyanate starting materials. The complexes formed are also susceptible to further isocyanate insertions leading to product mixtures.149 [M] NR2

O

O

C

N

Ph

NR2

[M] N Ph

Figure 1.10 Isocyanate insertion into a metal-amido bond to generate ureate complexes.

The protonolysis route with urea proligands, analogous to that utilized for preparing amidate complexes, is a much more attractive synthetic approach. Extensive work spearheaded by Dr. David Leitch of the Schafer group describe the synthesis, structure, and reactivity of titanium and zirconium bis(ureate) complexes bearing alkyl,150 chloro,101 and amido ligands (Scheme 1.6).100 The ureate ligands examined include both tethered and untethered bis(urea) proligands and studies show that the tethered ligand is often much more successful in generating well behaved coordination complexes by eliminating the fluxional behaviour and coordination isomerism observed with the untethered bis(ureate) complexes. The tethered motif also increases steric accessibility to the metal centre, which is particularly attractive for catalytic applications. The chloro complexes formed are electron deficient complexes and often retain a neutral ligand; either the dimethylamine by-product or coordinating solvent. The larger alkyl ligands allow for the synthesis of complexes without a coordinating neutral donor ligand by affording greater steric protection of the metal centre.

19

O

O iPr

Ti(NMe2)2Cl2 or Zr(NMe2)2Cl2(DME) iPr

2N

NH HN

ZrBn4

iPr

2N

M

N N

L

2N

iPr

O

iPr

2N

O NMe2

Ph N N

Zr O

O Cl iPr

Zr(NMe2)4

2N

O Cl N N

NiPr2

Ph

2N

iPr

2N

Zr

NMe2

O NMe 2 H 4

M = Ti, Zr L = HNMe2, THF

Scheme 1.6 Synthesis of bis(ureate) alkyl, chloro, or amido complexes of group 4 metals. DME = 1,2-dimethoxyethane.

The most extensively investigated class of metal ureate complexes are the bis(ureate) bis(amido) complexes and their associated hydroamination reactivity.151 These precatalysts have been extensively characterized and the metrical parameters in the solid-state molecular structures provide firm evidence that the ureate-ligands are more electron-rich than their amidate counterpart leading to tighter metal-ligand interactions. The ureate ligands are consistently bound in the κ2-chelating motif irrespective of the steric bulk of the ligand. The solid-state data for the tethered complexes reveals planar sp2 geometry of the backbone nitrogen, consistent with lonepair donation into the π-system. This is not always consistent with solution-phase NMR spectroscopy that shows magnetic equivalence of the iso-propyl methyl groups, suggesting weak electron donation by the distal nitrogen, allowing for free rotation about the C–N bond in solution.100

20

Reactivity studies indicate that the zirconium bis(ureate) precatalysts examined are more reactive for intramolecular hydroamination than the titanium analogues, consistent with what has been observed with the amidate systems.131 The tethered systems show drastically improved activity compared to the untethered systems. Complex 4 (Scheme 1.6) was identified, following extensive catalytic screening, as a highly active, broadly applicable hydroamination precatalyst.100 This system is applicable for the intramolecular hydroamination of alkynes as well as the intramolecular hydroamination of aminoalkenes. Heteroatoms are also tolerated, though unsaturated ester or amide functionalities are not, and this system functions well with a broad variety of primary and secondary amines substrates. The reactivity with secondary amines is particularly notable, as this implies that the bis(ureate) ligand is active via a different mechanistic pathway than that followed by most group 4 hydroamination catalysts. Extensive mechanistic studies support the catalytic cycle shown in Scheme 1.8, which has been independently corroborated by computational studies.152,153 The key step of this mechanism is a proton-assisted σ-bond insertion via transition state A (Scheme 1.8). This novel reactivity has been attributed to increased nucleophilicity of the equatorial amido ligand because of the electron-rich ureate ligand, as well as the presence of the coordinated neutral amine that participates in proton transfer.152,153

21

[Zr](NMe2)2HNMe2 R2 R2 excess

R1HN 3 HNMe2 NR2

R1 R1 R1HN

[Zr]

R2

R2

NR

R 2N H R2

NR2 [Zr]

[Zr] NR2 R2

NR1

R 2N H

R2

A R2

NR1

NR2 [Zr]

N R1

R2 R2

NR2

Scheme 1.7 Proposed catalytic cycle for hydroamination using precatalyst 4 in which the key step is a proton-assisted σ-bond insertion. [Zr] = tethered bis(ureate)

1.2.4

Pyridonate complexes for hydroamination and hydroaminoalkylation The 2-pyridone ligand and its derivatives are an important ligand motif in coordination

chemistry.102,154 In this (N,O)-ligand system (Figure 1.11), the nitrogen is a part of an aromatic ring and is, therefore, both sterically and electronically distinctive from the acyclic amidate and ureate ligands described in the above sections. R6

N

OH R3

R6

H N

O R3

Figure 1.11 The tautomeric equilibrium between 2-hydroxypyridine (2-pyridinol) and 2-pyridone.

22

2-Pyridonates have been reported as ancillary ligands in a wide variety of monometallic and polymetallic structures with diverse coordination modes102,154 and are most commonly found as bridging ligands in late transition metal systems. Monometallic complexes supported by κ1-O and N-bound pyridonates often follow the trend of softer metal centres preferring the N-bound hydroxypyridine, while the hard metals more commonly display the O-bound κ1-motif. Until recently, the use of this ligand motif with early transition metals has been scarce.155-157 Preliminary examples early transition metal systems include mixed pyridonate/cyclopentadienyl complexes that display both chelating and monodentate bonding, with the hard oxophilic early transition metals bound preferentially to the oxygen donor. Early transition metal complexes of group 4 have been investigation by the Schafer group for application in a variety of catalytic transformations. These include titanium alkoxide complexes supported by 3-substituted and 6-substituted pyridonates, which are discrete catalysts for the random copolymerization of rac-lactide and ε-caprolactone.115 Bis(pyridonate) zirconium complex 5, bearing sterically demanding pyridonate ligands, can be synthesized via protonolysis (Figure 1.12), and is an active precatalyst for the intramolecular hydroamination of aminoalkenes to generate pyrrolidine and piperidine products.158 Substrates that are more challenging for hydroamination, such as those without gem-disubstituents, can undergo both hydroamination and hydroaminoalkylation that results in undesirable product mixtures.159

23

R1 R1 H 2N

R1 ( )n

R2

R1

10 mol% 5 110 - 145 °C toluene

( )n

R2

N H hydroamination

Ph

( )n R1 R1

R2 NH2

hydroaminoalkylation

O Zr(NMe2)2 N 2

5

tBu

Figure 1.12 Bis(pyridonate) bis(dimethylamido) zirconium complex 5 that promotes both hydroamination and hydroaminoalkylation.

The C–C bond formation via hydroaminoalkylation has been proposed to occur via a zirconaziridine (Scheme 1.8), analogous to the tantalaziridine of the group 5 metal systems. However, extensive mechanistic studies of this system have not yet been reported and are the focus of Chapter 3. Since the report of this group 4 metal system, a variety of titanium systems have been reported,160-163 including a system supported by related aminopyridonate ligands (Figure 1.4),78 and their reactivity is discussed in Section 3.1.

24

2 [Zr]

NMe2 NMe2

NH2

2

4 HNMe2 [Zr]

N

( )4

N

( )4

Hydroamination

NH2 [Zr] [Zr]

2

NH2

[Zr]

[Zr]

N [Zr]

[Zr] N H

( )4

N N H

( )4

Scheme 1.8 Proposed mechanism for hydroaminoalkylation promoted by complex 5 via a bridging zirconaziridine. [Zr] = bis(pyridonate).

1.2.5

Sulfonamidate complexes for hydroamination Sulfonamidates, also termed sulfonamide and sulfonamido ligands, have a rich history as

ancillary ligands in titanium complexes for application in synthesis.103 The majority of these ligand motifs are the bis(tosyl) ligands based on the a chiral diamine backbone (Figure 1.13). These tethered ligands are chiral and, therefore, have the potential to afford enantioenriched products through catalytic asymmetric transformations. Indeed, these ligands have been extensively studied for the titanium-catalyzed asymmetric addition of dialkylzinc reagents to aldehydes.164-167 25

The initial studies used an in situ generated chiral catalyst and proposed an active bis(sulfonamidate) titanium species. Extensive solution phase and mechanistic investigations have since been performed, as well as extensive structural studies on the bonding of these bis(sulfonamidate) systems with titanium;103,168,169 common coordination motifs observed during these studies include κ3- and κ4-bound complexes (Figure 1.13). The addition of the sulfur into the backbone does significantly alter the coordination geometries of these complexes compared with the amidate, ureate, and pyridonate complexes discussed in the above sections. The most evident difference is the preferential binding to nitrogen in these systems; in almost all of these cases the Ti–N bond distance is shorter than that of the Ti–O.169 O Ph

N

R1 S O

R2 N

N O

N

O

S R1

O

O

O

O

S Ti(NMe2)2

Ti(NMe2)2 Ph

O S

Ti(NMe2)2 N 2

O

S R2

6

R1 = 4-tert-buyllphenyl

R2 = 4-methylphenyl

Gagné, 1998

Walsh, 1999

Nagashima

Mes

O S N

O

N O

O

O

Ta(NMe2)3

(NMe2)3Ti

O S Mes

Zi

S N Ph

Ti(NMe2)3 N Ph

Doye

Figure 1.13 Representative early transition metal systems containing sulfonamidate and sulfamide ligands. Mes = mesityl.

Bergman and co-workers investigated the application of complex 6 for the hydroamination of alkynes and allenes.170,171 These complexes are significantly more reactive 26

and regioselective compared with Ti(NMe2)4 or Cp2TiMe2 precatalysts, which can be attributed to the increased electron withdrawing properties of the sulfonamidate ligand. The sulfonamidate titanium complexes are proposed to proceed via the [2+2] cycloaddition mechanism shown in Scheme 1.3.170 Recently, the related sulfamide ligands have been used as ancillary ligands in a bridging dimeric titanium precatalyst for hydroaminoalkylation by Doye and co-workers.172 These preliminary studies demonstrate the potential of this ligand motif; the precatalyst is able to perform the first group 4 catalyzed α-alkylation of a dialkylamine substrate. There has been one report by Zi and co-workers that describes axially chiral bis(sulfonamidate) tantalum and niobium complexes for application as precatalysts for hydroamination and hydroaminoalkylation (Figure 1.13).143 Unfortunately, these complexes did not show any reactivity for either of these applications. This overview of the coordination behaviour of (N,O)-chelating ligands reveals their successful application for the generation of monomeric, well-defined organometallic precatalysts. The complexes generated have interesting coordination chemistry that follows trends based on the steric and electronic properties of the ligands, a quality that is very attractive for catalytic application. Though these (N,O)-ligands appear to be very similar, their applications are especially broad, and small changes to the ligand motif (eg. amidate vs. ureate vs. pyridonate) result in vast differences in selectivity and reactivity. Their chemistry and potential is by no means exhausted and early transition metals (N,O)-chelated complexes are deserving of extensive future research.

27

1.3

Scope of thesis The research presented within this thesis focuses on the structure and reactivity of early

transition metals of group 4 and 5 complexes as precatalysts for hydroamination and hydroaminoalkylation. Though the amidate, ureate, pyridonate, and sulfonamidate ligands are all related as modular classes of (N,O)-chelating ligands, their varied electronic and steric properties result in different reactivity patterns. The precedent for the successful application of (N,O)ligated early transition metal complexes has been established in the preceding sections, however, a myriad of studies investigating the synthesis, structure, and reactivity of these complexes is required to realize their full potential as precatalysts for the catalytic synthesis of amines. As outlined in Section 1.2.3, bis(ureate) zirconium precatalysts have shown impressive application in the synthesis of amines via hydroamination. The tethered bis(ureate) system shows vastly expanded substrate scope compared with its amidate counterparts. Chapter 2 focuses on the synthesis of a new chiral ureate ligand to provide insight into ligand design trends and highlight key principles to guide future catalyst development efforts. The asymmetric synthesis of α-chiral amines with high enantioselectivity is particularly attractive. Therefore, the ability of the C1-symmetric ureate ligand to support chiral zirconium precatalysts for the asymmetric hydroamination is explored. A related class of group 4 precatalysts supported by pyridonate ligands have been examined as both hydroamination and hydroaminoalkylation precatalysts.158,159 The mechanism for this precatalyst has not been extensively examined, though preliminary investigations have indicated that a bimetallic species might be the catalytically active species. In Chapter 3, investigations probe the solution phase behaviour of the precatalyst and the stoichiometric reaction of complex 5 with aryl and alkyl amines. Kinetic investigations have been performed to 28

provide insight into the mechanism that will aid in the development of future generations of catalysts for this reaction. Tantalum complexes supported by amidate ligands successfully promote the intermolecular α-alkylation of secondary amine substrates; however, high reaction temperatures, long reaction times, and substrate scope limitations restrict the broad application of this methodology. Chapter 4 details extensive ligand development efforts, including new amidate ligands with tethered neutral donors and altered steric and electronic parameters. The Schafer group and others have utilized electron-withdrawing ligands to promote improved reactivity and, therefore,

the

synthesis

and

reactivity

of

mixed

amidate

chloro

tantalum

and

mono(sulfonamidate) tantalum complexes were examined. Reliable synthetic routes to generate these new tantalum complexes, along with extensive characterization and catalytic screening data provide a basis for on-going catalyst development. In their 2009 communication, Schafer and co-workers reported the first example of the αalkylation of an N-heterocyclic substrate using a mono(amidate) tetrakis(dimethylamido) tantalum precatalyst.118 Due to the ubiquitous nature of these functionalized N-heterocycles in the agrochemical, fine chemical, and pharmaceutical industries, as well as the existing limitations of current synthetic methods, a broad substrate scope investigation into the αalkylation of this class of substrates is described in Chapter 5. A potential rational for the lack of reactivity observed with five-membered pyrrolidine substrates is presented based upon both in silico as well as deuterium labeling studies. Finally, a summary of the research is presented in Chapter 6 and potential avenues of future research are highlighted. These include new ligand systems to address substrate scope limitations of the mono(amidate) tantalum amido complex. New chiral urea proligands are also 29

proposed to generate enantioselective tantalum complexes based on simple mono(ureate proligands) for application in asymmetric hydroaminoalkylation. The work presented herein constitutes a broad investigation into the synthesis, structure, and reactivity of early transition metal complexes featuring (N,O)-ancillary ligands. The development of reliable synthetic routes, the fundamental understanding of the structure and bonding of these complexes, and their associated reactivity profiles is essential for the development of highly reactive and selective precatalysts for hydroamination and hydroaminoalkylation.

30

CHAPTER 2: C1-symmetric ureate complexes of zirconium for the asymmetric hydroamination of unactivated aminoalkenes

2.1 2.1.1

Introduction Asymmetric hydroamination of unactivated olefins Asymmetric inter- and intramolecular hydroamination of alkene substrates can generate

chiral amine products in a waste-free, highly atom-economical manner (Scheme 2.1). Therefore, an important focus of recent research is the development of catalyst systems supported by chiral ancillary ligands to promote enantioselective hydroamination.173-178 A particularly desirable, yet challenging transformation is the asymmetric hydroamination of simple unactivated olefinic and amine substrates.

R1

R1

R1 R2

H N

N R2

catalyst

∗

R3

H

NR2R3 H

∗ R1

R1 catalyst

H

∗

∗

R2 N

R

Scheme 2.1 Hydroamination of alkenes using a chiral catalyst to generate enantioenriched alkylamines.

Late-transition metal systems of iridium,179,180 palladium,181-184 gold,185-188 rhodium,189 and copper190 are almost exclusively applied for the intermolecular hydroamination of amines and alkenes. These catalyst systems often take advantage of C2-symmetric axially chiral phosphine ligands (Figure 2.1) to impart the desired enantioselectivity.180,189 The main disadvantage of these precatalysts is their requirement of activated olefinic substrates such as strained alkenes, dienes, styrenes, or allenes along with amines of reduced basicity such as 31

anilines, carboxamides, and sulfonamides. Alkali metal based systems are less populous in the literature; however, bis(amido) dinaphthyl lithium systems have been used with limited success for the intramolecular cyclization of aminopentenes.191-193 Alkaline earth metal systems have been more extensively studied but are often plagued with facile ligand exchange (Schlenk equilibrium) or ligand cleavage upon substrate addition.194-197 These damaging side-reactions result in low enantioselectivities and until recently the most selective catalysts, chiral pseudo C3symmetric monoanionic bulky tris(oxazolinyl)borato magnesium and calcium complexes, could only achieve ee’s of up to 36%.198 A recent innovation in the field by Hultzsch and co-workers describes chiral phenoxyamine magnesium catalysts, which, by circumventing ligand redistribution reactions, can generate pyrrolidine products with ee’s of 51 – 93% (Figure 2.1).199 Metal-free catalytic systems have also been examined, such as binaphthol-derived dithiophosphoric acid compounds as chiral Brønsted acids to generate vinyl-pyrrolidines,200 and (R)-glyceraldehyde for the intermolecular Cope-type hydroamination of allylic amines.201 Although metal and non-metal systems from across the periodic table have been developed, the most successful systems for the intramolecular hydroamination of simple, unactivated aminoalkenes are those based on rare-earth or group 4 metals.173,174 tBu

O O O

NMe2 O Mg Ph SiPh3

OCHPh2

PR2 PR2

PCy2 tBu

O Iridium

N

Rhodium

N NMe2 O Mg Ph SiPh3

Ar S O P O SH Ar Ar = 10-(2,4,6-trimethylphenyl)9-anthracenyl

d.r. = 9:1 at 25 °C

Figure 2.1 Representative late transition metal, alkaline metal, and Brønsted acid based catalytic systems.

32

2.1.2

Rare-earth metal systems Rare-earth metal complexes are the most extensively studied systems for the

intramolecular asymmetric hydroamination of unactivated olefins.123,174,175,177,202 The first report of asymmetric hydroamination describes the use of C1-symmetric ansa-lanthanocene complexes containing a pendant chiral group (eg. R*= (–)-menthyl, (+)-neomenthyl) on one of the cyclopentadienyl groups (Figure 2.2).203 However, these catalysts are plagued by facile catalyst epimerisation and the focus has since shifted to non-cyclopentadienyl based systems. SiPh3 Me2N Si

Ln X(SiMe3)2 X = N, CH ∗ Ln = La, Nd, Sm R Lu, Y

Marks up to 74% ee

O O

Ph B

Ln

O ON

Me2N SiPh3

Y

N

Ph

Hultzsch Sc; up to 95% ee

CH2SiMe3

tBu tBu

Sadow 89 − 96% ee

Figure 2.2 Chiral lanthanide precatalysts for asymmetric hydroamination.

Significant contributions from the research groups of Scott,204 Livinghouse,205 Marks,206 Schulz,207 and Hultzsch208,209 have described the use of C2-symmetric ligands, such as biphenolate, binaphtholate, biarylamido, and bis(oxazolinato) compounds, for the generation of chiral lanthanide precatalysts for intramolecular hydroamination. The binaphtholate catalysts of yttrium, scandium, and lutetium developed by Hultzsch (Figure 2.2) are particularly noteworthy as these remain the only rare-earth catalysts that are also active for the intermolecular asymmetric hydroamination of alkenes. The C2-symmetric catalyst systems have dominated the literature210 until a recent report by Sadow and co-workers that describes the synthesis and

33

reactivity of mixed cyclopentadienyl-bis(oxazolinyl)borate chiral yttrium complexes for the intramolecular hydroamination of primary aminoalkenes (Figure 2.2).211 These chiral lanthanide complexes are attractive systems due to their high catalytic activities without requiring protected or activated substrates. However, these precatalysts are restricted by very low functional group tolerance and extreme sensitivity to air and moisture. The focus has, therefore, shifted to group 4 metal systems as these complexes have low toxicity, low cost, and improved stability and functional group tolerance over lanthanide complexes while exhibiting excellent reactivity.

2.1.3

Group 4 metal systems Initial catalyst development with group 4 systems focused on the use of cationic

complexes,45,46 as these complexes are isoelectronic with the rare-earth systems.107,116 The first asymmetric hydroamination by a chiral group 4 metal catalyst, reported by Scott and co-workers, is an aminophenolate complex based on the axially chiral biaryl backbone (Figure 2.3).107 The cationic zirconium complex readily catalyzes the cyclization of secondary aminoalkenes to generate heterocyclic products in 14 – 82% ee. Neutral group 4 complexes are also active and selective catalysts for the intramolecular hydroamination of unactivated alkenes. Bergman and co-workers have developed chiral zirconium complexes, prepared in situ from the diphosphinic amide proligands and Zr(NMe2)4, for the formation of pyrrolidines and piperidines in up to 80% ee.132 Though these catalysts require heating (85 – 135 ºC), high catalyst loadings (20 mol%) and are limited by ligand redistribution into inactive homoleptic species, they demonstrate that stable, easier to handle, neutral zirconium complexes are effective catalysts for this transformation. Indeed, in a 2007 report, the Schafer group described the use of neutral biaryl bis(amidate) 34

zirconium complexes for this transformation.141,142 These complexes can be easily synthesized via a simple protonolysis reaction of the bis(amide) proligand with Zr(NMe2)4. The most efficient of the complexes screened is proficient at the cyclization of aminoalkenes with enantiomeric excesses up to 93% (Figure 2.3). The bis(amidate) complex is more reactive than the Bergman system requiring shorter reaction times and lower catalyst loadings (10 mol%). Since these reports, numerous neutral group 4 metal systems with chiral biaryl-based ligands have been reported,133,212 though these complexes also require high reaction temperatures and do not demonstrate significantly improved selectivity or reactivity. A recent report by Sadow and co-workers describes the notable application of a C1-symmetric mixed cyclopentadienyl bis(oxazolinyl)borate ligand (Figure 2.3), which catalyzes the hydroamination of primary aminoalkenes generating five-, six-, and seven-membered N-heterocyclic amines at room temperature with ee’s of up to 99%.213,214 tBu tBu

N

B(C6F5)4

O

N Zr

O Ph tBu

tBu C2-symmetric

Scott

Ar

R R P N O

O N

Zr(NMe2)2 O P R R

Zr(NMe2)2HNMe2

N

N

O

Ph B

N O ON

NMe2 Zr NMe 2

Ar

R = 3,5-dimethylphenyl

Ar = 2,4,6-trimethylphenyl

C2-symmetric

C2-symmetric

C1-symmetric

Bergman

Schafer

Sadow

Figure 2.3 Cationic and neutral zirconium systems for the asymmetric intramolecular hydroamination of aminoalkenes.

Untethered amidate ligands synthesized from enantiomerically pure chiral ketones, such as (–)-menthone, have been investigated for the intermolecular hydroamination of aminoalkenes 35

(Figure 2.4).215,216 The complex that is formed in situ using two equivalents of chiral acyclic amide 7, and Zr(NMe2)4 catalyzes the formation of pyrrolidine products in ee’s of up to 26%. The low selectivity of the complexes studied has been attributed to potential ligand redistribution reactions in solution resulting in achiral catalytic species. Ph Ph

O ∗

NH

N H 7

>98% yield, 26% ee 5 h, 110 °C

Figure 2.4 Untethered chiral amidate ligands for intramolecular asymmetric hydroamination.

Overall, group 4 catalytic systems show potential in mediating the asymmetric intramolecular hydroamination of aminoalkenes. However, apart from the system developed by Sadow and co-workers, excellent enantioselectivities (>90%) have only been reported for a select number of catalyst-substrate combinations, and have been restricted to catalyst systems supported by axially chiral tethered biaryl ligands. These systems also suffer from low tolerance to polar functional groups leading to restricted substrate scope. While these contributions illustrate the potential of group 4 catalytic systems, they are only a first step toward realizing the goal of a broadly applicable asymmetric alkene hydroamination catalyst.

2.1.4

Expansion of substrate scope using a tethered bis(ureate) zirconium catalyst A recent report from the Schafer group describes the reactivity of a tethered urea

proligand supporting an easily synthesized zirconium complex (4, Figure 2.5) that induces a promising scope of reactivity for both inter- and intramolecular hydroamination.100 This ureate precatalyst is arguably the most generally useful group 4 metal complex reported to date, 36

promoting reactions with both primary and secondary amine substrates, exhibiting vastly expanded substrate scope and tolerance of polar functional groups (eg. pyrrole, acid-sensitive protected catechol) as well as excellent chemoselectivity for hydroamination. iPr

2N

O NMe2 N N iPr

2N

Zr

NMe2

4 O HNMe2

Figure 2.5 Tethered bis(ureate) zirconium hydroamination precatalyst 4.

The chemoselectivity is of particular interest, as other group 4 systems can be plagued with unwanted hydroaminoalkylation side-reactions to give amine substituted carbocycles rather than heterocyclic products.159,161,162,172,217 The development of a chiral ligand system that imparts both the high activity and selectivity of 4 and instills excellent enantioselectivity is of the utmost importance for the efficient synthesis of chiral amines. Thus, the investigation of a new chiral urea proligand would provide invaluable insight regarding the role of the urea functionality and the tether in the reactivity accessed by such ureate zirconium complexes and potentially mediate asymmetric hydroamination.

2.1.5

Scope of chapter This chapter focuses on the reactivity and selectivity mediated by a new chiral ureate

ligand for the asymmetric hydroamination of aminoalkenes. The modular synthesis of the chiral proligand from inexpensive enantiopure amino acids is presented, and its ability to support complexes that mediate enantioselective intramolecular hydroamination is explored. This study 37

compares and contrasts the reactivity and selectivity of amidate versus ureate, cyclic versus acyclic, and tethered versus untethered ligands, while exploring the use of an untethered chiral ureate ligand for affecting enantioselectivity. Through the characterization of the resultant coordination complex in this system, critical features required for improved reactivity and selectivity are identified and discussed.

2.2 2.2.1

Results and discussion Ligand and complex Cyclic ureas, synthesized from the chiral pool of natural amino acid starting materials,

are an attractive modular class of proligands for the generation of group 4 asymmetric hydroamination precatalysts. While cyclic ureas have been investigated for their medicinal properties,218-220 examined for potential application as Lewis basic organocatalysts,221,222 used as chiral auxiliaries for asymmetric syntheses,223,224 and utilized as substrates for intermolecular hydroamination,225 these compounds represent a new proligand motif for coordination of early transition metals. The coordination of related chiral 2-aminopyrrolines to titanium has been studied226 and zirconium alkyl and amido complexes supported by related achiral imidazolone ligands have been shown to be effective catalysts for intramolecular hydroamination.227 Proligand 8 is synthesized as white crystalline needles in five steps from L-valine (Scheme 1). This modular synthetic route is attractive, as a wide variety of proligands can be generated from inexpensive enantiopure starting materials. A solid-state molecular structure has been obtained of the crystalline product and confirms the connectivity of proligand 8 (Figure 2.6).

38

O

O

O Boc2O, NEt3 1:1 dioxane:H2O

HO NH2

CyNH2, HOBt, DCC DIPEA, DCM

HO

Cy N H

NHBoc 74% yield

NHBoc 94% yield

TFA, DCM O N NH 8

O

Triphosgene DCM

Cy N H

84 %yield 46% yield overall

NH2

88% yield

LiAlH4 THF, ∆

Cy N H

NH2

90% yield

Scheme 2.2 Modular route for the synthesis of urea proligand 8. Boc = tert-butoxycarbonyl, HOBt = hydroxybenzotriazole, DCC = N,N'-dicyclohexylcarbodiimide, DIPEA = diisopropylethylamine, TFA = trifluoroacetic acid.

N3 O2

C13 N4 H99 O1

H98 N2 C1 N1

Figure 2.6 ORTEP depiction of the solid-state molecular structure of proligand 8. The ellipsoids are plotted at 50% probability and the majority of the hydrogen atoms are omitted for clarity. The hydrogens displayed (N–H) have been located from unassigned electron density and their positions refined.

The cyclic urea crystalizes in the chiral P1 space group and the asymmetric unit includes two urea molecules each displaying one of the two possible chair conformations of the Ncyclohexyl moiety. The urea proligand displays intermolecular hydrogen bonding between the N–H and the carbonyl oxygen (O1–H99, 1.96(2) Å; O2–H98, 2.09(2) Å, Table 2.1) of the two 39

urea molecules. The presence of this interaction is further supported by the short donor-acceptor distance (O1···N4, 2.874(2) Å; O2···N2, 2.883(2) Å) and the acceptor–H–donor angles that are close to 180° (O1–H99–N1, 177.1(19)°; O2–H98–N1, 170.6(18)°). Each atom of the urea fragment is sp2-hybridized consistent with signification delocalization of the nitrogen lone pairs and indicated by the planar nature of the NC(O)N fragment.

Table 2.1 Relevant bond lengths (Å) and angles (°) for 8.

N1–C1 N3–C13 O1–C1 O1–H99 O1···N4

1.371(2) 1.364(2) 1.233(2) 1.96(2) 2.874(2)

N2–C1 N3–C13 O2–C13 O2–H98 O2···N2

1.356(2) 1.358(2) 1.241(2) 2.09(2) 2.883(2)

O1–H99–N1 ∑ N1 ∑ C1

177.1(19) 355.8(2) 360.0(2)

O2–H98–N1 ∑ N2

170.6(18) 356.7(2)

Synthesis of the targeted bis(ureate) bis(amido) zirconium complex 9 via protonolysis of Zr(NMe2)4 with 8 results in crude material that has elemental analysis data consistent with the proposed complex (Scheme 2.3). However, the complicated 1H and

13

C NMR spectra for this

material suggest that multiple species are present in the solution phase. No free ligand remains, as is evident by loss of the diagnostic amide N–H signal at δ 7.18 ppm in the 1H NMR spectrum (Appendix B). iPr

iPr

2 Cy

NH N 8

O

Zr(NMe2)4

-n HNMe2 hexanes

N

Mixture of products, 9

N Cy

Zr(NMe2)2 O

2

Target

Scheme 2.3 Precatalyst formation by the protonolysis reaction of Zr(NMe2)4 with 8.

40

Unfortunately, high yielding crystalline material could not be obtained from the product mixture 9. Although rigorous characterization of the resultant coordination complexes could not be achieved, the in situ prepared complexes are suitable for screening for hydroamination reactivity.

2.2.2

Intramolecular hydroamination The ability of 9 to mediate enantioselective hydroamination has been investigated for the

cyclization of 2,2-diphenyl-1-amino-4-pentene, a standard test substrate202 (Table 2.2). The use of cyclic ureate 8 in isolated 9 (entry 1) and the in situ procedure (entry 2) generates the heterocyclic product in good yield, albeit with low enantioselectivity. One of the exceptional elements of the tethered urea zirconium catalyst 4 (Figure 2.5) is the broadened substrate scope it displays compared with other existing group 4 catalyst systems.100 In order to discern if this improved scope of reactivity can be attributed to the urea functionality or to the ligand tether, reactivity with a broader range of substrates was investigated using in situ prepared 9. The substrate scope using complexes prepared with proligand 8 is very limited; reaction progresses in good yield to generate five- and six-membered heterocycles (entries 2-6). However, no conversion to product is observed with secondary amines, internal olefins, or substrates without gem-disubstituents.

41

Table 2.2 Reactivity studies using proligand 8 for intramolecular hydroamination with alkenes. R2 R2 R2 NH2

( )n

Entry

Substrate

20 mol% 8 10 mol% Zr(NMe2)4 d8-toluene

Product

∗

(

)n

R2

N H

Conditions

Yield, ee (%)c

12 h, 100 ºC

92, 11

4 h, 100 ºC

>98b, 12

8 h, 100 ºC

89, 95b, nd

Ph

1

2

3

5

6

Ph

Ph Ph NH2

N H Ph Ph

Ph Ph NH2

N H

Ph Ph

Ph Ph NH2

NH2

NH2

N H

N H

N H

a

10 mol% in zirconium of crude product 9, reaction time not optimized bNMR yield

determined by 1H NMR using 1,3,5-trimethoxybenzene as internal standard due to product volatility cee determined by 1H NMR spectroscopy of derivitized product with (+)-(S)-αmethoxy-α-trifluoromethylphenylacetyl chloride dDerivitized with tosyl chloride to aid in isolation.

This limited reactivity profile is reminiscent of substrate scope limitations previously reported when Zr(NMe2)4 is used as a precatalyst for the intramolecular hydroamination of aminoalkenes.228,229 Ligand redistribution could be occurring under catalytic conditions to generate Zr(NMe2)4 in solution which could also act as a catalyst for the transformation resulting in the substrate scope limitations and poor enantioselectivities observed. Ligand redistribution 42

and disproportionation reactions have been observed previously with group 4 amidate systems230 and has been postulated to occur during the attempted synthesis of zirconium alkyl complexes supported by acyclic untethered ureate ligands.150 The enantiomeric excesses achieved (up to 12%) are lower than those of related amidate chiral acyclic ligands based on (–)-menthone such as 7 (up to 26%);215,216 this is consistent with increased ligand redistribution due to minimal steric bulk as well as the fact that the stereogenic centre is well removed from the metal centre of these cyclic ligands (Figure 2.7). O

O

R

Ph [Zr]

N N

vs.

iPr

Ph Ph

[Zr] ∗

N iPr

N H Proligand: 8 12% ee 7 26% ee

Figure 2.7 Proximity to the zirconium center of steric bulk and source of chirality in the cyclic ureate compared with the amidate ligand.

2.2.3

Isolated bimetallic complex The only crystalline material that was obtained from a solution of 9 were isolated from a

saturated pentane solution of the crude material that was cooled to -30 °C for several weeks. The solid state-molecular structure reveals a bimetallic species 10 with an unexpected ligand stoichiometry, containing three ligands for two metal centers (Figure 2.8). This suggests that ligand redistribution is indeed a complicating factor under catalytic conditions, and provides a rationale for both the poor ee’s and limited substrate scope observed.

43

N5 C25

N3

N6 O3

N4

N9

C13 N8

Zr1

Zr2 O2 O1

N7

N10 N11

N2 C1 N1

Figure 2.8 ORTEP depiction of the solid-state molecular structure of complex 10. The ellipsoids are plotted at 50% probability and the hydrogen atoms are omitted for clarity. In the simplified structure (right) the methyl groups of dimethylamido ligands (N7-11) and the cyclohexyl groups of the ureate ligands ((N1, N3, N5) are removed for clarity.

The isolated bimetallic complex 10 has three bridging ureate ligands, suggesting that the steric bulk of the ligand is insufficient to support monomeric species of these highly electrophilic metal centers. This compound is C1-symmetric with pseudo-octahedral geometry about each zirconium atom. Of the three bridging ureate ligands, two are bound in a simple bridging mode, and one (O1, N2) is also bound in a κ2-fashion to Zr1 while bridging Zr1 and Zr2. Selected bond lengths and angles for complex 10 are summarized in Table 2.3. The C–N2,4,6 (1.317 – 1.325 Å) and C–O (1.271 – 1.290 Å) bond lengths of the metal-bound ureate fragments are consistent with moderate delocalization within the ureate backbone. The respective C–N bonds to the tertiary amine (N1,3,5; 1.361 – 1.368 Å) are slightly longer than those of the bonding moiety, however these, and the sum of 353º for the subtending angles, are indicative of sp2 hybridization and 44

limited lone-pair donation to the π-system. Interestingly, the Zr–N bonds to the ureate ligands (2.2372(9) and 2.2459(9) Å) are slightly shorter than those of the tethered ureate complex (2.279(1) Å),100 the exception being the Zr–N6 bond which is presumably lengthened due to the presence of the strongly donating trans dimethylamido ligand (N11). Comparable bis(amidate) bis(amido) Zr–N(amidate) bonds are significantly longer (~2.32 Å).142,231 The Zr–O bond lengths in 10, excluding those related to O1 which is bridging two metal centers, are significantly shorter (by >0.1 Å), than those of related monomeric bis(amidate) bis(amido)231,232 and bis(ureate) bis(amido)100 zirconium complexes. Zr–NMe2 (2.035 – 2.130 Å) bond lengths in 10 are slightly longer on average than Zr-N double bonds reported in the literature (98%. 1H NMR (d8toluene, 600 MHz) δ 1.26-1.33 (m, 2H, CH2), 1.45 (dt, 3JH,H = 11.1, 5.5 Hz, 4H, CH2), 2.12-2.17 (m, 6H, CH2), 2.33 (s, 6H, CCH3), 2.65 (t, 3JH,H = 7.0 Hz, 2H, NCH2) 3.40 (br s, 24H, N(CH3)2), 6.94-6.97 (m, 1H), 7.04-7.06 (m, 2H);

13

C NMR (C6D6, 150 MHz) δ 18.2, 24.9, 26.3 (CH3),

137

30.5, 47.0 (br, N(CH3)2), 54.7 (NCH2), 55.0 (NCH2), 124.6, 128.3, 132.5 (CCH3), 143.5 (NCarom), 174.4 (C=O).

Mono(N-(2,6-diisopropylphenyl)-3(dimethylamino)propanamidate)

N

O Ta(NMe2)4

tetrakis(dimethylamido)

N iPr

iPr

tantalum: In situ screening procedure was used to evaluate catalytic activity as crystalline material could not be obtained. Proligand 26 (6.91 mg, 0.025 mmol) and Ta(NMe2)5 (10.0 mg, 0.025 mmol) were weighed into a small vial and d8-toluene was added (0.5 g). The solution was transferred to a J. Young tube and NMR spectroscopy is consistent with formation of the mono(amidate) tetrakis(dimethylamido) tantalum complex. Yield: >98%. 1H NMR (d8-toluene, 600 MHz) δ 1.26 (d, 3JH,H = 6.9 Hz, 6H, CH(CH3)2), 1.28 (d, 3JH,H = 6.8 Hz, 6H, CH(CH3)2), 1.95 (s, 6H, CH2N(CH3)2), 2.20-2.25 (m, 2H, CH2), 2.72-2.74 (m, 2H, CH2), 3.31 (br s, 24H, N(CH3)2), 3.40-3.47 (m, 2H, CH2), 7.05-7.11 (m, 3H, CHarom); 13C NMR (C6D6, 150 MHz) δ 25.1, 25.2, 27.0, 31.7, 45.3, 47.0 (br, N(CH3)2), 55.6, 123.9, 125.3, 140.2, 143.2, 174.3 (C=O).

Dichloro tris(dimethylamido) tantalum, pyridine adduct, 34: Synthesized following adapted literature procedure.290 1H NMR (d6-benzene, 600 MHz) δ

TaCl2(NMe2)3⋅ py 34

3.64 (br s,18H, N(CH3)2), 6.50 (t, 3JH,H = 6.7 Hz, 2H, CHpyr), 6.80 (t, 3JH,H = 6.8 Hz, 1H CHpyr), 8.78 (br s, 2H, CHpyr),;

13

C NMR (d6-benzene, 150 MHz) δ 49.3 (N(CH3)2), 124.4 (CH), 137.9

(CH), 151.7 (CH); MS (EI): m/z = 383 ([M-pyridine]+), 348 ([M-pyridine-Cl]+), 339 ([Mpyridine-NMe2]+); Anal. calcd. for C11H23Cl2N4Ta: C, 28.52; H, 5.01; N, 12.10; Found C, 28.13; H, 4.85; N, 11.75. 138

Bis(N-(2,6-diisopropylphenyl)pivalamidate)

dichloro

Cl

tBu

mono(dimethylamido)

tantalum,

38:

Ta(NMe2)5

(0.500

iPr

g,0.651) was weighed into a large vial equipped with a stir bar

O Ta

N

tBu

O

iPr

N

Cl iPr

and dissolved in benzene (5 mL). A suspension of N-(2,6-

iPr NMe2 38

diisopropylphenyl)pivalamide (0.340 g, 1.302 mmol) in benzene (2 mL) was added to the stirring solution at room temperature. The solution was left to stir for 1.5 hour and a colour change of yellow to orange-red was observed. The volatiles were removed under high vacuum. Recrystallization from hot toluene gave the title compound as a red crystalline solid. 1H NMR (d6-benzene, 600 MHz) δ 1.06 (s, 18H, C(CH3)3), 1.21 (d, 3JH,H = 6.9 Hz, 6H, CH(CH3)2), 1.26 (d, 3JH,H = 6.6 Hz, 6H, CH(CH3)2), 1.36 (d, 3JH,H = 6.9 Hz, 6H, CH(CH3)2), 1.47 (d, 3JH,H = 6.5 Hz, 6H, CH(CH3)2), 3.17 (dt, 3JH,H = 13.6, 6.9 Hz, 2H, CH(CH3)2), 4.12 (br s, 3H, N(CH3)2), 4.25 (br s, 3H, N(CH3)2), 4.42 (dt, 3JH,H = 13.4, 6.7 Hz, 2H, CH(CH3)2), 7.01-7.07 (m, 4H, CHarom), 7.10 (3JH,H = 6.4, 2.9 Hz, 2H, CHarom); 13C NMR (d6-benzene, 150 MHz) δ 23.6, 24.6, 27.1, 27.6, 28.0, 28.2, 28.6, 42.5 (C), 50.1 (N(CH3)2), 54.9 (N(CH3)2), 123.7 (CH), 124.9 (CH), 127.9 (CH), 138.1 (C), 143.9 (C=O), 146.2 (C=O). EI MS (m/z): 816 ([M]+). Characterized by X-ray crystallography (Appendix A).

(N-(2,6-diisopropylphenyl)pivalamido) tris(dimethylamino) tantalum chloride,

42:

N-(2,6-Diisopropylphenyl)pivalamide

(53.1 mg,

0.203 mmol) was dissolved in 2 mL of benzene and NaHMDS (37.2 mg,

O tBu

TaCl(NMe2)3 N

iPr

iPr

42

0.203 mmol) was added in one portion. After stirring for 1 hour the volatiles were removed under high vacuum. Pentane (2 mL) was added, the resultant suspension 139

was stirred for 5 minutes, and then the volatiles were removed under high vacuum. The proligand salt was then dissolved in benzene and added dropwise over 5 minutes to a toluene suspension of [TaCl2(NMe2)3]2 (78.0 mg, 0.102 mmol). The mixture was allowed to stir over night (18 h) to give a clear yellow solution. All volatiles were removed under high vacuum. The complex was dissolved in hexanes and filtered through a Celite™ plug to remove the sodium chloride by-product. Removal of the volatiles under high vacuum and recrystallization from hot hexanes or hot pentane at 25 °C, then -30 °C gave the title compound as a yellow crystalline solid. Yield: 0.124 g, 75%. 1H NMR (d6-benzene, 400 MHz) δ 1.11 (s, 9H, C(CH3)3), 1.22 (br s, 3H, CH(CH3)2), 1.23 (br s, 3H, CH(CH3)2), 1.36 (br s, 3H, CH(CH3)2), 1.43 (br s, 3H, CH(CH3)2), 3.22 (br s, 1H, CH(CH3)2), 3.49 (br s, 18H, N(CH3)2), 4.25 (br s, 1H, CH(CH3)2), 7.04-7.10 (m, 3H, CHarom.);

C NMR (d6-benzene, 100 MHz) δ 23.7, 24.9, 25.7, 27.2 (br s,

13

CH(CH3)2), 27.9, 28.0 (br s, CH(CH3)2), 28.5 (C(CH3)3), 42.3 (C(CH3)3), 48.9 (br s, N(CH3)2), 123.3, 124.6 (br s, CHarom.), 126.4 (CHarom.), 140.4 (Carom.N), 142.5, 144.8 (br s, Carom.CH), 181.2 (C=O). MS (EI): m/z = 564 ([M-NMe2]+), 348 ([M-amidate]+); Anal. Calcd. for C23H44ClN4OTa: C, 45.36; H, 7.28; N, 9.20; Found C, 45.00; H, 7.22; N, 8.80; Characterized by X-ray crystallography (Appendix A).

Monochloro bis(dimethylamido)

bis(N-(2,6-diisopropylphenyl)pivalamidate) tantalum,

40:

Sodium

N-(2,6-

tBu

O N

iPr

TaCl3(NMe2) 2 iPr

40

diisopropylphenyl)pivalamidate (0.1020 g, 0.360 mmol) and TaCl4(NMe2)2H2NMe2 (0.1847 g, 0.360 mmol) were dissolved in 10 mL of benzene. The resulting mixture was stirred for 18 hours at room temperature. The volatiles were removed under high vacuum. The reaction mixture was dissolved in hexanes and filtered through a 140

Celite™ plug to remove the sodium chloride by-product to give the title compound as a yelloworange solid. 1H NMR (d6-benzene, 600 MHz) δ 1.30 (m, 12H, CH(CH3)2), 1.41 (s, 9H, C(CH3)3), 2.33 (s, 6H, N(CH3)2), 3.10 (dt, 3JH,H = 13.8, 6.9 Hz, 2H, CH(CH3)2), 7.06-7.08 (m, 1H, CHarom), 7.16-7.19 (m, 2H, CHarom);

C NMR (d6-benzene, 150 MHz) δ 22.8 (CH(CH3)2),

13

24.3 (CH(CH3)2), 29.1 (CH(CH3)2), 30.8 (C(CH3)3), 40.4 (C(CH3)3), 41.6 (N(CH3)2), 121.7 (CHarom), 122.9 (CHarom), 136.7 (Carom), 147.4 (Carom), 161.3 (C=O); MS (EI): m/z = 330 tentative assignment ([M – amidate]+).

Mono(N-(tert-butyl)-4-methylbenzenesulfonamidate)

O

O S N

tetrakis(dimethylamido) tantalum, 43: Prepared following GP2

Ta(NMe2)4

43

using proligand 43 (0.170 g, 0.748 mmol) and Ta(NMe2)5 (0.300 g, 0.748 mmol). Recrystallization of the crude product from hot pentanes gave the title compound as a yellow solid. Yield: 0.419 g, 96%. 1H NMR (d6-benzene, 400 MHz) δ 1.35 (s, 9H, C(CH3)3), 1.90 (s, 3H, CTsCH3), 3.54 (br s, 24H, N(CH3)2), 6.81 (m, 3JH,H = 8.19 Hz, 2H, CHTs), 7.95 (m, 3

JH,H = 8.19 Hz, 2H, CHTs);

13

C (d6-benzene, 125 MHz) δ 21.5 (CH3), 32.8 (CH3), 48.34 (br,

N(CH3)2), 56.16 (NC), 129.4 (CHTs), 129.6 (CHTs), 142.6 (CTs), 143.3 (CTs); EI MS (m/z): 539 ([M-NMe2]+), 483 ([M-C(CH3)3-NMe2]+), 438 ([M-C(CH3)3-2NMe2]+), 395 ([M-C(CH3)33NMe2]+); Anal. calcd. for C19H40N5O2STa: C, 39.10; H, 6.91; N, 12.00; Found: C, 39.34; H, 6.78; N, 12.23. Characterized by X-ray crystallography (Appendix A).

Mono(4-methyl-N-phenylbenzenesulfonamidate)

O

O S

tetrakis(dimethylamido) tantalum, 44: Prepared following GP2

N

Ta(NMe2)4

44

141

using proligand 44 (0.185 g, 0.748 mmol) and Ta(NMe2)5 (0.300 g, 0.748 mmol). Recrystallization of the crude product from hot pentanes gave the title compound as a yellow solid. Yield: 0.399 g, 88%. 1H NMR (d6-benzene, 400 MHz) δ 1.82 (s, 3H, CTsCH3), 3.45 (s, 24H, N(CH3)2), 6.70 (d, 3JH,H = 8.0 Hz, 2H, CHTs), 6.86 (t, 3JH,H = 7.4 Hz, 1H, CHPh), 7.22 (dd, 3

JH,H = 8.5, 7.5 Hz, 2H, CHPh), 7.54 (m, 3JH,H = 7.7 Hz, 2H, CHTs), 7.99 (d, 3JH,H = 8.2 Hz, 2H,

CHTs),

13

C (d6-benzene, 125 MHz) δ 21.5 (CH3), 47.8 (N(CH3)2), 120.7, 122.2, 128.2, 128.3,

128.6, 129.7, 129.9, 139.9, 143.2 (CTs), 144.4 (CTs); EI MS (m/z): 603 ([M]+), 559 ([M-NMe2]+); Anal. calcd. for C23H40N5O2STa: C, 41.79; H, 6.01; N, 11.60; Found: C, 41.89; H, 6.22; N, 11.80. Characterized by X-ray crystallography (Appendix A).

Mono(N-(2,6-dimethylphenyl)-4-methylbenzenesulfonamidate) O

tetrakis(dimethylamido) tantalum, 45: Prepared following GP2

S O

using proligand 45 (0.135 g, 0.491 mmol) and Ta(NMe2)5 (0.197 g,

N

Ta(NMe2)4

45

0.491 mmol). Recrystallization of the crude product from hot pentanes gave the title compound as a yellow solid. Yield: 0.294 g, 95%. 1H NMR (d6-benzene, 400 MHz) δ 1.88 (s, 3H, CTsCH3), 2.51 (s, 6H, CH3), 3.32 (s, 24H, N(CH3)2), 6.72 (d, 3JH,H = 8.0 Hz, 2H, CHTs), 6.84 - 6.99 (m, 1H, CHPh), 7.06 (d, 3JH,H = 7.3 Hz, 2H, CHPh), 7.79 (d, 3JH,H = 8.2 Hz, 2H, CHTs);

13

C (d6-benzene, 125 MHz) δ 20.1, 21.5, 25.2, 28.7, 46.3, 46.9, 47.3 (br,

N(CH3)2), 124.2, 125.0, 127.6, 129.1, 129.2, 129.3, 137.7, 140.8, 141.8, 142.0; EI MS (m/z): 587 ([M-NMe2]+), 357 ([M-sulfonamidate]+); Anal. Calcd. for C23H40N5O2STa: C, 43.74; H, 6.38; N, 11.09; Found: C, 44.09; H, 6.53; N, 10.81. Characterized by X-ray crystallography (Appendix A).

142

Mono(N-(2,6-diisopropylphenyl)-4methylbenzenesulfonamidate)

tantalum, 46: Prepared following GP2 using proligand 46 (0.170 g, 0.748

mmol)

and

Ta(NMe2)5

O

tetrakis(dimethylamido)

(0.300

g,

0.748

S O iPr

N

Ta(NMe2)4 iPr

46

mmol).

Recrystallization of the crude product from hot pentanes gave the title compound as a yellow solid. Yield: 90.5 mg, 18%. 1H NMR (d6-benzene, 400 MHz) δ 1.27 (d, 3JH,H = 6.8 Hz, 12H, CH(CH3)2), 1.87 (s, 3H, CTsCH3), 3.22 (s, 24H, N(CH3)2), 4.02-4.08 (m, 2H, CH(CH3)2), 6.75 (d, 3

JH,H = 8.0 Hz, 2H, CHTs), 7.11-7.18 (m, 3H, CHPh), 7.83 (d, 3JH,H = 8.2 Hz, 2H, CHTs); 13C (d6-

benzene, 125 MHz) δ 21.4, 25.2, 28.7, 46.9 (br, N(CH3)2), 124.2, 125.0, 127.9, 129.3, 141.4, 146.8. In situ screening procedure was used to evaluate catalytic activity as crystalline material could only be obtained in low yield.

143

CHAPTER 5: Synthesis of α-alkylated N-heterocycles via hydroaminoalkylation

5.1

Introduction

5.1.1

Functionalized N-heterocycles Saturated N-heterocycles are key structural elements found in a wide variety of natural

products,321 agrochemicals,322,323 and pharmaceuticals324-327 (Figure 5.1). Familiar examples of this heterocyclic motif include (S)-nicotine and coniine, naturally occurring alkaloids containing a pyrrolidine and a piperidine core, respectively. Additional alkaloids containing a piperidine ring include (–)-lobeline and (–)-spectaline, which have been used to aid with smoking cessation328 and for analgesic applications.329 Saturated heterocyclic compounds containing either a pyrrolidine or piperidine represent 33 of the top 200 prescription drugs sold in the US in 2010.330 Key examples include the naturally occurring opiates morphine and codeine, Concerta (used to treat Attention Deficit Hyperactivity Disorder), and Vesicare (used to alleviate symptoms of an overactive bladder). H N

N N

N O

(S)-Nicotine RO

O O

Morphine, R = H Codeine, R = Me

OH (−)-Spectaline

O N

O O

N HO

O

OH

H N

H

( )9

(−)-Lobeline

Coniine

H N

Concerta Johnson & Johnson

N

Vesicare Astellas

Figure 5.1 Natural products and pharmaceuticals containing an α-alkyl/aryl N-heterocycle.

144

Owing to these extensive applications, there is significant value in the development of a selective and efficient route for the synthesis of a wide range of molecularly diverse Nheterocyclic compounds. A synthetic methodology which allows for the selective direct functionalization of the N-heterocyclic core to gain access to a variety of products is particularly attractive. A successful methodology would require simple, inexpensive, and readily available chemicals and would avoid the generation of stoichiometric by-products. To this end, catalytic C–H functionalization has emerged as a selective and powerful tool for the direct functionalization of sp3-hybridized C–H bonds adjacent to the nitrogen of amine substrates.330-334 The current state-of-the-art for this strategy include stoichiometric α-lithiation strategies,335-337 copper-catalyzed cross-dehydrogenative couplings,338-340 radical-based C–H activation,341-343 photoredox functionalization,344 metal-catalyzed carbene insertions,345,346 and ruthenium catalyzed α-alkylation and arylation strategies.347,348 The most relevant of these strategies are discussed in the following sections.

5.1.2

Stoichiometric α-lithiation and functionalization

Initially pioneered by the Beak335 and Hoppe336 groups, the organolithium-mediated αdeprotonation of cyclic amines, followed by functionalization with an electrophile, is one methodology that has undergone extensive research (Scheme 5.1). This transformation is a multiple-step process in which an N-protected amine heterocycle, classically pyrrolidine, is deprotonated with an organolithium reagent to generate an α-lithiated species. This α-amino anion intermediate can then react with a variety of electrophiles (eg. Me3SiCl, benzophenone, CO2) to generate the α-functionalized product. This transformation is restricted to protected

145

amine substrates, as secondary or primary amines undergo deprotonation of the more acidic N–H group over the desired C–H abstraction. PG N

1) RLi, L low temperature + 2) E

R = sBu, nBu, iPr PG = Boc, Et, nBu, CH2CH2OMe, trimethylallyl

PG N

H E

L=

N

N H

(−)-Sparteine

N

H N

(+)-Sparteine Surrogate

Scheme 5.1 Multiple-step organolithium-mediated α-deprotonation/functionalization methodology.

This transformation can be performed in an enantioselective fashion using chiral diamine ligands. Initial developments focused on the use of (–)-sparteine as the chiral ligand (Scheme 5.1),336 however, a broad range of chiral diamine ligands have been examined349 and have allowed for significant extension of the α-lithiation methodology. One such ligand is the (+)sparteine surrogate which mediates improved enantioselectivity and, more importantly, addresses the N-protected amine substrate scope limitations of preceding systems (Scheme 5.1).350 The majority of systems for the α-lithiation strategy are restricted to N-protected pyrrolidine substrates, and low yields are observed when the less reactive six-membered piperidines are used.330,351-354 This is attributed to the relatively lower reactivity of these larger ring substrates and a greater sensitivity to ligand sterics resulting in a negative impact on yield when ligand sterics are increased to improve enantioselectivity. In their 2010 report, Sanderson and coworkers described the use of sBuLi/(+)-sparteine surrogate in the first communication of enantioenriched 2-substituted piperidines produced in high yields.355 This methodology has since been extended to piperazines and azepanes, albeit with lower yields.356 The substrate scope of the electrophile has also been expanded via a transmetallation strategy to give an organozinc 146

intermediate suitable for subsequent Negishi cross-coupling to produce 2-aryl substituted products (Scheme 5.2).337,357-360 It is only recently that Fu and co-workers extended the scope to secondary alkyl iodide and bromide electrophiles for the enantioconvergent synthesis of 2-alkyl N-Boc-pyrrolidines (Scheme 5.2).361 The cross-coupling reaction of the α-zincated N-Bocpyrrolidine is catalyzed by NiCl2·DME with a chiral diamine ligand. Previous to this report only dimethyl sulfate and methyl iodide were useful coupling partners for the alkylation transformation.330 Currently, the substrate scope of the amine when using alkyl electrophiles remains restricted to N-Boc-pyrrolidine.361

Boc N

1) RLi, L low temperature 2) ZnCl2 or ZnI2

Boc N Zn proposed

RX Pd catalyst Negishi coupling RX Ni catalyst, L*

Boc N R

R = Aryl

Boc N R

R = Alkyl

Scheme 5.2 Organolithium-mediated α-deprotonation-transmetallation-Negishi coupling strategy.

While significant developments in the synthetic potential of these α-lithiation strategies have been achieved, this methodology is not without disadvantages. For example, this approach is only compatible with tertiary or protected amines, and the required protection and deprotection synthetic steps are costly. The α-deprotonation methodology is also a multiple-step procedure and generates stoichiometric amounts of lithium by-products, both of which are undesirable for application in large-scale synthesis.

147

5.1.3

Oxidative functionalization of α-C–H bonds In their 2003 report, Murahashi and co-workers described the RuCl3-catalyzed oxidative

α-cyanation of tertiary amines, using hydrogen peroxide or oxygen as the stoichiometric oxidant.362,363 While the majority of the substrates reported are acyclic tertiary amines, two examples of cyclic substrates, N-phenylpiperidine and N-phenyltetrahydroisoquinoline have been used successfully (Scheme 5.3). Ph N

CN

R1

R2 N

AcOH/NaCN

5 mol% RuCl3 H2O2 or O2 MeOH

R1

R2 N

R3

Ph CN R3

N 76% yield

CN

69% yield

Scheme 5.3 Ruthenium catalyzed α-cyanation of tertiary amines.

Since this seminal report, the majority of transition-metal catalyzed oxidative C–H functionalization methodologies have involved the use of less expensive copper catalysts.338-340 Developed initially by Li and co-workers, copper-catalyzed cross-dehydrogenative couplings require a tertiary amine, a copper salt (eg. CuBr, CuOTf), a stoichiometric oxidant (eg. tertbutylhydroperoxide), and a nucleophile (Scheme 5.4).331,340 Nu Ar

N

H Nu

R

HN

catalyst [Cu] tBuOOH

O 2N

H Nu = H

O

R H

H

Ar

N

O

RO

R

O

OR H

H

Scheme 5.4 Copper-catalyzed cross-dehydrogenative coupling reactions of tertiary amines.

148

While the initial report describes the use of terminal alkynes as the nucleophiles,364 subsequent investigations have extended the substrate scope to encompass indoles, nitroalkanes, and malonates.338 Ketones are also compatible with this strategy though they require the addition of a co-catalyst such as pyrrolidine. Unfortunately, the substrate scope of the amine partner remains limited and as a result, studies have mainly focused on the functionalization of the activated benzylic position of tertiary tetrahydroisoquinolines. Less harsh oxidants such as oxygen or air have been employed,365 and a variety of metals including iron,366-369 ruthenium,362,363,370-373 gold,374 and vanadium375 have been investigated for this approach. To date, the mechanism of this reaction has not been rigorously established. However, it is proposed to proceed initially via single-electron transfer (SET) from the tertiary amine to form an amino radical-cation (Scheme 5.5).376 Oxidative α-C–H activation at the benzylic position generates an iminium ion intermediate that can be trapped by the nucleophilic partner. These iminium cations can also be generated in a separate step via electrochemical methods. This “cation pool” approach consists of the generation of large concentration of N-acyliminium cations,343 which can be subsequently trapped by nucleophiles377 or electron deficient olefins.343 Nu R

N

SET

R

N

-H+

R

N

Nu-X

R

N

Scheme 5.5 α-Functionalization via an iminium cation.

The one-step copper-catalyzed cross-dehydrogenative coupling methodology shows a great deal of promise for the functionalization of α-C–H bonds of cyclic amines. However, numerous limitations currently restrict the application of this procedure, including a very narrow amine substrate scope, restricted to tertiary amines with activated C–H bonds. Additionally, 149

while progress has been made in the use of non-toxic and low cost molecular oxygen as the oxidant, stoichiometric amounts of tert-butylperoxide are required in most cases. The oxidative functionalization of α-C–H bonds has also been achieved using a photocatalysis for the generation of the iminium cation intermediates.344 This was first demonstrated by Stephenson and co-workers in 2010 using an iridium species as the photocatalyst.378 Visible light is proposed to initiate a series of single-electron transfers which result in the oxidation of the amine to an amino radical-cation (Scheme 5.6, top). An external oxidant regenerates the photocatalyst and abstracts the α-hydrogen of the amino radical-cation to generate the iminium ion. Trapping of the iminium cation with a nucleophile generates the αfunctionalized product. The use of radicophiles as trapping agents for α-amino radicals which can be formed following deprotonation of the amino radical cations have also been reported, initially by MacMillan and co-workers (Scheme 5.6, bottom).379,380 A series of metal-based photocatalysts

(eg.

[Ir(ppy)2(dtbbpy)]PF6,

[Ru(bpy)3](PF6)2)378-380

as

well

as

organic

photosensitizers (eg. Rose Bengal)341-343 are competent for this transformation, although this methodology is currently restricted by the need for highly-functionalized coupling partners and tertiary amine substrates.

150

Pn+ light R N

[O]

Photoredox cycle

Pn+ *

[O]

P(n-1)+

H

iminium ion intermediate R N

H -H

R≠H

R N

NuH

R1

R N

-H+ EWG

R N

Nu

R1 EWG

R N R N

NC

EWG

EWG α-amino radical intermediate

Scheme 5.6 Photoredox approach for the α-functionalization of tertiary amines via an iminium ion (top) or an α-amino radical cation (bottom).

5.1.4

Directed transition metal-catalyzed C–H activation Ruthenium complexes have been successfully applied for the α-alkylation and arylation

of heterocycles. This technique uses to advantage a heteroatom directing group for selective oxidative addition of the α-C–H bond of the N-heterocycle (Scheme 5.7).347,348 These directing groups (eg. 2-pyrrolinyl or 2-pyridinyl), are critical to the success of this reaction, but in turn limit substrate scope for this approach once more to tertiary amine substrates.

151

N Ar

N

bisarylation

Ar

( )n

N

=

N N

N [Ru]

Ar

0

N

( )n

N or

( )n

N

N

N Rull Ar

N

N

( )n

Ru3(CO)12 precatalyst n=1−3

[Ru]

0

( )n N

O H B O

N O Ar B O

Rull H

( )n

Scheme 5.7 Directed ruthenium-catalyzed α-arylation of N-substituted heterocycles.

The first application of α-arylation is described in a report by Sames and co-workers using Ru3(CO)12 and arylboronate esters as the coupling partner.381 Once more, the lessened reactivity of piperidine when compared with pyrrolidine is noted, and only one example of piperidine arylation is described with yields too low to be synthetically useful. Extension of this reactivity to piperidines was accomplished by Maes and co-workers using 2-pyridinyl as a directing group with Ru3(CO)12.352,382 In all cases a stoichiometric amount of a ketone or alcohol additive is required for productive reactivity and, along with the use of boronate esters as the coupling partner, results in stoichiometric amounts of waste. The high catalyst loadings required (6 – 8 mol% Ru3(CO)12) are also costly. In cases where the amine substrate contains two α-CH2 positions selective mono-arylation cannot be achieved and a mixture of the mono-arylated and bis-arylated products is obtained.383 152

The heteroatom-directed C–H activation strategy has also been used for the α-alkylation of N-heterocyclic substrates using olefinic substrates. The initial report of α-alkylation by Murai and co-workers details pyrrolidine alkylation (Scheme 5.8) to give a mixture of mono- and bisalkylated products.384 A more recent report by Maes and co-workers describes how the addition of a catalytic amount of trans-1,2-cyclohexanedicarboxylic acid is necessary to achieve high conversions with the more challenging piperidine substrates.385

N R

N (

8 mol% Ru3(CO)12, CO i PrOH, 140 °C n=1−3

)n

N

N N

R

R

N

( )n

R

( )n

Scheme 5.8 Directed ruthenium-catalyzed α-alkylation of N-substituted heterocycles.

The selective mono-alkylation of N-heterocycles using this methodology has not yet been achieved and the amine substrate scope remains restricted to tertiary amines. Two systems using the expensive late transition metals iridium386 and ruthenium387 have been reported for the αalkylation of secondary amines, however, these systems have not been extended to Nheterocyclic substrates.

5.1.5

Hydroaminoalkylation of N-heterocycles with early transition metals Hydroaminoalkylation is an atom-economic strategy for alkylation at the α-position of

unprotected secondary amines (Scheme 5.9).217 This strategy is very attractive as it uses simple, readily available feedstocks and, unlike the methodologies discussed above, is compatible with unprotected amine substrates. This avoids costly protection/deprotection sequences that generate waste without adding molecular complexity. 153

R1

H N

R2

catalyst

R3

R1

H N

R2

R1

H N

R3 H N

H N

catalyst

R

R2

R3 R

H N

R

Scheme 5.9 Intermolecular hydroaminoalkylation of alkenes with secondary amines.

The use of low cost, non-toxic early transition metal-based systems for this promising new approach has been the recent focus of much research. These reports describe numerous group 478,159,160,162,163,172,255 and 577,112,118,143-145,247,250,251,284 based catalyst systems for the efficient preparation of secondary arylalkyl and dialkyl acyclic amines. However, it is notable that substrate scope investigations of such systems have repeatedly shown that N-heterocycles are particularly challenging substrates. The majority of reports, which detail successful reactivity with

N-heterocycles,

focus

on

the

direct

C–H

functionalization

of

1,2,3,4-

tetrahydroquinoline.118,143,162,251 The challenge of cyclic dialkyl amine substrates has been explicitly conceded in several catalyst development reports.77,78,144,162 Even catalytic systems that display vastly expanded substrate scope in the olefinic substrate247 or show unprecedented reactivity at room temperature249 are not compatible with N-heterocyclic substrates. This difficulty with N-heterocyclic substrates is exemplified in a recent report by Doye and coworkers that discloses the reaction of pyrrolidine with styrene catalyzed by a titanium aminopyridinate complex.78 Even with forcing reaction conditions (96 h, 140 ºC) and high catalyst loading (20 mol%), the authors could only achieve a mixture of the regioisomers of the alkylated pyrrolidine product in 17% yield. The most promising system for these challenging 154

substrates has been reported by Schafer and co-workers and uses mono(amidate) tetrakis(dimethylamido) tantalum precatalyst 2 to access the mono-alkylated piperidine product in good yield with excellent selectivity (Scheme 5.10).118 O H N 3

10 mol% 2 n 134 hexyl h, 165 °C toluene

tBu

H N

n

hexyl

Ta(NMe2)4 N

iPr

iPr

2

74% yield

Scheme 5.10 Hydroaminoalkylation of piperidine with mono(amidate) tetrakis(dimethylamido) tantalum precatalyst 2.

5.1.6

Scope of chapter The focus of Chapter 5 is the investigation of the substrate scope for the α-alkylation of

N-heterocyclic amines using mono(amidate) tantalum complex 2. This involves the direct αalkylation of a variety of N-heterocycles including piperidines, piperazines, and azepanes. The αalkylated products are formed with remarkable chemo-, regio-, and stereoselectivity, and the mechanistic rational for the observed selectivity is discussed. Section 5.2.2 describes a potential reason for the unprecedented reactivity of precatalyst 2 and guiding principles that should be taken into account for subsequent catalyst development efforts. The observed reactivity differences between five- and six-membered amine substrates are examined by both experimental and in silico approaches in Section 5.2.3 – 5.2.4.

155

5.2 5.2.1

Results and discussion The hydroaminoalkylation of N-heterocyclic substrates Typically, pyrrolidines are preferred over piperidine substrates for direct C–H

functionalization strategies (eg. α-lithiation methodologies,330 ruthenium catalyzed α-alkylation and arylation352,381). Thus, the direct alkylation of a wide variety of larger N-heterocycles such as piperidines is complementary to other established approaches and an important target. The exploration of substrate scope in both the alkene and the amine partner builds upon the single successful report of the α-alkylation of piperidine with 1-octene (Table 5.1, entry 1).118 Using 5 mol% 2, the internal alkene norbornene can undergo hydroaminoalkylation with piperidine (entry 2), indicating that this strategy is not restricted to terminal olefins. Protected alcohols can be incorporated into the alkene substrate, providing a site for further functionalization of the resultant amine product and demonstrating good functional group tolerance for this oxophilic tantalum complex (entry 3). Interestingly, a piperidine with a protected carbonyl substituent is also compatible with this early transition metal catalyst (entry 4). With such functional group tolerance established for 2, olefinic silyl ethers can be used for the α-alkylation of 1,2,3,4tetrahydroquinoline (entries 5, 6). This aryl alkyl amine substrate is more reactive, and resulted in good yields at lower reaction temperatures (145 ºC). Impressively, larger heterocyclic rings such as azepane can also be used (entry 7), though decreased diastereoselectivity is observed. Related amine substrates such as morpholine and piperazine are not viable substrates; however, the direct alkylation of N-substituted piperazines is possible (entries 8 – 12). Remarkably, this reaction is tolerant to various substituents on the distal nitrogen, including p-methoxyphenyl and benzhydryl, which allow for subsequent deprotection. Good yields are also obtained with both alkyl and benzylic olefins (entries 11, 12). 156

Table 5.1 Hydroaminoalkylation of saturated N-heterocycles. H N R 1.5-2 equiv.

X

Entry 1

n

H H N

R

X

t (h)

Yield (%)b

d.r.c

143

76

>20:1

72

79

>20:1

96d

26

>20:1

69

59

>20:1

OTBS ( )3 48

165

64

>20:1

OTBS ( )4

118

78

>20:1

72

60

10:1

R = Me

72

43

>20:1

Ph

72

68

>20:1

PMP

69

69

>20:1

R = hexyl

72

46

>20:1

Bn

72

84

>20:1

hexyl

H H N

2

3

10 mol% 2 165 °C toluene

Product (+/-)a H H N

H H N

Ph

O ( )3

47

Ph

H H N

n

hexyl

4 O

O

e,f

H H N

e,f

H H N

5

6

H H N

7 8

H H N

9 10

n

n

hexyl

hexyl

N R

11

H H N

12

N

n

R

Bhyd a c

Stereochemistry assigned by analogy to 47. bIsolated after N-tosylation.

Determined by 1H NMR spectrum of isolated product. dReaction time was not

optimized. eReaction run at 145 °C and isolated as the free amine. f5 mol% 2.

157

Interestingly, selective monoalkylation is observed under these rather forcing reaction conditions, despite the presence of excess alkene. This is encouraging, as the directed αalkylation of unactivated olefinic substrates using Ru3(CO)12 results in a mixture of mono- and bis-alkylated products (Scheme 5.8).384 This selective monoalkylation can be rationalized based upon the sensitivity of this catalyst system to steric bulk. For example, no reaction is observed when using 2-methylpiperidine or 3-methylpiperidine and 1-octene with these reaction conditions. In all cases the regio- and diastereoselectivity of this transformation are excellent and typically only one isomer is detected when monitoring the reaction by NMR spectroscopy. Based upon the mechanistic proposal for hydroaminoalkylation (Scheme 5.11),118,144,145 excellent selectivity is anticipated due to the proposed formation of metallacyclic intermediates. The regioand diastereoselectivity of the α-alkylation are established during the olefin insertion step of the catalytic cycle, generating intermediate B. For the hydroaminoalkylation of N-heterocycles, alkene insertion occurs in an orientation that positions the alkene R substituent away from the steric bulk of the ligands on the tantalum metal center. This regioselective alkene insertion into the strained metallaziridine intermediate, generating the branched regioisomer, has been consistently observed for group 5 hydroaminoalkylation.77,112,118,143,144,250,251 Only one example in a recent communication reports the generation of mixtures of linear and branched regioisomers using styrene or trimethylvinylsilane as the olefinic substrate and

mono(phosphoramidate)

tantalum complex 21.249

158

[Ta] H N

NMe2

O [Ta] =

NMe2

tBu

Ta(NMe2)2 N

iPr

iPr

2 HNMe2 [Ta] H H N

N A

R

R

N

R [Ta]

[Ta] N

H

N

H R H N

B

Scheme 5.11 Postulated mechanism for the tantalum-catalyzed hydroaminoalkylation of N-heterocycles.

The diastereoselectivity of this reaction arises from the approach of the alkene to the less hindered face of the tantalaziridine A. This selectivity proposal and the resultant diastereomer have been confirmed by X-ray crystallography, in which derivatization of the crude product of entry 3 with tosyl chloride generates a white solid 47 that can be recrystallized for rigorous analysis (Scheme 5.11). The solid-state molecular structure of 47 shows the formation of the anticipated diastereomer of the branched, monoalkylated product (Figure 5.2).

159

Ts H N

Ph

O Ph

Figure 5.2 ORTEP representation of the solid-state molecular structure of compound 47. The ellipsoids are plotted at 50% probability and the majority of the hydrogen atoms are omitted for clarity.

As shown in Table 5.1, hydroaminoalkylation precatalyst 2 can be used to efficiently prepare a variety of α-alkylated N-heterocycles. The functional group tolerance and excellent selectivity displayed by this precatalyst suggests that interesting bicyclic or tricyclic Nheterocyclic products could be synthesized following the α-alkylation

via further

functionalization of the secondary amine moiety (Scheme 5.12). The products afforded from 1,2,3,4-tetrahydroquinoline are particularly attractive for these investigations as the presence of the chromophore aids in product isolation and purification via column chromatography. H H N

( )n

( )n ( )n

X

N

N

H

H

( )n N H

N R

R

Scheme 5.12 Proposed synthesis of bicyclic and tricyclic compounds following α-alkylation of an Nheterocyclic amine.

One approach that has been investigated utilizes an olefin containing a silyl protected alcohol, which, after α-alkylation, could be converted into a suitable leaving group for nucleophilic substitution and ring closure. A recently developed method of Gembus and co160

workers describes the one-pot conversion silyl ethers to tosylates using diazabicycloundecene (DBU) and tosyl fluoride.388 This methodology has been extended by Dr. Pierre Garcia of the Schafer group to a one-pot methodology for the synthesis of β-substituted N-heterocycles following hydroaminoalkylation of acyclic N-arylamine substrates with olefins containing silyl protected alcohols (Scheme 5.13, top).284 The analogous reaction conditions have been examined using 1,2,3,4-tetrahydroquinoline and tert-butyldimethyl(pent-4-en-1-yloxy)silane, which are a successful substrate combination for hydroaminoalkylation (entry 5, Table 5.1). Unfortunately, the ring-closure step has not been successful, though monitoring by 1H NMR spectroscopy confirms that α-alkylation reaches full conversion. Attempts at applying a multiple-step approach consisting of hydroaminoalkylation to produce 48 (entry 5, Table 5.1), deprotection and purification of the intermediate alcohol, followed by derivitization with mesyl or tosyl chloride and heating to promote ring closure have also not been productive.

Ar

H N 5 mol% 2 130 °C toluene

+

OTBS ( )n

TBSO ( )n H N Ar

( )n N Ar

n = 1-3

H N OTBS ( )3

+

H N

TsF, DBU 130 °C toluene

OTBS

( )3

48, Table 5.1

1) 5 mol% 2 165 h, 145 °C

N

2) TsF, DBU 48 h, 130 °C

TBAF, AcOH 1:1 H2O:THF 2 h, 25 °C

H N

OH MsCl or TsCl ( )3 toluene, reflux

N

90% yield

Scheme 5.13 α-Alkylation and ring-closure methodology using N-arylalkyl amines (top)284 and attempts at the analogous methodology with 1,2,3,4-tetrahydroquinoline.

161

Transformation of the alcohol to the aldehyde followed by reductive amination is another potential approach; however, oxidation of the alcohol without decomposition of the product could not be achieved. A piperidine with a protected carbonyl substituent has been shown to be compatible with the reaction conditions (entry 4, Table 5.1), and therefore acetal olefin 49 could be a potential candidate for this reaction, as this would generate the desired aldehyde moiety desired upon deprotection. Unfortunately, the hydroaminoalkylation reactions using piperidine and 1,2,3,4-tetrahydroquinoline amine substrates have not been successful. Thermal stability studies indicate that the lack of reactivity is due to decomposition of the olefinic substrate in the presence of tantalum precatalyst 2 at elevated temperatures.

OEt 49

OEt

H N

5 mol% 2 48 h, 165 °C toluene

H N

OEt OEt

Scheme 5.14 Hydroaminoalkylation with olefin substrate containing a protected carbonyl moiety.

Precatalyst 2 has previously been shown to selective for the monoalkylation of diene substrates.118 Therefore, a tandem hydroaminoalkylation/hydroamination methodology could potentially be used to generate the cyclic compounds. Indeed, the α-alkylation of p-methoxy-Nmethylaniline can be selectively achieved to generate the aminoalkene intermediate. Unfortunately, both one-pot or multiple-step procedures for the cyclization of this substrate using precatalyst 4 have not been effective (Scheme 5.15).

162

iPr

H N

O NMe2 10 mol% 2 40 h, 130 °C

MeO

2N

Ar

H N

10 mol% 4 48 h, 130 °C

Ar

N N

N iPr

2N

Zr

NMe2

O HNMe2 4

Scheme 5.15 Attempted hydroaminoalkylation-hydroamination procedure catalyzed by tantalum and zirconium precatalysts, respectively.

5.2.2

Urea proligands for the hydroaminoalkylation of N-heterocycles The substrate scope described in Table 5.1 demonstrates the potential of this

methodology for the late-stage synthesis of a variety of N-heterocyclic compounds from readily available olefin starting materials without the need for protecting or directing groups. One intriguing element of this methodology is that this reactivity is unique to precatalyst 2. No other reported system is capable of the α-alkylation of piperidine, piperazine, or azepane substrates despite a great deal of catalyst development.77,112,143,144,160,172,247 This achievement highlights the uniqueness of the amidate ligand, and indicates that the asymmetric binding of the mixed (N,O)ligand system and the potential for hemi-lability could be the source of this vastly expanded substrate scope accessible by precatalyst 2 above all other (O,O)-77 and (N,N)-supported catalytic systems.78 If the asymmetric binding and the potential for hemi-lability is critical for this reactivity, then other (N,O)-chelating ligands may be suitable for the α-alkylation of N-heterocycles. To test this proposal, two urea proligands have been investigated as supporting ligands for tantalum precatalysts, 50 and 8 (Scheme 5.16). These in situ formed catalyst systems are indeed successful at promoting the α-alkylation of piperidine under analogous reaction conditions as those used 163

with 2. Gratifyingly, the chiral proligand 8 even mediates the enantioselective α-alkylation of this six-membered heterocycle with 21% ee.389 This is the only example of enantioselective hydroaminoalkylation mediated by a non-C2-symmetric ligand system, and highlights an avenue worthy of further research. H N

1.5

1) 5 mol% 50 or 8 5 mol% Ta(NMe2)5 143 h, 165 °C, toluene 2) TsCl, 2M NaOH n

hexyl

Ts H N

O n

N

hexyl (+/-) 50: 82% yield d.r. > 20:1 : 8 68% yield, 21% ee d.r. > 20:1

N H 50 O NH

CyN 8

iPr

Scheme 5.16 α-Alkylation of piperidine catalyzed by in situ generated tantalum ureate complexes.

5.2.3

Attempted α-alkylation with pyrrolidine and indoline Despite the unprecedented reactivity with six and seven-membered N-heterocycles,

efforts to directly alkylate five-membered substrates such as pyrrolidine and indoline have not been successful thus far. This is particularly surprising because pyrrolidines are typically preferred over piperidine substrates for α-lithiation strategies and ruthenium catalyzed αalkylation.334,352,353,385 Indoline, which as an aryl alkyl amine substrate is expected to be more reactive than pyrrolidine, does not undergo any α-alkylation reactivity. In fact, upon addition of the amine to a solution of 2, a precipitate is observed concurrent with a dramatic colour change of the solution from pale yellow to bright red-orange (Scheme 5.17).

164

N

O tBu

Ta(NMe2)4 N

iPr

H N 10

N